Off-line identification method and system for vehicle active damping parameter of permanent magnet synchronous motor

Abstract

The invention discloses an off-line identification method and system for the active damping parameter of a whole vehicle of a permanent magnet synchronous motor. The off-line identification method comprises the steps that the end angular speed # imgabs0 # and the required torque TR of the motor under the current working condition are obtained in real time; the motor end angular speed # imgabs1 # is input into the pre-calibrated state space model to obtain the real-time vehicle wheel end angular speed # imgabs2 #, and a gain coefficient K under the current working condition is inquired from a pre-calibrated two-dimensional gain coefficient table according to the motor end angular speed # imgabs3 # and the required torque TR; according to the motor end angular speed # imgabs4 #, the vehicle wheel end angular speed # imgabs5 # and the gain coefficient K, the equivalent damping compensation torque Tcomp is calculated; and superposing the equivalent damping compensation torque Tcomp to a required torque TR to generate a total torque Ttotal, and adjusting the output of the motor according to the total torque Ttotal. According to the method and the system, the whole vehicle calibration workload is reduced, the vibration of the transmission system can be suppressed in a high-precision manner without an additional sensor, and the driving and riding comfort is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle motor control, and particularly relates to a method and system for offline identification of active damping parameters of a permanent magnet synchronous motor for a whole vehicle. Background Technique

[0002] The rapid development of new energy vehicles in the domestic passenger vehicle market has made the driving experience the focus of competition. Among them, the vibration problem of the transmission system directly affects the user comfort. Due to the direct coupling of the power system and the transmission system of new energy vehicles, combined with the fast torque response characteristics of the permanent magnet synchronous motor, underdamped vibration phenomena are likely to occur under conditions such as starting, accelerating, decelerating, and energy recovery. Currently, the industry generally adopts active damping control technology to suppress vibration by improving the equivalent damping through software algorithms, but the existing technical solutions have obvious limitations. The multivariable feedback control method based on Kalman filtering adopted by hybrid electric vehicles has the problem of complex parameter solution, and its control effect has only been verified under the tip in condition, lacking adaptability research on other typical conditions. For the longitudinal shudder in pure electric mode, the existing solutions mainly add a frequency compensator based on the stator voltage vector orientation coordinate system, but it is only applicable to low-speed and small-load conditions and cannot cover the vibration suppression requirements of the full operating conditions. Due to the lack of the damping buffer effect of the traditional clutch in pure electric vehicles and the rapid torque response characteristics of the motor, the longitudinal jitter phenomenon is more prominent under various conditions.

[0003] In addition, the technical bottleneck of wheel end speed measurement restricts the active damping control effect. Currently, it is necessary to indirectly calibrate the speed difference between the wheel end and the reducer end to estimate the equivalent damping parameters. This method has the problem of insufficient accuracy, which directly affects the actual effect of vibration control. These technical defects seriously restrict the further improvement of the driving comfort of new energy vehicles. Summary of the Invention

[0004] To solve the technical problems in the background technique, the present invention proposes a method and system for offline identification of active damping parameters of a permanent magnet synchronous motor for a whole vehicle.

[0005] A method for offline identification of active damping parameters of a permanent magnet synchronous motor for a whole vehicle proposed by the present invention includes:

[0006] Obtain the angular velocity of the motor end under the current condition in real time and the required torque T R ;

[0007] Input the angular velocity of the motor end into the pre-calibrated state space model to obtain the real-time angular velocity of the vehicle wheel end

[0008] According to the angular velocity of the motor end and the required torque TR Query the gain coefficient K under the current working condition from the pre-calibrated two-dimensional gain coefficient table;

[0009] According to the angular velocity at the motor end The angular velocity at the vehicle wheel end And the gain coefficient K, calculate the equivalent damping compensation torque T comp ;

[0010] Superimpose the equivalent damping compensation torque T comp To the required torque T R To generate the total torque T total And adjust the output of the motor according to the total torque T total .

