A method for suppressing compound disturbances in a permanent magnet synchronous motor steer-by-wire system

Abstract

This invention discloses a method for suppressing complex disturbances in a permanent magnet synchronous motor-controlled steering system. The method includes: designing a non-singular fast terminal sliding mode angle tracking controller; the controller receives the desired steering angle from an external input; subtracts the desired steering angle from the actual steering angle to obtain the difference value; differentiates the difference value to obtain the derivative value; adjusting the controller parameters to ensure that the difference value and its derivative converge to zero, thereby obtaining the output current; and designing a complex disturbance suppression method based on a spatial domain iterative learning algorithm with a forgetting factor and a linearly extended state observer, which compensates for time-varying periodic torque pulsations and aperiodic disturbances by providing current to the output current of the angle tracking controller, and inputting this current into the steering motor to provide an ideal steering angle. This method effectively suppresses various disturbances in the steering system, exhibiting better angle tracking performance and robustness, and therefore has broad market prospects.

Description

Technical Field

[0001] This invention belongs to the field of vehicle steer-by-wire control, and specifically relates to a method for suppressing complex disturbances in a permanent magnet synchronous motor steer-by-wire system. Background Technology

[0002] With the development trends of vehicle intelligence, electrification, and connectivity, the shortcomings of traditional mechanical steering systems and power steering systems have become increasingly prominent, including poor safety and a poor driving experience. The emergence of steer-by-wire (SbW) technology has solved many of these shortcomings and has become a popular application system in automobiles today; however, the permanent magnet synchronous motor (PMSM) used as the actuator of steer-by-wire will generate torque pulsation, resulting in a deterioration in the tracking performance and control stability of steer-by-wire.

[0003] Current methods for suppressing disturbances in steer-by-wire systems mainly include the observer method and robust control method. The observer method uses a disturbance observer to monitor frictional disturbances, tire return torque, and unknown random disturbances experienced by the steer-by-wire system. The robust control method marks the maximum value of the disturbance and suppresses it by modifying the control gain. While existing methods can suppress non-periodic disturbances such as frictional disturbances, tire return torque, and unknown random disturbances, they cannot suppress the time-varying periodic torque pulsations within the permanent magnet synchronous motor.

[0004] Studies have shown that time-domain iterative learning can effectively suppress periodic torque ripples generated inside a motor under a fixed period. However, the period of torque ripples in permanent magnet synchronous motors (PMSMs) of steer-by-wire systems is constantly changing, which is not conducive to the application of time-domain iterative learning. Therefore, the combined disturbances between the disturbances introduced by the PMSMs to the steer-by-wire system and the disturbances inherent in the steer-by-wire system itself have not been effectively resolved. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method for suppressing compound disturbances in a permanent magnet synchronous motor steer-by-wire system, so as to solve the problem of deterioration in control accuracy and system stability caused by compound disturbances in the prior art.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The present invention provides a method for suppressing complex disturbances in a permanent magnet synchronous motor steer-by-wire system. The steer-by-wire system includes: an angle controller, an inverter, a steering motor, an angle sensor, a steering gear, a steering rack, a steering tie rod, and a front wheel.

[0008] The steering angle controller is electrically connected to the inverter and sends control commands to the inverter; the inverter is electrically connected to the steering motor, which generates three-phase current to drive the steering motor, which is a permanent magnet synchronous motor; the output end of the steering motor is mechanically connected to the steering gear, with the connection method being a shaft and key fit; the steering gear meshes with the steering rack, and the steering rack is mechanically connected to one end of the steering tie rod, with the connection method being a threaded connection, and the other end of the steering tie rod is mechanically connected to the front wheel, with the connection method being a threaded connection; the steering tie rod drives the front wheel to generate a steering angle; the steering angle sensor is installed between the output shaft of the steering motor and the steering gear, and outputs the motor steering angle signal; the steering angle sensor is electrically connected to the steering angle controller and transmits the steering angle signal to the steering angle controller through a signal line.

[0009] The method steps are as follows:

[0010] 1) The steering angle controller receives the vehicle's desired steering angle from an external input, and subtracts the desired steering angle from the vehicle's actual steering angle to obtain the difference between the vehicle's desired steering angle and the vehicle's actual steering angle.

[0011] 2) Establish an uncertainty model for the steer-by-wire system that includes non-periodic disturbances and time-varying periodic torque pulsations. Design an error dynamic equation based on the uncertainty model of the steer-by-wire system. Input the difference between the desired steering angle and the actual steering angle of the vehicle and its differential value into the error dynamic equation. Design a non-singular fast terminal sliding surface. Design a non-singular fast terminal sliding mode (NFTSM) angle tracking controller based on the non-singular fast terminal sliding surface and the error dynamic equation. By adjusting the parameters of the non-singular fast terminal sliding mode angle tracking controller, ensure that the difference between the desired steering angle and the actual steering angle of the vehicle and its differential value converge to 0. The output current is calculated by the non-singular fast terminal sliding mode angle tracking controller.

