A harmonic optimization and torque compensation integrated electric drive system collaborative vibration suppression method

Abstract

The application discloses a kind of harmonic optimization and torque compensation fusion electric drive system collaborative vibration suppression method, belong to electric drive system collaborative vibration suppression field, this method includes the half shaft speed difference as state variable input ADRC controller, obtains torque compensation control quantity;Torque compensation control quantity and original target torque instruction are superimposed, obtain final target torque instruction;Get d-q axis target current instruction;According to d-q axis target current instruction, obtain d-q axis harmonic current injection quantity, and with d-q axis target current instruction is superimposed, obtain final d-q axis target current instruction, and with current d-q axis current difference after input current loop, and by vector pulse width modulation algorithm SVPWM, obtain the driving signal of motor at current time.The application solves the problem that existing method only relies on single control strategy, lack of steady-state, transient state Working condition electric drive system multi-physical field coupling vibration collaborative optimization, overall suppression effect is not good.

Description

Technical Field

[0001] This invention belongs to the field of collaborative vibration suppression of electric drive systems, and particularly relates to a collaborative vibration suppression method for electric drive systems that integrates harmonic optimization and torque compensation. Background Technology

[0002] During steady-state operation of the electric drive system, the continuous coupling effect of electromagnetic vibration excitation from the motor and mechanical vibration excitation from the gear transmission system leads to significant torsional vibration, severely deteriorating the NVH performance of the electric drive system. Furthermore, under transient impact conditions, sudden increases or decreases in external load cause violent oscillations in the dynamic torque of the drive shaft, resulting in a substantial increase in the dynamic load on the gears and severely affecting the reliability of the electric drive system.

[0003] Existing technologies mainly reduce torsional vibration in electric drive systems from two aspects: structural design and active control.

[0004] In terms of structural design: the main focus is on the excitation source, optimizing structural parameters to reduce the impact of electromagnetic vibration excitation of the motor or mechanical vibration excitation of the gear transmission system. For example, slotting the motor stator and rotor, adopting a rotor skewed pole structure to reduce magnetic field harmonics and cogging torque, and micro-modifying the gears to reduce gear meshing transmission errors.

[0005] In terms of active control: under steady-state conditions, methods such as suppressing electromagnetic torque fluctuations are mostly used to reduce the torsional vibration of the electric drive system, while under transient conditions, control methods such as filters, PI control, linear quadratic regulators, and damping compensation are mostly used.

[0006] Structural design-based methods can reduce torsional vibration in electric drive systems to some extent, but they are usually costly and difficult to manufacture and improve. Furthermore, optimizing the motor structure will reduce the motor's power density and efficiency, making it difficult to obtain an effective design result.

[0007] Most methods for suppressing electromagnetic torque fluctuations under steady-state conditions focus on suppressing winding harmonic currents, neglecting the effects of motor body magnetic field harmonics and cogging torque, resulting in poor suppression effectiveness. Methods for actively suppressing electromagnetic torque fluctuations under transient conditions by controlling the dynamic load of the gears via the motor are less common. Furthermore, no active vibration reduction methods have been found for electric drive systems under complex time-varying conditions. Summary of the Invention

[0008] To address the aforementioned shortcomings in existing technologies, this invention provides a collaborative vibration suppression method for electric drive systems that integrates harmonic optimization and torque compensation. This method solves the problem that existing methods rely solely on a single control strategy, lack collaborative optimization of multi-physics field coupled vibrations of electric drive systems under steady-state and transient conditions, and thus have poor overall suppression effects.

[0009] To achieve the aforementioned objectives, the technical solution adopted by this invention is: a collaborative vibration suppression method for an electric drive system integrating harmonic optimization and torque compensation, comprising:

[0010] The encoder is used to track the speed difference of the half-shaft of the electric drive system.

[0011] The half-shaft speed difference is input as a state variable into the ADRC controller to obtain the torque compensation control quantity;

[0012] Obtain the original target torque command and superimpose the torque compensation control quantity with the original target torque command to obtain the final target torque command;

[0013] Obtain the current motor temperature and current dq axis current, and solve for the optimal current operating point based on the current motor temperature, current dq axis current and final target torque command to obtain the target current command for the dq axis;

[0014] Obtain the motor rotor position angle, and based on the motor rotor position angle, the current motor temperature, and the target current command for the dq axis, solve for the amplitude and phase of the harmonic current to obtain the amount of harmonic current injection for the dq axis;

[0015] The dq-axis harmonic current injection amount is superimposed with the dq-axis target current command to obtain the final dq-axis target current command;

[0016] The final target current command for the dq axis is subtracted from the current dq axis current and input into the current loop. The signal is then processed by the Vector Pulse Width Modulation (SVPWM) algorithm to obtain the motor drive signal at the current moment, thus completing the coordinated vibration suppression of the electric drive system.

