DE102015110867B3 - Condition monitoring of an electromechanical actuator - Google Patents

Abstract

The invention relates to a method and a device for monitoring an electromechanical actuator (1), which comprises a drive (2) and an output (3). The method comprises the following steps: detecting (101) a time series of a state Z (t) of the actuator (1), providing (102) a state estimator ZS for determining an estimated state ZG (t) of the actuator (1), determining (103) a time series of an estimated state ZG (t) of the actuator (1), determining (104) a time interval Δt = [ta, te] in which: ṖAn (t) ≠ 0 and / or ṖAb (t) ≠ 0 and / or ṖGAn (t) ≠ 0 and / or ṖGAb (t) ≠ 0, for the time interval Δt determining (105) at least one residual Ri (t '), for the time interval Δt transforming (106a, b) of the at least one residual Ri (t ') in a frequency range and / or in a path range, determining (107a) error frequencies ωF of the transformed residual Ri, Freq (ω) and associated amplitudes A (ωF) and / or determining (107b) error positions sF of the transformed residual Ri, Path (s) and associated amplitudes A (sF), classifying (108a) the at least one transformed residual Ri, Freq (ω ) into a class KFreq * of several predetermined classes KFreq, and / or classifying (108b) the at least one transformed residuum Ri, path (s) into a class KWeg * of several predetermined classes KWeg, and comparing (109a) the amplitudes A ( ωF) with standard amplitudes ANorm, KFreq (ωF) specified for the class KFreq *, and / or comparisons (109b) of the amplitudes A (sF) with standard amplitudes ANorm, KWeg (sF) given for the class KWeg * if the amplitudes A (ωF) of the transformed residual Ri, Freq (ω) do not satisfy a condition B1 dependent on the normal values Anorm, KFreq (ωF) of the class KFreq *, or if the amplitudes A (sF) of the transformed residue Ri, Way (s) does not satisfy a condition B2 dependent on the normal values ANorm, KWeq (sF) of the class KWeq *, generating (110a) and outputting (100b) a warning.

Description

The invention relates to a method and a device for monitoring an electromechanical actuator comprising a drive and an output. Furthermore, the invention relates to a vehicle, in particular an aircraft, a motor vehicle, a watercraft, a spacecraft with such a device, and a computer program product, a computer system, a digital storage medium, and a computer program.

The term "electromechanical actuator" is to be interpreted broadly. In the present case, it comprises an electric drive (for example an electric motor, a piezomotor or a linear motor, etc.), an output (for example an output shaft, a power take-off, etc.) and optionally a coupling member (for example a transmission) connected between drive and output ) with a transmission ratio, are communicated via the movements of the drive to the output. In addition, all associated actuator components, such as ball bearings, ball screw, seals, etc. are included.

Since in many applications a failure of such an actuator can lead to security-related problems, such as in aviation, the earliest possible detection of imminent errors in the actuator, so-called "health monitoring" is required.

For this purpose, for example, it is known to detect mechanical vibrations of the actuator by means of acceleration sensors and to evaluate them in terms of imminent errors.

From the US 9,002,678 B1 For example, a method for detecting and isolating faults is known. The method uses a Kalmanfiter-Resiuden approach. From the US 5,602,761 A For example, a method for error classification is known in which a statistic of expotential moving average values is compared with a limit value. From the US 6,590,362 B2 For example, a method for detecting incipient faults of an electric motor is known. From the US 7,024,335 B1 For example, a method for state detection as well as a lifetime estimation of a device is known. From the US 2009/0043447 A1 is a method for model-based detection and isolation of a sensor error bekannnt.

The object of the invention is to specify a method and a device with which error detection in an electromechanical actuator is more reliable and more accurate.

The invention results from the features of the independent claims. Advantageous developments and refinements are the subject of the dependent claims. Other features, applications and advantages of the invention will become apparent from the following description, as well as the explanation of embodiments of the invention, which are illustrated in the figures.

A first aspect of the object is achieved by a method for monitoring an electromechanical actuator having a drive and an output, comprising the following steps.

• – Position P_An(t) des Antriebs und Position P_Ab(t) des Abtriebs,
• – optional zumindest eine der daraus ermittelte zugehörige erste Zeitableitung Ṗ_An(t) oder Ṗ_Ab(t), und
• – optional die zwei daraus ermittelten zugehörigen zweiten Zeitableitungen P .._An(t) und P .._Ab(t) .

In one step, a time series of a state Z (t) of the actuator is detected. This state Z (t) comprises at least the following state components:

• Position P _An (t) of the drive and position P _Ab (t) of the output,
• Optionally at least one of the associated first time derivation determined therefrom Ṗ _An (t) or Ṗ _Ab (t), and
• Optionally, the two associated second time derivatives derived therefrom P .. _An (t) and P .. _Ab (t) ,

• – elektrische Stromaufnahme I_Mot(t) des Elektromotor,
• – elektrische Spannung U_Mot(t) am Elektromotor
• – Temperatur T_Mot(t) des Elektromotors
• – Temperatur T_Luft(t) der Luft im Innenvolumen V_Akt(t)
• – Temperatur T_Schm(t) eines Schmiermittels im Innenvolumen V_Akt(t)
• – Last L_Mot(t) des Elektromotors
• – Last L_Zyl(t) des Zylinders
• – Druck P_Akt(t) im Innenvolumen V_Akt(t) des Aktors
• – Menge M_Schm(t) eines Schmiermittels im Aktor
• – Volumen V_Schm(t) des Schmiermittels im Aktor.

The positions P _An (t) and P _Ab (t) are measured by means of suitable sensors. The first and second time derivatives Ṗ _An (t), Ṗ _Ab (t), P .. _An (t) . P .. _Ab (t) can also be measured by means of suitable sensors, or they are calculated from the detected positions P _An (t) and P _Ab (t). Advantageously, the detected state Z (t) further comprises the following state components:

• Electric current consumption I _Mot (t) of the electric motor,
• - Electrical voltage U _Mot (t) on the electric motor
• - Temperature T _Mot (t) of the electric motor
• Temperature T _air (t) of the air in the internal volume V _Akt (t)
• Temperature T _Schm (t) of a lubricant in the internal volume V _Akt (t)
• - Last L _Mot (t) of the electric motor
• - Last L _Zyl (t) of the cylinder
• - Pressure P _Akt (t) in the internal volume V _Akt (t) of the actuator
• - Quantity M _Schm (t) of a lubricant in the actuator
• - Volume V _Schm (t) of the lubricant in the actuator.