[0011] Preferably, the process of calculating the equivalent damping compensation torque T according to the angular velocity at the motor end The angular velocity at the vehicle wheel end And the gain coefficient K specifically includes: comp :

[0012] Calculate the difference between the angular velocity at the motor end And the angular velocity at the vehicle wheel end ;

[0013] Multiply the gain coefficient K by the difference To calculate the equivalent damping compensation torque T comp .

[0014] Preferably, the calibration process of the pre-calibrated state space model specifically includes:

[0015] Under the conditions of real vehicle acceleration, coasting and energy recovery, collect real vehicle data, and the real vehicle data includes the angular velocity at the motor end The required torque T R And the actual output torque T L ;

[0016] Build a dynamic model, input the real vehicle data into the dynamic model, and adjust the dynamic model parameters through open-loop simulation until the root mean square error between the simulation speed and the actual speed is less than the preset threshold to obtain the pre-calibrated dynamic model. The dynamic model parameters include the equivalent moment of inertia I at the vehicle body wheel end L , The equivalent moment of inertia I at the motor output end after passing through the double planetary row end R , The equivalent elastic coefficient k TI And the equivalent damping coefficient c TI ;

[0017] Convert the pre-calibrated dynamic model into state space form to obtain the pre-calibrated state space model.

[0018] Preferably, the kinetic model is specifically:

[0019]

[0020] where θ R is the angle value of the motor through the double planetary gear set; θ L is the angle value of the vehicle wheel end; is the angular velocity of the motor end; is the angular velocity of the vehicle wheel end; is the angular acceleration of the motor end; is the angular acceleration of the vehicle wheel end; I L is the equivalent moment of inertia of the vehicle body wheel end; I R is the equivalent moment of inertia of the motor output end through the double planetary gear set end; T L is the equivalent vehicle body load end torque; T R is the equivalent motor output end torque; k TI is the equivalent elastic coefficient of the transmission system; c TI is the damping coefficient of the transmission system.

[0021] Preferably, the pre-calibrated state space model is specifically:

[0022]

[0023] where A is the system matrix, B is the input matrix, C is the output matrix, C = [1 0 0 0]; u is the input vector, u = T R ; is the state variable; is the system output, is the differential of the state variable.

[0024] Preferably, the pre-calibrated two-dimensional gain coefficient table is pre-configured with a one-to-one mapping relationship between the angular velocity of the motor end and the required torque T R and the gain coefficient K.

[0025] Preferably, the pre-calibrated two-dimensional gain coefficient table is obtained by calibrating the vehicle jitter under various rotational speeds and required torque conditions of the whole vehicle.

[0026] Preferably, it further includes:

[0027] During the process of obtaining the angular velocity of the motor end under the current working condition, the real-time collected motor speed signal is filtered to obtain the angular velocity

[0028] An off-line identification system for active damping parameters of a permanent magnet synchronous motor vehicle proposed by the present invention includes:

[0029] A data acquisition module for real-time obtaining the angular velocity of the motor end under the current working condition and the required torque T R ;

[0030] A first processing module for inputting the angular velocity of the motor end into a pre-calibrated state space model to obtain the real-time angular velocity of the vehicle wheel end

[0031] A data query module for querying the gain coefficient K under the current working condition from a pre-calibrated two-dimensional gain coefficient table according to the angular velocity of the motor end and the required torque T R ;

[0032] A second processing module for calculating the equivalent damping compensation torque T according to the angular velocity of the motor end the angular velocity of the vehicle wheel end comp ;

[0033] An output module for superimposing the equivalent damping compensation torque T comp onto the required torque T R to generate the total torque T total and adjusting the output of the motor according to the total torque T total ;

[0034] In the present invention, the proposed off-line identification method and system for active damping parameters of a permanent magnet synchronous motor vehicle reduce the complexity and time cost of vehicle commissioning by off-line calibrating the dynamic model parameters and the two-dimensional gain coefficient table, and avoid the problem of solving multi-variable coupling parameters in the traditional method; combined with a pre-calibrated state space model, it can accurately estimate the rotational speed difference without relying on a wheel-end rotational speed sensor, generate a dynamic compensation torque, effectively suppress the longitudinal jitter of the drive system under conditions such as starting, acceleration, deceleration, and energy recovery, and improve the ride comfort. At the same time, through the two-dimensional look-up table mechanism of the gain coefficient, multi-condition adaptive control is realized, taking into account both the algorithm efficiency and the vibration suppression effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a schematic structural diagram of the working process of an off-line identification method for active damping parameters of a permanent magnet synchronous motor vehicle proposed by the present invention;