[0017] The beneficial effects of this invention are as follows: For steady-state conditions, this invention employs an electromagnetic torque fluctuation suppression method based on proportional-integral-resonant (PI-Res) optimal harmonic current injection; for transient conditions, it employs a gear dynamic load suppression method based on active disturbance rejection (ADRC) torque compensation. This enables effective suppression of torsional vibration in electric drive systems under both steady-state and transient conditions. Furthermore, this invention utilizes an active vibration reduction method based on the coordinated integration of harmonic current optimization and active disturbance rejection torque compensation, enabling rapid and effective suppression of multi-field coupled vibration in electric drive systems under complex time-varying conditions.

[0018] Furthermore, the ADRC controller includes a tracking differentiator TD, an extended state observer ESO, and a nonlinear state error feedback NLSEF;

[0019] The tracking differentiator TD is used to obtain the tracking signal after the reference signal transition and the differential of the tracking signal:

[0020]

[0021] Where ω1(k+1) is the tracking signal after the reference signal transition at time k+1; ω1(k) is the tracking signal after the reference signal transition at time k; T0 is the sampling step size; ω2(k) is the derivative of the tracking signal after the reference signal transition at time k; ω2(k+1) is the derivative of the tracking signal after the reference signal transition at time k+1; fhan() is the fastest control synthesis function; Δω tar is the reference signal; r is the velocity factor; h0 is the filter factor; a is the intermediate control quantity; sign() is the sign function; x1 is the tracking error; x2 is the derivative of the tracking error;

[0022] The extended state observer (ESO) is used to estimate the system state and total disturbance in real time based on the half-shaft speed difference, obtaining the state estimate of the half-shaft speed difference, the derivative of the state estimate of the half-shaft speed difference, and the estimate of the total system disturbance:

[0023]

[0024] Where, e(k) is the state estimation error output at time k; z1(k) is the state estimate of the half-shaft speed difference by the extended state observer ESO at time k; Δω is the half-shaft speed difference; z1(k+1) is the state estimate of the half-shaft speed difference by the extended state observer ESO at time k+1; z2(k) is the derivative of the state estimate of the half-shaft speed difference by the extended state observer ESO at time k; β1, β2, and β3 are error correction coefficients; fal() is a nonlinear power function; a1, a2, and a3 are nonlinear factors; δ is the interval width of the nonlinear power function; z2(k+1) is the derivative of the state estimate of the half-shaft speed difference by the extended state observer ESO at time k+1; z3(k) is the estimate of the total system disturbance at time k; b0 is the compensation factor of the extended state observer ESO; T eADRC (k) represents the torque compensation control quantity output by the ADRC controller at time k; z3(k+1) represents the estimated value of the total disturbance of the system at time k+1;

[0025] The nonlinear state error feedback (NLSEF) is used to calculate the initial compensation torque based on the tracking signal after the reference signal transition, the derivative of the tracking signal, the state estimate of the half-shaft speed difference, the derivative of the state estimate of the half-shaft speed difference, and the estimate of the total system disturbance. The initial compensation torque is then combined with the feedforward compensation term to generate the torque compensation control quantity T. eADRC :

[0026]

[0027] Where e1(k) is the difference between ω1(k) and z1(k) at time k; e2(k) is the difference between ω2(k) and z2(k) at time k; T e0The initial compensation torque; a4 and a5 are both nonlinear factors; β4 is the proportional factor of the nonlinear state error feedback (NLSEF); β5 is the differential factor of the nonlinear state error feedback (NLSEF); T eADRC This is the torque compensation control variable.

[0028] The beneficial effects of the above-mentioned further scheme are as follows: taking the half-shaft speed difference = 0 as the control target, the ADRC controller generates a compensating torque control quantity that is opposite in phase to the half-shaft speed difference, which is superimposed with the original target torque command to form the final target torque command, and then controlled by the motor vector control system to achieve the purpose of suppressing gear dynamic load and rapidly attenuating the free vibration of the system.

