CA2114835A1

CA2114835A1 - Battery powered electric vehicle and electrical supply system

Info

Publication number: CA2114835A1
Application number: CA002114835A
Authority: CA
Inventors: Ross Martin Green; Michael John Kellaway
Original assignee: Individual
Current assignee: Wavedriver Ltd
Priority date: 1991-08-01
Filing date: 1992-08-03
Publication date: 1993-02-18
Also published as: EP0596988B1; DE69220228D1; DK0596988T3; AU2380592A; GR3024130T3; EP0596988A1; ES2106192T3; AU669853B2; US5642270A; WO1993002887A1; JPH09508256A; ATE153928T1; DE69220228T2

Abstract

A charging system for a battery powered electric vehicle has means (2, 3, 4, 6, 13, 7) for passing electric power to or from its battery in either direction, to or from a utility/mains (100), whereby load-levelling on the mains can be effected. Control of the procedure can be effected by the utility through a suitable communications link. A traction drive system for an electrically driven vehicle having this type of charging system includes an AC
traction motor (5) and a pulse width modulated converter (2) controllable to convert a DC electrical signal from a battery (1) into an AC drive signal for the motor (5). A switch means (4) is connected to a port of the converter, for switching the second port between the traction motor and an AC input port, so that the mains supply can be used to charge the battery (1).

Description

211~835 BATTERY POWERED ELECTRIC VEHICLE
AND E~ECTFUC~AL SYPPI,Y SYS~

Th0 pre~;ent invention relates to electric vehic:le~
S and, more particularly, to battery powered vehic~es having power trans f er syster~s, either integral or ~eparate, aapable o~ transferring power between ut~lity supplies and th~ vehi le~;.
The present invention further relates to battery 10 chzlrging ~;ystems~ required for el~ctric vehicle~ and so-called 'lelectric-hybrid" vehicles of the type having a main electric drive and an auxiliary internal combustion ~ngine ~AICE) dri~e~
Enviros~mental issues have heigh1:ened the interest in 15 ree:ent years in alternative means for providing per~;onal and commercial transpor~ationO Economic and rçgulatory issues have co~ ned to promote the view that electric p~wered vehicleE; will, over the next ten years, appear in signif icant nw~ers . It is possible that the number of 20 elec:tric or electric-hy~rid (E~I) vehicles in 3;ey areas ~ay be around 100,000 or ~or~ by the y@ar 2000.
Aside fro:m the known modest p~rfonaance level~; of elec:tri :: and hybrid ~rehicles, a laajor issue is that of cost . Pr¢sently, EH vahicles with perf ormance levels : 25 ac:aeptabl~ to personal users are expected to ~;~11 at -~- considerably greater prices than functionally comparable onv~nt~ona 1 v~hicles. Of par~icular importanc~ here is the cost- 03E ~he traction battary which is alway~ much mor~
expensive t~an a fuel tank. There are unlik~ly to be any : ;: - 30 compensating C05t sa~ings elsewhere within the vehicle, at - l~ast ~for - the next few years. This ~ans that a bat~ery puwered el~ctric vehicle will alway~; have a considerable - ( - purcha~;e co. t disadv~ntage in co~pari~on with conventional vehicle. The day to day running cost is 3 s do2~linated by the need to replace the }: attery every f ew year~;. Without this need, the running cos~s of an electric v~hicle can be very low.

W093/0~87 PCT/GB92/01435 1 2 ~ 3 5 Eleotric vehicles of this type require charging of their ~atteries from time to time, most commonly overnight.
The charging process can take from less than 1 hour to perhaps 10 hours or more depending on the initi~l state of charge and the charging rate possible. Electric utilities are keen to sponsor the use of large numbers of electric vehicles, primarily to increase their sales of electricity.
A major problem for utility electricity suppliers i8 th¢ matching of supply to instantaneous demand. This involves a considerable amount of planning including estimates of TV audience patterns and weather forecasting.
The load can also vary significant~y due to unforeseen circumstances.
The utility companies attempt to ensure continuity of supply whilst minimising total cost of generation. Several means are used to do this. Firstly, a ut~lity will prefer to keep its most economical generators running where possible, though these typically take longer to bring up to speed and are expensive to use at very low loads. In parallel with this it wiil keep some 'spinning' reserve available. This reserve consists of generators running at - ; no or part load~b~t synchronised with the grid. These can generate additional electricity very rapidly. The cost of ~ maintaining ~such reserve is significant however. In ~ Z5 addition to ~his, utilities maintain a number of other - - generators in a state-where they can be started and brought into use in ~psrhaps l to 30 minute~. In general the smaller systems ~an~be~brought on-line faster but generate electricity less economically than-larger sy~tems.
30 - Other systems are in -use,- or being considered, including pumped storage systems~which use surplus, low-cost electricity to pump water from a one resexvoir to a second reservoir located physically higher. This procedure can then be reversed when additional power is needed.
Battery storage is also becomi~g more attracti~e, including the use of high temperature batteries, though this requires additional rectification and in~erter equipment to W093/02887 P~T/GB92/01435 , . , . . ~