[0012] 3) Design a linear extended state observer to differentiate the actual steering angle of the vehicle collected by the steering angle sensor to obtain the differential value of the actual steering angle. Input the actual steering angle of the vehicle, the differential value of the actual steering angle, and the output current of the non-singular fast terminal sliding mode steering angle tracking controller to the linear extended state observer. After calculation, the estimated value of the steering angle, the estimated value of the differential value of the steering angle, and the estimated value of the non-periodic disturbance are obtained. Input the above estimated values ​​to the non-singular fast terminal sliding mode steering angle tracking controller to obtain the compensation current for the non-periodic disturbance.

[0013] 4) Design a torque observer, input the steering motor speed and the output current of the non-singular fast terminal sliding mode angle tracking controller to the torque observer, and calculate the observed value of the steering motor output torque;

[0014] 5) Combine the torque observation value from the torque observer with the actual q-axis current i of the steering motor. q The output current of the non-singular fast terminal sliding mode corner tracking controller is substituted into the spatial domain iterative learning algorithm with a forgetting factor to calculate the compensation current that suppresses time-varying periodic torque pulsation.

[0015] 6) The output current of the non-singular fast terminal sliding mode angle tracking controller is added to the compensation current of non-periodic disturbance and time-varying periodic torque pulsation to obtain the steering motor input current. The steering motor is driven by the input current to provide the vehicle with an ideal steering angle.

[0016] Further, step 1) specifically includes:

[0017] The desired steering angle of the vehicle is provided by the driver, and the actual steering angle is obtained by the steering angle sensor. The difference e1 between the desired steering angle and the actual steering angle is:

[0018]

[0019] In the formula, δ represents the desired steering angle. f This is the actual steering angle.

[0020] Further, the uncertainty model of the steer-by-wire system in step 2) includes aperiodic disturbances and time-varying periodic torque pulsations. The aperiodic disturbances, composed of unknown model parameter mismatch disturbances, unknown modeling dynamic disturbances, tire self-centering torque, and other external unknown disturbances, are modeled as bounded uncertainty disturbances. The time-varying periodic torque pulsations, composed of cogging torque, flux harmonic torque, inverter nonlinear torque, and current sampling error disturbances, are modeled as bounded time-varying periodic disturbances, thus obtaining the uncertainty model of the steer-by-wire system. Based on the uncertainty model of the steer-by-wire system, the error dynamic equation is obtained, and the step... In step 1), the difference between the desired steering angle and the actual steering angle of the vehicle is differentiated to obtain the differential value of the steering angle difference. The difference between the desired steering angle and the actual steering angle of the vehicle and its differential value are substituted into the error dynamic equation to design a non-singular fast terminal sliding surface. Based on the non-singular fast terminal sliding surface and the error dynamic equation, a non-singular fast terminal sliding angle controller is designed. By adjusting the parameters of the non-singular fast terminal sliding angle controller, the difference between the desired steering angle and the actual steering angle of the vehicle and its differential value are converged to 0. The input current required by the angle controller is calculated.

[0021] Further, step 2) specifically includes:

[0022] 21) The non-periodic disturbance composed of unknown model parameter mismatch disturbance, unknown modeling dynamic disturbance, tire self-aligning torque and other external unknown disturbances is modeled as a bounded uncertainty disturbance, as follows:

[0023]

[0024]

[0025]

[0026]

[0027] In the formula, t is time, and δ f τ is the front wheel steering angle. f Let τ be the Coulomb friction torque. a τ is the tire return torque. d For non-periodic torques generated by uncertain model parameters, unmodeled dynamics, and other external random disturbances, F f Let t be the Coulomb friction coefficient. p The aerodynamic trajectory (the distance between the wheel center and the point of application of the lateral force), t m C represents the mechanical trajectory (the distance between the wheel center and the road surface point where the wheel pivot is located). αf β is the front wheel slip stiffness coefficient, α is the slip angle, a is the distance from the front axle to the center of gravity, r is the yaw rate at the center, V is the longitudinal velocity component of the center of gravity, and F is the front wheel slip stiffness coefficient. d It is a non-periodic perturbation function;

[0028] 22) The time-varying periodic torque pulsation, composed of cogging torque, flux harmonic torque, inverter nonlinear torque, and current sampling error disturbance, is modeled as a bounded time-varying periodic disturbance, as follows:

[0029]

[0030]

[0031]

[0032]