[0029] Furthermore, the step of obtaining the current motor temperature and solving for the optimal current operating point based on the current motor temperature and the final target torque command to obtain the dq-axis target current command specifically involves:

[0030] Obtain the current motor temperature and current dq axis current, and use the interior point method to interpolate and look up the dq axis inductance and dq axis permanent magnet flux linkage to obtain the current dq axis inductance value and the current dq axis permanent magnet flux linkage value.

[0031] The motor speed is obtained. Based on the final target torque command, motor speed, current dq-axis inductance value, and current dq-axis permanent magnet flux linkage value, field weakening control is performed in combination with the maximum torque current ratio and the maximum torque voltage ratio to obtain the dq-axis target current command.

[0032] The beneficial effects of the above-mentioned further scheme are as follows: the optimal current operating point is solved based on the current motor temperature and the final target torque command, and the target current command of the dq axis is obtained, which prepares for the solution of harmonic current injection and motor vector control.

[0033] Further, the process of obtaining the motor rotor position angle and, based on the motor rotor position angle, the current motor temperature, and the target current command for the dq axis, calculating the harmonic current amplitude and phase to obtain the dq axis harmonic current injection amount is as follows:

[0034] Obtain the motor rotor position angle. Based on the target current command of the dq axis, the motor rotor position angle, and the current motor temperature, use the current-magnetic flux linkage mapping relationship to obtain the instantaneous magnetic flux linkage and the average magnetic flux linkage of the dq axis.

[0035] Subtracting the average flux linkage along the dq axis from the instantaneous flux linkage along the dq axis yields the harmonic flux linkage along the dq axis:

[0036]

[0037] Where, ψ dhar ψ is the d-axis harmonic flux linkage;qhar ψ is the q-axis harmonic flux linkage; dins ψ is the instantaneous magnetic flux linkage along the d-axis; qins ψ is the instantaneous magnetic flux linkage along the q-axis; dave ψ is the d-axis average flux linkage; qave m is the q-axis average flux linkage; -1 This represents the d-axis current-magnetic flux mapping relationship; n -1 This represents the q-axis current-magnetic flux mapping relationship; The target current command for the d-axis; For the q-axis target current command; θ e The rotor position angle is T; the current motor temperature is T.

[0038] Based on the dq-axis harmonic flux linkage, the motor rotor position angle, and the current motor temperature, the dq-axis harmonic current before filtering is obtained using the flux linkage-current mapping relationship.

[0039] The dq-axis harmonic current injection amount is obtained by filtering the unfiltered dq-axis harmonic current using a first-order high-pass filter:

[0040]

[0041] Among them, i dhar The d-axis harmonic current injection amount; filter[·] is a first-order high-pass filter; f is the d-axis harmonic current before filtering; -1 (·) represents the flux linkage-d-axis current mapping relationship; i qhar This refers to the amount of q-axis harmonic current injected. For the q-axis harmonic current before filtering; g -1 () represents the flux linkage-q-axis current mapping relationship; y[k] is the output signal of the first-order high-pass filter at time k; K com f is the filter gain; c T is the cutoff frequency; s y[k] is the sampling time; x[k] is the input signal of the first-order high-pass filter at time k; x[k-1] is the input signal of the first-order high-pass filter at time k-1; y[k-1] is the output signal of the first-order high-pass filter at time k-1.

[0042] The beneficial effects of the above-mentioned further scheme are as follows: based on the motor rotor position angle, the current motor temperature and the target current command of the dq axis, the amplitude and phase of the current harmonic current injection can be accurately extracted and stripped out by utilizing the current-magnetic flux and magnetic flux-current mapping relationship.

[0043] Furthermore, the step of subtracting the final dq-axis target current command from the current dq-axis current and inputting the result into the current loop, followed by a Vector Pulse Width Modulation (SVPWM) algorithm, to obtain the pulse signal controlling the start and stop of the three-phase inverter IGBT switches, specifically involves:

[0044] Get the current dq axis current and electric angular velocity;

[0045] The final target current command for the dq axis is subtracted from the current dq axis current to obtain the dq axis current difference signal;

[0046] Input the electrical angular velocity and the difference signal between the dq-axis currents into the current loop to obtain the first dq-axis voltage command;

[0047] Based on the current dq-axis current and electric angular velocity, voltage feedforward is performed to obtain the voltage feedforward result;

[0048] The first feedback result is obtained by subtracting the first dq axis voltage command from the voltage feedforward result.

