Method and device for controlling an electric motor

Abstract

Method for controlling and / or regulating an electric motor, wherein it is provided that - the electric motor (180, 220) is supplied with electrical energy from a battery (130, 230) and / or a generator (150, 190, 210) and - for control and / or regulation, a high-frequency pulse-width modulated voltage is applied to the electric motor, wherein the motor current (IM) supplied from the battery and / or the generator when the electric motor starts up to operate the electric motor is continuously increased, wherein the increase in the motor current is limited to a predetermined setpoint value within a predeterminable time, and wherein the electric motor is operated at a maximum of a predetermined setpoint speed (ω i j+1 ), whereby it is intended that the target speed is dependent on - the battery voltage, and / or - the load torque on the electric motor is specified.

Description

State of the art

[0001] The invention is based on a method and a device for operating an electric motor controlled by pulse width modulation on a DC voltage network according to the independent claims.

[0002] Such a method is known from DE 199 44 194 A1, in which the output stage of an electronically commutable motor is controlled via an electronic control unit using pulse-width modulated signals. The electric motor is supplied with voltage pulses from a direct current network according to the control voltage pulses specified by a setpoint stage. Further methods are described in the published patent applications DE 43 04 960 A1, DE 101 02 885 A1, and EP 1 039 623 A2. Advantages of the invention

[0003] The present invention describes a method and a device for controlling and / or regulating an electric motor. Such electric motors are used, for example, in motor vehicles in the form of pump motors. Generally, the electric motor is supplied with electrical energy from a battery and / or by means of a generator. Control and / or regulation is achieved by means of high-frequency pulse width modulation (PWM). The core of the invention is that when the electric motor starts up, the PWM continuously increases the motor current required to operate the electric motor, for example, starting from a value of 0.

[0004] The advantage of controlling the electric motor in this way is that it avoids current spikes caused by the high starting currents that occur when electric motors start up. Such current spikes can damage the battery. Furthermore, these current spikes can cause a dynamic voltage drop, which can lead to failure in other loads that are also powered by the battery and / or generator.

[0005] According to the invention, the current gradient with which the motor current is continuously increased is therefore limited to a predetermined maximum value.

[0006] The maximum value, which is configured as a target value according to the invention, can be specified depending on the operating parameters of the battery and / or the generator. Typical operating parameters include the battery discharge current and / or the current rise generated by the generator. However, the structural design of the generator can also be used to specify the maximum value, for example, by determining the maximum current rise rate achievable by the generator. The latter is defined as the current that can be generated by the generator within a specified time.

[0007] To limit the increase in motor current or the current gradient, the invention provides for limiting the speed of the electric motor. A maximum target speed is used, which is specified as a function of the battery voltage. Another possibility according to the invention is to take into account the load torque applied to the electric motor or which must be generated by the electric motor when determining the maximum target speed.

[0008] In one embodiment of the invention, it is further provided that the battery or generator provided for operating the electric motor supplies additional electrical consumers with energy. The present invention prevents these additional consumers from failing in the event of current peaks during start-up of the electric motor. In a further development of the invention, it can also be provided that the continuous increase in the motor current or the current gradient to which this increase is limited is made dependent on the supply of electrical energy to the additional consumers. For example, the power requirement for additional consumers in a vehicle such as the heating, the lights or ACC (Adaptive Cruise Control) can be taken into account in order to determine the permissible current gradient.

[0009] In a special development of the invention, it is provided that only the electric motor is supplied with electrical energy by the generator, while the other electrical consumers are largely supplied by the electrical energy supplied by the battery. It is also conceivable that the supply is only disconnected during the start-up phase of the electric motor.

[0010] Further advantages emerge from the following description of embodiments and from the dependent patent claims. Drawings

[0011] Fig. Figure 1 shows a schematic block diagram of an inventive device for controlling or regulating an electric motor. A corresponding circuit diagram is shown in Fig. 2. In Fig. 3 shows an equivalent circuit diagram of an electric motor. A comparison of the motor current with and without current gradient limitation is shown in Fig. 4. The Fig. 5a to 5c show illustrations to explain the derivation of the speed limitation. Example

[0012] When operating an electric motor, a high power requirement occurs during startup. If the electric motor is powered by a battery, the supply voltage provided by the battery can drop. This dynamic voltage drop can affect other consumers that are also connected to the battery. Especially when starting up an electric (pump) motor in a vehicle, for example, when generating pressure in an electro-hydraulic braking system, such a temporarily reduced supply voltage can lead to problems with other electrical components in the vehicle.