These are used in the further course of the proposed method, inter alia, to a classification described below and in particular allow a more accurate estimate of the estimates given below.

In a further step, provision is made of a state estimator ZS for determining an estimated state ZG (t) of the actuator. The state estimator ZS is based on a corresponding state model of the actuator and allows the determination of an estimated state when specifying required for the state model, typically measured by sensors input variables ZG (t) of the actuator. Basically, the more inputs that are also considered by the state model, the more accurate the estimation of state ZG (t). The estimated state is thus based on a combination of measured values by the state estimator ZS. The state estimator is advantageously formulated on the basis of the state model as a Kalman filter.

• – Position PG_An(t) des Antriebs und Position PG_Ab(t) des Abtriebs,
• – zumindest eine der zugehörigen ersten Zeitableitungen: ṖG_Ant) oder ṖG_Ab(t), und
• – optional die zwei zugehörigen zweiten Zeitableitungen P ..G_An(t) , P ..G_Ab(t)

In a further step, the state estimator is used to determine a time series of an estimated state ZG (t) of the actuator, comprising the following estimated state components:

• Position PG _An (t) of the drive and position PG _Ab (t) of the output,
• At least one of the associated first time derivatives: ṖG _An t) or ṖG _Ab (t), and
• Optionally, the two associated second time derivatives P ..G _An (t) , P ..G _Ab (t)

on the basis of the detected states Z (t) and / or on the basis of provided manipulated variables SG (t) for controlling the drive. In an electric motor as a drive of the actuator, manipulated variables are, for example, an electrical voltage U _Stell (t) and / or an electric current I _Stell (t).

In a further step includes determining a time interval .DELTA.t is carried = [t _a, t _e] in which: P _{A n} (t) ≠ 0 and / or p _{a b} (t) ≠ 0 and / or PG _An (t) ≠ 0 and / or ṖG _Ab (t) ≠ 0. In this step, based on the time series of the state Z (t) already recorded, a time interval Δt is determined in which the actuator has moved continuously, ie the measured and / or or estimated speeds are different from zero. The time interval .DELTA.t is advantageously selected such that the state Z (t) of the actuator in the time interval .DELTA.t is largely constant, ie that also other state parameters such as a constant actuator temperature, actuator load, or drive speed, drive Voltage, drive current, etc. (essentially) are constant.

Is also advantageous for the time interval At = [t _a, t _e] determined such that in .DELTA.t for PG _to (t ') is valid and / or P _at (t'): ṖG _An (t ')> 0.9Ṗ _{On, Max} Ṗ _An (t ')> 0.9Ṗ _{On, Max} with Ṗ _{On, Max} = maximum speed of the drive. The factor 0.9 can alternatively be chosen from the range of 0.2 to 0.95, from 0.2 to 0.3, from 0.3 to 0.4, from 0.4 to 0.5, from 0.5 to 0.6, from 0.6 to 0.7, from 0.7 to 0.8 or from 0.8 to 0.9 and is advantageous: 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.95.

Further advantageously, the time interval .DELTA.t = [t _a , t _e ] is determined such that in .DELTA.t ṖG _An (t ') and / or _An (t') vary less than 5%.

In a further step, at least one of the following residuals R _i (t ') is determined for the previously determined time interval Δt: R _{i = 1} (t ') = P _Ab (t') * UV - P _An (t ') R _{i = 2} (t ') = PG _Ab (t') * UV - PG _An (t ') R _{i = 3} (t ') = Ṗ _Ab (t') Ü ÜV - Ṗ _An (t ') R _{i = 4} (t ') = Ṗ G _Ab (t') Ü ÜV - Ṗ G _An (t ') R _{i = 5} (t ') = P .. _Ab (t') * ÜV - P .. _An (t ') R _{i = 6} (t ') = P ..G _Ab (t') * UV - P ..G _An (t ') R _{i = 7} (t ') = P _Ab (t') - PG _Ab (t ') R _{i = 8} (t ') = P _An (t') - PG _An (t ') R _{i = 9} (t ') = Ṗ _Ab (t') - Ṗ G _Ab (t ') R _{i = 10} (t ') = Ṗ _An (t') - Ṗ G _An (t ') R _{i = 11} (t ') = P .. _An (t') -P.G _An (t ') R _{i = 12} (t ') = P .. _Ab (t') -P.G _Ab (t ') with: t 'ε [t _a , t _e ], and ÜV = transmission ratio between drive and output. The at least one residual R _i (t ') is thus again a time series for the time interval t _a to t _e .

In a further step, for the time interval Δt, the at least one residual R _i (t ') is transformed into a frequency range for generating a transformed residual R _{i, Freq} (ω), where ω indicates an angular frequency. The transformation of the at least one residual R _i (t ') into the frequency domain advantageously takes place with a recursively defined discrete Fourier transformation, since this requires less computing power and storage capacity. Alternatively or additionally, the at least one residual R _i (t ') is transformed into a _{path region} for generating a transformed residual R _{i, path} (s), where s indicates a path or an angle.

In a further step, determination of error frequencies ω _{F of} the transformed residual R _{i, Freq} (ω) and associated amplitudes A (ω _F ) and / or determination of error positions s _{F of} the transformed residual R _{i, path} (s) and associated ones takes place Amplitudes A (s _F ). The error frequencies ω _F or the error positions s _{F of} the transformed residuals R _{i, Freq} (ω) and R _{i, path} (s) can be determined by means of known analysis methods (eg frequency analysis / period analysis) are determined. The frequency analysis is preferably carried out with an FFT analysis or a recursively defined discrete FFT analysis.

In a further step, the at least one transformed residual R _{i, Freq} (ω) is classified into a class K _Freq * of several predetermined classes K _Freq . Alternatively or additionally, the at least one transformed residual R _{i, path} (s) is classified into a class K _path * of a plurality of predetermined classes K _path . The classification is carried out in each case on the basis of given for the given classes K _Freq and K _way parameters. Advantageously, a number T temperature-dependent classes, a number L load-dependent classes, and a number S speed-dependent classes are given. The classification of the residuals R _{i, Freq} (ω) and R _{i, path} (s) is therefore advantageously based on the respectively applicable for the time interval .DELTA.t characteristics (eg. Temperature, load and speed of the actuator, or its drive or output). The classification continues to be advantageous depending on the determined error frequencies ω _F and associated amplitudes A (ω _F ) or depending on the determined error positions s _F and associated amplitudes A (s _F ).