[0049] Based on the first feedback result and the motor rotor position angle, perform inverse Park transformation to obtain the α-β axis voltage command at the current moment;

[0050] Based on the α-β axis voltage command at the current moment, the drive signal of the motor at the current moment is obtained by using the Vector Pulse Width Modulation (SVPWM) algorithm.

[0051] The beneficial effects of the above-mentioned further scheme are: the torque compensation control quantity and harmonic current injection quantity are synergistically integrated, and the motor drive signal at the current moment is obtained after motor vector control, so as to achieve the purpose of multi-physical field coupling vibration synergistic optimization of electric drive system under steady-state and transient conditions.

[0052] Furthermore, the current loop employs a PI-Res controller; the PI-Res controller includes a current loop PI controller and a quasi-resonant controller connected in parallel.

[0053] The beneficial effect of the above-mentioned further scheme is that a quasi-resonant controller is connected in parallel on the basis of the d-axis and q-axis current loop PI controller, so as to achieve the purpose of real-time control of the DC and AC components in the current.

[0054] Furthermore, the current loop PI controller is used to control the DC current component to obtain the DC voltage command component; the quasi-resonant controller is used to control the 6th harmonic current and the 12th harmonic current respectively to obtain the voltage command component corresponding to the 6th harmonic current and the voltage command component corresponding to the 12th harmonic current.

[0055] The beneficial effect of the above-mentioned further scheme is that, based on the d-axis and q-axis current loop PI controller, a quasi-resonant controller is connected in parallel to control the 6th and 12th harmonic currents, so as to suppress the dominant 6th and 12th order electromagnetic torque fluctuation amplitudes.

[0056] Furthermore, the transfer function of the PI-Res controller is:

[0057]

[0058] Wherein, H(s) PIR For the transfer function of the PI-Res controller; K P K is the proportional gain coefficient. I K represents the integral gain coefficient; s represents the complex frequency variable; K represents the integral gain coefficient. R ω is the quasi-resonant gain coefficient; ω0 is the resonant frequency, i.e., the harmonic current frequency; ω c This is the cutoff frequency.

[0059] The beneficial effect of the above-mentioned further scheme is that a quasi-resonant controller is connected in parallel on the basis of the d-axis and q-axis current loop PI controller to achieve rapid response and accurate tracking of harmonic currents of specific frequencies. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating the framework of the present invention.

[0061] Figure 2 This is a schematic diagram of the ADRC controller framework in this invention.

[0062] Figure 3 This is a schematic diagram illustrating the solution for the dq-axis harmonic current injection in this invention. Detailed Implementation

[0063] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0064] like Figure 1 As shown, in one embodiment of the present invention, a method for coordinated vibration suppression of an electric drive system that integrates harmonic optimization and torque compensation includes:

[0065] The encoder is used to track the speed difference of the half-shaft of the electric drive system.

[0066] The half-shaft speed difference is input as a state variable into the ADRC controller to obtain the torque compensation control quantity;

[0067] Obtain the original target torque command and superimpose the torque compensation control quantity with the original target torque command to obtain the final target torque command;

[0068] Obtain the current motor temperature and current dq axis current, and solve for the optimal current operating point based on the current motor temperature, current dq axis current and final target torque command to obtain the target current command for the dq axis;

[0069] Obtain the motor rotor position angle, and based on the motor rotor position angle, the current motor temperature, and the target current command for the dq axis, solve for the amplitude and phase of the harmonic current to obtain the amount of harmonic current injection for the dq axis;

[0070] The dq-axis harmonic current injection amount is superimposed with the dq-axis target current command to obtain the final dq-axis target current command;

[0071] The final target current command for the dq axis is subtracted from the current dq axis current and input into the current loop. The signal is then processed by the Vector Pulse Width Modulation (SVPWM) algorithm to obtain the motor drive signal at the current moment, thus completing the coordinated vibration suppression of the electric drive system.