3 2 1 ~
interface to the grid. The capital and running costs of this reserve capacity are high. Such techniques as are described above are generally known as "load levelling"
techniques. Recently there has been increased i~terest in providing means for carrying out this function by giving utilities some control over consumers' load pattern, usuall~ by allowing utilities to remotely turn off or reduce the power consumption of large power consuming loads. This technique is known as "demand side load-10 managementn.
According to a first aspect of the present invention,a charging system for a battery powered electric vehicle has means for passing electric power to or from its battery in either direction, to or from a utility (mains).
15The charging system may be stationary, ie located at the normal point of charge, or else t preferably, may be integral with an on-board electric traction drive system.
Preferably, in this latter case at least, the battery charging/supply sy tem includes ~ number of semiconductor switches.
~ ~ ~;` In order~that the direction and amount of power being ; passed can be controlled locally or remotely by the electric utility company in such a way 5 to match the electricity supply and demand, either on a grid or a local basis, the vehicle charging system needs, preferably, to - include either a timing means for arranging a connected ~;:` charging system to supply the mains grid at appropriate `:tim~s, or else includes a communication facility in order that the utility can send signals to vehicles to cause them 30 ` to vary the power being taken during charging, or to r~ver~e the p~wer flow to supply the grid.-This system of~ers utility companies a potentially`large rapid~ response reserve capacity without having to have spinning reserve and at little or no capital CQSt.
This may make it attractive for them to subsidise electric vehicle ~atteries, thereby removing a barrier to the spread W093/0~87 PCT/GB92/0143~
21~48~5 of electric vehicles and bringing forward the sales of electricity which they need for charging.
The invention thu~ also includes a mains/grid utility electric supply system having means for communicating with S an electric vehicle, which is connected with the grid/~ains supply for charging, to cause the vehicle to transfer electric power from the vehicle battery to the supply.
Minor modifications from the basic charging system as described above are required, for example, to allow communications to be received, and optionally tran~mitted, to/from a local or remote command centre controlled by the utility company, or others. Such a communication link ~ay be based on signals superimposed on the utility supply or on a separate channel such as a cable or fibre-optic llnk lS which may additionally be used for billing purposes.
The charging system described above inherently has the capability to transfer power bidirectionally, that is from the utility supply to the ~ehicle battery or vice versa as it is required both to drive the vehicle (battery to motor) and to charge the battery regeneratively (~otor to battery). The system is based on semiconductor technology.
-~-The selection of the direction and size of the power flow ~; ~ to/from the mains grid can be based on commands sent to the drive from the utility company. It is thu~ possible for ^ 25 ~he utility company to command electric vehicles, which are onnected for charging, to supply or ~ake varying am~unts of power tolfrom the utility supply network. This can be ;- done with both single and ~ulti-phase lines. The amount and direction can be changed very rapidly. The utility company would have the --ability to command connected electric ~ehicles singly, or in batches, to ~atch total - demand and supply on-thei~ network. An advantage of ~he present inYention is that this balance can be both net~ork wide and local, thereby minimising transmission losses.
- 35 A further preferable feature of the present invention - as for the operation of a delta charging strategy - i~
~he use of an accurate state-of-charge estimator used to W093/02887 ' ' , PCT/GB92/01435 .
5 21~483~
control the maximum charge and discharge states of the battery. In one possible implementation this estimator can ea~ily be integrated into the charging co~trol 6y~tem at little additional cost.
S A further advantage of the present system is that it can be controlled in such a way as to operate at different power factors and/or draw/supply controllable harmonic currents. These features are very beneficial to correct for other loads on a network which cause a utility company extra costs in the generation of reactive and/or harmonic power.
A further ad~antage of the present system is that it can be used to supply power to key installations, or to the local ho~e, commercial or industrial location in the event lS of interruption to the main utility supply. In this event the charging syst~m controllers could either be commanded to supply power using the communications link or could : automatically detect the loss of the utilîty supply and initiate generation in response to this. In this mode of -20 oparation t~e charger would generate its own internal frequency reference,and supply an AC voltage based on thi~.
. Where several~electric vehicle chargers are used in this way driving a common network, some means of synchron~sation is required.~ ,In the ~ystem to which ~he present invention relates the.synchronisation would be carried out using the ~ . communicaeion link or by using a master.reference generator i ~ ` ' ``~,h~ .in eachslocal~:group- Some means of preventing~generation when.there:is~a,fault on the incoming line i~ needed, but this~can be implemented by detecting voltage and current at -30~ the interface to the utility supply and-terminatins supply , , if there i a~fault.
:: ~- It .~s al~o necessary to consider the return of the ~ utility supply, and the changeover that must,take place ,~;-- when this occurs. It is likely that a practical system 3S must ha~e the synchronisation signal supplied by the utility company. If this is done it will be possible to break and restore the utility supply with little or no ~11~ 8 3 S PCT/GB92/01435 L