[0033] In the formula, T cog For cogging torque; i smag T represents the stator current amplitude. n N represents the Fourier coefficients; s θ is the least common multiple of the number of tooth indentations and the number of pole pairs; n is a positive integer; θ m T is the rotor angle; ψf T is the magnetic flux harmonic torque. 6kω represents the harmonic coefficients of the magnetic flux for each order, k = 1, 2, 3, ..., n; e Electric angular velocity; Δu α and Δu β θ represents the inverter nonlinear voltages along the α and β axes, respectively; Δu is the inverter voltage value; θ e ΔT is the electrical angle of the motor; offset For current bias error disturbance; ΔT scalig For current scaling error perturbation; K t This is the current error coefficient; For the offset phase; Δi as and Δi bs These are the measured current offsets for phases a and b, respectively; the measured current scaling factors for phases a and b are respectively expressed as K. a and K b t represents time;

[0034] 23) The uncertainty model of the steer-by-wire system is described as follows:

[0035]

[0036] In the formula, J is the moment of inertia of the steering system; B is the damping coefficient of the steering system; τ m κ represents the output torque of the steering motor; κ is the transmission coefficient of the steering system.

[0037] make For non-periodic torque disturbances, we get:

[0038]

[0039] In the formula, f a This is a non-periodic total disturbance;

[0040] The output torque τ of the steering motor m Represented as:

[0041]

[0042] In the formula, P n ψ is the number of pole pairs of the steering motor; f i is the flux linkage value of the steering motor. q τ is the q-axis current of the steering motor. ripple It is a periodic torque pulsation;

[0043] The electromagnetic output torque containing periodic torque pulsations is expressed as:

[0044]

[0045] In the formula, ψ dk Here are the flux linkage values ​​for each order; Δi offset , Δiscaling , Δi inverter These are the current errors caused by the offset and scaling errors of the current sensor and the nonlinearity of the inverter, respectively.

[0046] The output torque of the steering motor consists of a DC component and harmonic components of the 1st, 2nd, 6th, 12th, and multiples of 6th orders. The output torque of the steering motor can be simplified as follows:

[0047]

[0048] In the formula, τ ek and φ k These are the amplitude and phase angle of the k-th harmonic component, respectively;

[0049] 24) Establish a steer-by-wire angle tracking controller;

[0050] Due to the compensation effect of the linearly extended state observer, the error between the vehicle's expected steering angle and its actual steering angle is expressed as:

[0051]

[0052] In the formula, δ represents the vehicle's desired steering angle. f This refers to the vehicle's actual steering angle. These are the observed values ​​of the steering angle; The observed value is the time derivative of the steering angle;

[0053] The error dynamic equation is obtained as follows:

[0054]

[0055] In the formula, t1 = [e1, e2] Τ ; Linear extended state observer observes disturbances D represents the maximum disturbance value; u represents the control input. Let be the expected value of the second time derivative of the steering angle;

[0056] The non-singular fast terminal sliding surface is designed as follows:

[0057]

[0058] In the formula, γ, p, and q are the coefficients of the non-singular fast terminal sliding mode controller; γ > 0; p and q are positive odd numbers, and p > q;

[0059] The output current u of the non-singular fast terminal sliding mode angle controller is:

[0060]

[0061] In the formula, is the second derivative of the desired steering angle; 1 < p / q < 2; D is the controller parameter; η is the coefficient of the non-singular fast terminal sliding mode controller, η > 0; sgn is the sign function.

[0062] Further, in step 3), the steering angle sensor measures the actual steering angle of the vehicle, and differentiates it to obtain the differential value of the actual steering angle. A linear extended state observer is designed, and the actual steering angle of the vehicle, the differential value of the actual steering angle, and the output current of the non-singular fast terminal sliding mode steering angle tracking controller are used as inputs to the linear extended state observer. By adjusting the observer bandwidth, the observation error is ensured to converge to 0. The linear extended state observer outputs the estimated value of the steering angle, the estimated value of the differential value of the steering angle, and the estimated value of the aperiodic disturbance. The steering angle, the differential value of the steering angle, and the aperiodic disturbance are input into the error dynamic equation, and the compensation current for the aperiodic disturbance is obtained using the non-singular fast terminal sliding mode steering angle controller.

[0063] Furthermore, step 3) specifically includes:

[0064] A linear extended state observer (ESO) is designed to compensate for non-periodic disturbances in the uncertainty model of the steer-by-wire system. The design of the linear extended state observer specifically involves:

[0065] Let the non-periodic torque disturbance f a Continuous and differential, linearly extended state observer extends the aperiodic torque disturbance f a As the new state variable x3; let For different state variables, b = κr / J;

[0066] Described as:

[0067]

[0068] The linearly extended state observer is designed as follows:

[0069]

[0070] In the formula, For state variable x i The estimated values, i = 1, 2, 3; β 01 ,β 02 and β 03 These are the parameters to be designed;

[0071] make This represents the estimation error of the state variables, combined with the formula

[0072] get:

[0073]