[0072] like Figure 2 As shown, the ADRC controller includes a tracking differentiator TD, an extended state observer ESO, and a nonlinear state error feedback NLSEF;

[0073] The tracking differentiator TD is used to obtain the tracking signal after the reference signal transition and the differential of the tracking signal:

[0074]

[0075]

[0076] Where ω1(k+1) is the tracking signal after the reference signal transition at time k+1; ω1(k) is the tracking signal after the reference signal transition at time k; T0 is the sampling step size; ω2(k) is the derivative of the tracking signal after the reference signal transition at time k; ω2(k+1) is the derivative of the tracking signal after the reference signal transition at time k+1; fhan() is the fastest control synthesis function; Δω tar is the reference signal; r is the velocity factor; h0 is the filter factor; a is the intermediate control quantity; sign() is the sign function; x1 is the tracking error; x2 is the derivative of the tracking error;

[0077] The extended state observer (ESO) is used to estimate the system state and total disturbance in real time based on the half-shaft speed difference, obtaining the state estimate of the half-shaft speed difference, the derivative of the state estimate of the half-shaft speed difference, and the estimate of the total system disturbance:

[0078]

[0079] Where, e(k) is the state estimation error output at time k; z1(k) is the state estimate of the half-shaft speed difference by the extended state observer ESO at time k; Δω is the half-shaft speed difference; z1(k+1) is the state estimate of the half-shaft speed difference by the extended state observer ESO at time k+1; z2(k) is the derivative of the state estimate of the half-shaft speed difference by the extended state observer ESO at time k; β1, β2, and β3 are error correction coefficients; fal() is a nonlinear power function; a1, a2, and a3 are nonlinear factors; δ is the interval width of the nonlinear power function; z2(k+1) is the derivative of the state estimate of the half-shaft speed difference by the extended state observer ESO at time k+1; z3(k) is the estimate of the total system disturbance at time k; b0 is the compensation factor of the extended state observer ESO; T eADRC (k) represents the torque compensation control quantity output by the ADRC controller at time k; z3(k+1) represents the estimated value of the total disturbance of the system at time k+1;

[0080] The nonlinear state error feedback (NLSEF) is used to calculate the initial compensation torque based on the tracking signal after the reference signal transition, the derivative of the tracking signal, the state estimate of the half-shaft speed difference, the derivative of the state estimate of the half-shaft speed difference, and the estimate of the total system disturbance. The initial compensation torque is then combined with the feedforward compensation term to generate the torque compensation control quantity T. eADRC :

[0081]

[0082] Where e1(k) is the difference between ω1(k) and z1(k) at time k; e2(k) is the difference between ω2(k) and z2(k) at time k; T e0 The initial compensation torque; a4 and a5 are both nonlinear factors; β4 is the proportional factor of the nonlinear state error feedback (NLSEF); β5 is the differential factor of the nonlinear state error feedback (NLSEF); T eADRC This is the torque compensation control variable.

[0083] In this embodiment, the half-shaft speed difference is used as the real-time monitoring object. An encoder tracks the half-shaft speed difference and inputs it as a state variable into the ADRC controller. ESO estimates the system states z1(k), z2(k) (z2(k) is the derivative of z1(k)) and the total disturbance z3(k) based on the input Δω, and compensates for the disturbance through feedforward. TD transmits the target signal Δω... tar Smoothing is performed to obtain ω1(k) and ω2(k) (ω2(k) is the derivative of ω1(k)) to avoid overshoot and oscillation. NLSEF calculates T based on the differences e1(k) and e2(k) between the estimated signals z1(k) and z2(k) and the transition signals ω1(k) and ω2(k). e0The torque compensation control quantity T is obtained by superimposing it with the feedforward compensation term of the total system disturbance z3(k). eADRC Then, the motor vector control system is used for control, which aims to suppress gear dynamic load and rapidly dampen the system's free vibration.

[0084] The PI-Res optimal harmonic current is injected in real time under steady-state and transient conditions. ADRC torque compensation is only triggered when the half-shaft speed difference is greater than the speed trigger value.

[0085] The process of obtaining the current motor temperature and, based on the current motor temperature and the final target torque command, solving for the optimal current operating point to obtain the dq-axis target current command is as follows:

[0086] Obtain the current motor temperature and current dq axis current, and use the interior point method to interpolate and look up the dq axis inductance and dq axis permanent magnet flux linkage to obtain the current dq axis inductance value and the current dq axis permanent magnet flux linkage value.

[0087] The motor speed is obtained. Based on the final target torque command, motor speed, current dq-axis inductance value, and current dq-axis permanent magnet flux linkage value, field weakening control is performed in combination with the maximum torque current ratio and the maximum torque voltage ratio to obtain the dq-axis target current command.

[0088] like Figure 3 As shown, the process of obtaining the motor rotor position angle and, based on the motor rotor position angle, current motor temperature, and dq-axis target current command, calculating the harmonic current amplitude and phase to obtain the dq-axis harmonic current injection amount is as follows:

[0089] Obtain the motor rotor position angle. Based on the target current command of the dq axis, the motor rotor position angle, and the current motor temperature, use the current-magnetic flux linkage mapping relationship to obtain the instantaneous magnetic flux linkage and the average magnetic flux linkage of the dq axis.