[0013] As a remedy, according to the present invention, the energy supplied to the motor or the motor current supplied is slowly increased in order to avoid unwanted high current peaks and to protect the vehicle's electrical system.

[0014] Fig. 1 schematically shows a possible device with which a slow or continuous increase in the motor current at the electric motor 180 can be achieved. A processing unit 110, for example in the form of a microprocessor or an ASIC, is provided in a control unit 100. This processing unit evaluates external data and derives control signals therefrom, which control or regulate the electric motor 180 and, if applicable, also a generator 190 present in the vehicle.

[0015] Operating parameters of the electric motor itself can be recorded as external data, which are recorded by a suitable means 130 and forwarded to the processing unit 110. Operating parameters can also be understood as the pressure ratio in the hydraulic branch upstream and downstream of a pump motor, which can be determined using pressure sensors or models. Furthermore, predefined control parameters of the electric motor can also be recorded as operating parameters, which can be stored either in a suitable memory directly on the electric motor or in the control unit. Such a memory is schematically represented by block 120.

[0016] To initiate the continuous current increase, it is also necessary that a request to start up the electric motor is detected by a corresponding device 140. This device can also query the required load torque for the electric motor and transmit it to the processing unit 110 so that it can be taken into account during control.

[0017] If a generator 190 is present in the vehicle in which the control unit 100 is used, its operating parameters can also be used to generate the control signals of the electric motor 180. For this purpose, the current operating state of the generator 190 is detected, for example, using a suitable means 150. Furthermore, it is also conceivable that specific design parameters of the generator 190 can be taken into account, which can be stored in a memory on the generator or in the control unit. A specific design parameter can be understood, for example, as the maximum current gradient that can be generated by the generator, i.e., the current that can be generated within a predetermined time.

[0018] To control the electric motor 180, the state of the battery 160 is also detected and taken into account. The primary focus here is on the state of charge or the degree of load on the battery by other consumers. However, it is also possible to specifically detect the required supply power of the other consumers using suitable means 170 in order to be able to create a forecast of the utilization of the battery 160. Such a forecast can also be incorporated into the control of the electric motor 180, for example, if the battery provides the only supply voltage for the considered consumers.

[0019] In a further embodiment, it can also be provided that the generator 190 can also be operated depending on the detected data and / or depending on the control of the electric motor 180.

[0020] In Fig. 2 shows a schematic block diagram in which the control unit 200 controls the electric motor 220 by means of a motor voltage U M or a motor current I M Typically, such control is carried out by (high-frequency) pulse width modulation. The power supply of the control unit 200 or the electric motor 220 is, in this case, provided by a battery 230, which has a battery voltage U Bat and a generator 210. By combining the battery and the generator, the control unit 200 or the electric motor 220 can be supplied with a higher current I Zul be supplied.

[0021] Since the generator and battery in the vehicle are usually connected in parallel and the generator is only regulated with a certain current gradient (e.g. 300 to 1000 A / s), if the engine is switched on suddenly the current is supplied from the battery. These high currents can damage the battery. By specifying a maximum current gradient, i.e. a defined current increase in a predetermined time when controlling the motor during start-up, this type of damage can be avoided. If the current gradient is also adapted to the current gradient of the generator, the motor current required to start up the engine can be generated entirely by the vehicle's generator. This protects the battery from large current jumps. In addition, with this type of control, dynamic voltage bridges in the on-board network are avoided because the load on the battery is reduced.

[0022] The comparison of the stress of a motor start with and without current gradient limitation is shown in Fig. 4. The motor current I required for starting the motor is shown. M over time t. The curve according to 420 shows the motor current I M without current limitation, with a very high current peak shortly after the start of the ramp-up. The dashed curve according to 400, on the other hand, shows a continuous curve that, like the curve according to 420, transitions after a certain time into a constant current requirement for operating the motor.