[0036] Figure 2 is a simplified equivalent damping logic structure diagram of an off-line identification method for active damping parameters of a permanent magnet synchronous motor vehicle proposed by the present invention;

[0037] Figure 3 Structural schematic diagram of a simplified model of a pure electric drive system for an active damping parameter off-line identification method of a permanent magnet synchronous motor vehicle proposed by the present invention;

[0038] Figure 4 Observation logic block diagram of the real model for an active damping parameter off-line identification method of a permanent magnet synchronous motor vehicle proposed by the present invention;

[0039] Figure 5 Schematic diagram of the demand torque and motor speed data curve for an active damping parameter off-line identification method of a permanent magnet synchronous motor vehicle proposed by the present invention;

[0040] Figure 6 Effect diagram of the open-loop adjustment control parameters for an active damping parameter off-line identification method of a permanent magnet synchronous motor vehicle proposed by the present invention;

[0041] Figure 7 Effect diagram of the comparison between the observed speed and the motor speed for an active damping parameter off-line identification method of a permanent magnet synchronous motor vehicle proposed by the present invention;

[0042] Figure 8 System architecture schematic diagram of an active damping parameter off-line identification system for a permanent magnet synchronous motor vehicle proposed by the present invention. Detailed implementation method

[0043] Refer to Figure 1-8 , an active damping parameter off-line identification method of a permanent magnet synchronous motor vehicle proposed by the present invention includes the following steps:

[0044] S1. Real-time acquisition of the motor terminal angular velocity and the demand torque T R .

[0045] In this embodiment, in the process of acquiring the motor terminal angular velocity under the current working condition, the motor speed signal collected in real time is filtered to obtain the motor terminal angular velocity

[0046] S2. Input the motor terminal angular velocity into the pre-calibrated state space model to obtain the real-time vehicle wheel terminal angular velocity

[0047] In this embodiment, the calibration process of the pre-calibrated state space model specifically includes:

[0048] Under the conditions of real vehicle acceleration, coasting and energy recovery, collect real vehicle data, and the real vehicle data includes the motor terminal angular velocity demand torque T R and the actual output torque TL ;

[0049] Build a dynamic model, input the real vehicle data into the dynamic model, and adjust the parameters of the dynamic model through open-loop simulation until the root mean square error between the simulated speed and the actual speed is less than a preset threshold to obtain a pre-calibrated dynamic model. The parameters of the dynamic model include the equivalent moment of inertia I of the vehicle wheel end L , the equivalent moment of inertia I of the motor output end passing through the double planetary gear set R , the equivalent elastic coefficient k TI and the equivalent damping coefficient c TI ;

[0050] Convert the pre-calibrated dynamic model into the state space form to obtain a pre-calibrated state space model.

[0051] In this embodiment, the dynamic model is specifically:

[0052]

[0053] where θ R is the angle value of the motor passing through the double planetary gear set; θ L is the angle value of the vehicle wheel end; is the angular velocity of the motor end; is the angular velocity of the vehicle wheel end; is the angular acceleration of the motor end; is the angular acceleration of the vehicle wheel end; I L is the equivalent moment of inertia of the vehicle wheel end; I R is the equivalent moment of inertia of the motor output end passing through the double planetary gear set; T L is the equivalent torque of the vehicle load end; T R is the equivalent torque of the motor output end; k TI is the equivalent elastic coefficient of the transmission system; c TI is the damping coefficient of the transmission system.

[0054] Specifically, to calculate the wheel end speed, a simplified model of the pure electric drive system as shown in Figure 3 is proposed, which only includes a two-mass model with a single-stage main reducer.