[0074] In the formula,

[0075] Let ω0 be the bandwidth of the linearly extended state observer, and choose the characteristic polynomial (s+ω0). 3 =s 3 +β 01 s 2 +β 02 s+β 03 , represented as β 01 =3ω0, Ensure that matrix A is a Hurwitz matrix;

[0076] For the estimated value of non-periodic torque disturbance Compensation current for non-singular fast terminal sliding mode for:

[0077]

[0078] Further, in step 4), the state equation of the permanent magnet synchronous motor is constructed, and a torque observer is designed. The torque observer receives the differential value of the steering angle measured by the angle sensor and the output current of the non-singular fast terminal sliding mode angle tracking controller, and calculates the observed value of the motor output torque. The observed value of the motor output torque is then input into the iterative learning calculation to calculate the iterative learning tracking error.

[0079] Furthermore, step 4) specifically includes:

[0080] Design a torque observer to obtain torque observation values; construct the state equations of the permanent magnet synchronous motor:

[0081]

[0082] In the formula, B m =[1 / J m 0] Τ C m =[1 0];x′=[ω m T e ] Τ For state variables; y′=ω m T represents the motor speed. e J is the output torque of the motor. m b is the moment of inertia of the motor. m A is the motor damping coefficient; m B mWith C m represents the coefficients of the state equation; u′ represents the observer input;

[0083] The torque observer is represented as:

[0084]

[0085] In the formula, the desired state variable is The actual control output is y′=C m x′; the desired control output is K1 = [k1 k2] Τ And K2 = [k3 k4] Τ Here, k is the feedback matrix; k1, k2, k3, and k4 are the torque observer coefficients;

[0086] Depend on and We can obtain:

[0087]

[0088] In the formula, This is the observation error; The differential value of the observation error; I is the identity matrix; s is the differential operator;

[0089] The characteristic equation of the observer is expressed as:

[0090]

[0091] In the formula, λ is the independent variable of the characteristic equation; det represents the product of the eigenvalues ​​of the matrix corresponding to the linear transformation of the square matrix.

[0092] Based on the desired pole σ, and the expected observer expression Then we can obtain:

[0093]

[0094] Assume b m =0, and design k1=k3=0, then the state feedback coefficient can be obtained as:

[0095]

[0096] The observed torque value of the permanent magnet synchronous motor can be obtained as follows:

[0097]

[0098] Received That is, the torque observation value T m .

[0099] Further, in step 5), the uncertainty model of the steer-by-wire system is simplified to an approximate linear integral model, and the state space of the approximate linear integral model is obtained. The state space of the approximate linear integral model is rewritten as an iterative learning state space. Based on the iterative learning state space, a spatial domain iterative learning algorithm with a forgetting factor is designed. The input of the algorithm includes the output of the previous iteration and the iterative learning tracking error of the previous iteration. The iterative learning tracking error is the difference between the actual torque value and the observed torque value. The learning law of the spatial domain iterative learning algorithm with a forgetting factor is designed. The learning law is the compensation current of the permanent magnet synchronous motor for periodic torque pulsation.

[0100] Furthermore, step 5) specifically includes:

[0101] Designing iterative learning with a forgetting factor: Choosing the bandwidth value to make... make Uncertainty model of steer-by-wire system Converting to an approximate linear integral model, the state space is obtained as follows:

[0102]

[0103] In the formula:

[0104] State space of approximate linear integral model Rewritten:

[0105]

[0106] In the formula, x k and y k These represent the system states in the k-th iteration; u k This is the input for the k-th iteration; k is the iteration number.

[0107] The spatial domain iterative learning algorithm with a forgetting factor is represented as:

[0108]

[0109] In the formula: Γ and Φ are the proportional learning gain and differential learning gain, respectively; ν is the forgetting factor; It is bounded; The error is the (k-1)th iteration error of the motor output torque; each error is stored in memory during the iteration process; T m The torque observation value obtained from the torque observer in step 4);

[0110] The output of iterative learning is the q-axis compensation current Δi of the permanent magnet synchronous motor. q,k Then the learning law is expressed as:

[0111]

[0112] The iterative learning tracking error is represented as:

[0113]

[0114] In the formula, e k-1 The torque error is the value after the (k-1)th iteration; i q,k-1 This is the compensation current after the (k-1)th iteration.

[0115] Furthermore, in step 6), the ideal input current of the steering motor is the sum of the control output current of the non-singular fast terminal sliding mode steering angle controller, the compensation current output by the spatial domain iterative learning algorithm with forgetting factor, and the compensation current for aperiodic disturbances. The compensation current output by the spatial domain iterative learning algorithm with forgetting factor includes the compensation current for periodic torque pulsations. The sum of the compensation current for aperiodic disturbances and the compensation current of the spatial domain iterative learning algorithm with forgetting factor is the compensation for the composite disturbance of aperiodic disturbances and periodic torque pulsations. At this time, the steering motor converts the ideal input current into electromagnetic torque to provide steering power for the steer-by-wire system. The rotation of the steering motor drives the steering gear, and the rotation of the steering gear causes the steering rack to move laterally, which in turn moves the steering tie rod, thereby driving the front wheels to perform steering work.

