CA2659087C

CA2659087C - Method, apparatus, signals, and media, for selecting operating conditions of a genset

Info

Publication number: CA2659087C
Application number: CA2659087A
Authority: CA
Inventors: Nicolas Louis Bouchon; Wei Liu
Original assignee: Conversant Intellectual Property Management Inc
Current assignee: GE Hybrid Technologies LLC
Priority date: 2006-06-26
Filing date: 2007-06-26
Publication date: 2014-12-09
Anticipated expiration: 2027-06-26
Also published as: US20140121879A1; CA2659087A1; EP2620341B1; WO2008000071A1; US20130066503A1; US20130090864A1; US8655570B2; US9020734B2; EP2620341A2; EP2620342B1; US20080122228A1; EP2035270B1; EP2620342A3; EP2035270A1; US9403529B2; EP2620342A2; EP2620341A3; EP2035270A4; ES2433373T3; US20150203101A1

Abstract

A method and apparatus for selecting optimal operating conditions of a genset is disclosed. The genset includes an engine coupled to an electrical power generator, the genset having a plurality of operating points each including an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points. The method involves selecting a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in the set is minimized and such that the engine speed and generator electrical power output values of the operating points in the set increase or decrease monotonically.

Description

METHOD , APPARATUS, S I GNALS . AND MEDIA, FOR SELECTING
OPERATING CONDITIONS OF A GENSET
BACKGROUND OF THE INVENTION
1. Field of Invention This invention relates generally to selecting operating conditions of an engine coupled to an electrical power generator for generating electrical power.

2. Description of Related Art Advances in engine and hybrid vehicle technology have continually reduced harmful pollutant emissions. Hybrid vehicles having an engine driving a generator (a genset) cause emissions such as nitrogen oxides, carbon monoxide, hydrocarbons, and particulate matter. Each emission has a characteristic dependence on a power output level of the generator, and a rotational speed of the engine. Furthermore, given the present climate of higher prices of fossil fuels, there is a corresponding desire to reduce fuel consumption of the engine, thereby reducing the cost of operating the genset.
In particular, in electric hybrid vehicles, where the genset may supply electrical power to a drive motor and/or a charger for charging a storage battery, choosing operating points for the vehicle that minimize fuel consumption while simultaneously minimizing emissions or other operating conditions is a multi-variable problem and selecting operating points for the genset usually involves trading off between various emissions and fuel consumption to meet a desired criterion.
There remains a need for better methods and apparatus for selecting operating points of a genset, for use in hybrid vehicles and other applications.

SUMMARY OF THE INVENTION
In accordance with one aspect of the invention there is provided a method for selecting optimal operating conditions of a genset, the genset including an engine coupled to an electrical power generator, the genset having a plurality of operating points each including an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points. The method involves selecting a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in the set is minimized and such that the engine speed and generator electrical power output values of the operating points in the set increase or decrease monotonically.
The method may involve assigning a weight to each of the power output values such that cost values corresponding to more frequently demanded power output values are assigned a greater weight in the sum of cost values than cost values corresponding to less frequently demanded electrical power output values.
Assigning the weight may involve assigning greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.
The method may involve generating a record of received demands for power outputs during operation of the genset over a period of time, and assigning the weights may involve assigning greater weight to more frequently used power output values.

-3-Generating the record may involve generating the record and updating the weights while operating the genset, and selecting may involve selecting a new set of operating points when the weights have been updated.
Selecting may involve successively applying a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power outputs the generator is capable of supplying, and selecting a set of operating points corresponding to a last power output value in the range as the minimized set of operating points for the genset.
Applying the dynamic programming algorithm may involve producing a first plurality of the sums of cost values at a first power output value and memoizing a result of at least one of the first plurality of the sums. The method may further involve using at least some of the memoized results for producing a further sum of cost values at a subsequent power output value.
Selecting may involve locating a minimum cost value at each power output value and successively applying the dynamic programming algorithm may involve first producing the sums of cost values corresponding to the minimum cost value at each successive power output value.
Locating the minimum cost value may involve locating an operating point corresponding to the minimum cost value using a golden section search technique.
The method may involve receiving a demand to supply power at a demanded power output value and operating the genset at an operating point in the set of operating points corresponding to the demanded power output value.
The genset may be used to generate electrical energy for use in a hybrid electrical vehicle and receiving the demand may involve receiving a demand

-4-to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.
Receiving the demand to supply power to the charger may involve receiving a demand to supply power to the charger while the hybrid electric vehicle remains stationary.
Receiving the demand to supply power to the drive motor may involve receiving a drive signal from an operator foot pedal representing a desired drive power to be supplied by the drive motor to wheels of the hybrid vehicle.
The storage element may be operable to supply at least a first portion of the desired drive power, and receiving the demand may involve receiving a demand for a second portion of the desired drive power.
Receiving the demand to supply power to the charger may involve receiving a charge signal from a storage element controller, the charge signal being produced in response to a state of charge associated with the storage element.
The method may involve receiving a plurality of cost values.
The method may involve selecting a new set of operating points in response to receiving the plurality of cost values.
Selecting may involve selecting the set of genset operating points prior to operating the genset.
In accordance with another aspect of the invention there is provided an apparatus for selecting operating conditions of a genset, the genset including an engine coupled to an electrical power generator, the genset having a

-5-plurality of operating points each including an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points. The apparatus includes provisions for selecting a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in the set is minimized and such that the engine speed and generator electrical power output values of the operating points in the set increase or decrease monotonically.
The apparatus may include provisions for assigning a weight to each of the power output values such that cost values corresponding to more frequently demanded power output values are assigned a greater weight in the sum of cost values than cost values corresponding to less frequently demanded electrical power output values.
The provisions for assigning the weight may include provisions for assigning greater weight to electrical power output values proximate a midpoint of a range of power output values that the generator is capable of supplying.
The apparatus may include provisions for generating a record of received demands for power outputs during operation of the genset over a period of time, and the provisions for assigning the weights may include provisions for assigning greater weight to more frequently used power output values.
The provisions for generating the record may include provisions for updating the weights while operating the genset, and the provisions for selecting may be operably configured to select a new set of operating points when the weights have been updated.
The provisions for selecting may include provisions for successively applying a dynamic programming algorithm to select sets of operating points at

-6-successive power output values within a range of power outputs the generator is capable of supplying and provisions for selecting a set of operating points corresponding to a last power output value in the range as the minimized set of operating points for the genset.
The provisions for applying the dynamic programming algorithm may include . provisions for producing a first plurality of the sums of cost values at a first power output value, provisions for memoizing a result of at least one of the first plurality of the sums, and provisions for using at least some of the memoized results for producing a further sum of cost values at a subsequent power output value.
The provisions for selecting may include provisions for locating a minimum cost value at each power output value and the provisions for successively applying the dynamic programming algorithm may include provisions for first producing the sums of cost values corresponding to the minimum cost value at each successive power output value.
The provisions for locating the minimum cost value may include provisions for performing a golden section search technique.
The apparatus may include provisions for receiving a demand to supply power at a demanded power output value and provisions for operating the genset at an operating point in the set of operating points corresponding to the demanded power output value.
The genset may be used to generate electrical energy for use in a hybrid electrical vehicle and the provisions for receiving the demand may include provisions for receiving a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.

-7-The provisions for receiving the demand to supply power to the charger may include means for receiving a demand to supply power to the charger while the hybrid electric vehicle remains stationary.
The storage element may include at least one of a storage battery, a capacitor, and an electrically coupled flywheel.
The provisions for receiving the demand to supply power to the drive motor may include provisions for receiving a drive signal from an operator foot pedal representing a desired drive power to be supplied by the drive motor to wheels of the hybrid vehicle.
The storage element may be operable to supply at least a first portion of the desired drive power, and the demand may include a demand for a second portion of the desired drive power.
The provisions for receiving the demand to supply power to the charger may include provisions for receiving a charge signal from a storage element controller, the charge signal being produced in response to a state of charge associated with the storage element.
The apparatus may include provisions for receiving a plurality of cost values.
The apparatus may include provisions for selecting a new set of operating points in response to receiving the plurality of cost values.
The provisions for selecting may include provisions for selecting the set of genset operating points prior to operating the genset.