[0055] In the process of converting the pre-calibrated dynamic model into the state space form, the dynamic model is converted into the state space form as shown below. Since the load end torque, as an external disturbance variable of the system, is related to the vehicle mass and the slope it is on and does not change suddenly, it is taken as a state variable. Therefore, there are state vector x, system matrix A, input matrix B, output matrix C, direct transmission matrix D, input vector u, and the initial state space model is specifically:

[0056]

[0057] After obtaining the initial state - space model, since the initial state - space model is obtained through a simplified transmission model, it is different from the actual vehicle. The initial state - space model is converted into a simplified model, and the simplified model is used as the pre - calibrated state - space model.

[0058] In this embodiment, the pre - calibrated state - space model is specifically:

[0059]

[0060] Among them, \(A\) is the system matrix, \(B\) is the input matrix, \(C\) is the output matrix, \(C = [1\ 0\ 0\ 0]\); \(u\) is the input vector, \(u = T\) R ; is the state variable; is the system output, is the differential of the state variable.

[0061] For the above - mentioned pre - calibrated state - space model, theoretically, when the parameters of the system matrix \(A\), input matrix \(B\), and output matrix \(C\) are determined accurately enough, the output of the vehicle in reality can be obtained through real - time calculation. However, in actual situations, it is impossible to fully simulate the physical characteristics of the actual vehicle. Therefore, for the same input in the open - loop case, the outputs of the real model and the pre - calibrated state - space model are different. To make the output of the simplified model more accurate, the difference between the output of the pre - calibrated state - space model and the output of the real model is introduced as feedback for closed - loop. The logic block diagram for observing the real model through the pre - calibrated state - space model is as Figure 4 shown.

[0062] As Figure 4 shown, by observing the output of the real model through the pre - calibrated state - space model, only the parameters \(I\) R , \(I\) L , \(k\) TI , \(c\) TI related to the state - space model need to be determined. Since the motor - end speed can be obtained in real - time through resolver decoding, the parameters \(I\) R , \(I\) L , \(k\) TI , \(c\) TI can be confirmed by comparing the speeds of the state model and the actual vehicle under the same input. Since the model parameters related to the brake are not considered in the transmission - system state model, in fact, only the data of the vehicle during acceleration and coasting phases are collected for model simulation, so as to confirm the state - space model parameters. Figure 5 is the required torque and motor - speed data collected from the actual vehicle.

[0063] Perform simulation experiments through Matlab and adjust parameter I R 、I L 、k TI 、c TI Under the same input conditions, without considering the closed-loop control gain L, that is, under open-loop observation, the rotational speed results almost coincide with the actual rotational speed data. The adjustment results are as Figure 6 shown

[0064] Substitute the parameters adjusted in the above open-loop into the state-space equation of the real model, and after adjusting the design of the gain L, perform simulation to make the observed output of the pre-calibrated state-space model consistent with the real model, that is, the observed rotational speed confirmed by simulation is consistent with the motor rotational speed collected from the real vehicle under the same input conditions. After reasonably designing the gain L, the motor rotational speed observed by the pre-calibrated state-space model can be almost consistent with the motor rotational speed output by the real model, as Figure 7 shown

[0065] Through the above steps, each parameter in the pre-calibrated state-space model can be confirmed offline. Combining with the dynamic model, the vehicle wheel-end rotational speed can be obtained by observing the rotational speed at the motor end collected in real time. Finally, through calibration, the appropriate gain coefficient K under each torque and rotational speed condition can be confirmed. Multiply the gain coefficient K by the difference between the vehicle wheel-end rotational speed and the motor-end rotational speed to obtain the appropriate equivalent damping compensation torque T comp .

[0066] S3. According to the angular velocity at the motor end and the required torque T R , query the gain coefficient K under the current working condition from the pre-calibrated two-dimensional gain coefficient table

[0067] In this embodiment, the pre-calibrated two-dimensional gain coefficient table is pre-configured with a one-to-one mapping relationship between the angular velocity at the motor end, the required torque T R and the gain coefficient K

[0068] In this embodiment, the pre-calibrated two-dimensional gain coefficient table is obtained by calibrating the vehicle jitter under each rotational speed and required torque condition of the whole vehicle

[0069] S4. According to the angular velocity at the motor end, the angular velocity at the vehicle wheel end, and the gain coefficient K, calculate the equivalent damping compensation torque T comp .