In a further step, the amplitudes A (ω _F ) are compared with the standard amplitudes A _{Norm, KFreq} (ω _F ) specified for the class K _Freq *, and / or a comparison of the amplitudes A (s _F ) with for the class K _path * given standard amplitudes A _{norm, KWeg} (s _F ), where, provided the amplitudes A (ω _F ) of the transformed residual R _{i, Freq} (ω) one of the normal _values A _{norm, KFreq} (ω _F ) of the class K _Freq * dependent condition B1, and / or if the amplitudes A (s _F ) of the transformed residual R _{i, path} (s) have a condition B2 dependent on the normal _values A _{Norm, KWeq} (s _F ) of the class K _path * do not meet, a warning is generated and issued.

The standard amplitudes A _{norm, KFreq} (ω) and A _{norm, KWeg *} (s) advantageously indicate the normal behavior of the actuator, ie without any actuator error, and are predefined for each of the classes K _Freq * or K _path *. The standard amplitudes A _{norm, KFreq} (ω) and A _{norm, KWeg *} (s) for the different classes K _Freq and K _path can advantageously be determined based on measurements on faultless actuators. The warning is advantageously transmitted after its generation wired or wirelessly to an output unit for output.

The proposed method allows a robust, detailed, and accurate monitoring of an actuator already with regard to suggestive errors. By using the method, it is thus possible on the basis of the issued warning very early to take precautions or initiate measures that prevent the progression of an actuator error up to the actuator failure. The warning can advantageously cause a changed activation of the actuator. For example, the control can be changed manually or automatically after a warning in such a way that the maximum power that can be output by the actuator is reduced, or the actuator is completely switched off. If such actuators are installed in aircraft, the warning generated in the cockpit is preferably at least optically output and / or forwarded to an ECAM system (ECAM = Electronic Centralized Aircraft Monitor) of the aircraft or a comparable system.

According to the invention, alternatively or additionally at least one of the following residuals R _i (t ') can be determined for the time interval Δt: R _{i = 13} (t ') = ENV (P _Ab (t')) * ÜV - ENV (P _An (t ')) R _{i = 14} (t ') = ENV (PG _Ab (t')) * ÜV - ENV (PG _An (t ')) R _{i = 15} (t ') = ENV (Ṗ _Ab (t')) × ÜV - ENV (Ṗ _An (t ')) R _{i = 16} (t ') = ENV (Ṗ G _Ab (t')) Ü ÜV - ENV (Ṗ G _An (t ')) R _{i = 17} (t ') = ENV (P .. _Ab (t')) × ÜV - ENV (P .. _An (t ')) R _{i = 18} (t ') = ENV (P..G _Ab (t')) · ÜV - ENV (P..G _An (t ')) R _{i = 19} (t ') = ENV (P _Ab (t')) - ENV (PG _Ab (t ')) R _{i = 20} (t ') = ENV (P _An (t')) - ENV (PG _An (t ')) R _{i = 21} (t ') = ENV (Ṗ _Ab (t')) - ENV (ṖG _Ab (t ')) R _{i = 22} (t ') = ENV (Ṗ _An (t')) - ENV (ṖG _An (t ')) R _{i = 23} (t ') = ENV (P .. _An (t')) - ENV (P .. G _An (t ')) R _{i = 24} (t ') = ENV (P .. _Ab (t')) - ENV (P .. G _Ab (t '))

Where ENV (...) is the envelope or envelope of the respective time-dependent quantity given in brackets. Advantageously, the envelope ENV () is generated by approximating a Hilbert transform by means of a finite impulse response filter, a so-called FIR filter.

Another aspect of the invention relates to a computer program product having program code stored on a machine-readable carrier for carrying out the method as described above when the program code is executed on a data processing device.

Another aspect of the invention relates to a computer system having a data processing device, wherein the data processing device is configured such that a method as described above is performed on the data processing device.

Another aspect relates to a digital storage medium having electronically readable control signals, wherein the control signals may interact with a programmable computer system to perform a method as described above.

Another aspect of the invention relates to the invention of a computer program with program codes for carrying out the method, as described above, when the program runs on a data processing device. For this, the data processing device can be configured as any known from the prior art computer system.

Another aspect of the invention relates to a device for monitoring an electromechanical actuator comprising a drive and an output.

• – Position P_An(t) des Antriebs und Position P_Ab(t) des Abtriebs,
• – optional zumindest eine daraus ermittelte zugehörige erste Zeitableitung Ṗ_An(t) oder Ṗ_Ab(t), und
• – optional die zwei daraus ermittelten zugehörigen zweiten Zeitableitungen P .._An(t) und P .._Ab(t) .

The proposed device comprises a first means for detecting a time series of a state Z (t) of the actuator, comprising the following state components:

• Position P _An (t) of the drive and position P _Ab (t) of the output,
• Optionally, at least one associated first time derivative daraus _An (t) or Ṗ _Ab (t), and
• Optionally, the two associated second time derivatives derived therefrom P .. _An (t) and P .. _Ab (t) ,

The first means advantageously comprises a suitable sensor system for detecting at least the positions P _An (t) and P _Ab (t), an evaluation unit for evaluating sensor raw data of the sensor system and a memory unit for storing the time series of the detected states Z (t) of the actuator , with t: = time. Advantageously, the state components comprise the following data: temperature of the actuator T _act (t), temperature of the drive T _Ant (t), load of the output L _Ab (t); Load of the drive L _An (t), el. Voltage of the drive U (t), el. Current of the drive I (t), which are also detected by a sensor system with suitable sensors.

The proposed device further comprises an interface for providing a state estimator ZS for determining an estimated state ZG (t) of the actuator. The interface is advantageously a data interface for transmitting the state estimator ZS as a software program to a processor unit of the device for executing the software program.