-8-In accordance with another aspect of the invention there is provided an apparatus for selecting operating conditions of a genset, the genset including an engine coupled to an electrical power generator, the genset having a plurality of operating points each including an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points. The apparatus includes a processor circuit operable to select a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in the set is minimized and such that the engine speed and generator electrical power output values of the operating points in the set increase or decrease monotonically.
The processor circuit may be operably configured to assign a weight to each of the power output values such that cost values corresponding to more frequently demanded power output values are assigned a greater weight in the sum of cost values than cost values corresponding to less frequently demanded electrical power output values.
The processor circuit may be operably configured to assign greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.
The processor circuit may be operably configured to generate a record of received demands for power outputs during operation of the genset over a period of time, and to assign greater weight to more frequently used power output values.
The processor circuit may be operably configured to update the weights while operating the genset and to select a new set of operating points when the weights have been updated.

-9-The processor circuit may be operably configured to successively apply a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power outputs the generator is capable of supplying and select a set of operating points corresponding to a last power output value in the range as the minimized set of operating points for the genset.
The processor circuit may be operably configured to produce a first plurality of the sums of cost values at a first power output value, memoize a result of at least one of the first plurality of the sums, and use at least some of the memoized results for producing a further sum of cost values at a subsequent power output value.
The processor circuit may be operably configured to locate a minimum cost value at each power output value and to apply the dynamic programming algorithm to first produce the sums of cost values corresponding to the minimum cost value at each successive power output value.
The processor circuit may be operably configured to locate the minimum cost value using a golden section search technique.
The processor circuit may be operably configured to receive a demand to supply power at a demanded power output value and to operate the genset at an operating point in the set of operating points corresponding to the demanded power output value.
The genset may be used to generate electrical energy for use in a hybrid electrical vehicle and the processor circuit may be operably configured to receive a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.

-10-The processor circuit may be operably configured to receive the demand to supply power to the charger while the hybrid electric vehicle remains stationary.
The storage element may include at least one of a storage battery, a capacitor, and an electrically coupled flywheel.
The processor circuit may be operably configured to receive a drive signal from an operator foot pedal representing a desired drive power to be supplied by the drive motor to wheels of the hybrid vehicle.
The storage element may be operable to supply at least a first portion of the desired drive power, and the demand may include a demand for a second portion of the desired drive power.
The processor circuit may be operably configured to receive a charge signal from a storage element controller, the charge signal being produced in response to a state of charge associated with the storage element.
The processor circuit may be operably configured to receive a plurality of cost values.
The processor circuit may be operably configured to select a new set of operating points in response to receiving the plurality of cost values.
The processor circuit may be operably configured to select the set of genset operating points prior to operating the genset.
In accordance with another aspect of the invention there is provided a computer readable medium encoded with codes for directing a processor

-11 -circuit to perform a method for selecting optimal operating conditions of a genset, the genset including an engine coupled to an electrical power generator, the genset having a plurality of operating points each including an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points. The method involves selecting a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in the set is minimized and such that the engine speed and generator electrical power output values of the operating points in the set increase or decrease monotonically.
In accordance with one aspect of the invention there is provided an apparatus for selecting operating conditions of a genset. The apparatus includes an engine coupled to an electrical power generator and includes a processor circuit configured to select a set of operating points from a plurality of operating points of the genset. Each operating point includes an engine speed value and a generator electrical power value, the set of operating points being selected such that a sum of cost values associated with the operating points in the set is minimized and such that the engine speed increases or decreases monotonically with monotonically increasing or decreasing electrical power output values.
The processor circuit may be operably configured to assign a weight to each of the power output values such that cost values corresponding to more frequently demanded power outputs are assigned a greater weight in the sum of cost values than cost values corresponding to less frequently demanded electrical power output values.

- 12 -The processor circuit may be operably configured to assign greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.
The processor circuit may be operably configured to generate a record of received demands for power outputs during operation of the genset over a period of time and to assign greater weight to more frequently used power output values.
The processor circuit may be operably configured to update the weights while operating the genset and to select a new set of operating points when the weights have been updated.
The processor circuit may be operably configured to receive a demand to supply power at a demanded power output value and to operate the genset at an operating point in the set of operating points corresponding to the demanded power output value.
The genset may be used to generate electrical energy for use in a hybrid vehicle and the processor circuit may be operably configured to receive a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.

- 13 -The processor circuit may be operably configured to successively apply a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power outputs the generator is capable of supplying, and select a set of operating points corresponding to a last power output value in the range as the set of operating points for the genset.
The processor circuit may be operably configured to produce a first plurality of sums of cost values at a first power output value, memoize a result of at least one of the first plurality of the sums, and use at least some of the memoized results for producing a further sum of cost values at a subsequent power output value.
The processor circuit may be operably configured to locate a minimum cost value at each power output value and to apply the dynamic programming algorithm to first produce the sums of cost values corresponding to the minimum cost value at each successive power output value.
The processor circuit may be operably configured to receive a plurality of cost values.
The processor circuit may be operably configured to produce the plurality of cost values by assigning weights to each of a plurality of operating conditions associated with operating the genset, the weights representing a desired trade-off between the plurality of operating conditions, receiving operating condition values corresponding to each of the plurality of operating conditions, the operating condition values associated with operating the genset at each of a plurality of operating points, the operating points each may include an

- 14 -engine speed value and a generator electrical power output value, at least one generator electrical power output value having more than one possible operating point, and combining the operating condition values in accordance with the weights for each respective operating point to produce the cost values for each operating point.
The processor circuit may be operably configured to select a new set of operating points in response to receiving the plurality of cost values.
The processor circuit may be operably configured to select the set of genset operating points prior to operating the genset.
In accordance with another aspect of the invention there is provided a method for selecting optimal operating conditions of a genset including an engine coupled to an electrical power generator. The method involves selecting a set of operating points from a plurality of operating points of the genset. Each operating point includes an engine speed value and a generator electrical power value, the set of operating points being selected such that the sum of cost values associated with the operating points in the set is minimized such that the engine speed increases or decreases monotonically with monotonically increasing or decreasing electrical power output values.
The method may involve assigning a weight to each of the power output values such that cost values corresponding to more frequently demanded power outputs are assigned a greater weight in the sum of cost values than cost values corresponding to less frequently demanded electrical power output values.

- 15 -The method may involve assigning greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.
The method may involve generating a record of received demands for power outputs during operation of the genset over a period of time and assigning greater weight to more frequently used power output values.
The method may involve updating the weights while operating the genset and selecting a new set of operating points when the weights have been updated.
The method may involve receiving a demand to supply power at a demanded power output value and operating the genset at an operating point in the set of operating points corresponding to the demanded power output value.
The method may involve using the genset to generate electrical energy for use in a hybrid vehicle and receiving a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.
The method may involve successively applying a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power output values the generator is capable of supplying, and selecting a set of operating points corresponding to a last power output value in the range as the minimized set of operating points for the genset.

- 16 -The method may involve producing a first plurality of sums of cost values at a first power output value, memoizing a result of at least one of the first plurality of the sums, and using at least some of the memoized results for producing a further sum of cost values at a subsequent power output value.
The method may involve locating a minimum cost value at each power output value and applying the dynamic programming algorithm to first produce the sums of cost values corresponding to the minimum cost value at each successive power output value.
The method may involve receiving a plurality of cost values.
The method may involve producing the plurality of cost values by assigning weights to each of a plurality of operating conditions associated with operating the genset, the weights representing a desired trade-off between the plurality of operating conditions, receiving operating condition values corresponding to each of the plurality of operating conditions, the operating condition values associated with operating the genset at each of a plurality of operating points, the operating points each may involve an engine speed value and a generator electrical power output value, at least one generator electrical power output value having more than one possible operating point, and combining the operating condition values in accordance with the weights for each respective operating point to produce the cost values for each operating point.
The method may involve selecting a new set of operating points in response to receiving the plurality of cost values.