disturbance to consumers. ~his clearly needs a substantial infrastructure but is well within the bounds of existing technology for telephony and cable television and is within the scope of proposed integrated communication services to domestic, commercial and industrial properties.
The way in which this capability is used will clearly depend on the circumstances of the particular utility, network and number of electric vehicles available.
However, it is likely that the usage will be along the following lines: When buying an electric vehicle customers will have a dedicated charging interface installed at the point where the vehicle is to be charged, typically overnight. Similar sites will be available for charging at other locations, such as shopping centres, of'fices etc. At each of these interfaces will be a power conneotion and a co D unications link. }n exchange for some incentive, such as reduced~pricing for the electricity used, the consumer will agree~ to t~e utility controlling the direction of power flow,~subject~to criteria including a minimum state of charge of the vehicle's batteries.' For example, it may be appropriate~to only use vehicles which have batteries at more than 75%, sa~y, of full capacity. The electric vehicle driver can override this function, for example if a very fast charge is~ required for operational reasons, though - 25~ ~this~will lose~ the incentive for that charging cycle.
;'Co -unication links to charging stations are already under discussion~for remote billing based on an identification code stored in each electric vehicle. However, implementations-can be envisaged~where simple operation is ^30 possible without a communications link in place, whereby a connected vehicle is progra D ed to take or deliver set amounts of power at specific times. The programmed times - can, for example, be adjusted when the electricity consumption of the vehicle is recorded by the supplying authority. In a strategy like this, the vehicle logs the exact electricity consumption and the consumer is billed 2 1 ~ 5 regularly on an estimated basis which is confirmed by periodic readings directly from the vehicle.
The utility ~upplier will be capable of estimating the potential power available either from the knowled~e that he already has about the typical electric vehicle charging patterns, or by direct signals from the electric vehicle along the com~unlcation link for billing purpose~. The utility can then sig~al cingle, or more likely groups, of electric vehicle~ to control their power take off or supply from/to the utility network. By scheduling and responding to changing demands the electric vehicles will act as a flexible load leYelling system which eliminates, or reduces, the need to have spinning reserve or to build dedicated load levelling systems, as i5 current practice.
15Although the cost differential between conventional and electric ve~icle~ is, in a major part, due to the cost of the batteries needed, it is also due to the many support systems needed in an electric vehicle. The support systems problem is even ~ore severe in an EH vehicle.
- 20Such support systems include the supply for vehicle - ancillaries and the means for charging the traction motor battery or batteries.
For the first of the~e, means must be provided that deliver a power supp~y for float char~ing of a battery which supports a~ on-board low voltage electrical 8ystem.
-Although an electric or EH vehicle has substantial levels of accessibIe electrical energy, thi~ is invariably at ~oltages inco~patible wi~h conventional vehicle components ~- such~ as bulbs, switches, fuses, relays etc. ~ The main electrical supply for traction purposes is invariably above - ;-- the Safe Extra Low Voltage (SELV) li~it o~ 42V, and may be ~ - higher than 300V, so tha~ reticulation of traction voltages to pecially rated vehicle ancillaries is dangerous as well - as inco~venient and costly.
- In conventional vehicles, support for the low voltage supply is provided by an alternator or generator - rated at several hundred Watts - driYen by the int~rnal combus ion W093/0 ~ 7 1 1 4 8 3 5 PCT/CB92/0143s ' engine, together with regulating and control equipment.
,~ This method is expensive, and although it can be applied to t3 electric and EH vehicle~, it is typical with electric andEH vehicles to use a DC-DC converter to provide the low :!' 5 voltage supply directly from the main traction-battery-, Although the traction battery is conventionally made up of cell blocks, it is not possible to utilise one of these individually. Aside from the safety aspect, using one traction block, or a number of traction blocks, to support lo the low voltage supply cause~ unbalanced discharge and eventual battery damage. The DC-DC converter is also ,s expensive, costing S200 to $400 or more, and adds weight.
~; Traction batteries in EH vehicles are recharged fromthe normal AC mains current using a charging system, but 15 such conventional charging systems are very bulky, weighing ' many kilograms, and are (in the case of industrial electric ~;~ vehicles)~mounted off the vehicle, at the home location.
~.
For personal~electric vehicles, it is common practice to ;c~rry the~charger with the vehicle, 80 that charging can be 20 carried out~at~destinations where there is a convenient AC
supply~.~Such~on-board chargers normally use high frequency switching~technology to provide a charging unit of '~ acceptable~weight, in the tens of kilograms.
' A second~problem~arises when energy flow rates are 25 ~ considered.~A typical vehicle~traction battery will be rated~at (say) 20kW-hr. A'single-phase AC line can deliver energy~ at'~the~'rate'~ of -2 'to ~3kW~, or more in some circucstances, 80 that full recharging is possible in ` around ten ~hs: if perfect ~efficiency at -unity power 30 - factor is~achieved. In practice, the-charge efficiency of - the batteries~;may~ be as Iow as 80%,~the energy efficiency of the charger'may-be 85S, and the-charger may draw line ~ - currents ~with a -power factor of 0.47 or lower.
¦~ Consequently, charging in adeguate time can be a problem,35 and energy costs can become unattractive with simple systems. Furthermore, the injection of harmonic currents to the AC network caused by simple charging systems may also J

W~ 93/02881 ~ PCT/GB92/01435 ~
211g~35 raise both technical and legislative difficulties, particularly if very large numbers of chargers come into use.
Solutions to this second problem include provi~ion of higher-rated single-phase lines, and utilisation of multiphase supplies. Although acceptable, availability of these supplies is not universal. Alternative technical approaches include using special charging systems that draw line currents at near-unity power factor, and the adoption of "delta" or charge/discharge methods that permit batteries to ac~ept energy at higher rates, where an appropriate supply is available. Such approaches are expensive - with a high component cost - and add weight to the vehicle.
15The traction system itself utilises well-known pulse-width modulation (PWM) inverter methods to synthesize a closely-controlled~AC supply, from the DC traction battery.
The controlled AC supply is used to drive conventional induction,~ permanent-magnet synchronous or other motors under-variable speed and torque regimes to meet the demand~
; of th~ vehicle user. Suc~ systems are applicable from small ratings, up to ratings ~f many hundreds of kilowatts.
InYerters of this type are not new, and have been used in small numbers for industrial drive systems from the mid . ~ ~
nineteen sixtie~. ~Early work in the field is described in Schonung, A.~ and Stemmler, ~.; "Static. Freguency Chan~er~
- .w~th - ~Subha~onic~ Control in Conjunction with ~Reversible ariable-Speed AC Drives~, Brown Boveri Review, Vol 51, p.555 (1964).
30The major advantage-of such systems is the brushless nature of the motor, which has mar~edly lower cost - and - higher environmental tolerance - than the brushed DC motors normally used for controllable drives.
Since 1985, dri~e systems of the AC typ~ have come 3S into more general use. An industrial inve~ter drive normally operates by first conver~ing the normal three phase or single phase line supply to an intermediate D~

211~8~5 ln voltage, prior to "inverting" the DC back to AC with the desired param~ters for driving the target motor. Thi8 intermediate rectification process complicates the drive, adding to cost, and has played a part in slowing the spread S of AC inverter drives for industrial applications. An AC
drive is, however, well suited to vehicle traction applications where the primary energy source is DC
batterie~. VehicIe applications of AC cy~tems are ~till in the minority compared to conventional DC brushed traction systems, primarily due to the sophistication of the control systems necessary to achieve satisfactory operation with the AC system - and the. costs of such systems when conventional ~ethods are used.
In the system to which the present invention relates, lS the inverter i based on insulated-gate bipolar transistors,: operating under the control of a ~icrocomputer. Many other device technologies are also ~ applicable, and other control methods aside from PWM can :; also be us~ed. An example is the Load Commutated ~nversion ~ethod, relying on natural commutation of the inverter devices, which~:is~ particularly applicable with permanent magnet ~achines~:
The inverter has an energy efficiency of approximately : 96~ at full load, so that when 50kW are being delivered to th~ traction motor or motors, 2kW is dissipated in the -`:: inverter. ~:At lower power levels, the losses are not as r ' ' substantial.~ :However,- it is-rare for the losses to be ~ ` lower than lkW.
:~ ~o~t inverter drives inherently offer regenerati~e op0ration. Consequently, energy is reco~ered to the batteries when "engine braking" occurs, and this ef~ect is ` ~ exploited in the majority of personal electric and EH
vehicle drives.
:It is possible also to utilise the bidirectional energy flow capabilities of the system which i~ the subject :: of this invention in further way~.