• – Position PG_An(t) des Antriebs und Position PG_Ab(t) des Abtriebs,
• – zumindest eine der zugehörigen ersten Zeitableitungen: ṖG_An(t) oder ṖG_Ab(t), und
• – optional die zwei zugehörigen zweiten Zeitableitungen P ..G_An(t) , P ..G_Ab(t) durch den Zustandsschätzer ZS auf Basis der erfassten Zustände Z(t) und/oder auf Basis von bereitgestellten Stellgrößen SG(t) zur Ansteuerung des Antriebs.

The proposed device further comprises a second means for determining a time series of an estimated state ZG (t) of the actuator, comprising the following estimated state components:

• Position PG _An (t) of the drive and position PG _Ab (t) of the output,
• At least one of the associated first time derivatives: AnG _An (t) or ṖG _Ab (t), and
• Optionally, the two associated second time derivatives P ..G _An (t) . P .. G _Ab (t) by the state estimator ZS on the basis of the detected states Z (t) and / or on the basis of provided manipulated variables SG (t) for controlling the drive.

The proposed device further comprises a third means for determining a time interval Δt = [t _a , t _e ] in which: Ṗ _An (t) ≠ 0 and / or Ṗ _Ab (t) ≠ 0 and / or ṖG _An (t) ≠ 0 and / or ṖG _Ab (t) ≠ 0.

The proposed device further comprises a fourth means for determining for the time interval Δt at least one of the following residuals R _i (t '): R _{i = 1} (t ') = P _Ab (t') * UV - P _An (t ') R _{i = 2} (t ') = PG _Ab (t') * UV - PG _An (t ') R _{i = 3} (t ') = Ṗ _Ab (t') Ü ÜV - Ṗ _An (t ') R _{i = 4} (t ') = Ṗ G _Ab (t') Ü ÜV - Ṗ G _An (t ') R _{i = 5} (t ') = P .. _Ab (t') * ÜV - P .. _An (t ') R _{i = 6} (t ') = P ..G _Ab (t') * UV - P ..G _An (t ') R _{i = 7} (t ') = P _Ab (t') - PG _Ab (t ') R _{i = 8} (t ') = P _An (t') - PG _An (t ') R _{i = 9} (t ') = Ṗ _Ab (t') - Ṗ G _Ab (t ') R _{i = 10} (t ') = Ṗ _An (t') - Ṗ G _An (t ') R _{i = 11} (t ') = P .. An (t') -P.G _An (t ') R _{i = 12} (t ') = P .. Ab (t') - P .. G _Ab (t ') R _{i = 13} (t ') = ENV (P _Ab (t')) * ÜV - ENV (P _An (t ')) R _{i = 14} (t ') = ENV (PG _Ab (t')) * ÜV - ENV (PG _An (t ')) R _{i = 15} (t ') = ENV (Ṗ _Ab (t')) × ÜV - ENV (Ṗ _An (t ')) R _{i = 16} (t ') = ENV (Ṗ G _Ab (t')) Ü ÜV - ENV (Ṗ G _An (t ')) R _{i = 17} (t ') = ENV (P .. _Ab (t')) × ÜV - ENV (P .. _An (t ')) R _{i = 18} (t ') = ENV (P..G _Ab (t')) · ÜV - ENV (P..G _An (t ')) R _{i = 19} (t ') = ENV (P _Ab (t')) - ENV (PG _Ab (t ')) R _{i = 20} (t ') = ENV (P _An (t')) - ENV (PG _An (t ')) R _{i = 21} (t ') = ENV (Ṗ _Ab (t')) - ENV (ṖG _Ab (t ')) R _{i = 22} (t ') = ENV (Ṗ _An (t')) - ENV (ṖG _An (t ')) R _{i = 23} (t ') = ENV (P .. _An (t')) - ENV (P .. G _An (t ')) R _{i = 24} (t ') = ENV (P .. _Ab (t')) - ENV (P .. G _Ab (t ')) With:
t 'ε [t _a , t _e ],
ÜV = gear ratio between drive (2) and output (3),
ENV () = envelope or envelope of (...),

The proposed device further comprises a fifth means for transforming the at least one residual R _i (t ') into a frequency range for the time interval Δt for generating a transformed residual R _{i, Freq} (ω) and / or for transforming the at least one residual R _i (t ') into a _{path region} for generating a transformed residual R _{i, path} (s).

The proposed device further comprises a sixth means for determining error frequencies ω _{F of} the transformed residual R _{i, Freq} (ω) and associated amplitudes A (ω _F ) or for determining error positions s _{F of} the transformed residual R _{i, path} (s) and associated amplitudes A (s _F ).

The proposed device further comprises a seventh means for classifying the at least one transformed residual R _{i, Freq} (w) into a class K _Freq * of several predetermined classes K _Freq , and / or for classifying the at least one transformed residual R _{i, Path} ( s) in a class K _way * of several predetermined classes K _way .

The proposed device furthermore comprises an eighth means for comparing the amplitudes A (ω _F ) with standard amplitudes A _{Norm, KFreq} (ω _F ) specified for the class K _Freq * and / or for comparing the amplitudes A (s _F ) for the class K _path * predetermined standard amplitudes A _{Norm, KWeg} (s _F ), wherein the eighth means is designed and arranged such that the amplitudes A (ω _F ) of the transformed residual R _{i, Freq} (ω) one of do not satisfy the normal _values A _{norm, KFreq} (ω _F ) of the class K _Freq * dependent condition B1, and / or if the amplitudes A (s _F ) of the transformed residual R _{i, path} (s) are one of the normal _values A _{Norm, KWeq} (s _F ) of the class K _Weq * dependent condition B2 do not meet, a warning is generated and issued. For this purpose, the eighth means advantageously comprises an output means for outputting the warning in optical, acoustic, haptic or electrical form.

Advantages and preferred developments of the proposed device are obtained by an analogous and analogous transmission of the statements made to the proposed method.

Finally, a last aspect of the invention relates to a vehicle, in particular a motor vehicle, an aircraft, a spacecraft or a watercraft with a device described above.

Further advantages, features and details will become apparent from the following description in which - where appropriate, with reference to the drawings - at least one embodiment is described in detail. The same, similar and / or functionally identical parts are provided with the same reference numerals.

Show it:

1 a schematized actuator,

2 a schematic flow of a variant of the proposed method, and

3 a schematic structure of a variant of the proposed device.