- 17 -The method may involve selecting the set of genset operating points prior to operating the genset.
Other aspects and features of the present invention will become apparent to those ordinarily skilled in the art upon review of the following description of specific embodiments of the invention in conjunction with the accompanying figures.

- 18 -BRIEF DESCRIPTION OF THE DRAWINGS
In drawings which illustrate embodiments of the invention, Figure 1 is a schematic diagram of an apparatus for selecting operating conditions of a genset in accordance with a first embodiment of the invention;
Figure 2 is a processor circuit for use in the apparatus shown in Figure 1;
Figure 3 is a table, depicting a data structure for use in the apparatus shown in Figures 1 and 2;
Figure 4 is a schematic diagram of an apparatus for selecting operating conditions of a genset in accordance with a second embodiment of the invention;

-19-Figure 5 is a flowchart of a process for producing cost values executed by the processor circuit shown in Figure 2;
Figure 6 is a table of operating condition values used in producing the cost values in the process shown in Figure 5;
Figure 7 is a flowchart of a process for selecting a set of operating conditions executed by the processor circuit shown in Figure 2;
Figure 8 is a graphical depiction of the operating points selected in the process shown in Figure 7; and Figure 9 is a schematic diagram of a hybrid electric vehicle.
DETAILED DESCRIPTION
Selecting operating points Referring to Figure 1, an apparatus for selecting operating conditions of a genset 10 is shown generally at 12. The genset 10 includes an engine 14 and a generator 16, the engine being coupled to the generator by a shaft 20. The genset 10 has a plurality of operating points, each including an engine speed value and a generator electrical power output value. The genset 10 also has a plurality of cost values associated with operating the genset at respective operating points. The apparatus 12 further includes a processor circuit 18, which is operable to select a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in the set is minimized and such that the engine speed and generator electrical power output values of the operating points in the set increase or decrease monotonically.

-20-The generator 16 further includes an output 22 for delivering the electrical power output to be used, for example, in operation of a hybrid electric vehicle.
In the embodiment shown the processor circuit 18 also includes an output 28 for producing control signals and the engine 14 includes an interface 30 for receiving an engine control signal from the processor circuit. Optionally, the generator 16 may include an interface 32, for receiving a generator control signal from the processor circuit 18 (for example, to set a field current level in a DC electric generator).
The processor circuit 18 also includes an input 24 for receiving data values such as cost values, and an input 26 for receiving a demand to supply electrical power.
The processor circuit 18 is shown in greater detail in Figure 2. Referring to Figure 2, the processor circuit 18 includes a central processing unit (CPU) 40, a program memory 42, a random access memory (RAM) 44, a media reader/writer 46, and an input output port (I/O) 48. The program memory 42, the RAM 44, the media reader/writer 46, and the I/O 48, are all in communication with the CPU 40.
The I/O 48 includes the input 26 for receiving the power demand. In the embodiment shown the I/O 48 also includes the output 28 for producing an engine/generator control signal to control the engine 14. The I/O 48 may optionally include an input 62 for receiving user input, as described later herein.
The media reader/writer 46 facilitates loading program codes into the RAM
memory 44 from a computer readable medium, such as a CD ROM disk 52, or a computer readable signal 54 such as may be received from a network such as a telephone network or the internet, for example. Optionally, the plurality of cost values may be encoded on the computer readable medium

-21-50, and the CPU 40 may be configured to load the cost values into the processor circuit 18, and store the cost values in a location 56 in the RAM
44.
The RAM 44 further includes locations 58, 60, and 64 for storing operating condition values, weights .91, and operating points (power output values and corresponding speed values) respectively, as described later herein.
Data structure-cost values Referring to Figure 3, the plurality of cost values are shown in a tabular format at 70. The table 70 includes a plurality of power output value rows 71, each row corresponding to a generator power output value that the generator is capable of supplying. In the embodiment shown in Figure 3, the genset 10 is capable of supplying power in 5kW increments, up to a maximum power of 35kW. In other embodiments the power increment may be greater than or less than 5kW, and the maximum power may be higher than 35kW. The plurality of rows 71 includes a row 72 corresponding to a no-load power output value, a row 74 corresponding to a 5kW power output value, a row 76 corresponding to a 10kW power output value, etc. Each row 71 includes a plurality of rotational speeds 80 at which the genset 10 is capable of supplying power. For example, the row 76, corresponding to a 10kW power output value, may be supplied at any of the rotational speeds 80.
Values 82, in the table 70, represent cost values associated with operating the genset 10 at various operating points, each operating point being defined by a power output value and a corresponding rotational speed 80. For example, the value 84 represents the operating cost associated with operating the genset 10 at a speed of 2800rpm while supplying a 10kW power output.
It may not be practical to supply some power output values at some of the listed rotational speeds 80. For example, the region 90 in the table 70 includes high rotational speeds 80 at which it may not be practical to supply low, or no load, power output values. Similarly, the region 92 in the table 70

-22-includes low rotational speeds 80 at which it may not be practical to supply high power output values. Accordingly, in this embodiment, no cost values are assigned to operating points in the regions 90 and 92.
The plurality of values 82 in the table 70, therefore represent operating costs associated with operating at a plurality different operating points at which the genset 10 is capable of supplying electrical power.
In one embodiment data representing the table 70 is encoded on a CD-ROM
disk and loaded into the location 56 in the RAM 44, shown in Figure 2.
Alternatively, the data representing the table 70 may be encoded on a computer readable signal and loaded into the location 56. The computer readable signal may be received over an internet connection, over a serial or parallel cable, a wireless connection, or any other medium for transferring data, for example.
Generating cost values Referring to Figure 4, an apparatus for producing the plurality of cost values in accordance with one embodiment of the invention is shown generally at 110.
The apparatus 110 includes the genset 10, shown in Figure 1, and also includes the processor circuit 18, shown in Figure 1 and Figure 2. However, it should be understood that the apparatus 110 may be implemented using a separate processor circuit.
As described above the output 28 of the processor circuit 18 is in communication with the interface 30 of the engine 14 for controlling the engine, and optionally the output 28 may also be in communication with the interface 32 of the generator 16, for controlling the generator. The processor circuit 18 also includes an input 112 for receiving operating condition values.

-23-The engine 14 further includes a plurality of sensors 114 for sensing operating conditions of the engine. The sensors 114 may include, for example, a fuel sensor 116, a hydrocarbon (HC) sensor 118, a carbon monoxide (CO) 120, and a nitrogen-oxide (N0x) sensor 122. Other sensors may be included in the plurality of sensors 114, such as a particulate matter sensor (not shown). The sensors 114 ¨ 122 each have an associated output 124 ¨ 130 respectively.
Each of the outputs 124 ¨ 130 produce a signal representing the respective operating conditions. The outputs 124 ¨ 130 are in communication with the input 112 of the processor circuit 18.
Operation ¨ peneratinp cost values The operation of the apparatus 110 is described with reference to Figure 2, Figure 4, Figure 5, and Figure 6. Referring to Figure 5, a flow chart depicting blocks of code for directing the processor circuit 18 in Figure 2 to produce the plurality of cost values, is shown generally at 160. The blocks generally represent codes that may be read from the computer readable medium 50, and stored in the program memory 42, for directing the CPU 40 to perform various functions related to producing the plurality of cost values. The actual code to implement each block may be written in any suitable program language, such as C, C++ and/or assembly code, for example.
The process begins with a first block of codes 162, which directs the CPU 40 to assign weights a; to the operating conditions and to store the operating condition weights in the location 60 in the RAM 44.
In one embodiment the processor circuit may implement a user interface (not shown) and block 164 may optionally direct the CPU 40 to receive user input at the I/O input 62 (shown in Figure 2). The user input may be received from a keyboard or other user input device, for example.