W093J02*87 PCT/GB92/01435 11 211~8~
According to a further aspect of the present invention, a traction drive system, for an electrically driven vehicle, includes an AC traction motor; a pul~e width modulated (PWM) converter controllable to convert a DC electrical signal, fed from its battery to a first portl into an AC drive signal for the motor fed out from a second port; an AC input port; and switch mean~ connected to the second converter port, for switching the second port between the ~raction motor and the AC input port, whereby the AC input port, on connection to a suitable AC source can be connected to the converter to charge the battery.
Preferably, the drive system comprises a traction battery of lead-acid or NaS type (although other types may be e~ually usable), plural DC link capacitors for sourcing and sinking high frequency current pulses that result from the operation of the power converter, a PWM converter consisting of six unidirectional self-commutating : semiconductor switches wi~h anti-parallel diodes (insulated gate bipolar transistors are preferable, but any suitably : ~0 rated self-commutating switch can be utilised) the switches - being arranged in a standard full-wave six-pulse bridge configuration, a` microprocessor based controller one function of which is to generate six drive pulses to :: - provide ~ating signals to ~he static switches, an 2S electromechanical changeover swi~ch consisting of two mechanically interlocked contactors which allow connection of the PWM converter to either the AC traction ~otor or the AC utility supply inlet port. The position of the -:- changeover ~witch is controlled by the microprocessor, but an additional interlock feature will be included which inhibits changeover occurring whilst the vehicle or the traction motor is in motion. The traction motor may be of - the synchronous, asynchronous or switched reluctance type.
The invention also includes apparatus for pro~iding an 3~ auxiliary DC supply for vehicle equipment, fro~ a symetrical multi-phase traction supply which employs a PWM
controlled converter, the apparatus comprising means for W093/02887 PCT/GB92~01435 !
21i~8~5 offsetting a neutral point of the multi-phase supply by off etting the PWM sequences for each inverter output pole.
One example of a grid sy~tem and a ve hicle employing a traction drive and charging system in accordance with the S present invention will now be described with reference to the accompanying drawings, in which:-Figures 1 to 6 are diagrams illustrating the principles of the invention;
Figure 7 is a basic circuit diagram illustrating an inverter drive system embodying the principles of the invention;
Figure 8:illustrates the vectorial relationship of the converter pole voltages, leading to a neutral point offset, according to the second aspect of the invention;
Figure 9 is a flowchart which illustrates a charging regime which may be employed in a vehicle according to the invention;
Figure 10 illustrates a line current waveshape associated with~a charging regime of the type described in ~:~ 20 relation to::~figure 9;
Figures~ llA and B illustrate devices which may be used for handling low ~battery voltage conditions when re-~ :~
:~charging~
Figure:::l2:::illustrates the alternative use of an ~ additional:converter which may be used for handling lowbattery voltage co nditions when re-charging;
FigU re 1~3 i 8: a diagramm atic illustration of a . ~ ~
' grid~lconsum er system according to the invention;
. Figure~14 is~a diagramm atic illustration of a vehicle 30 . charging/supply system; and, .. Figùres 15A and B are diagrams illustrating details of filters which~may employed in the circuit of figure 7.
- ~
- The basic circuit elements for a three phase inverter drive are shoHn in figure 1. The basic inverter circuit 35 comprises a pulse width modulated converter consisting of :
six self-commutating semiconductor switches T~-T6with anti-parallel diodes Dl-D6 arranged in a standard full-wave W093/02887 PCT/GB92/01~35 13 2I 1~
bridge configuration. The diodes D~-D6 are required where the switches T1-T6 are capable of unidirectio~al current conduction only. A microprocessor based controlle~ ~mC ha~
the function of generating the six drive pulses necessary to provide gating signals to the switches Tl-~6. The PWM
converter thus con~erts a DC supply UD (usually a battery source) and supplies the resulting AC signals to an AC
. traction motor M. The traction motor M may be of the ! synchronous, asynchronous or switched reluctance type.
Substantially identi~al circuitry is so~etimes used to give "regenerative rectifier" operation, interchanging power between an AC source and a DC source. A modification of this principle is utilis~-d in the proposed vehicle drive, so that the traction inverter components are utilised - when the vehicle is stationary - as a regenerative rectifier.
The operating principles are best illustrated in the sinqle phase context. As illustrated diagrammatically by figure 2, an AC utility supply (or "mainsn) U is separated : 20 from an AC/DC converter C by a line i~pedance Z. The , , ~ .
: converter in turn is connected to a DC supply YL~. ~here the converter is sinusoidally PWM controlled, an AC source : E is generated at its terminals derived from the DC supply.
The syætem cin then be represented as two independent AC
q ~ . .
~ sources separated:by a line imp~dance Z (figure 3). The impedance Z is normally predo~inantly or wholly inductive in character.
~ ~ . .............. ...
3 ,. ~" ~, . , Taking the mains or utility voltage as a reference, the vectorial representation of the system is ac in figure ;~ 30 4, and pwm~oontrol of the ~onverter C enables variation of the phase and magnitude of the converter-generated AC
. voltage E with respect to ~he utility voltage U, as shown : in figure 5. This control enables forcing of the current , vector in the AC network, and it can be shown that the 35 current ~ector may be forced to lead/lag or be in phase with the utility voltage. Unity power factor operation may WO 93/02887 ' PCl/GB92/01435 211~835 14 be achieved in both rectification and regeneration. In the si~plified case where the line impedance is wholly reactive, the locus of the voltage vector is in quadrature with the utility reference voltage U. This is illustrated by figure 6.
Figure 7 illustrates the basic circuit diagram of one possible system according to the present invention. A
battery 1 supplies a DC voltage to a PWM converter 2 which i8 in the form of six self-commutating semiconductor switches Q~-Q6 with anti-parallel diodes Dl-D6 arranged in a standard full-wave bridge configuration. A
microprocessor based controller 3 controlæ a drive board 8 which generates the required drive pulses necessary to provide gating signals to the switches Q~-Q6. The PWM
converter thus converts the DC supply and supplies the resulting AC signals to the contacts on one side 4a of a solenoid operated multi-contactor switch 4. To the other :
side 4b of the switch 4 are connected the AC traction motor 5 (of the synchronous, asynchronous or switched reluctance type) and, through~appropriate impedances 6, the terminals 7a,b,c of an input~port 7, to which a mains supply may be connected when~the vehicle is stationary.
In normal~operation, the switch 4 connects the converter 2 to the motor 5, the microcontroller 3, through ; 25 the drive board~8, controlling the converter switches Q--Q6 as required to drive the motor. The microcontroller 3 also ,, . . ~ .. ~ ~
' controls' the'''solenoid 4c of the switch 4 through an ~` ` amplifier 9~and is arranged to actuate the switch when '' called upon to do` so by the operator ~o that a mains supply 30 ~' can be connected to the converter 2 through the input port ~; i 7 to charge the battery 1. The controller includes an .
interlock function such that the switch cannot be actuated ' when the vehicle is moving or the traction motor is operating. To this end a sensor 10 monitors the motor's speed and connects to the microcontroller to provide speed signals. In'order to enable charging of the battery, the microcontroller actuates ~he'converter switches Q1-Q6 as W093/0~7 PCT/GB92/01435 required to convert the AC mains ~ignal~ to a DC ~oltage acros~ the terminals of the battery, along the same lines as a "regenerative rectifier~, as described in Desto~beleer, E. and Seguier, G.; ~Use of Pulse Width 5 ~odulated Techniques to Improve the Performances of Rectif~rs~, Proc. 2nd European Conf. Power Electronic~
Vol. 1, Grenoble, 22-24 September 1987, and briefly as outlined below.
This system thus has the advantage, for the addition of a small numker of inexpensive components to the dri~e, of providing sophisticated charging compatible with many input sources - without a separate charging unit as such.
Delta charging is also possible with this system, so that the batteries can be rapidly charged, and indeed discharged economically into the AC source if necessary to prevent memory effects.
A further advantage of the present system is that it minimizes the complexity of the physical infrastructure needed off the vehicle for charging purposes, thereby reducing the cost of providing facilities for charging electric and EH Yehicles, particularly where high-rate : charging services are needed. This will encourage the spread of power points suitable for charging, and hence increase the attractiveness of electric and/or EH vehicles : - ~25 ~ to users.
While:in most circumstances a utility ~ains supply at a~frequency of 6QHz, or a similar low-frequency, will be used as the charging source, it is a further preferable :~ fe~ture of the inYention that a wide range of ~ource ~30 frequencies can ke-- accepted,- including DC and/or frequencies of several kilohertz. The upper frequency limit - is principally determined by econo~ic factors such as the ~ cost of the switching devices Q-Q6 shown in figure 7.
: - : However, with normal designs a source frequency in t~e region of three kilohertz can be accommodated, and ultrasonic frequencies in the region of thirty kilohertz are possible if switching devices optimized for speed are W093/0~7 ` PCT/GB92/01435;
., .
. 211~83~ 16 used in the system illustrated by figure 7. This aspect of ~ the invention is important in situations wherein for 3 legislative or other reason~ a non-galvanic ~eg inductive1 couplinq between the charging source is required. Where S inductive coupling (say) i8 required, power is reticulated to the vehicles to be charged at a freguency of several kilohertz ~and at the appropriate voltage level). In multivehicle charging stations thi~ reticulation would advantageougly be from a centrally mountéd large converter, 10 which transforms power from the normal low freguency source, to the frequency required for charging. To accommodat- an inductive interface, the line impedance 6 in figure 7 is replaced by a two part transformer. m e secondary windin~ of the transformer is carried as part of 15 the vehicIe and connected to the switch 4 shown in figure 7, while the~mating primary of the transformer forms the ~; utility connector ~interface 13 and is held a~ part of the , ~ stationary charging infrastructure. The high frequency of 3~ reticulation is important in reducing the phyFical size and weight of~the~the~ coupling transformer required.
A further~important aspect of the invention is that delta~ charging regimes are easily implemented, wherein ~ ~:
~ the batteries can be discharged into the connected utility .
network without~-di~sipation of energy, as well a~
~ 25 recharged.; This prccedure is important where very rapid chaxging (for exa~ple, at the ~ or ~ hour rates) is needed, and involve~-using~brief or sustained periods of discharge current to ~prevent electrode - polarization effects.
El~ctrode polarization occurs during charging through the - 30 finite time; needed for ion migration-from -the-electrode surfaces.~ Periods of discharge reduce these effects mar~edly, and~per~it the total time required to reach full charge to be significantly reduced, with lowered levels of - battery self heating during charging. Discharging is also used for conditioning of new and partially aged batteries (for the avoidance, for example, of so-called "memory"
:

~ .

W093/0~87 PCT/GB92/01435 211~835 effects3. The operation of a delta charging regime is described with reference to the flowchart of figure 9.
According to the delta charging regi~e exemplified by figure 9, energy is delivered from the charging source to the batteries for a brief period (ranging from less than one second to several seconds or longer). Following thifi delivery period, energy is then withdrawn from the batteries for a period shorter than the delivery period, 80 that there is a nett energy flow from the charging source to the batteries. Resting periods, where energy is neither delivered or extracted, are also possible. Such strategies are well known for rapid charging of batteries. In the improved regime which is used as part of this invention, information regarding the battery state of charge is used to define the parameters (periods, energy delivery and extraction levels etc) of the delta charging regime. In this fashion, specific increments of charge can be added to, or extracted from, the batteries, with the increments being selected to be the most appropriate with regard to the prevailing battery state of charge. Instantaneous measurements of the battery terminal voltage, and ~eagurements of other parameters, are also used to control the charging processes. For exa~ple, if during a charging period (where~energy is being added to the batteries) the te D inal voltage is observed to rise above a predefined level, ch~rging is terminated, and a resting or discharging -period is co D enced. -~
The discharging abilities of the system described here` ~ can also be used in a variety of ways to correct line - 30 1 curren~ oonditions on the A~ network which ~re non-ideal.
For example, the -preponderance of uncontrolled diode-`~ capacitor rectifiers in electronic products that are in - general use today cause a well known depression of the crest of the utility AC ~oltage waveform, such rectifiers only drawing current at, or near to, the crect of the utility waveform. This depression is reflected to the utility as a requirement to supply principally third, some W093/02~87 PCT/GB92/01435 211423~ 1~
fifth, and other harmonic currents, raisi~g the costs of generation and transmission of electricity. The syste~
which is the object of this invention can, howev~r, bæ
contro~led to draw current of a preset fornL from the S utility network. One such waveshape is shown in figure 10.
Under this regime, the energy drawn from the network by the system described here is shifted from the crest of the waveform, and indeed energy is deli~ered back to the network for a brief period at the waveform crest. This strategy has the effect of supporting the network during the period where uncontrolled rectifiers are drawing current, and also offers a brief depolarization period to the batteries under charge. The principal drawback of such - a scheme is the magnitude of the high frequency currents that are induced to flow within the batteries. The rectifier can be programmed if required to draw other waveshapes with different harmonic structures if desired, such programming also being possible by remote communications means if these are available.
Where the traction drive in~erter is used as a regenerative rectifier in the manner suggested by thi~
invention, additional means are required to limit source currents in the ~situation where the DC batte~y voltage level has fallen below the cre~t of the connected voltage source. While it is rare for the voltage of t~e tract~on ; batteries to ~fall below the source voltage crest, such events are ~observed and can reasonably be expected in -~ vehicles which are (a) made available for use by the - general-public,-~(b) equipped with ~igh voltage traction batteries which have voltages, in any case, with a normal ~; value only slightly above the voltage of typical incoming sources, (-c) equipped with high temperature batteries or (d) heavily discharged. Where the vehicle is left idle and not connected to a charging source for a period varying in length from a few days to perhaps several weeks, t~e charge level of the batteries will become slowly depleted, either from supporting on-board systems that require continuous W093/0~87 PCT/GB92/01435 power, and/or from battery sel~ discharge effects. In the case of high temperature batterie~, cooling of the battery will occur over this period. When such batterie~ reash the fxozen state, the voltage level falls essentially to zero.
S In most high temperature battery management scheme~, the battery energy is consumed to maintain battery temperature before the frozen state is reached, so that the frozen state will be coincident with a low level of charge.
Consequently, a regenerative rectifier charging system is not practical for use in normal vehicles without additional means for handling the low battery voltage condition, and indeed unless special steps are taken to account for this condition, large surge currents will result from the closure of the switch 4 (see fi~ure 7) to connect a vehic~e in this state to the normal charging source, with possible battery damage,~and protective systemæ such as fuses being called into play.
Such means could be implemented externally. For examp}e, in the situation where (say) the vehicles are forklifts or material handling devices used within a constrained area-such as a factory, technicians can bring i special power converters and chargers to the discharged vehicle for purposes of providing an initial boost charge.
However, for vehicles used by the public, it is le~s convenient for intervention by service technicians to be necessary,~ and the -equipment should be provided on the ` ~ vehicle itself so that charging can be-carried out by the normall charging source, without any difference being :~ visible to the user.
These means can be implemented in various ways; either in series with the regenerative rectifier on the source or battery sides, or in parallel with the main regenerative rectifier pa~h. An additional converter (as described later) may be proYided, in which case power for the additional converter is drawn from the source side of the switch 4 (figure 7). The series limiting means take the fnrm of a variable impedance which may be variously W093/0~87 ' PCT/GB92/01435 -2 11~ ~ ~ emented using resistors, inductors, controlled semiconductor devices, or other methods.
For example, triac devices may be inserted in ~erie~
with each incoming AC source line. The triac de~ices 17 in figure llA are controlled by the inverter/rectifier 2 and microcontroller 3, under the well-known phase-controlled regime, to reduce the effective AC input voltage - and therefore control the battery currents - during the period until the battery voltage rises above the crest voltage of the AC lines. Drawbacks of this system include the non-sinusoidal nature of the AC line currents during the period when the triacs are in operation. An alternative, illustrated in figure llB, is the placement of a series element between the inverter DC output terminals and the battery. The element 18 may be inserted in either the positive pole, as shown, or the negative pole, as required.