[0070] In this embodiment, step S4 specifically includes

[0071] Calculate the difference between the angular velocity at the motor end and the angular velocity at the vehicle wheel end

[0072] Multiply the gain coefficient K by the difference to calculate the equivalent damping compensation torque T comp .

[0073] S5. Add the equivalent damping compensation torque T comp to the demand torque T R to generate the total torque T total , and adjust the output of the motor according to the total torque T total .

[0074] Specifically, as Figure 2 shown Figure 2 , is a simplified logic architecture diagram for equivalent damping generation; the real-time motor speed can be obtained through resolver decoding, and the gain coefficient K is queried by using the calculated motor speed and the demand torque received by the motor controller in the current calculation cycle. The real-time motor speed obtained by resolver decoding is multiplied by the reduction gear ratio to confirm the motor-end speed, that is, the motor-end angular velocity The vehicle wheel-end speed, that is, the vehicle wheel-end angular velocity, is obtained by observing the real-time calculated output actual torque of the motor-end speed and the motor controller Finally, multiply the gain coefficient K by the difference between the vehicle wheel-end speed and the motor-end speed to obtain the real-time equivalent damping, and add the real-time equivalent damping as the compensation torque to the torque closed-loop control link, so as to suppress the longitudinal jitter in the underdamped transmission system

[0075] In this embodiment, the demand torque is sent in real time by the VCU (vehicle control unit), and received in real time by the MCU (motor controller). The motor speed signal can be obtained in real time through resolver decoding. At the same time, to avoid the influence of random loads and high-frequency noises on the speed signal, the real-time collected motor speed signal is filtered. The gain coefficient K can be determined by calibrating the equivalent damping under different torque and speed conditions. Since the wheel-end speed, as the input of the entire control link, cannot be directly sampled by the MCU due to its physical characteristics, the active damping control parameters can be calibrated offline, and the wheel-end speed can be calculated in real time through the calibrated parameters, so as to obtain the compensation torque

[0076] Referring to Figure 1-8 , an off-line identification system for active damping parameters of a permanent magnet synchronous motor vehicle proposed by the present invention includes:

[0077] A data acquisition module for obtaining the motor-end angular velocity and the demand torque T R in real time under the current working condition;

[0078] A first processing module for inputting the motor-end angular velocity into a pre-calibrated state space model to obtain the real-time vehicle wheel-end angular velocity

[0079] A data query module, configured to query, according to the angular velocity at the motor end and the required torque T R the gain coefficient K under the current working condition from a pre-calibrated two-dimensional gain coefficient table;

[0080] A second processing module, configured to calculate, according to the angular velocity at the motor end the angular velocity at the vehicle wheel end and the gain coefficient K, the equivalent damping compensation torque T comp ;

[0081] An output module, configured to superimpose the equivalent damping compensation torque T comp onto the required torque T R to generate the total torque T total , and adjust the output of the motor according to the total torque T total .

[0082] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered by the protection scope of the present invention.

Claims

1. An off-line identification method for the active damping parameters of a permanent magnet synchronous motor vehicle, characterized in that, including: Obtain the angular velocity at the motor end under the current working condition in real time and the required torque T R ; Input the angular velocity at the motor end into the pre-calibrated state space model to obtain the real-time angular velocity at the vehicle wheel end According to the angular velocity at the motor end and the required torque T R query the gain coefficient K under the current working condition from the pre-calibrated two-dimensional gain coefficient table; According to the angular velocity at the motor end The angular velocity at the vehicle wheel end and the gain coefficient K, calculate the equivalent damping compensation torque T comp ; Add the equivalent damping compensation torque T comp to the required torque T R to generate the total torque T total , and adjust the output of the motor according to the total torque T total .