[0116] Furthermore, step 6) specifically includes:

[0117] Ideal input current for steering motor This is represented as the control input u in step 2) and the non-periodic torque ripple compensation motor in step 3). And the q-axis compensation current Δi output by iterative learning in step 5) q,k Add:

[0118]

[0119] According to the dynamic equations of the steer-by-wire system, the steering motor provides steering power to the steer-by-wire system, and the dynamic equations of the steer-by-wire system are expressed as follows:

[0120]

[0121] In the formula, f a For non-periodic torque disturbance; τ ripple It is a periodic torque pulsation.

[0122] To address the time-varying periodic torque ripples and aperiodic disturbances in steer-by-wire systems, this invention proposes a composite disturbance suppression strategy combining a spatial domain iterative learning algorithm with a forgetting factor and a linearly extended state observer, along with a non-singular fast end sliding mode angle tracking strategy. The composite disturbance suppression strategy fully leverages the advantages of both approaches. The linearly extended state observer observes aperiodic disturbances and compensates for them in the non-singular fast end sliding mode controller, improving accuracy and robustness. The spatial domain iterative learning algorithm with a forgetting factor can better compensate for time-varying periodic torque ripples compared to the time domain iterative learning algorithm, and the forgetting factor enhances the system's robustness to noise and initial errors. The addition of the linearly extended state observer compensates for the spatial domain forgetting factor iterative learning algorithm's inability to suppress aperiodic disturbances.

[0123] The beneficial effects of this invention are:

[0124] The steer-by-wire system composite disturbance suppression method and angle tracking control of the present invention can effectively suppress composite steering disturbances, and have better angle tracking performance and robustness under composite disturbances. It can minimize the driver's operating burden and the vehicle's handling stability, and achieve safety and handling performance control of steer-by-wire vehicles. Therefore, it has broad market application prospects. Attached Figure Description

[0125] Figure 1 This is a structural diagram of the steer-by-wire system in this invention.

[0126] Figure 2 This is a structural diagram of the composite disturbance suppression method for the steer-by-wire system in this invention.

[0127] Figure 3 This is a tracking performance diagram of the method of the present invention when the desired steering angle is a sinusoidal signal.

[0128] Figure 4 This is a tracking performance diagram of the method of the present invention at a desired steering angle of a swept frequency signal.

[0129] Figure 5 This is a tracking performance diagram of the method of the present invention when the desired steering angle is a ramp signal. Detailed Implementation

[0130] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0131] This invention discloses a method for suppressing complex disturbances in a permanent magnet synchronous motor steer-by-wire system. The steer-by-wire system includes: an angle controller, an inverter, a steering motor, an angle sensor, a steering gear, a steering rack, a steering tie rod, and a front wheel. Figure 1 As shown;

[0132] The steering angle controller is electrically connected to the inverter and sends control commands to the inverter; the inverter is electrically connected to the steering motor, which generates three-phase current to drive the steering motor, which is a permanent magnet synchronous motor; the output end of the steering motor is mechanically connected to the steering gear, with the connection method being a shaft and key fit; the steering gear meshes with the steering rack, and the steering rack is mechanically connected to one end of the steering tie rod, with the connection method being a threaded connection, and the other end of the steering tie rod is mechanically connected to the front wheel, with the connection method being a threaded connection; the steering tie rod drives the front wheel to generate a steering angle; the steering angle sensor is installed between the output shaft of the steering motor and the steering gear, and outputs the motor steering angle signal; the steering angle sensor is electrically connected to the steering angle controller and transmits the steering angle signal to the steering angle controller through a signal line.

[0133] The method steps are as follows:

[0134] 1) The steering angle controller receives the desired steering angle of the vehicle from external input, and subtracts the desired steering angle from the actual steering angle of the vehicle to obtain the difference between the desired steering angle and the actual steering angle of the vehicle; specifically including:

[0135] The desired steering angle of the vehicle is provided by the driver, and the actual steering angle is obtained by the steering angle sensor. The difference e1 between the desired steering angle and the actual steering angle is:

[0136]

[0137] In the formula, δ represents the desired steering angle. f This is the actual steering angle.

[0138] 2) Establish an uncertainty model for the steer-by-wire system that includes non-periodic disturbances and time-varying periodic torque pulsations. Design an error dynamic equation based on the uncertainty model of the steer-by-wire system. Input the difference between the desired steering angle and the actual steering angle of the vehicle and its differential value into the error dynamic equation. Design a non-singular fast terminal sliding surface. Design a non-singular fast terminal sliding mode (NFTSM) angle tracking controller based on the non-singular fast terminal sliding surface and the error dynamic equation. By adjusting the parameters of the non-singular fast terminal sliding mode angle tracking controller, ensure that the difference between the desired steering angle and the actual steering angle of the vehicle and its differential value converge to 0. The output current is calculated by the non-singular fast terminal sliding mode angle tracking controller.