The series-device may con~eniently ~e an arrangement of IGBT devices, or similar semiconductor ~witches. The series device 18 is modulated by the inverter microcontroller according to a pulse width modulated or other regime, to reduce the inverter DC output voltage until the battery voltage rises as before. A parallel device l9 is provided for the control of currents flowing in inductances 20 on the DC side. A bypass switch 21 is used so that high rate charging is possible during normal ~ operation of the syste~ as a regenerative rectifier. Thi~
`1 - method has considerable advantages over the triac method.
-In particular, AC line currents of high quality are still drawn during the initial charging period, and the fioftware required for the control of this series device is similar in many respects to the regenerative rertifier software used for the bulk of the charging period.
In so~e circumstances, it is preferable to provide an additional converter, placed in parallel with the main 35 regenerative rec*ifier, to cover the situation in which the vol~age of both the auxiliary and traction batteries is very low. The placement of the additional converter is . W093/02887 PCT/GB92/01435 .
~' 21 211~83~
illustrated by figure 12. This additional converter 24 is normally controlled by the main inver~er microcontroller under a similar regime as described sbove, but i~ capable of limited operation on its own account. It is connected 1 S so as to charge both the main traction battery 1 and the ; auxiliary battery or batteries 12 (see figure 7 al~o).
Such limited operation invol~es conversion of power at a low rate (say 200W). The additional converter contains I sufficient circuitry to provide this form of operation, j 10 which is invoked only by connecting it to a suitable ; charging source such as a single phase supply. Energy is then convertsd from the charging source by the additional ~'! , converter until the voltage of the auxiliary battery rises , sufficiently that operation of the main inverter microcontroller becomes possible. The ~ain inverter microcontroller then modulates the additional converter according to a~pulse width modulated or other regime, to improve the quality of the currents drawn from the charging source. Charging of the main traction battery is then ~, : 20 possible via the additional converter at higher rate (say ; lkW) without undesirable levels of har~onic pollution being imposed on the connected charging ~ource. With typical : traction batteries it is necessary for 2% or mor~ of full ~h~: charge to be~supplied ~efore the open circuit voltage rises close to the~normal levels. Consequently, at the lkW rate, ` i* ~is necéssary to operate the additional converter for approximately 30- minutes before the trac~ion battery voltage rises~to the levels where normal charging, via the regenerative rectifier, can commence. Suc~ operation is also compatible~with the reheating requirements of high ~ temperature-batteries, such as sodium sulphur or sodium ;~ ~~nickel chloride-batteries for example, which must be heated to the molten state before charging can commence.
.~
:In this way, it is possible to re-energize a vehicle ~1 35 which has become completely "dead" ~even to the extent that the main inverter microcontroller ~as ceased operation), merely by connection of the normal, or another suitable, , , W0~3/0~87 PCT/GB92/01435 211~8~5 22 charging ~ource. Such operation occurs completely without interYention fro~ the vehicle user - although the longer charging time will be apparent to the user - which i~
important for vehicles which are to be operated by the general public. Provided the equipment described here is provided on the vehicle, it is unnecessary for any technician or service organization to be involved. The user can reenergize a deeply discharged vehicle in the normal way.
Drawbacks include the extra cost of the additional converter, although in practice this cost is minimised if the major control functions are delegated to the microcontroller already used for controlling the inverter/regenerative rertifier.
The operation of the vehicle traction drive sy~em shown in Figure 14 of the drawings is generally as described above, but the system will now be further described in relation to figure 13, specifically in relation to the use of the vehicle charging system in reverse, to supply power from the vehicle battery to the utility grid or Dains.
The particular arrangement of consumer units, meters, communication facilities will depend upon individual utility requirements, as will the control software and com~unication link(s) to enable operation.
- - In particular,- alternatives of a dedicated ; communications link or a ripple signal decoder are illustrated. In other respects, the charging system is as described above.~
30 - Figure 13 is a simplified diagram of the grid (mains utility~ ~ystem and associated vehicle components. The diagram shows plural utility generators lO~,102 normally supplying electrical power to a grid, generally indicated at lO0, ~hrough conventional transformers 103 which step up the voltage as normal for transmission. Further trans~ormers 104,lOS step the voltage down to local and then domestic and industrial user lev~ls and the supply is W093/0~87~ PCT~GB92/0l435 fed to individual meters 106,107 and then distributed through conventional domestic or industrial consumer units 108,109. one dedicated outlet is an EV charging outlet 110 which can be connected through a power cord 111 to an electric vehicle 112, the components of which are generally as described abo~e, but include a charging system 113 and traction battery 114.
A ripple signalling unit 115, located at the utility generator can be used to provide signals to corresponding vehicle units 14 to cause connected vehicles to re-supply the grid as desired, but, alternative communications systems such as radio, cable, fibreooptic may be utilised.
These links are indicated schematically at 117,118 and the vehicle-end units for receiving the siqnals are indicated at 15 (see figure 7).
In actual operation of the system to supply the grid or mains from~a given vehicle, in the vehicle the switch 4 (see figure~ 7) connects the converter 2 to the utility supply through an impedance 6 as for charging. The operation i~ substantially the same as for charging, in that the converter-generated voltage is controlled to have a desired vectorial relationship with the utility grid/network voltage. In the case whe~e power is supplied to the qrid~or mains, this relationship is different from the charging case in that the converter-generated voltage is controlled so~t~at power is forced to flow from the attery throug~ the converter into the utility network.
e`vectorial relationship depends on the requirements for powér level~, power factor, and harmonic content. In one implementation of the present system, this requirement is transmitted, along either a dedicated communication link 15 or alternatively a ripple communication system 14, from the utility company or other authorised control authority.
In the context of the type of vehicle to which the present invention primarily relates, where a star connected three phase traction motor is being used, there is a W093/02887 PCTtGB92/01435.i.