-24-Block 166 directs the processor circuit 18 to receive the operating condition values from the sensors 114 (shown in Figure 4) at the input 112. Block 166 further directs the processor circuit 18 to cause the operating condition values to be stored in the location 58 in the RAM 44.
Referring to Figure 6 the operating condition values are presented in tabular format at 200. The table 200 includes operating condition values 224 for a plurality of operating conditions 210, including a fuel consumption operating condition 202, a hydrocarbon emission operating condition 204, a carbon monoxide emission operating condition 206, and a nitrogen oxide emission operating condition 208.
The operating condition values received from the plurality of sensors 114 may include operating condition values expressed in a number of different units of measurement. For example, the hydrocarbon, carbon monoxide, and nitrogen oxide operating condition values, which represent engine emissions, may be expressed in parts per million (ppm), while the fuel consumption operating condition value may be expressed in kilograms of fuel consumed per hour of operation (kg/h). Accordingly, block 168 directs the processor circuit 18 to normalize the operating condition values stored in the location of the RAM 44.
The operating condition values 224 in the table 200 corresponding to each of the operating conditions 210 are separately normalized such that each operating condition includes values ranging from 0.000 to 1.000. For example the fuel consumption operating condition values may generally range from 0 kg/h to about 8.4 kg/h for one particular engine 14, and the values in the table 200 are normalized so that the no load, 1600rpm fuel consumption value is 0.000 while the 35kW, 3400rpm consumption value is 1.000. Operating condition values for other operating conditions are similarly normalized over the operating power range of values 222 and the engine speed range of

-25-values 220. Note that the operating condition values in the Table 200 were determined by experiment and may include noise and/or experimental error and are used herein for illustrative purposes only.
Referring back to Figure 5, the process continues at block 170, which directs the processor circuit 18 to read the RAM 44 to retrieve the normalized operating condition values from the location 58 and to read the weights a;
stored in the location 60. Block 170 further directs the processor circuit 18 to apply the weights to the operating condition values 224 to produce the cost values 82 in the table 70 (shown in Figure 3). The cost values 82 in the table 70 were generated by applying the function below to the normalized operating condition values in the table 200:
Of = Ea,.C, (Equation 1) where 0; are the cost values associated with the plurality of operating points, C; are the operating condition values 224 corresponding to the plurality of operating conditions 210. For sake of simplicity, in generating the cost values in the table 70 the weights a; were set to unity, which assigns equal weight to each of the operating conditions 210.
Other cost value combination functions may also be used to combine the cost values C; in other ways.
The operating condition values 224 in the table 200 may be produced by operating the genset 10 (shown in Figure 4) at the power output values 222 while receiving signals from the sensors 114, which represent the real-time values of the operating conditions 210.
In other embodiments the operating condition values may be produced before operating the genset 10. For example a manufacturer of the engine 14 may

-26-provide expected operating condition values for a specific type of engine.
Alternatively the operating condition values may be established by performing tests on an engine of type and/or performance similar to the engine 14. Such values may be provided on the computer readable medium 50 and may be loaded into the processor circuit 18 using a media reader, such as the media reader/writer shown at 46 in Figure 2.
Referring back to Figure 5, in embodiments where at least one of the operating condition values is a real-time value provided by a sensor 114 the processor circuit may periodically repeat execution of the blocks 166, 168, and 170, thereby producing a new set of cost values based on changed operating condition values.
In other embodiments, it may desirable to alter the weights ai over time to compensate for changes in environment or the genset 10, in which case the blocks 166 to 170 may be repeated with the updated weights, thus producing new cost values 82.
Finding local minimum cost values Referring back to Figure 3, a minimum cost value for each power output row 71 may easily be identified by inspection of the values 82, or by performing a simple linear search for a minimum value in the row. However, in some embodiments the genset 10 may include a larger plurality of possible operating points, in which case a simple linear search for a minimum value for each power output row 71 may be slow or impractical, particularly in real-time implementations.
In one embodiment, a golden section search is used to find the minimum cost value at each successive power output value. The golden section search is a bracketing technique, which may be applied to a set of values to find a single minimum value in the set between an upper bound bracket value and a lower

-27-bound bracket value. The search begins by selecting upper and lower bound brackets at end points of the range of engine speeds 80 in the table 70. The upper and lower bound brackets are then successively narrowed until a minimum is found. The technique derives its name from the golden ratio, which has been found to be an effective bracketing ratio. Applying the golden ratio involves selecting an intermediate engine speed 80 between the upper bound bracket and the lower bound bracket that is 0.38197 from one end and 0.61803 from the other end, and then moving the bracket having a greater corresponding cost value 82, to the intermediate engine speed, which then becomes the new upper or lower bound bracket. The process is then repeated until the minimum cost value coincides with either the upper bound bracket or the lower bound bracket, in which case the lesser of the cost values corresponding to the upper and lower bound brackets is the minimum cost value.
The application of the golden section search technique to finding the minimum cost value for each power output row 71 is described with reference to the row 76 of the table 70. The first step in the application of the technique is to select 1600rpm as the lower bound bracket and 3400rpm as the upper bound bracket, and to calculate an intermediate point between the upper bound bracket and the lower bound bracket using the golden section ratio of 0.38197, yielding an intermediate point of 2287rpm. The speed value 2200rpm the closest value to the calculated intermediate value of 2287rpm.
Since the cost value at 3400rpm is 0.791, which is larger than the cost values at 2200rpm and 1600rpm, a new upper bound bracket of 2200rpm is selected.
Using the new upper bound bracket of 2200rpm and the lower bound bracket of 1600rpm, the golden ratio is again applied to find an intermediate point, which in this case evaluates to 1829rpm. This intermediate point is closest to the 1800rpm speed value. Again the upper bound bracket value of 2200rpm has a corresponding cost value of 0.550 which is larger than the cost values at 1800 and 1600rpm. Accordingly the new upper bound bracket is chosen at

-28-180Orpm. Because there are no intermediate values between the 1800rpm upper bound bracket value and the 1600rpm lower bound bracket value, the minimum of these two values represents the minimum cost value in the row 76, which in this case is 0.526 or 1800rpnn.
Advantageously the golden section search allows quick convergence on a minimum value in a plurality of values having a single minimum between an upper bound and a lower bound. Referring to Figure 3, the minimum cost value for each power output row 71 is indicated in bold italic typeface in the table 70.
Selecting operating points A set of operating points corresponding to each of the bolded costs values in the table 70, would minimize the overall genset operating cost. However such a set of values would result in a rotational speed of the engine increasing in a somewhat sporadic manner in response to monotonically increasing power output values 71. For example, if the output power were to increase from a no load value to 15kW, the operating speed at no load would start at 1600rpm, and then jump to 2000rpm as the power output is increased to 10kW, and then the speed would reduce to 1800rpm again at 15kW. While such a set of operating points may optimize the sum of the cost values shown in the table 70, other operating conditions such as the lifetime of the engine, mechanical stresses in the engine, drivability for hybrid vehicle implementations, and other factors, may not be optimized by selecting these local minimum cost values at each power output value.
Accordingly an additional constraint is placed on the selection of the set of operating values. The additional constraint requires that the engine speed increase (or decrease) monotonically with monotonically increasing (or decreasing) electrical power output value. Accordingly, in response to each