[0023] In a further embodiment, different current gradients can also be set, which are required depending on the hydraulic power requirement of an ESP controller present in the vehicle.

[0024] Since the electric motor typically also has a speed control, the maximum current gradient can be set by specifying a target speed. In a particular embodiment, the maximum target speed used in this way is calculated according to ωij+1=1k8(Δibattmax−k9)

[0025] The maximum target speed can be derived as follows: As shown by an equivalent circuit diagram of an electric motor (see Fig. 3), the motor voltage U M in different partial voltage U R (Assignment to the ohmic resistance 300 of the motor), U Ind (Assignment to the inductive resistance 320 of the motor) and U Gen (Assignment to the generator part 340 of the motor).

[0026] The system equations of the engine can be derived from stress balance and angular momentum theorem and are: Jdωdt=Ki−TloadLdidt=U−Ri−Kω

[0027] The states are the armature current i and the speed ω. The parameters are the inductance L, the moment of inertia J, the motor constant K and the resistance R. The system inputs are the manipulated variable supply voltage U and the load torque T load The load torque depends on the pressures applied to the pump elements and the speed. As a first approximation, the speed dependence can be neglected.

[0028] Thus, the load moment consists of a constant friction component and a pressure-dependent load component: Tload=Tfric+kload(pds−pss).

[0029] Here, p ds the pressure side and p ss represents the suction-side pressure, the values ​​of which can be taken, for example, from HIM (the hydraulic model of the ESP) or directly measured by pressure sensors. Thus, the load torque can be assumed to be known.

[0030] In the following, the supply voltage U is calculated to achieve the desired target speed. This is based on the theory of flat systems.

[0031] If the speed ω is selected as the output y=ω, By taking the first derivative and inserting it into the system equations (1.1) we obtain a relationship between the armature current i and the rotational acceleration y˙=ω˙=1J(Ki−Tload)⇒i=1K(Jω˙+Tload).

[0032] If further derived, the supply voltage U occurs: y¨=ω¨=KJL(U−Ri−Kω)+T˙loadJ.

[0033] Solving for U we get U=JLKω¨+RJKω˙+Kω+RKTload+LKT˙load.

[0034] This equation can be simplified due to the small influence of the inductance. By inserting the desired values, one obtains a control that moves the system along the desired trajectory ω d moves: Ud=RJKω˙d+Kωd+RKTload.

[0035] The first term takes into account the inertia of the motor, the second the voltage induced by rotation and the third the required voltage due to the load torque. Planning the target trajectory

[0036] The sampling rate of the control unit is 5 ms. Therefore, both the supply voltage U and the target trajectory ω d (t) are modified in these time steps. In a first step, the trajectory is planned over a sampling cycle. This is shown in the Fig. 5a.

[0037] At time j, the setpoint speed at time j+1 is selected such that both the manipulated variable constraints are met and the current gradient does not exceed the required maximum. The calculation of the new setpoint ωdj+1 will be discussed in more detail below. First, the control system must be calculated. Calculation of the regulation

[0038] A linear PI controller is used to control the target speed. Thus, the controller component consists of a proportional and integral component. In discretized notation, the general controller equations can be written as follows: UPIj=kp(ωdj−ωmeasj)︸Ujp+ki∑l=1j(ωdj−ωmeasj)︸Uji

[0039] However, the fact that the supply voltage U belonging to time j is only output at time j+1 must be taken into account, and therefore the target trajectory is shifted by one sampling cycle. This fact is explained in the Fig. 5b illustrates this.

[0040] Therefore, the controller compares the current speed with the target speed of the last cycle. This results in the following equation for the controller: UPIj=(kp+ki)(ωdj−1−ωmassj)+I. Control and regulation

[0041] The manipulated variable U consists of feedforward control U d and regulation U PI together: U=Ud+UPI=RJKω˙d+Kωd+RKTload+UPI.