21148~5 24 straightf orward method for providing a 12V supply - or indeed any other appropriate voltage that may be de~ired.
~ he method used in this invention is to off~et each motor pole voltage by a constant amount, as illustrated by figure 8. In a symmetrical multiphase system, such offsets which ~re vectorially concurrent have no nett effect on the individual phases~ Such concurrent disturbances cause the neutral point of a 6ymetrical multiphase system to be displaced. The e~fect is commonly obs,erved in multipha~e, inverter-fed,systems as voltage oscillation of the neutral point, relative to the ground terminal of the-DC supply, particularly where an asymetrical off~et occur8 per phase.
In the inverter of the present system, as illustrated by figure 7, an offset is applied by the microcontroller 3 to the PWM sequences generated for each inverter output pole. The offset is such that the the output PWM sequ,ences ~ contain a DC component which - when time-averaged by the : f iltering ef~ect of the motor windings and an additional filter 11 - equates to a neutral point of f set voltage of nominally 14V. The neutral point of a star-connected motor is directly accessible and can.be used twith, appropriate : : protection circuitry) for charging an auxiliary 12V battery 12. Feedback via the traction drive microcontroller 3 is : used to regulate the offset voltage, controlling the auxilia~y battery charging processes.
. Figure 8 illustrates ~he vectorial re~ationship of the pole voltages, leading to th,e neutral point offset.
Several operational points are, important for the practical utilisation of this~ neutral point offset technique. In particular, a periodic re~ersal of the offset, with respect to the centrepoint of the traction battery, is required so that a balanced discharge of both halves of the traction battery is achieved. Diodes 22 as fihown in figure 15A can be added to the filtering circuitry 11 of figure 7 so that unidirectional current flow through the auxiliary battery 12 is maintained in these circumstances. Furthermore, precise control of the pole to W093/0~87 PCT/GB92/01435 21~8~5 psle switching periods of the pul~e width modulated i~verter is required, with the desixed pul~e period d~fferentials required for the creat~on of the neutral point voltage offset being maintained through the power staqe to the motor terminals. In practice, the energy delivered to the auxiliary battery from the main traction battery via the inverter power stage is in ~hort pulses, so that the effects of source impedance are important.
Further~ore, the traction motor must be rated to withstand the additional winding losses that result from th~ extra currents that flow as a consequence of this neutral point of f set technique. In practice, this loss increase is of the order of 1% of the rated machine power, depending on the level of low voltage power delivery, the specific characteristics of the machine, and other factors, so that in typical circumstances no change to the machine design is required.
Where delta-co~nected motors Are used, the neutral point is not directly accessible. An artificial neutral for charging purposes can be conveniently created by a number of star-connected reactive impedances. T~e impedances need only be rated to carry the anticipated auxiliary battery charging currents.
In some systems, it is possible to modulate the pole offset at a high frequency of several hundred Hertz, ~o that $he charging source derived from the neutral point off~et can itself be passed through a small high freque~cy transformer 23 (see figure l5B) to provide isolation if required. A drawback of this secondary-isolated system in motor drive~ intended for wide speed ranges is that the effecti~e carrier frequency available to support the neutral point oscillation becomes low at high motor speeds, as the number and time duration of the PWM pulses used to form each cycle of the basal motor frequency becomes reduced. The isolation tranformer must therefore be designed to support the lowest frequency encountered over the full range of in~erter operation, which has the effect W093/0~7 ~ PCT/GB92/01435 21 i ~3 ~ncreasing the transformer size and weight. The transformer positioning is illustrated by figure lSB. In most systems, however, the effective frequency does not drop below lOOHz, and problems do not therefore arise.
Alternatively, the offset can be temporarily removed if the effective frequency falls below a set point. This means that charging the auxiliary battery will not occur during periods of high speed vehicle operation, but that a high frequency isolation transformer is still usable. Thi~
is unlikely to be a disadvantage for these vehicles, and if necessary (i~ charging is essential) the offset can be restored by introducing additional PWM pulses, at the expense of~underfluxing the traction motor slightly.