-29-increase in power output value, the engine speed value should either stay the same or should increase.
In general, selecting an optimal set of operating points involves selecting a set of operating points from the table 70. This involves selecting a particular rotational speed at each power output value, and then computing a sum of the cost values for the set of operating points. A set of operating points that has the lowest sum of cost values, while meeting the constraint of monotonically increasing speed, will be the optimal set of operating points.
The sum of cost values for a set of operating points may be written as:
Sum= E0, (Equation 2) J=1 where 0; are the cost values 82 at each of n power output values 71 (in Figure 3 n = 8).
In general, when operating the genset 10, some power outputs may be more frequently used than other power outputs. For example, when operating a hybrid electric vehicle, power output values proximate a midpoint of the range of power outputs that the generator is capable of supplying, may be more frequently used than power output values at a lower end and a higher end of the range. Accordingly, in one embodiment, the sums may be computed using the function below:
Sum =Ebi.0 (Equation 3) J=1 where bi are weights that cause cost values corresponding to more frequently demanded power output values to be assigned greater weight in the sum than cost values corresponding to less frequently demanded electrical power output values. In a hybrid electric vehicle, the weights b; may be used to reflect real world drive cycles when the vehicle is operated in a particular terrain. The weight factors 131 may further be modified when genset operating

-30-conditions change, thus facilitating adaptation of the genset to a changing environment.
Dynamic programming selection technique In one embodiment a dynamic programming technique is used to select the set of operating points for the genset 10. In dynamic programming an optimization problem is divided into sub-optimization problems, which are successively optimized to obtain an overall optimized solution. In the following description, for sake of clarity it is assumed that the weights by are all unity (i.e. all power output values have the same importance as per Equation 2). In practice, higher weights may be set for some operating conditions that other operating conditions in accordance with the relative impact of the various conditions on the environment, and/or the cost of operating the genset, for example.
The selection of the set of operating points using a dynamic programming algorithm is described with reference to Figure 3, Figure 7, and Figure 8.
Referring to Figure 7 a flow chart depicting blocks of code for directing the processor circuit 18 (shown in Figure 2) to select the set of operating conditions, is shown generally at 230. The blocks generally represent codes that may be read from the computer readable medium 50, and stored in the program memory 42, for directing the CPU 40 to perform various functions related to producing the plurality of cost values. The actual code to implement each block may be written in any suitable program language, such as C, C++
and/or assembly code, for example.
The process begins with a first block of codes 232, which directs the processor circuit 18 to retrieve the first power output value from the location 64 of the RAM 44.

-31-Figure 8 shows a graphical depiction of the operating points 82 shown in Figure 3, where each circle 250 represents an operating point having a particular power output value 252 (x-axis), and a particular rotational speed 254 (y-axis). The first power output value 252 is OkW. However in other embodiments the optimization may start at a maximum power output value, for example.
Referring back to Figure 7, the process continues at block 234 which directs the processor circuit 18 to calculate the sum of cost values at the first power output value of OkW.
Referring to Figure 3, at a no-load power output value (0kW), the minimum cost for the power output value row 72 in the table 70 occurs at 1600rpm.
Block 236 thus directs the processor circuit 18 to select the engine speed 1600rpm as the first operating point, since this point has a minimum cost value. The operating point is shown in Figure 8 at 260. Since the minimum cost value in the row 72 is selected, the sum of cost values is the minimum operating cost at OkW (i.e. Sum = 1.221).
Block 238 directs the processor circuit 18 to determine whether the present power output value is the last power output value. If "NO", then the process continues at block 240, where the next power output value is read from the RAM 44 at location 64. The process then returns to block 234.
The next power output value (5kW) is read from the operating points stored in the location 64 in the RAM 44. Block 234 again directs the processor circuit to calculate the sum of cost values. At a 5kW power output, any of the speeds 1600rpm ¨ 3400rpm would satisfy the constraint of monotonically increasing the speed as the power output is increased. Since the minimum cost (0.819) meets the constraint, the sum of this cost value and the cost value at OkW will result in minimum sum of operating costs up to and including

-32-the 5kW power output value (i.e. Sum = 1.221 + 0.819 = 2.040). Block 236 thus directs the processor circuit to add the operating point 262 to the set of operating points. The process continues as described above through 10kW to 35kW power output values.
At 10kW power output, the speeds 2000 rpm ¨ 3400rpm would all satisfy the constraint of monotonically increasing the speed. However, even though the minimum cost for the power output value row 76 occurs at 1800rpm, the speed 1800rpm (and 1600rpm) would not satisfy the constraint. At this point in the process, an operating point 264 corresponding to a speed of 2000rpm, which meets the constraint but is not a minimum cost operating point, may be selected.
Alternatively, one or more of the previously selected operating points in the set may be changed, such that the minimum cost value in row 76 may be selected while meeting the constraint. The decision of which alternative to select made by calculating a plurality of cost value sums corresponding to possible sets of operating points leading to the minimum cost value (0.309 @
1800rpm) and other cost values such as 0.317 @ 1600rpm and 0.333 @2000rpm, for example. From the plurality of cost value sums, a set of operating points that yields a minimum sum is selected. For the values of cost values indicated in the table 70, the minimum sum occurs for the operating point set (Okw,1600rpm), (5kW,2000rpm), (10kW,2000rpm). If however, the minimum sum of cost values were to correspond to a different set of operating points, this may have resulted in changing a previously selected operating point in the set. For example, if the minimum sum had corresponded to the set of operating points (Okw;1600rpm), (5kW,1800rpm), (10kW,1800rpm), then the operating point at 5kW would have been changed from 2000rpm to 1800rpm.

-33-If at block 238 the power output value is at the last power output value, then the process ends at block 242.
It may be appreciated that for each value in the power output value rows 71, when calculating the sums of cost values, some portions of the sums may already have been calculated at a previous power value. For example, when optimizing at 5kW output power, at least the sum from 1600rpm to 2000rpm would have been calculated (Sum = 221 + 0.819 = 2.040). Thus, when optimizing at 10kW, in calculating the sum corresponding to the set of operating points (Okw;1600rpm), (5kW,2000rpm), (10kW,2000rpm), the sum from 1600rpm to 2000rpm has already been calculated as 2.040.
Accordingly, the number of calculations that are required at each power output value may be reduced by memoizing results of previous calculations.
Memoization is a technique used to speed up computer programs by storing the results of previous calculations for re-use, rather than re-computing them each time. Advantageously, memoizing has a significant effect in reducing the number of operations required in selecting a set of operating points, thus facilitating computationally efficient selection of operating points for the genset 10, when faced by a changing operating condition or change in either of the weights ai or bi.
Referring again to Figure 8, the process is successively applied at each of the power output values 252 up to a 35kW power output value until a complete set of operating points has been selected as indicated by the bold line 258.
In the above description, for simplicity it was initially assumed that the weights 191 were all set to unity in accordance with Equation 2. The weights bi are simply taken into account when applying the dynamic programming technique by calculating the sums in accordance with Equation 3 using weights bi having differing values to account for real-world conditions of use of the genset.

-34-Hybrid electric vehicle application Referring to Figure 9, a hybrid electric vehicle embodiment of the invention is shown generally at 300. The hybrid vehicle 300 includes a first pair of driven wheels 302 and a second pair of wheels 304. The hybrid vehicle 300 further includes an electric drive motor 306, a gearbox 308 and a coupling shaft 310.
The drive motor 306 is coupled to the driven wheels 302 through the gearbox 308 and the coupling shaft 310, to provide mechanical power to the driven wheels.
The hybrid vehicle 300 also includes a storage element 312, which in this embodiment includes a plurality of storage batteries 314 and a plurality of capacitor storage elements. The storage element 312 is in communication with a first energy bus 318. In one embodiment the batteries 314 are nickel metal hydride cells (NiMH), and the capacitor 316 includes one or more ultracapacitors. In other embodiments the storage element 312 may include batteries only, ultracapicitors only, or may include a flywheel energy storage element. Flywheel energy storage elements generally operate by accelerating an electrically coupled flywheel rotor in a motor/generator to a high speed thus storing inertial energy in the flywheel for later use.
The hybrid vehicle 300 further includes a genset 320, which as described above includes an engine 322 coupled to an electrical generator 324. The generator is in communication with a second energy bus 326. The genset 320 also includes a processor circuit 332, which may be the processor circuit 18 shown in Figures 1, 2, and 4.
The engine 320 may be any type of internal or external combustion engine, e.g. Otto, Atkinson and diesel cycle engines, Stirling engines, gas turbine engines etc. The engine 320 may also run on any type of fuel such as gasoline, diesel, biogas or other bio-fuels including cellulosic and other ethanols, propane etc.