1 shows a schematic electromechanical actuator 1 with an electric motor as drive 2 and a driven by the electric motor via a transmission (not shown) with a transmission ratio ÜV piston as the output 3 , On the piston 3 grabs the load 4 at. The piston position 5 results along the arrow direction shown.

2 shows a schematic flow of a variant of the proposed method for monitoring an electromechanical actuator 1 according to 1 that a drive 2 and a downforce 3 includes.

• – Position P_An(t) des Antriebs 2 und Position P_Ab(t) des Abtriebs 3,
• – daraus ermittelte zugehörige erste Zeitableitungen Ṗ_An(t) und Ṗ_Ab(t),
• – daraus ermittelte zugehörige zweite Zeitableitungen P .._An(t) und P .._Ab(t) ,
• – Temperatur des Aktors,
• – Strom des Elektromotors 2 I_Ant(t), und
• – Abtriebslast L_Ab(t) 4 des Aktors.

In one step 101 a time series of a state Z (t) of the actuator is detected, comprising the following state components:

• - Position P _An (t) of the drive 2 and position P _Ab (t) of the output 3 .
• - associated first time derivatives daraus _An (t) and Ṗ _Ab (t) determined therefrom,
• - derived second time derivatives derived therefrom P .. _An (t) and P .. _Ab (t) .
• Temperature of the actuator,
• - Electric motor current 2 I _Ant (t), and
• - output load L _Ab (t) 4 of the actor.

The position P _at (t) of the drive 2 corresponds to an angle. The position P _Ab (t) of the output 3 corresponds to the piston position 5 ,

In one step 102 there is provided a state estimator ZS for determining an estimated state ZG (t) of the actuator. The state estimator is based on an actuator model and a Kalman filter.

• – Position PG_An(t) des Antriebs und Position PG_Ab(t) des Abtriebs,
• – die zugehörigen ersten Zeitableitungen: ṖG_An(t) und ṖG_Ab(t), und
• – die zugehörigen zweiten Zeitableitungen P ..G_An(t) und P ..G_Ab(t)

durch den Zustandsschätzer ZS auf Basis der erfassten Zustände Z(t).In one step 103 the time series of an estimated state ZG (t) of the actuator is determined, comprising the following estimated state components:

• Position PG _An (t) of the drive and position PG _Ab (t) of the output,
• - the associated first time derivatives: AnG _An (t) and ṖG _Ab (t), and
• - the associated second time derivatives P ..G _An (t) and P .. G _Ab (t)

by the state estimator ZS on the basis of the detected states Z (t).

In one step 104 carried determining a time interval At = [t _{a, e} t] in which the following applies: P _at (t) ≠ 0 and p _Ab (t) ≠ 0. In addition, the time interval is determined .DELTA.t such that the estimated values with an estimated engine speed PG _At (t) less than 10% of the maximum speed of the motor Ṗ _{An, Max} can not be used.

In one step 105 For the time interval Δt, the following residuals R _i (t ') are determined: R _{i = 1} (t ') = P _Ab (t') * UV - P _An (t '), R _{i = 8} (t ') = P _An (t') - PG _An (t '), R _{i = 13} (t ') = ENV (P _Ab (t')) * ÜV - ENV (P _An (t ')), and R _{i = 19} (t ') = ENV (P _Ab (t')) - ENV (PG _Ab (t ')) with: t 'ε [t _a , t _e ], and ÜV = transmission ratio between drive and output.

In one step 106a For the time interval Δt, the residuals R _i (t ') are transformed by means of a recursively defined FFT into a frequency range for generating transformed residuals R _{i, Freq} (ω).

In one step 107a there is a determination of error frequencies ω _{F of} the transformed residual R _{i, Freq} (ω) and associated amplitudes A (ω _F ).

In one step 108a classifying the transformed residuals R _{i, Freq} (ω) into a class K _Freq * of several predetermined classes K _Freq . The classifying takes place in each case on the basis of previously determined error frequencies ω _{F of} the respective transformed residual R _{i, Freq} (ω) and the associated amplitudes A (ω _F ). The error frequencies ω _F are determined by a period analysis method. The classification is based on the classification in tmperatur-, load- and speed-dependent classes, where present under temperature of the actuator, under load the load acting on the output, and at speed, that of the drive 2 to be understood. The number of classes is A = T * L * S where T is the number of temperature-dependent classes, L is the number of load-dependent classes and S is the number of speed-dependent classes.

In a further step 109a in each case, the amplitudes A (ω _F ) are compared with the standard amplitudes A _{Norm, KFreq} (ω _F ) specified for the respective class K _Freq *, where the amplitudes A (ω _F ) of the transformed residual R _{i, Freq} ( _FIG. ω) does not satisfy a condition B1 dependent on the normal _values A _{norm, KFreq} (ω _F ) of the class K _Freq *, generates a warning 110a and output as an audible and visual signal 110b becomes.

Additionally or alternatively, after step 105 following steps are performed: for the time interval Δt transform 106b of the residuals R _i (t ') into a _{path region} for generating transformed residuals R _{i, path} (s), respectively 107b of error positions s _{F of} the transformed residuals R _{i, path} (s) and associated amplitudes A (s _F ), classify 108b the transformed residuals R _{i, way} (s) in each case a class K _way * of several predetermined classes K _way , respectively comparing 109b the amplitudes A (s _F ) with standard amplitudes A _{Norm, KWeg} (s _F ) specified for the respective class K _path *, where the amplitudes A (ω _F ) of one of the transformed residuals R _{i, path} (s) are one of the normal _values A _{Norm, KWeq} (s _F ) of the respective class K _Weq * dependent condition B2 do not satisfy a generating 110a and spending 100b a warning occurs.