[0139] The uncertainty model of the steer-by-wire system includes aperiodic disturbances and time-varying periodic torque pulsations. The aperiodic disturbance, composed of unknown model parameter mismatch disturbances, unknown modeling dynamic disturbances, tire self-centering torque, and other external unknown disturbances, is modeled as a bounded uncertainty disturbance. The time-varying periodic torque pulsation, composed of cogging torque, flux harmonic torque, inverter nonlinear torque, and current sampling error disturbances, is modeled as a bounded time-varying periodic disturbance, thus obtaining the uncertainty model of the steer-by-wire system. Based on the uncertainty model of the steer-by-wire system, the error dynamic equation is obtained, and the vehicle's expected value in step 1) is... The difference between the desired steering angle and the actual steering angle of the vehicle is differentiated to obtain the differential value of the steering angle difference. This differential value is then substituted into the error dynamic equation to design a non-singular fast-end sliding surface. Based on the non-singular fast-end sliding surface and the error dynamic equation, a non-singular fast-end sliding angle controller is designed. By adjusting the parameters of the non-singular fast-end sliding angle controller, the difference between the desired steering angle and the actual steering angle of the vehicle, along with its differential value, converges to zero. The required input current for the angle controller is then calculated. Specifically, this includes:

[0140] 21) The non-periodic disturbance composed of unknown model parameter mismatch disturbance, unknown modeling dynamic disturbance, tire self-aligning torque and other external unknown disturbances is modeled as a bounded uncertainty disturbance, as follows:

[0141]

[0142]

[0143]

[0144]

[0145] In the formula, t is time, and δ f τ is the front wheel steering angle. f Let τ be the Coulomb friction torque. a τ is the tire return torque. d For non-periodic torques generated by uncertain model parameters, unmodeled dynamics, and other external random disturbances, F f Let t be the Coulomb friction coefficient. p The aerodynamic trajectory (the distance between the wheel center and the point of application of the lateral force), t m C represents the mechanical trajectory (the distance between the wheel center and the road surface point where the wheel pivot is located). αf β is the front wheel slip stiffness coefficient, α is the slip angle, a is the distance from the front axle to the center of gravity, r is the yaw rate at the center, V is the longitudinal velocity component of the center of gravity, and F is the front wheel slip stiffness coefficient. d It is a non-periodic perturbation function;

[0146] 22) The time-varying periodic torque pulsation, composed of cogging torque, flux harmonic torque, inverter nonlinear torque, and current sampling error disturbance, is modeled as a bounded time-varying periodic disturbance, as follows:

[0147]

[0148]

[0149]

[0150]

[0151] In the formula, T cog For cogging torque; i smag T represents the stator current amplitude. n N represents the Fourier coefficients; s θ is the least common multiple of the number of tooth indentations and the number of pole pairs; n is a positive integer; θ m T is the rotor angle; ψf T is the magnetic flux harmonic torque. 6k ω represents the harmonic coefficients of the magnetic flux for each order, k = 1, 2, 3, ..., n; e Electric angular velocity; Δu α and Δu β θ represents the inverter nonlinear voltages along the α and β axes, respectively; Δu is the inverter voltage value; θ e ΔT is the electrical angle of the motor; offset For current bias error disturbance; ΔT scalig For current scaling error perturbation; K t This is the current error coefficient; For the offset phase; Δi as and Δi bs These are the measured current offsets for phases a and b, respectively; the measured current scaling factors for phases a and b are respectively expressed as K. a and K b t represents time;

[0152] 23) The uncertainty model of the steer-by-wire system is described as follows:

[0153]

[0154] In the formula, J is the moment of inertia of the steering system; B is the damping coefficient of the steering system; τ m κ represents the output torque of the steering motor; κ is the transmission coefficient of the steering system.

[0155] make For non-periodic torque disturbances, we get:

[0156]

[0157] In the formula, f a This is a non-periodic total disturbance;

[0158] The output torque τ of the steering motor m Represented as:

[0159]

[0160] In the formula, P n ψ is the number of pole pairs of the steering motor; f i is the flux linkage value of the steering motor. q τ is the q-axis current of the steering motor. ripple It is a periodic torque pulsation;

[0161] The electromagnetic output torque containing periodic torque pulsations is expressed as:

[0162]

[0163] In the formula, ψ dk Here are the flux linkage values ​​for each order; Δi offset , Δi scaling , Δi inverter These are the current errors caused by the offset and scaling errors of the current sensor and the nonlinearity of the inverter, respectively.