-35-The hybrid vehicle 300 also includes a controller 328, which is in communication with the first and second energy busses 318 and 326, for receiving electrical energy, and is also in communication with the drive motor 306 for supplying energy to the drive motor to drive the wheels 302. The controller may include various voltage converters such as a DC-DC converter for example to convert the electrical energy received on the busses 318 and 326 into a voltage suitable for supplying the drive motor 306. The controller 328 further includes an output 342 for producing power demand signals. The output 342 is in communication with the processor circuit 332.
In this embodiment the hybrid vehicle 300 includes a storage element controller 340, which is in communication with the storage element 312 for monitoring a state of charge of the storage element 312. The storage element controller 340 is also in communication with the controller 328 for providing a state of charge signal associated with a state of charge of the storage element 312 to the controller.
The hybrid vehicle also includes a foot pedal 329 for producing a signal representing a requested power to the drive motor for driving the wheels. The foot pedal 329 is in communication with the controller 328 and is disposed in a driving compartment (not shown) of the hybrid vehicle 300.
In this embodiment the hybrid vehicle also includes an accessory 330, such as compartment lighting, a wiper motor, a compartment cooling fan, etc. The controller 328 is in communication with the accessory 330 for supplying electrical energy to the accessory at a suitable voltage & current (e.g. at DC).

-36-Operation of the hybrid vehicle When an operator of the hybrid vehicle 300 depresses the foot pedal 329, a signal representing a requested power associated with an amount of depression of the foot pedal is produced. The controller 328 receives the requested power signal from the foot pedal 329, and the state of charge signal from the storage element controller 340, and responds by producing a power demand signal at the output 342 for demanding a power output from the genset 320. The power output demanded from the genset 320 by the controller 328 depends on the state of charge of the storage element 312 and the power request signal received from the foot pedal 329. The power demand signal produced by the controller 328 may also include an additional demand for supplying power to the accessory 330.
The power requested by the operator's depression of the foot pedal 329 may thus be supplied by the storage element 312 over the bus 318, by the genset 320 over the bus 326, or by a combination of both. The request for drive power to be delivered to the wheels 203 is thus satisfied by power supplied by the genset 320 alone, or by the storage element 312 alone, or by a combination of a first portion of power supplied by the storage element and a second portion of power supplied by the genset.
Alternatively, if the storage element 312 is discharged, then the storage element controller produces a state of charge signal indicating that charging of the storage element 312 is required. The controller 328 then responds by producing a power demand signal to the genset 320 that includes an additional demand for power to charge the storage element 312, and the genset 320 is required to generate power for satisfying both the power request from the foot pedal 329, and a power demand for charging the storage element 312.

-37-When the operator removes releases the foot pedal 329 without disengaging the gearbox 308, causing the hybrid vehicle 300 to coast, power from the wheels 302 drive the motor 306 thus generating electrical power, which advantageously may be coupled through the controller 328 to charge the storage element 312. If the storage element is fully charged this power may need to be dissipated in a resistor, for example.
The processor circuit 332 and genset 320 operates as described above in connection with the genset 10 and the processor circuit 18.
In the embodiment shown, the processor circuit 332 receives a set of operating points (for example the operating points shown in Figure 8 at 258) that have been determined externally to the vehicle 300 and stored on a computer readable medium, such as the computer readable medium 50 shown in Figure 2. The set of operating points is then read into the processor circuit 332 using a media reader, such as the media reader/writer 46 shown in Figure 2. When operating the hybrid vehicle 300, the genset 320 receives the power demand signal from the controller 328, and responds by running the engine 322 at an engine speed corresponding to the operating point in the set of operating points that corresponds to the demanded power. The generator 324 responds by supplying the demanded level of power to the energy bus 326.
Alternatively, in another embodiment, the cost values 82 (shown in Figure 3) may be determined externally to the hybrid vehicle 300, stored on a computer readable medium, and read into the processor circuit 332 using a media reader. The processor circuit 332 may further be configured to generate a record of received demands for power outputs during operation of the genset 320 over a period of time. The processor circuit 332 then calculates the weights bj, assigning greater weight to more frequently used power output values, according to the record of received demands. When the weights 131 have been updated, the processor circuit 332 is configured to select the set of operating points using the externally supplied cost values, as described

-38-above. Advantageously, in this embodiment the vehicle 300 is able to adapt to a change in environment (for example, when moving the vehicle from a substantially flat terrain, to mountainous terrain) by changing the weights bp In another embodiment, the operating condition values (shown in Figure 6) may be read into the vehicle by a media reader into the processor circuit 332, and the cost values may be calculated on-board the hybrid vehicle 300, thereby facilitating on-board changes to the weights ai in accordance with Equation 1 for altering a desired-trade off between the various operating conditions. Advantageously, in this embodiment the vehicle 300 is able to select new operating points reflecting a different desired trade off between the operating conditions. For example, when operating the vehicle in a city environment, emissions may have to meet stricter criteria than in more rural environments, and the weights ai may be changed to reflect this difference.
In yet another embodiment, at least some of the operating condition values are provided by on-board sensors, such as the sensors 114 shown in Figure 4. In this embodiment, the operating condition values, cost value values and the selection of a set of operating points for the genset 320 all take place in on-board the vehicle 300 in the processor circuit 332. Furthermore, in this embodiment, since the operating conditions are represented by real-time signals from the sensors 114, it may be more important to be able to quickly select a new set of operating points in response to changing operating condition values. Advantageously, the use of a dynamic programming technique provides a computationally efficient method for selecting the operating points, thus reducing a hardware speed requirement for the processor circuit 332. The sensing of operating condition values thus facilitates a quick response to changing conditions such as changes in fuel consumption patterns or emissions, due to changes in engine operation over time, for example.

-39-The hybrid vehicle 300 shown in Figure 9 represents an example of what is often referred to as a "serial hybrid vehicle", in that the engine 322 is not mechanically coupled to the wheels 302. Other configurations of hybrid vehicles may include engines that are coupled to both a generator and to a drive wheel of the vehicle, and may be referred to as "parallel hybrid vehicles".
At least some of the above described embodiments of the invention may be applicable to various configurations of parallel hybrid vehicles, where an engine is coupled to a generator. For example, when a hybrid vehicle is at a standstill and no power is being provided to the wheels, the engine and generator may be operated using operating points selected as described above.
Other optimization methods Other optimization methods may also be used to select the set of operating points for the gensets 10 or 320. In general an optimization method is a search method that is able effectively to find a condition, which minimizes the function in Equation 1 or Equation 2, subject to the constraint of monotonically increasing speed. One possible alternative is to compute every possible sum in the range of power output values and engine speeds. Such a "brute force"
technique is easy to implement but has the potential downside of being slow in execution. Where quick response is required for onboard selection of operating points, more efficient techniques such as steepest descent method, conjugate gradients method, variable metric method, Newton's method, linear programming (including golden search), may be used in place of dynamic programming.
The method of steepest descent approaches the minimum in a zig-zag manner, where the new search direction is orthogonal to the previous search direction. However this method converges slowly, since after each step the method searches in an opposite direction. The method of conjugate gradients attempts to alleviate this problems with steepest descent methods to by

-40-"learning" from experience. Newton's method results in faster convergence, but not necessarily less computing time compared against steepest descent and conjugate gradient methods.
While specific embodiments of the invention have been described and illustrated, such embodiments should be considered illustrative of the invention only and not as limiting the invention as construed in accordance with the accompanying claims.

Claims

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A method for selecting operating conditions of a genset, the genset comprising an engine coupled to an electrical power generator, the genset having a plurality of operating points each comprising an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points, the method comprising:
selecting a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in said set is minimized and such that the engine speed and generator electrical power output values of the operating points in said set increase or decrease monotonically.