In summary, the method comprises the following steps: Detecting 101 a time series of a state Z (t) of the actuator 1 , Provide 102 a state estimator ZS for determining an estimated state ZG (t) of the actuator 1 , Determine 103 a time series of an estimated state ZG (t) of the actuator 1 , Determine 104 of a time interval Δt = [t _a , t _e ] in which: Ṗ _An (t) ≠ 0 and / or Ṗ _Ab (t) ≠ 0 and / or ṖG _An (t) ≠ 0 and / or ṖG _Ab (t) ≠ 0, for the time interval Δt 105 at least one residual R _i (t '), for the time interval Δt transform 106a , B of the at least one residual R _i (t ') in a frequency range and / or in a path range, determining 107a of error frequencies ω _{F of} the transformed residual R _{i, Freq} (ω) and associated amplitudes A (ω _F ) and / or determining 107b of error positions s _{F of} the transformed residual R _{i, path} (s) and associated amplitudes A (s _F ), classify 108a of the at least one transformed residual R _{i, Freq} (ω) into a class K _Freq * of several predetermined classes K _Freq , and / or classifying 108b of the at least one transformed residue R _{i, path} (s) into a class K _path * of several predetermined classes K _way , and compare 109a the amplitudes A (ω _F ) with given for the class K _Freq * standard amplitudes A _{norm, KFreq} (ω _F ), and / or comparison 109b the amplitudes A (s _F ) with standard amplitudes A _{Norm, KWeg} (s _F ) specified for the class K _path *, where the amplitudes A (ω _F ) of the transformed residual R _{i, Freq} (ω) are one of the normal _values A _{norm, KFreq} (ω _F ) of class K _Freq * dependent condition B1, or if the amplitudes A (s _F ) of the transformed residue R _{i, path} (s) is one of the normal _values A _{Norm, KWeq} (s _F ) class K _Weq * do not satisfy dependent condition B2, generating 110a and spending 100b a warning.

• – Position P_An(t) des Antriebs 2 und Position P_Ab(t) des Abtriebs 3,
• – die zugehörigen ersten Zeitableitungen Ṗ_An(t) und Ṗ_Ab(t).

3 shows a schematic structure of a variant of the proposed device for monitoring an electromechanical actuator 1 that a drive 2 and a downforce 3 includes. The device comprises a first means 201 for detecting a time series of a state Z (t) of the actuator 1 comprising the following state components:

• - Position P _An (t) of the drive 2 and position P _Ab (t) of the output 3 .
• - The associated first time derivatives Ṗ _An (t) and Ṗ _Ab (t).

• – Position PG_An(t) des Antriebs 2 und Position PG_Ab(t) des Abtriebs 3,
• – die zugehörigen ersten Zeitableitungen: ṖG_An(t) und ṖG_Ab(t)

durch den Zustandsschätzer ZS auf Basis der erfassten Zustände Z(t) und/oder auf Basis von bereitgestellten Stellgrößen SG(t) zur Ansteuerung des Antriebs.The device further comprises an interface 202 for providing a state estimator ZS for determining an estimated state ZG (t) of the actuator 1 , a second remedy 203 for determining a time series of an estimated state ZG (t) of the actuator 1 comprising the following estimated state components:

• - Position PG _An (t) of the drive 2 and position PG _Ab (t) of the output 3 .
• - the associated first time derivatives: ṖG _An (t) and ṖG _Ab (t)

by the state estimator ZS on the basis of the detected states Z (t) and / or on the basis of provided manipulated variables SG (t) for controlling the drive.

The device further comprises a third means 204 for determining a time interval Δt = [t _a , t _e ] in which: Ṗ _An (t) ≠ 0 and Ṗ _Ab (t) ≠ 0 and ṖG _An (t) ≠ 0 and ṖG _Ab (t) ≠ 0.

The device further comprises a fourth means 205 for determining the time interval Δt of the following residuals R _i (t '): R _{i = 1} (t ') = P _Ab (t') * UV - P _An (t '), R _{i = 8} (t ') = P _An (t') - PG _An (t '), R _{i = 13} (t ') = ENV (P _Ab (t')) * ÜV - ENV (P _An (t ')), and R _{i = 19} (t ') = ENV (P _Ab (t')) - ENV (PG _Ab (t ')).

The device further comprises a fifth means 206 for transforming the residuals R _i (t ') into a frequency range for the time interval Δt to produce a respective transformed residual R _{i, Freq} (ω).

The device further comprises a sixth means 207 in each case for determining error frequencies ω _{F of} the transformed residuals R _{i, Freq} (ω) and associated amplitudes A (ω _F ).

The device further comprises a seventh means 208 for classifying the transformed residuals R _{i, Freq} (ω) into respectively one class K _Freq * of several predetermined classes K _Freq .

The device further comprises an eighth means 209 for comparing the amplitudes A (ω _F ) with standard amplitudes A _{Norm, KFreq} (ω _F ) given for the class K _Freq *, wherein the eighth means is designed and arranged such that, provided the amplitudes A (ω _F ) of one the transformed residuals R _{i, Freq} (ω) do not satisfy a condition B1 dependent on the normal _values A _{norm, KFreq} (ω _F ) of the respective class K _Freq *, a warning is generated and output. For this purpose, the eighth means comprises a unit for generating and a unit for outputting the warning.

Although the invention has been further illustrated and explained in detail by way of preferred embodiments, the invention is not limited by the disclosed examples, and other variations can be derived therefrom by those skilled in the art without departing from the scope of the invention. It is therefore clear that a multitude of possible variations exists. It is also to be understood that exemplified embodiments are really only examples that are not to be construed in any way as limiting the scope, applicability, or configuration of the invention. Rather, the foregoing description and description of the figures enable one skilled in the art to practice the exemplary embodiments, and those of skill in the knowledge of the disclosed inventive concept may make various changes, for example as to the function or arrangement of individual elements recited in an exemplary embodiment, without Protection area defined by the claims and their legal equivalents, such as further explanations in the description.

LIST OF REFERENCE NUMBERS

1

electromechanical actuator

2

Drive, electric motor

3

Downforce, piston

4

load

5

Piston position, position output

101-110a, b

steps

201

first means

202

interface

203

second means

204

third means

205

fourth means

206

fifth remedy

207

sixth means

208

seventh means

209

Eighth remedy

Claims (9)