[0042] The speed and acceleration used in the control system must still be specified. The acceleration is determined by linear interpolation of the target speed at time j and j+1. ω˙d=ωdj+1−ωdjΔt

[0043] This results in the manipulated variable U j+I at time j+1 by Uj+1=(RJKΔ t)︸k1ωdj+1+(K−RJKΔ t)︸k2ωdj+RKTloadi+UPIj

[0044] It should be noted that only the control component of the new target value ω d j+I This equation is used below to prevent manipulated variable limitations Compliance with manipulated variable restrictions

[0045] The speed control receives a target speed from higher-level functions. However, it cannot be guaranteed that this can be achieved by the pump. Therefore, the target speed should be modified at the next time (j+1) so that the manipulated variable constraints are met and the speed can therefore be achieved by the pump. In the following, an upper limit ω is defined for this purpose using (3.4). j+I max and lower bound ω j+1 min for the setpoint speed within which the manipulated variable constraints are maintained. The setpoint speed is limited by the limits: ωmodj+1={ωminj+1ωdj+1ωmaxj+1 for ωdj+1<ωminj+1ωminj+1≤ωdj+1≤ωmaxj+1ωdj+1>ωmaxj+1

[0046] The two bounds can be calculated using (3.4). By solving (3.4) for ω j+1 and inserting the maximum voltage, the upper limit is obtained: ωmaxj+1=1k1(Umax−k2ωdj−RKTloadj−UPij)

[0047] By inserting the minimum voltage U=U min you get the lower bound. ωminj+1=1k1(Umin−k2ωdj−RKTloadj−UPij)

[0048] Thus, (4.1)-(4.3) ensure that the manipulated variable constraints are met. However, very large current gradients can still occur. The limitation of the current gradients is described below. Limiting the current gradient. Limiting the motor current gradient

[0049] From the moment balance (1.1) the current at time m can be expressed as a function of the load moment and the rotational acceleration. im=1K(Tloadm+Jω˙m)

[0050] To limit the current gradient, the change in current between sampling steps is crucial. Using (5.1), the current change can be determined as a function of the change in load torque and the change in angular acceleration: Δij+1=ij+1−ij=JK(ω˙j+1−ω˙j)+1K(Tloadj+1−Tloadj)

[0051] By inserting the maximum allowed current change Δi max This gives the maximum rotational acceleration: ω˙maxj+1=KJΔimax+ω˙j−1J(Tloadj+1−Tloadj)︸≈Tloadj−Tloadj−1.

[0052] It should be noted that the load torque change due to the unknown load torque at time j+1 is approximated by the load torque change of the last cycle. Thus, by linearly approximating the angular acceleration, a further condition for the new target speed ω is obtained. d j+1 : ω˙dj+1=ωj+1−ωjΔt≤ω˙maxj+1 ωij+1=ωj+Δtω˙maxj+1.

[0053] The new target speed is therefore subject to the restrictions in (4.1) and the restriction ωdj+1≤ωij+1.

[0054] Under nominal conditions (no model errors, ideal load torque estimation), the permissible motor current gradient can be maintained by condition (5.6). It should be noted that both model errors and incorrect load torque estimations can lead to deviations. b. Limiting battery current gradient

[0055] The motor current i is not directly decisive for the on-board load. mot , but the battery current required by the PWM generator i bat . In the Fig. Figure 5c shows the PWM generator and its interfaces.

[0056] Due to the smoothing of the battery current by filters and the high clock frequency, averaged DC voltages are assumed here. The motor voltage is calculated based on the PWM duty cycle and the battery voltage: Umot=PWM Ubat.

[0057] Considering the power balance of the PWM generator Ubatibat=Umotimot you get for the battery current ibat=imotPWM.

[0058] The current gradient of the battery voltage is therefore: i˙bat=i˙motPWM+imotPWM⋅ or written with differences follows Δibatj+1=Δimotj+1PWMj+imotjΔPWMj+1.

[0059] The motor current of the last cycle i mot j and the PWM j can be calculated from known values ​​of the last cycle: imotj=Uj−KωdjR PWMj=UjUbatt.

[0060] The motor current difference can be expressed using the equation of motion as follows Δij+1=ij+1−ij=JK(ω˙j+1−ω˙j)+1K(Tloadj+1−Tloadj)︸≈(Tloadj−Tloadj−1)=ΔT

[0061] If we replace the rotational acceleration of the j+1 cycle by the corresponding difference approximation we get Δij+1=JK(ωj+1−ωjΔt−ω˙j)+1KΔT=JKΔt︸k3ωj+1−JKΔtωj−JKω˙j+ΔTK︸k4.