2. The method of claim 1 further comprising assigning a weight to each of the power output values such that cost values corresponding to more frequently demanded power outputs values are assigned a greater weight in said sum of cost values than cost values corresponding to less frequently demanded electrical power output values.

3. The method of claim 2 wherein assigning said weight comprises assigning greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.

4. The method of claim 2 further comprising generating a record of received demands for power outputs during operation of the genset over a period of time, and wherein assigning said weights comprises assigning greater weight to more frequently used power output values.

5. The method of claim 4 wherein generating said record comprises generating said record and updating said weights while operating the genset and wherein selecting further comprises selecting a new set of operating points when said weights have been updated.

6. The method of claim 1 wherein selecting comprises:
successively applying a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power outputs the generator is capable of supplying;
and selecting a set of operating points corresponding to a last power output value in said range as said minimized set of operating points for said genset.

7. The method of claim 6 wherein applying said dynamic programming algorithm comprises:
producing a first plurality of said sums of cost values at a first power output value;
memoizing a result of at least one of said first plurality of said sums; and using at least some of said memoized results for producing a further sum of cost values at a subsequent power output value.

8. The method of claim 6 wherein selecting comprises locating a minimum cost value at each power output value and wherein successively applying said dynamic programming algorithm comprises first producing said sums of cost values corresponding to said minimum cost value at each successive power output value.

9. The method of claim 8 wherein locating said minimum cost value comprises locating an operating point corresponding to said minimum cost value using a golden section search technique.

10. The method of claim 1 further comprising receiving a demand to supply power at a demanded power output value and operating the genset at an operating point in said set of operating points corresponding to said demanded power output value.

11. The method of claim 10 wherein the genset is used to generate electrical energy for use in a hybrid electrical vehicle and wherein receiving said demand comprises receiving a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.

12. The method of claim 11 wherein receiving said demand to supply power to said charger comprises receiving a demand to supply power to said charger while said hybrid electric vehicle remains stationary.

13. The method of claim 11 wherein receiving said demand to supply power to said drive motor comprises receiving a drive signal from an operator foot pedal representing a desired drive power to be supplied by said drive motor to wheels of said hybrid vehicle.

14. The method of claim 13 wherein said storage element is operable to supply at least a first portion of said desired drive power, and wherein receiving said demand comprises receiving a demand for a second portion of said desired drive power.

15. The method of claim 13 wherein receiving said demand to supply power to said charger comprises receiving a charge signal from a storage element controller, said charge signal being produced in response to a state of charge associated with said storage element.

16. The method of claim 1 further comprising receiving a plurality of cost values.

17. The method of claim 16 further comprising selecting a new set of operating points in response to receiving said plurality of cost values.

18. The method of claim 1 wherein selecting comprises selecting said set of genset operating points prior to operating said genset.

19. An apparatus for selecting operating conditions of a genset, the genset comprising an engine coupled to an electrical power generator, the genset having a plurality of operating points each comprising an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points, the apparatus comprising:
means for selecting a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in said set is minimized and such that the engine speed and generator electrical power output values of the operating points in said set increase or decrease monotonically.

20. The apparatus of claim 19 further comprising means for assigning a weight to each of the power output values such that cost values corresponding to more frequently demanded power output values are assigned a greater weight in said sum of cost values than cost values corresponding to less frequently demanded electrical power output values.

21. The apparatus of claim 20 wherein said means for assigning said weight comprises means for assigning greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.

22. The apparatus of claim 20 further comprising means for generating a record of received demands for power outputs during operation of the genset over a period of time, and wherein said means for assigning said weights comprises means for assigning greater weight to more frequently used power output values.

23. The apparatus of claim 22 wherein said means for generating said record comprises means for updating said weights while operating the genset and wherein said means for selecting is operably configured to select a new set of operating points when said weights have been updated.

24. The apparatus of claim 19 wherein said means for selecting comprises:

means for successively applying a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power output values the generator is capable of supplying; and means for selecting a set of operating points corresponding to a last power output value in said range as said minimized set of operating points for said genset.

25. The apparatus of claim 24 wherein said means for applying said dynamic programming algorithm comprises:
means for producing a first plurality of said sums of cost values at a first power output value;
means for memoizing a result of at least one of said first plurality of said sums; and means for using at least some of said memoized results for producing a further sum of cost values at a subsequent power output value.

26. The apparatus of claim 24 wherein said means for selecting comprises means for locating a minimum cost value at each power output value and wherein said means for successively applying said dynamic programming algorithm comprises means for first producing said sums of cost values corresponding to said minimum cost value at each successive power output value.

27. The apparatus of claim 26 wherein said means for locating said minimum cost value comprises means for implementing a golden section search technique.

28. The apparatus of claim 19 further comprising means for receiving a demand to supply power at a demanded power output value and means for operating the genset at an operating point in said set of operating points corresponding to said demanded power output value.

29. The apparatus of claim 28 wherein the genset is used to generate electrical energy for use in a hybrid electrical vehicle and wherein said means for receiving said demand comprises means for receiving a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.

30. The apparatus of claim 29 wherein said means for receiving said demand to supply power to said charger comprises means for receiving a demand to supply power to said charger while said hybrid electric vehicle remains stationary.

31. The apparatus of claim 29 wherein said storage element comprises at least one of a storage battery, a capacitor, and an electrically coupled flywheel.

32. The apparatus of claim 29 wherein said means for receiving said demand to supply power to said drive motor comprises means for receiving a drive signal from an operator foot pedal representing a desired drive power to be supplied by said drive motor to wheels of said hybrid vehicle.

33. The apparatus of claim 32 wherein said storage element is operable to supply at least a first portion of said desired drive power, and wherein said demand comprises a demand for a second portion of said desired drive power.

34. The apparatus of claim 32 wherein said means for receiving said demand to supply power to said charger comprises means for receiving a charge signal from a storage element controller, said charge signal being produced in response to a state of charge associated with said storage element.

35. The apparatus of claim 19 further comprising means for receiving a plurality of cost values.

36. The apparatus of claim 35 further comprising means for selecting a new set of operating points in response to receiving said plurality of cost values.

37. The apparatus of claim 19 wherein said means for selecting comprises means for selecting said set of genset operating points prior to operating said genset.

38. An apparatus for selecting operating conditions of a genset, the genset comprising an engine coupled to an electrical power generator, the genset having a plurality of operating points each comprising an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points, the apparatus comprising:

a processor circuit operable to select a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in said set is minimized and such that the engine speed and generator electrical power output values of the operating points in said set increase or decrease monotonically.

39. The apparatus of claim 38 wherein said processor circuit is operably configured to assign a weight to each of the power output values such that cost values corresponding to more frequently demanded power outputs are assigned a greater weight in said sum of cost values than cost values corresponding to less frequently demanded electrical power output values.

40. The apparatus of claim 39 wherein said processor circuit is operably configured to assign greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.

41. The apparatus of claim 39 wherein said processor circuit is operably configured to generate a record of received demands for power outputs during operation of the genset over a period of time and to assign greater weight to more frequently used power output values.

42. The apparatus of claim 41 wherein said processor circuit is operably configured to update said weights while operating the genset and to select a new set of operating points when said weights have been updated.

43. The apparatus of claim 38 wherein said processor circuit is operably configured to:
successively apply a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power outputs the generator is capable of supplying;
and select a set of operating points corresponding to a last power output value in said range as said minimized set of operating points for said genset.

44. The apparatus of claim 43 wherein said processor circuit is operably configured to:
produce a first plurality of said sums of cost values at a first power output value;
memoize a result of at least one of said first plurality of said sums;
and use at least some of said memoized results for producing a further sum of cost values at a subsequent power output value.

45. The apparatus of claim 43 wherein said processor circuit is operably configured to locate a minimum cost value at each power output value and to apply said dynamic programming algorithm to first produce said sums of cost values corresponding to said minimum cost value at each successive power output value.

46. The apparatus of claim 45 wherein said processor circuit is operably configured to locate said minimum cost value using a golden section search technique.