[0090] Subtracting the average flux linkage along the dq axis from the instantaneous flux linkage along the dq axis yields the harmonic flux linkage along the dq axis:

[0091]

[0092] Where, ψ dhar ψ is the d-axis harmonic flux linkage; qhar ψ is the q-axis harmonic flux linkage; dins ψ is the instantaneous magnetic flux linkage along the d-axis; qins ψ is the instantaneous magnetic flux linkage along the q-axis; dave ψ is the d-axis average flux linkage; qave m is the q-axis average flux linkage; -1 This represents the d-axis current-magnetic flux mapping relationship; n -1 This represents the q-axis current-magnetic flux mapping relationship; The target current command for the d-axis; For the q-axis target current command; θ e The rotor position angle is T; the current motor temperature is T.

[0093] Based on the dq-axis harmonic flux linkage, the motor rotor position angle, and the current motor temperature, the dq-axis harmonic current before filtering is obtained using the flux linkage-current mapping relationship.

[0094] The dq-axis harmonic current injection amount is obtained by filtering the unfiltered dq-axis harmonic current using a first-order high-pass filter:

[0095]

[0096] Among them, i dhar The d-axis harmonic current injection amount; filter[·] is a first-order high-pass filter; f is the d-axis harmonic current before filtering; -1 (·) represents the flux linkage-d-axis current mapping relationship; i qhar This refers to the amount of q-axis harmonic current injected. For the q-axis harmonic current before filtering; g -1 () represents the flux linkage-q-axis current mapping relationship; y[k] is the output signal of the first-order high-pass filter at time k; K com f is the filter gain; c T is the cutoff frequency; s y[k] is the sampling time; x[k] is the input signal of the first-order high-pass filter at time k; x[k-1] is the input signal of the first-order high-pass filter at time k-1; y[k-1] is the output signal of the first-order high-pass filter at time k-1.

[0097] The process of subtracting the final dq-axis target current command from the current dq-axis current and inputting the result into the current loop, followed by a Vector Pulse Width Modulation (SVPWM) algorithm, yields the pulse signal controlling the start and stop of the three-phase inverter IGBT switches. Specifically:

[0098] Get the current dq axis current and electric angular velocity;

[0099] The final target current command for the dq axis is subtracted from the current dq axis current to obtain the dq axis current difference signal;

[0100] Input the electrical angular velocity and the difference signal between the dq-axis currents into the current loop to obtain the first dq-axis voltage command;

[0101] Based on the current dq-axis current and electric angular velocity, voltage feedforward is performed to obtain the voltage feedforward result;

[0102] The first feedback result is obtained by subtracting the first dq axis voltage command from the voltage feedforward result.

[0103] Based on the first feedback result and the motor rotor position angle, perform inverse Park transformation to obtain the α-β axis voltage command at the current moment;

[0104] Based on the α-β axis voltage command at the current moment, the drive signal of the motor at the current moment is obtained by using the Vector Pulse Width Modulation (SVPWM) algorithm.

[0105] The current loop employs a PI-Res controller; the PI-Res controller comprises a current loop PI controller and a quasi-resonant controller connected in parallel.

[0106] The current loop PI controller is used to control the DC current component to obtain the DC voltage command component; the quasi-resonant controller is used to control the 6th harmonic current and the 12th harmonic current respectively to obtain the voltage command component corresponding to the 6th harmonic current and the voltage command component corresponding to the 12th harmonic current.

[0107] In this embodiment, based on the vector control system, the d-axis and q-axis current loop controller adopts a PI-Res controller, and an optimal harmonic current injection calculation module is added to solve the amplitude and phase of the dq-axis harmonic current injection online. The calculated dq-axis harmonic current injection is compensated into the current loop and superimposed with the dq-axis target current command output by the optimal current operating point algorithm to obtain the final dq-axis target current command. The difference between the final dq-axis target current command and the current dq-axis current is then input into the current loop PI-Res. By controlling the DC and AC components in the current in real time, the goal is to suppress electromagnetic torque fluctuations without affecting steady-state output performance.