:, .

, :

, . . -~ ' ' ' ~ .

Claims

1. A charging system for a battery powered electric vehicle having means (2,3,4,6,13,7) for passing electric power to or from its battery in either direction, to or from a utility/mains (100).

2. A system according to claim 1, which is integral with an on-board vehicle electric traction drive system.

3. A system according to claim 1, wherein the battery charging/supply system includes a number of semiconductor switches (Q1-Q6).

4. A system according to any of claims 1 to 3, which, in order that the direction and amount of power being passed can be controlled locally or remotely by the electric utility company in such a way as to match the electricity supply and demand, either on a grid or a local basis, includes a timing means for arranging a connected charging system to supply the mains grid at appropriate times.

5. A system according to any of claims 1 to 3, which, in order that the direction and amount of power being passed can be controlled locally or remotely by the electric utility company in such a way as to match the electricity supply and demand, either on a grid or a local basis, includes a communication facility (14/15,115/117,118) in order that the utility can send signals to vehicles to cause them to supply the grid.

6. A vehicle according to any of claims 1 to 5, further including a state-of-charge estimator (16) to control the maximum charge and discharge states of the battery (1).

7. A mains/grid utility electric supply system having means (115) for communicating with an electric vehicle, which is connected with the grid/mains supply for charging, to cause the vehicle selectively either to transfer electric power from the vehicle battery to the supply or alter or reduce the charging current.

8. A system according to claim 7, wherein the communication link (115) is based on signals superimposed on the utility supply or on a separate channel (117,118) such as a radio, cable or fibre-optic link which may additionally be used for billing purposes.

9. A traction drive system, for an electrically driven vehicle, the system having a charging system according to any of claims 1 to 6, and including an AC traction motor (S); a pulse width modulated converter (2) controllable to convert a DC electrical signal, fed from a battery (1) to a first port, into an AC drive signal for the motor fed out from a second port (4a); an AC input port (7); and switch means (4) connected to the second converter port, for switching the second port between the traction motor and AC
input port, whereby the AC input port, on connection to a suitable AC source can be connected to the converter (2) to charge the battery (1).

10. A drive system according to claim 9, which further comprises plural DC link capacitors for sourcing and sinking high frequency current pulses that result from the operation of the power converter (2), and a PWM converter consisting of unidirectional self-commutating semiconductor switches (Q1-Q6) with anti-parallel diodes (D1-D6), the switches being arranged in a standard full-wave bridge configuration.

11. A system according to claim 10, including a microprocessor based controller (3), one function of which is to generate drive pulses to provide gating signals to the switches (Q1-Q6).

12. A system according to any of claims 9 to 11, including an electromechanical changeover switch (4) consisting of two mechanically interlocked contactors which allow connection of the PWM converter (2) to either the AC
traction motor (5) or the AC utility supply inlet port (7).

13. A system according to claim 12, wherein the position of the changeover switch is controlled by the microprocessor (3).

14. A system according to any of claims 9 to 13, including an interlock (3,10) which inhibits changeover occurring whilst the vehicle or the traction motor is in motion.

15. A system according to any of claims 1 to 6, wherein the battery (1) is of lead-acid, sodium sulphur, sodium nickel chloride or other type suitable for vehicle traction.

16. A system according to any of claims 9 to 15, including apparatus for providing an auxiliary DC voltage supply (11, 22/23) for vehicle equipment, from the PWM controlled converter (2), the apparatus comprising means (3) for offsetting a neutral point of a multi-phase supply by offsetting the PWM sequences for each inverter output pole.

17. A system according to any of claims 9 to 16, including means (17/18-21) for enabling the battery (1) to be charged through the PWM converter (2) even when battery voltage is very low.

18. A system according to claim 17, wherein said means (17/18-21) comprises a switching device (17/18-21) positioned in series between the battery (13 and converter (2) or between the converter (2) and the charging AC supply and under the control of the microcontroller (3).

19. A system according to claim 17, wherein an auxiliary converter (12) is provided, in parallel with the main converter (2), to enable charging of the battery (1) to be effected independently of the main converter (2).