[0164] The output torque of the steering motor consists of a DC component and harmonic components of the 1st, 2nd, 6th, 12th, and multiples of 6th orders. The output torque of the steering motor can be simplified as follows:

[0165]

[0166] In the formula, τ ek and φ k These are the amplitude and phase angle of the k-th harmonic component, respectively;

[0167] 24) Establish a steer-by-wire angle tracking controller;

[0168] Due to the compensation effect of the linearly extended state observer, the error between the vehicle's expected steering angle and its actual steering angle is expressed as:

[0169]

[0170] In the formula, δ represents the vehicle's desired steering angle. f This refers to the vehicle's actual steering angle. These are the observed values ​​of the steering angle; The observed value is the time derivative of the steering angle;

[0171] The error dynamic equation is obtained as follows:

[0172]

[0173] In the formula, t1 = [e1, e2] Τ ; Linear extended state observer observes disturbances D represents the maximum disturbance value; u represents the control input. Let be the expected value of the second time derivative of the steering angle;

[0174] The non-singular fast terminal sliding surface is designed as follows:

[0175]

[0176] In the formula, γ, p, and q are the coefficients of the non-singular fast terminal sliding mode controller; γ > 0; p and q are positive odd numbers, and p > q;

[0177] The output current u of the non-singular fast terminal sliding mode angle controller is:

[0178]

[0179] In the formula, is the second derivative of the desired steering angle; 1 < p / q < 2; D is the controller parameter; η is the coefficient of the non-singular fast terminal sliding mode controller, η > 0; sgn is the sign function.

[0180] 3) Design a linear extended state observer to differentiate the actual steering angle of the vehicle collected by the steering angle sensor to obtain the differential value of the actual steering angle. Input the actual steering angle of the vehicle, the differential value of the actual steering angle, and the output current of the non-singular fast terminal sliding mode steering angle tracking controller to the linear extended state observer. After calculation, the estimated value of the steering angle, the estimated value of the differential value of the steering angle, and the estimated value of the non-periodic disturbance are obtained. Input the above estimated values ​​to the non-singular fast terminal sliding mode steering angle tracking controller to obtain the compensation current for the non-periodic disturbance.

[0181] The steering angle sensor measures the actual steering angle of the vehicle. Differentializing this angle yields the derivative value. A linear extended state observer is designed, using the actual steering angle, the derivative value of the actual steering angle, and the output current of the non-singular fast terminal sliding mode steering angle tracking controller as inputs. The observer bandwidth is adjusted to ensure the observation error converges to zero. The linear extended state observer outputs estimated steering angle, estimated value of the derivative of the steering angle, and estimated value of the aperiodic disturbance. These values ​​are then input into the error dynamic equation. The non-singular fast terminal sliding mode steering angle controller is used to obtain the compensation current for the aperiodic disturbance. Specifically, this includes:

[0182] A linear extended state observer (ESO) is designed to compensate for non-periodic disturbances in the uncertainty model of the steer-by-wire system. The design of the linear extended state observer specifically involves:

[0183] Let the non-periodic torque disturbance f a Continuous and differential, linearly extended state observer extends the aperiodic torque disturbance f a As the new state variable x3; let For different state variables, b = κr / J;

[0184] Described as:

[0185]

[0186] The linearly extended state observer is designed as follows:

[0187]

[0188] In the formula, For state variable x i The estimated values, i = 1, 2, 3; β 01 ,β 02 and β 03 These are the parameters to be designed;

[0189] make This represents the estimation error of the state variables, combined with the formula with formula get:

[0190]

[0191] In the formula,

[0192] Let ω0 be the bandwidth of the linearly extended state observer, and choose the characteristic polynomial (s+ω0). 3 =s 3 +β 01 s 2 +β 02 s+β 03 , represented as β 01 =3ω0, Ensure that matrix A is a Hurwitz matrix;

[0193] For the estimated value of non-periodic torque disturbance Compensation current for non-singular fast terminal sliding mode for:

[0194]

[0195] 4) Design a torque observer, input the steering motor speed and the output current of the nonsingular fast terminal sliding mode angle tracking controller into the torque observer, and calculate the observed value of the output torque of the steering motor;

[0196] Construct the state equation of the permanent magnet synchronous motor and design a torque observer. The torque observer receives the differential value of the steering angle measured by the angle sensor and the output current of the nonsingular fast terminal sliding mode angle tracking controller, and calculates the observed value of the motor output torque; and input the observed value of the motor output torque into the iterative learning to calculate the iterative learning tracking error. Specifically, it includes:

[0197] Design a torque observer to obtain the observed torque value; construct the state equation of the permanent magnet synchronous motor:

[0198]