[0062] The change in PWM is calculated from the quotient of supply voltage U to battery voltage U bat : ΔPWMj+1=PWMj+1−PWMj=Uj+1Ubatt−PWMj.

[0063] The calculation of the supply voltage U has already been derived above (see (3.4)): Uj+1=(RJKΔ t)︸k1ωj+1+(−RJKΔ t+K)︸k2ωj+RKTloadj︸k5

[0064] In order to prevent feedback effects of the control on the trajectory planning, the control component is neglected.

[0065] If you put this into the change of the PWM: ΔPWMj+1=k1ωj+1+k5Ubatt−PWMj=k1Ubatt︸k6ωj+1+k5Ubatt−PWMj︸k7

[0066] By inserting (5.12) and (5.15) into (5.9) we obtain Δibattj+1=(k3ωj+1+k4)PWMj+imolj(k6ωj+1+k7)=ωj+1(k3PWMj+k6imotj)︸k8+k4PWMj+k7imotj︸k9

[0067] Or solved for the speed ω j+1 : ωj+1=1k8(Δibattj+1−k9)

[0068] If you set the maximum allowed value for the battery current change, you get the corresponding target speed: ωij+1=1k8(Δibattmax−k9)

Claims

[1] Method for controlling and / or regulating an electric motor, wherein it is provided that - the electric motor (180, 220) is supplied with electrical energy from a battery (130, 230) and / or a generator (150, 190, 210) and - for control and / or regulation, a high-frequency pulse-width modulated voltage is applied to the electric motor, wherein the motor current (IM) supplied from the battery and / or the generator when the electric motor starts up to operate the electric motor is continuously increased, wherein the increase in the motor current is limited to a predetermined setpoint value within a predeterminable time, and wherein the electric motor is operated at a maximum of a predetermined setpoint speed (ω i j+1 ), whereby it is intended that the target speed is dependent on - the battery voltage, and / or - the load torque on the electric motor is specified. [2] Method according to claim 1, characterized by that the setpoint depends on operating parameters - the generator and / or - the battery is specified, It is particularly provided that the setpoint value is dependent - from the current increase generated by the generator and / or - is determined by the discharge current of the battery. [3] Method according to one of the preceding claims, characterized by that it is provided that further consumers (160) are supplied with electrical energy by the battery and / or the generator, wherein the continuous increase is provided as a function of the supply of the further consumers with electrical energy. [4] Method according to claim 3, characterized by that the other consumers are supplied by the battery, whereas the electric motor is supplied by the generator, at least during the start-up phase. [5] Device for controlling and / or regulating an electric motor, wherein - the electric motor (180, 220) is supplied with electrical energy by means of a battery (130, 230) and / or a generator (150, 190, 210) and - a means (100, 200) for controlling and / or regulating the electric motor by means of a high-frequency pulse-width modulated voltage applied to the electric motor, wherein the means continuously increases the motor current (IM) supplied from the battery and / or the generator for operating the electric motor when the electric motor starts up, wherein the means limits the increase in the motor current to a predetermined setpoint value within a predeterminable time, and wherein the electric motor, in order to limit the current gradient, operates at a maximum of a predetermined setpoint speed (ω i j+1 ), whereby it is intended that the target speed is dependent on - the battery voltage, and / or - the load torque on the electric motor is specified. [6] Device according to claim 5, characterized by that the agent determines the target value depending - operating parameters of the generator and / or - the battery, whereby it is particularly provided that - the setpoint depending on the generated current increase of the generator and / or - is determined by the discharge current of the battery. [7] Device according to one of claims 5 to 6, characterized bythat it is provided that further consumers (160) are supplied with electrical energy by the battery and / or the generator, wherein the continuous increase is provided as a function of the supply of electrical energy to the further consumers, wherein it is provided in particular that the further consumers are supplied by the battery, whereas the electric motor is supplied by the generator at least during the start-up phase.

Images