Method for monitoring an electromechanical actuator ( 1 ), which has a drive ( 2 ) and an output ( 3 ) comprising the following steps: - detecting ( 101 ) a time series of a state Z (t) of the actuator ( 1 ), comprising the following state components: position P _An (t) of the drive ( 2 ) and position P _Ab (t) of the output ( 3 ), Optionally an associated associated first time derivative Ṗ _An (t) or Ṗ _Ab (t), and optionally the two associated second time derivatives derived therefrom P .. _An (t) and P .. _Ab (t) - Provide ( 102 ) of a state estimator ZS for determining an estimated state ZG (t) of the actuator ( 1 ), - Determine ( 103 ) a time series of an estimated state ZG (t) of the actuator ( 1 ), comprising the following estimated state components: position PG _An (t) of the drive ( 2 ) and position PG _Ab (t) of the output ( 3 ), - at least one of the associated first time derivatives: AnG _An (t) or ṖG _Ab (t), and - optionally the two associated second time derivatives P ..G _An (t) . P .. G _Ab (t) by the state estimator ZS on the basis of the detected states Z (t) and / or on the basis of provided manipulated variables SG (t) for controlling the drive ( 2 ), - Determine ( 104 ) of a time interval Δt = [t _a , t _e ] in which: Ṗ _An (t) ≠ 0 and / or Ṗ _Ab (t) ≠ 0 and / or ṖG _An (t) ≠ 0 and / or ṖG _Ab (t ) ≠ 0, - for the time interval Δt ( 105 ) at least one of the following residuals R _i (t '): (1) R _{i = 1} (t') = P _Ab (t ') * UV-P _An (t') (2) R _{i = 2} (t ') = PG _Ab (t ') * ÜV - PG _An (t') (3) R _{i = 3} (t ') = Ṗ _Ab (t') * ÜV - Ṗ _An (t ') (4) R _{i = 4} (t ') = ṖG _Ab (t') Ü ÜV - ṖG _An (t ') (5) R _{i = 5} (t ') = P .. _Ab (t') * ÜV - P .. _An (t ') (6) R _{i = 6} (t ') = P ..G _Ab (t') * UV - P ..G _An (t ') (7) R _{i = 7} (t ') = P _Ab (t') - PG _Ab (t ') (8) R _{i = 8} (t') = P _An (t ') - PG _An (t') (9) R _{i = 9} (t ') = Ṗ _Ab (t') - Ṗ G _Ab (t ') (10) R _{i = 10} (t') = Ṗ _An (t ') - Ṗ G _An (t') (11) R _{i = 11} (t ') = P .. _An (t') -P.G _An (t ') (12) R _{i = 12} (t ') = P .. _Ab (t') -P.G _Ab (t ') (13) R _{i = 13} (t ') = ENV (P _Ab (t')) * ÜV - ENV (P _An (t ')) (14) R _{i = 14} (t') = ENV (PG _Ab ( t ')) · ÜV - ENV (PG _An (t')) (15) R _{i = 15} (t ') = ENV (Ṗ _Ab (t')) · ÜV - ENV (Ṗ _An (t ')) ( 16) R _{i = 16} (t ') = ENV (ṖG _Ab (t')) Ü ÜV - ENV (ṖG _An (t ')) (17) R _{i = 17} (t ') = ENV (P .. _Ab (t')) × ÜV - ENV (P .. _An (t ')) (18) R _{i = 18} (t ') = ENV (P..G _Ab (t')) · ÜV - ENV (P..G _An (t ')) (19) R _{i = 19} (t ') = ENV (P _Ab (t')) - ENV (PG _Ab (t ')) (20) R _{i = 20} (t') = ENV (P _An (t ') ) - ENV (PG _An (t ')) (21) R _{i = 21} (t') = ENV (Ṗ _Ab (t ')) - ENV (ṖG _Ab (t')) (22) R _{i = 22} (t ') = ENV (Ṗ _An (t')) - ENV (ṖG _An (t ')) (23) R _{i = 23} (t ') = ENV (P .. _An (t')) - ENV (P .. G _An (t ')) (24) R _{i = 24} (t ') = ENV (P .. _Ab (t')) - ENV (P .. G _Ab (t ')) with: t 'ε [t _a , t _e ], ÜV = transmission ratio between drive ( 2 ) and downforce ( 3 ), ENV () = Envelope or envelope of (...) - for the time interval Δt Transform ( 106a ) of the at least one residual R _i (t ') into a frequency range for generating a transformed residual R _{i, Freq} (ω) and / or transforming ( 106b ) of the at least one residuum R _i (t ') into a _{path region} for generating a transformed residual R _{i, path} (s), - determining ( 107a ) of error frequencies ω _{F of} the transformed residual R _{i, Freq} (ω) and associated amplitudes A (ω _F ) and / or determining ( 107b ) of error positions s _{F of} the transformed residual R _{i, path} (s) and associated amplitudes A (s _F ), - classifying ( 108a ) of the at least one transformed residual R _{i, Freq} (ω) into a class K _Freq * of several predetermined classes K _Freq , and / or classifying ( 108b ) of the at least one transformed residuum R _{i, path} (s) into a class K _path * of several predetermined classes K _path , - comparison ( 109a ) of the amplitudes A (ω _F ) with standard amplitudes A _{Norm, KFreq} (ω _F ) and / or comparison (for the class K _Freq * 109b ) of the amplitudes A (s _F ) with standard amplitudes A _{Norm, KWeg} (s _F ) given for the class K _path *, where, if the amplitudes A (ω _F ) of the transformed residual R _{i, Freq} (ω) is one of the normal _values A _{norm, KFreq} (ω _F ) of the class K _Freq * dependent condition B1, or if the amplitudes A (s _F ) of the transformed residual R _{i, path} (s) one of the normal _values A _{Norm, KWeq} (s _F ) the class K does not satisfy _Weq * dependent condition B2, generating ( 110a ) and output ( 100b ) of a warning. Method according to claim 1, wherein said transforming ( 106a ) of the at least one residual R _i (t ') into the frequency domain with a recursively defined discrete Fourier transform. Method according to one of Claims 1 to 2, in which the time interval Δt = [t _a , t _e ] is determined in such a way that in Δt for ṖG _An (t ') and / or _An (t'): (25) ṖG _An (t ')> 0.9Ṗ _{On, Max} (26) Ṗ _An (t')> 0.9Ṗ _{On, Max} with Ṗ _{On, Max} = maximum speed of the drive. Method according to Claim 3, in which the time interval Δt = [t _a , t _e ] is determined in such a way that in Δt ṖG _An (t ') and / or _An (t') varies less than 5%. Method according to Claim 1, in which the envelope ENV () is produced by approximating a Hilbert transformation by means of a finite-impulse-response filter, a so-called FIR