2. The off-line identification method for the active damping parameters of the permanent magnet synchronous motor vehicle according to claim 1, characterized in that, The equivalent damping compensation torque T is calculated based on the angular velocity at the motor end the angular velocity at the vehicle wheel end and the gain coefficient K, specifically including: comp , specifically including: Calculate the angular velocity at the motor end and the angular velocity at the vehicle wheel end difference Multiply the gain coefficient K by the difference to calculate the equivalent damping compensation torque T comp .

3. The off-line identification method for the active damping parameters of the whole vehicle of the permanent magnet synchronous motor according to claim 1, characterized in that, The calibration process of the pre-calibrated state space model specifically includes: Collect real vehicle data under real vehicle acceleration, coasting, and energy recovery conditions. The real vehicle data includes the angular velocity at the motor end Required torque T R and the actual output torque T L ; Build a dynamic model, input the real vehicle data into the dynamic model, and adjust the parameters of the dynamic model through open-loop simulation until the root mean square error between the simulated speed and the actual speed is less than a preset threshold to obtain a pre-calibrated dynamic model. The dynamic model parameters include the equivalent moment of inertia I of the vehicle body wheel end L , the equivalent moment of inertia I of the motor output end passing through the double planetary gear set end R , the equivalent elastic coefficient k TI , and the equivalent damping coefficient c TI ; Converting the pre-calibrated dynamic model into state space form to obtain the pre-calibrated state space model.

4. The off-line identification method for the active damping parameters of the permanent magnet synchronous motor vehicle according to claim 3, characterized in that The dynamic model is specifically: Among them, θ R is the angle value of the motor passing through the double planetary gear sets; θ L is the angle value of the vehicle wheel end; is the angular velocity of the motor end; is the angular velocity of the vehicle wheel end; is the angular acceleration of the motor end; is the angular acceleration of the vehicle wheel end; I L is the equivalent moment of inertia of the vehicle body wheel end; I R is the equivalent moment of inertia of the motor output end passing through the double planetary gear sets end; T L is the equivalent vehicle body load end torque; T R is the equivalent motor output end torque; k TI is the equivalent elastic coefficient of the transmission system; c TI is the damping coefficient of the transmission system.

5. The off-line identification method for the active damping parameters of the permanent magnet synchronous motor vehicle according to claim 3, characterized in that, The pre-calibrated state space model is specifically: where A is the system matrix, B is the input matrix, C is the output matrix, C = [1 0 0 0]; u is the input vector, u = T R ; is the state variable; is the system output, is the differential of the state variable.

6. The off-line identification method for the active damping parameters of the permanent magnet synchronous motor vehicle according to claim 1, characterized in that, The pre-calibrated two-dimensional gain coefficient table is pre-configured with the angular velocity at the motor end and the required torque T R as well as the one-to-one mapping relationship with the gain coefficient K.

7. The off-line identification method for the active damping parameters of the permanent magnet synchronous motor vehicle according to claim 6, characterized in that, The pre-calibrated two-dimensional gain coefficient table is obtained from the vehicle jitter conditions under various rotational speeds and various demand torque conditions during vehicle calibration.

8. The off-line identification method for the active damping parameters of the permanent magnet synchronous motor vehicle according to claim 1, characterized in that, It also includes: In the process of obtaining the angular velocity at the motor end under the current working condition the motor speed signal collected in real time is filtered to obtain the angular velocity at the motor end 9. An off-line identification system for the active damping parameters of a permanent magnet synchronous motor vehicle, characterized in that, including: A data acquisition module, which is used to obtain the angular velocity at the motor end under the current working condition in real time and the required torque T R ; The first processing module is used to input the angular velocity at the motor end into the pre-calibrated state space model to obtain the real-time angular velocity at the vehicle wheel end A data query module for querying, according to the angular velocity at the motor end, and the required torque T R the gain coefficient K under the current working condition from a pre-calibrated two-dimensional gain coefficient table; The second processing module is used to calculate the equivalent damping compensation torque T according to the angular velocity at the motor end The angular velocity at the vehicle wheel end and the gain coefficient K comp ; An output module for adding the equivalent damping compensation torque T comp to the required torque T R to generate the total torque T total and adjusting the output of the motor according to the total torque T total .

Images