47. The apparatus of claim 38 wherein said processor circuit is operably configured to receive a demand to supply power at a demanded power output value and to operate the genset at an operating point in said set of operating points corresponding to said demanded power output value.

48. The apparatus of claim 47 wherein the genset is used to generate electrical energy for use in a hybrid electrical vehicle and wherein said processor circuit is operably configured to receive a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.

49. The apparatus of claim 48 wherein said processor circuit is operably configured to receive said demand to supply power to said charger while said hybrid electric vehicle remains stationary.

50. The apparatus of claim 48 wherein said storage element comprises at least one of a storage battery, a capacitor, and an electrically coupled flywheel.

51. The apparatus of claim 48 wherein said processor circuit is operably configured to receive a drive signal from an operator foot pedal representing a desired drive power to be supplied by said drive motor to wheels of said hybrid vehicle.

52. The apparatus of claim 51 wherein said storage element is operable to supply at least a first portion of said desired drive power, and wherein said demand comprises a demand for a second portion of said desired drive power.

53. The apparatus of claim 51 wherein said processor circuit is operably configured to receive a charge signal from a storage element controller, said charge signal being produced in response to a state of charge associated with said storage element.

54. The apparatus of claim 38 wherein said processor circuit is operably configured to receive a plurality of cost values.

55. The apparatus of claim 54 wherein said processor circuit is operably configured to select a new set of operating points in response to receiving said plurality of cost values.

56. The apparatus of claim 38 wherein said processor circuit is operably configured to select said set of genset operating points prior to operating the genset.

57. A computer readable medium encoded with codes for directing a processor circuit to perform a method for selecting optimal operating conditions of a genset, the genset comprising an engine coupled to an electrical power generator, the genset having a plurality of operating points each comprising an engine speed value and a generator electrical power output value, and having a plurality of cost values associated with operating the genset at respective operating points, the method comprising:
selecting a set of operating points from the plurality of operating points such that a sum of cost values associated with operating points in said set is minimized and such that the engine speed and generator electrical power output values of the operating points in said set increase or decrease monotonically.

58. An apparatus for selecting operating conditions of a genset comprising an engine coupled to an electrical power generator, the apparatus comprising:
a processor circuit configured to select a set of operating points from a plurality of operating points of the genset, each operating point comprising an engine speed value and a generator electrical power value, the set of operating points being selected such that a sum of cost values associated with the operating points in said set is minimized and such that the engine speed increases or decreases monotonically with monotonically increasing or decreasing electrical power output values.

59. The apparatus of claim 58 wherein said processor circuit is operably configured to assign a weight to each of the power output values such that cost values corresponding to more frequently demanded power outputs are assigned a greater weight in said sum of cost values than cost values corresponding to less frequently demanded electrical power output values.

60. The apparatus of claim 59 wherein said processor circuit is operably configured to assign greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.

61. The apparatus of claim 59 wherein said processor circuit is operably configured to generate a record of received demands for power outputs during operation of the genset over a period of time and to assign greater weight to more frequently used power output values.

62. The apparatus of claim 61 wherein said processor circuit is operably configured to update said weights while operating the genset and to select a new set of operating points when said weights have been updated.

63. The apparatus of claim 58 wherein said processor circuit is operably configured to receive a demand to supply power at a demanded power output value and to operate the genset at an operating point in said set of operating points corresponding to said demanded power output value.

64. The apparatus of claim 63 wherein the genset is used to generate electrical energy for use in a hybrid vehicle and wherein said processor circuit is operably configured to receive a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.

65. The apparatus of claim 58 wherein said processor circuit is operably configured to:
successively apply a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power outputs the generator is capable of supplying; and select a set of operating points corresponding to a last power output value in said range as said set of operating points for said genset.

66. The apparatus of claim 65 wherein said processor circuit is operably configured to:
produce a first plurality of sums of cost values at a first power output value;
memoize a result of at least one of said first plurality of said sums;
and use at least some of said memoized results for producing a further sum of cost values at a subsequent power output value.

67. The apparatus of claim 66 wherein said processor circuit is operably configured to locate a minimum cost value at each power output value and to apply said dynamic programming algorithm to first produce said sums of cost values corresponding to said minimum cost value at each successive power output value.

68. The apparatus of claim 58 wherein said processor circuit is operably configured to receive a plurality of cost values.

69. The apparatus of claim 68 wherein said processor circuit is operably configured to produce said plurality of cost values by:
assigning weights to each of a plurality of operating conditions associated with operating the genset, said weights representing a desired trade-off between said plurality of operating conditions;
receiving operating condition values corresponding to each of the plurality of operating conditions, said operating condition values associated with operating the genset at each of a plurality of operating points, said operating points each comprising an engine speed value and a generator electrical power output value, at least one generator electrical power output value having more than one possible operating point; and combining said operating condition values in accordance with said weights for each respective operating point to produce the cost values for each operating point.

70. The apparatus of claim 68 wherein said processor circuit is operably configured to select a new set of operating points in response to receiving said plurality of cost values.

71. The apparatus of claim 58 wherein said processor circuit is operably configured to select said set of genset operating points prior to operating the genset.

72. A method for selecting optimal operating conditions of a genset comprising an engine coupled to an electrical power generator, the method comprising:
selecting a set of operating points from a plurality of operating points of the genset, each operating point comprising an engine speed value and a generator electrical power value, the set of operating points being selected such that the sum of cost values associated with the operating points in said set is minimized such that the engine speed increases or decreases monotonically with monotonically increasing or decreasing electrical power output values.

73. The method of claim 72 further including assigning a weight to each of the power output values such that cost values corresponding to more frequently demanded power outputs are assigned a greater weight in said sum of cost values than cost values corresponding to less frequently demanded electrical power output values.

74. The method of claim 73 further including assigning greater weight to electrical power output values proximate a midpoint of a range of power outputs that the generator is capable of supplying.

75. The method of claim 73 further including generating a record of received demands for power outputs during operation of the genset over a period of time and assigning greater weight to more frequently used power output values.

76. The method of claim 75 further including updating said weights while operating the genset and selecting a new set of operating points when said weights have been updated.

77. The method of claim 72 further including receiving a demand to supply power at a demanded power output value and operating the genset at an operating point in said set of operating points corresponding to said demanded power output value.

78. The method of claim 77 further including using the genset to generate electrical energy for use in a hybrid vehicle and receiving a demand to supply power to at least one of a drive motor, a charger operable to charge a storage element, and an accessory associated with the hybrid vehicle.

79. The method of claim 72 further including:
successively applying a dynamic programming algorithm to select sets of operating points at successive power output values within a range of power output values the generator is capable of supplying; and selecting a set of operating points corresponding to a last power output value in said range as said minimized set of operating points for said genset.

80. The method of claim 79 further including:
producing a first plurality of sums of cost values at a first power output value;
memoizing a result of at least one of said first plurality of said sums; and using at least some of said memoized results for producing a further sum of cost values at a subsequent power output value.

81. The method of claim 80 further including locating a minimum cost value at each power output value and applying said dynamic programming algorithm to first produce said sums of cost values corresponding to said minimum cost value at each successive power output value.

82. The method of claim 72 further including receiving a plurality of cost values.

83. The method of claim 82 further including producing said plurality of cost values by:
assigning weights to each of a plurality of operating conditions associated with operating the genset, said weights representing a desired trade-off between said plurality of operating conditions;
receiving operating condition values corresponding to each of the plurality of operating conditions, said operating condition values associated with operating the genset at each of a plurality of operating points, said operating points each comprising an engine speed value and a generator electrical power output value, at least one generator electrical power output value having more than one possible operating point; and combining said operating condition values in accordance with said weights for each respective operating point to produce the cost values for each operating point.

84. The method of claim 82 further including selecting a new set of operating points in response to receiving said plurality of cost values.

85. The method of claim 72 further including selecting said set of genset operating points prior to operating the genset.