[0108] The transfer function of the PI-Res controller is:

[0109]

[0110] Wherein, H(s) PIR For the transfer function of the PI-Res controller; K P K is the proportional gain coefficient. I K represents the integral gain coefficient; s represents the complex frequency variable; K represents the integral gain coefficient. R ω is the quasi-resonant gain coefficient; ω0 is the resonant frequency, i.e., the harmonic current frequency; ω c This is the cutoff frequency.

[0111] In this embodiment, the present invention employs a quasi-resonant controller connected in parallel with a d-axis and q-axis current loop PI controller to control the 6th and 12th harmonic currents, thereby suppressing the dominant 6th and 12th order electromagnetic torque fluctuation amplitudes.

Claims

1. A method for coordinated vibration suppression of an electric drive system that integrates harmonic optimization and torque compensation, characterized in that, include: The encoder is used to track the speed difference of the half-shaft of the electric drive system. The half-shaft speed difference is input as a state variable to the ADRC controller to obtain the torque compensation control quantity; the ADRC controller includes a tracking differentiator TD, an extended state observer ESO, and a nonlinear state error feedback NLSEF. The tracking differentiator TD is used to obtain the tracking signal after the reference signal transition and the differential of the tracking signal: in, for The tracking signal after the reference signal transition at time +1; for The tracking signal after the reference signal transition; This is the sampling step size; for The derivative of the tracking signal after the transition of the reference signal at any given moment; for The differential of the tracking signal after the transition of the reference signal at time +1; This is the fastest control synthesis function; For reference signal; The velocity factor; The filter factor; For intermediate control quantities; It is a symbolic function; For tracking error; The derivative of the tracking error; The extended state observer (ESO) is used to estimate the system state and total disturbance in real time based on the half-shaft speed difference, obtaining the state estimate of the half-shaft speed difference, the derivative of the state estimate of the half-shaft speed difference, and the estimate of the total system disturbance: in, for The state estimation error at each time step; for The state estimate of the half-shaft speed difference by the time-extended state observer (ESO); This is the difference in half-shaft speed; for The state estimate of the half-shaft speed difference by the extended state observer ESO at time +1; for The derivative of the state estimate of the half-shaft speed difference by the time-extended state observer ESO; , and All are error correction coefficients; It is a non-linear power function; , and All are nonlinear factors; The interval width of the nonlinear power function; for The derivative of the state estimate of the half-shaft speed difference by the extended state observer ESO at time +1; for The estimated value of the total system disturbance at any given time; The compensation factor for the extended state observer ESO; for The torque compensation control quantity output by the ADRC controller at any given time; for The estimated total disturbance of the system at time +1; The nonlinear state error feedback (NLSEF) is used to calculate the initial compensation torque based on the tracking signal after the reference signal transition, the derivative of the tracking signal, the state estimate of the half-shaft speed difference, the derivative of the state estimate of the half-shaft speed difference, and the estimate of the total system disturbance. The initial compensation torque is then combined with the feedforward compensation term to generate the torque compensation control quantity. : in, for time and The difference; for time and The difference; This is the initial compensation torque; and All are nonlinear factors; This is the scaling factor for the nonlinear state error feedback (NLSEF). The differentiating factor of the nonlinear state error feedback (NLSEF); This is the torque compensation control quantity; Obtain the original target torque command and superimpose the torque compensation control quantity with the original target torque command to obtain the final target torque command; Obtain the current motor temperature and current dq axis current, and solve for the optimal current operating point based on the current motor temperature, current dq axis current and final target torque command to obtain the target current command for the dq axis; Obtain the motor rotor position angle, and based on the motor rotor position angle, current motor temperature, and dq-axis target current command, calculate the harmonic current amplitude and phase to obtain the dq-axis harmonic current injection amount: Obtain the motor rotor position angle. Based on the target current command of the dq axis, the motor rotor position angle, and the current motor temperature, use the current-magnetic flux linkage mapping relationship to obtain the instantaneous magnetic flux linkage and the average magnetic flux linkage of the dq axis. Subtracting the average flux linkage along the dq axis from the instantaneous flux linkage along the dq axis yields the harmonic flux linkage along the dq axis: in, For d-axis harmonic flux linkage; For q-axis harmonic flux linkage; The instantaneous flux linkage along the d-axis; The instantaneous flux linkage along the q-axis; The average flux linkage along the d-axis; The q-axis average flux linkage; m -1 This represents the d-axis current-magnetic flux mapping relationship. n -1 This represents the q-axis current-magnetic flux mapping relationship; The target current command for the d-axis; This is the target current command for the q-axis. This refers to the rotor position angle of the motor; This refers to the current motor temperature. Based on the dq-axis harmonic flux linkage, the motor rotor position angle, and the current motor temperature, the dq-axis harmonic current before filtering is obtained using the flux linkage-current mapping relationship. The dq-axis harmonic current injection amount is obtained by filtering the unfiltered dq-axis harmonic current using a first-order high-pass filter: in, This refers to the amount of harmonic current injected along the d-axis. It is a first-order high-pass filter; This refers to the d-axis harmonic current before filtering. This represents the magnetic flux linkage-d-axis current mapping relationship. This refers to the amount of q-axis harmonic current injected. This refers to the q-axis harmonic current before filtering. This represents the magnetic flux linkage-q-axis current mapping relationship. A first-order high-pass filter Output signal at time; For filter gain; The cutoff frequency; Sampling time; A first-order high-pass filter The input signal at any given moment; A first-order high-pass filter The input signal at any given moment; A first-order high-pass filter Output signal at time; The dq-axis harmonic current injection amount is superimposed with the dq-axis target current command to obtain the final dq-axis target current command; The final target current command for the dq axis is subtracted from the current dq axis current and input into the current loop. The signal is then processed by the Vector Pulse Width Modulation (SVPWM) algorithm to obtain the motor drive signal at the current moment, thus completing the coordinated vibration suppression of the electric drive system.