filter. Method according to one of claims 1 to 5, wherein the warning is transmitted after its generation wired or wirelessly to an output unit for output. A computer program product having program code stored on a machine-readable medium carrying out the method of any one of claims 1 to 6 when the program code is executed on a data processing device. A computer system having a data processing device, wherein the data processing device is configured such that a method according to one of claims 1 to 6 is performed on the data processing device. Device for monitoring an electromechanical actuator ( 1 ), which has a drive ( 2 ) and an output ( 3 ) and for carrying out a method according to claims 1 to 6, comprising: - a first means ( 201 ) for detecting a time series of a state Z (t) of the actuator ( 1 ), comprising the following state components: position P _An (t) of the drive ( 2 ) and position P _Ab (t) of the output ( 3 ), Optionally an associated associated first time derivative Ṗ _An (t) or Ṗ _Ab (t), and optionally the two associated second time derivatives derived therefrom P .. _An (t) and P .. _Ab (t) An interface ( 202 ) for providing a state estimator ZS for determining an estimated state ZG (t) of the actuator ( 1 ), - a second means ( 203 ) for determining a time series of an estimated state ZG (t) of the actuator ( 1 ), comprising the following estimated state components: position PG _An (t) of the drive ( 2 ) and position PG _Ab (t) of the output ( 3 ), - at least one of the associated first time derivatives: AnG _An (t) or ṖG _Ab (t), and - optionally the two associated second time derivatives P ..G _An (t) . P .. G _Ab (t) by the state estimator ZS on the basis of the detected states Z (t) and / or on the basis of provided manipulated variables SG (t) for controlling the drive, - a third means ( 204 ) for determining a time interval Δt = [t _a , t _e ] in which: Ṗ _An (t) ≠ 0 and / or Ṗ _Ab (t) ≠ 0 and / or ṖG _An (t) ≠ 0 and / or ṖG _Ab (t) ≠ 0, - a fourth means ( 205 ) for determining for the time interval Δt at least one of the following residuals R _i (t '): (25) R _{i = 1} (t') = P _Ab (t ') * ÜV - P _An (t') (26) R _{i = 2} (t ') = PG _Ab (t') × ÜV - PG _An (t ') (27) R _{i = 3} (t') = Ṗ _Ab (t ') × ÜV - Ṗ _An (t') ( 28) R _{i = 4} (t ') = Ṗ G _Ab (t') Ü ÜV - Ṗ G _An (t ') (29) R _{i = 5} (t ') = P .. _Ab (t) · ÜV - P .. _An (t') (30) R _{i = 6} (t ') = P ..G _Ab (t') * UV - P ..G _An (t ') (31) R _{i = 7} (t ') = P _Ab (t') - PG _Ab (t ') (32) R _{i = 8} (t') = P _An (t ') - PG _An (t') (33) R _{i = 9} (t ') = Ṗ _Ab (t') - Ṗ G _Ab (t ') (34) R _{i = 10} (t') = Ṗ _An (t ') - Ṗ G _An (t') (35) R _{i = 11} (t ') = P .. _An (t') -P.G _An (t ') (36) R _{i = 12} (t ') = P .. _Ab (t') -P.G _Ab (t ') (37) R _{i = 13} (t ') = ENV (P _Ab (t')) * ÜV - ENV (P _An (t ')) (38) R _{i = 14} (t') = ENV (PG _Ab ( t ')) · ÜV - ENV (PG _An (t')) (39) R _{i = 15} (t ') = ENV (Ṗ _Ab (t')) · ÜV - ENV (Ṗ _An (t ')) ( 40) R _{i = 16} (t ') = ENV (ṖG _Ab (t')) * ÜV - ENV (ṖG _An (t ')) (41) R _{i = 17} (t ') = ENV (P .. _Ab (t')) × ÜV - ENV (P .. _An (t ')) (42) R _{i = 18} (t ') = ENV (P..G _Ab (t')) · ÜV - ENV (P..G _An (t ')) (43) R _{i = 19} (t ') = ENV (P _Ab (t')) - ENV (PG _Ab (t ')) (44) R _{i = 20} (t') = ENV (P _An (t ') ) - ENV (PG _An (t ')) (45) R _{i = 21} (t') = ENV (Ṗ _Ab (t ')) - ENV (ṖG _Ab (t')) (46) R _{i = 22} (t ') = ENV (Ṗ _An (t')) - ENV (ṖG _An (t ')) (47) R _{i = 23} (t ') = ENV (P .. _An (t')) - ENV (P .. G _An (t ')) (48) R _{i = 24} (t ') = ENV (P .. _Ab (t')) - ENV (P .. G _Ab (t ')) with: t 'ε [t _a , t _e ], ÜV = transmission ratio between drive (2) and output (3), ENV () = envelope or envelope of (...), - a fifth means ( 206 ) for transforming the at least one residual R _i (t ') into a frequency range for the time interval Δt to produce a transformed residual R _{i, Freq} (ω) and / or transforming the at least one residuum R _i (t') into a path region for generating a transformed residue R _{i, path} (s), - a sixth means ( 207 ) for determining error frequencies ω _{F of} the transformed residual R _{i, Freq} (ω) and associated amplitudes A (ω _F ) and / or for determining error positions s _{F of} the transformed residual R _{i, path} (s) and associated amplitudes A (s _F ), - a seventh means ( 208 ) for classifying the at least one transformed residual R _{i, Freq} (ω) into a class K _Freq * of a plurality of predetermined classes K _Freq , and / or for classifying the at least one transformed residuum R _{i, path} (s) into a class K _path * of several predetermined classes K _way , - an eighth means ( 209 ) for comparing the amplitudes A (ω _F ) with standard amplitudes A _{Norm, KFreq} (ω _F ), predetermined for the class K _Freq *, and / or for comparing the amplitudes A (s _F ) with standard amplitudes A _{Norm, KWeg} (s _F ) predetermined for the class K _path *, wherein the eighth means ( 209 ) is designed and arranged such that, provided that the amplitudes A (ω _F ) of the transformed residual R _{i, Freq} (ω) do not satisfy a condition B1 dependent on the normal _values A _{norm, KFreq} (ω _F ) of the class K _Freq *, or if the amplitudes A (s _F ) of the transformed residual R _{i, path} (s) do not satisfy a condition B2 dependent on the normal _values A _{Norm, KWeq} (s _F ) of the class KWeq *, a warning is generated and output.

Images