2. The method for coordinated vibration suppression of an electric drive system integrating harmonic optimization and torque compensation according to claim 1, characterized in that, The process of obtaining the current motor temperature and, based on the current motor temperature and the final target torque command, solving for the optimal current operating point to obtain the dq-axis target current command is as follows: Obtain the current motor temperature and current dq axis current, and use the interior point method to interpolate and look up the dq axis inductance and dq axis permanent magnet flux linkage to obtain the current dq axis inductance value and the current dq axis permanent magnet flux linkage value. The motor speed is obtained. Based on the final target torque command, motor speed, current dq-axis inductance value, and current dq-axis permanent magnet flux linkage value, field weakening control is performed in combination with the maximum torque current ratio and the maximum torque voltage ratio to obtain the dq-axis target current command.

3. The method for coordinated vibration suppression of an electric drive system integrating harmonic optimization and torque compensation according to claim 1, characterized in that, The process of subtracting the final dq-axis target current command from the current dq-axis current and inputting the result into the current loop, followed by a Vector Pulse Width Modulation (SVPWM) algorithm, yields the pulse signal controlling the start and stop of the three-phase inverter IGBT switches. Specifically: Get the current dq axis current and electric angular velocity; The final target current command for the dq axis is subtracted from the current dq axis current to obtain the dq axis current difference signal; Input the electrical angular velocity and the difference signal between the dq-axis currents into the current loop to obtain the first dq-axis voltage command; Based on the current dq-axis current and electric angular velocity, voltage feedforward is performed to obtain the voltage feedforward result; The first feedback result is obtained by subtracting the first dq axis voltage command from the voltage feedforward result. Based on the first feedback result and the motor rotor position angle, perform inverse Park transformation to obtain the α-β axis voltage command at the current moment; Based on the α-β axis voltage command at the current moment, the drive signal of the motor at the current moment is obtained by using the Vector Pulse Width Modulation (SVPWM) algorithm.

4. The method for coordinated vibration suppression of an electric drive system integrating harmonic optimization and torque compensation according to claim 3, characterized in that, The current loop employs a PI-Res controller; the PI-Res controller comprises a current loop PI controller and a quasi-resonant controller connected in parallel.

5. The method for coordinated vibration suppression of an electric drive system integrating harmonic optimization and torque compensation according to claim 4, characterized in that, The current loop PI controller is used to control the DC current component to obtain the DC voltage command component; the quasi-resonant controller is used to control the 6th harmonic current and the 12th harmonic current respectively to obtain the voltage command component corresponding to the 6th harmonic current and the voltage command component corresponding to the 12th harmonic current.

6. The method for coordinated vibration suppression of an electric drive system integrating harmonic optimization and torque compensation according to claim 5, characterized in that, The transfer function of the PI-Res controller is: in, For the transfer function of the PI-Res controller; This is the proportional gain coefficient; This is the integral gain coefficient; It is a complex frequency variable; The quasi-resonant gain coefficient; This is the resonant frequency, i.e., the harmonic current frequency; This is the cutoff frequency.

Images