[0199] In the formula, B m =[1 / J m 0] Τ ; C m =[1 0]; x′ = [ω m T e Τ is the state variable; y′ = ω m is the motor speed; T e is the motor output torque; J m is the motor inertia; b m is the motor damping coefficient; A m , B m and C m are the coefficients of the state equation; u′ is the observer input;

[0200] The torque observer is expressed as:

[0201]

[0202] In the formula, the desired state variable is The actual control output is y′ = C m x′; the desired control output is K1 = [k1 k2] Τ and K2 = [k3 k4] Τ are the feedback matrices; k1, k2, k3, and k4 are the torque observer coefficients;

[0203] From and it can be obtained:

[0204] ​

[0205] In the formula, This is the observation error; The differential value of the observation error; I is the identity matrix; s is the differential operator;

[0206] The characteristic equation of the observer is expressed as:

[0207]

[0208] In the formula, λ is the independent variable of the characteristic equation; det represents the product of the eigenvalues ​​of the matrix corresponding to the linear transformation of the square matrix.

[0209] Based on the desired pole σ, and the expected observer expression Then we can obtain:

[0210]

[0211] Assume b m =0, and design k1=k3=0, then the state feedback coefficient can be obtained as:

[0212]

[0213] The observed torque value of the permanent magnet synchronous motor can be obtained as follows:

[0214]

[0215] Received That is, the torque observation value T m .

[0216] 5) Combine the torque observation value from the torque observer with the actual q-axis current i of the steering motor. q The output current of the non-singular fast terminal sliding mode corner tracking controller is substituted into the spatial domain iterative learning algorithm with a forgetting factor to calculate the compensation current that suppresses time-varying periodic torque pulsation.

[0217] The uncertainty model of the steer-by-wire system is simplified to an approximate linear integral model, yielding its state space. This state space is then rewritten as an iterative learning state space. Based on this state space, a spatial domain iterative learning algorithm with a forgetting factor is designed. The algorithm's input includes the output of the previous iteration and the iterative learning tracking error, which is the difference between the actual torque value and the observed torque value. The learning law for this spatial domain iterative learning algorithm with a forgetting factor is designed, and it is the compensation current for periodic torque pulsations in the permanent magnet synchronous motor. Specifically, it includes:

[0218] Designing iterative learning with a forgetting factor: Choosing the bandwidth value to make... make Uncertainty model of steer-by-wire system Converting to an approximate linear integral model, the state space is obtained as follows:

[0219]

[0220] In the formula:

[0221] State space of approximate linear integral model Rewritten:

[0222]

[0223] In the formula, x k and y k These represent the system states in the k-th iteration; u k This is the input for the k-th iteration; k is the iteration number.

[0224] The spatial domain iterative learning algorithm with a forgetting factor is represented as:

[0225]

[0226] In the formula: Γ and Φ are the proportional learning gain and differential learning gain, respectively; ν is the forgetting factor; It is bounded; The error is the (k-1)th iteration error of the motor output torque; each error is stored in memory during the iteration process; T m The torque observation value obtained from the torque observer in step 4);

[0227] The output of iterative learning is the q-axis compensation current Δi of the permanent magnet synchronous motor. q,k Then the learning law is expressed as:

[0228]

[0229] The iterative learning tracking error is represented as:

[0230] e k-1 (θ m ) = ri q,k-1 (θ m )-T m (θ m )

[0231] In the formula, e k-1 The torque error is the value after the (k-1)th iteration; i q,k-1 This is the compensation current after the (k-1)th iteration.

[0232] 6) Add the output current of the nonsingular fast terminal sliding mode angle tracking controller to the compensation currents for the aperiodic disturbance and the time-varying periodic torque ripple to obtain the input current of the steering motor. Drive the steering motor with the input current to provide an ideal steering angle for the vehicle;

[0233] As Figure 2 shown, the ideal input current of the steering motor is the sum of the control output current of the nonsingular fast terminal sliding mode angle controller, the compensation current output by the spatial domain iterative learning algorithm with a forgetting factor, and the compensation current for the aperiodic disturbance. Among them, the compensation current output by the spatial domain iterative learning algorithm with a forgetting factor includes the compensation current for the periodic torque ripple. The sum of the compensation current for the aperiodic disturbance and the compensation current of the spatial domain iterative learning algorithm with a forgetting factor is the composite disturbance compensation for the aperiodic disturbance and the periodic torque ripple. At this time, the steering motor converts the ideal input current into electromagnetic torque to provide steering power for the steer-by-wire system. The rotation of the steering motor drives the steering gear, and the rotation of the steering gear causes the steering rack to move horizontally, and the tie rod moves accordingly, thereby driving the front wheels to perform steering work; specifically including:

[0234] The ideal input current of the steering motor is expressed as the control input u in step 2), the aperiodic torque ripple compensation motor in step 3) and the q-axis compensation current Δi output by the iterative learning in step 5) q,k added together:

[0235]

[0236] According to the dynamics equation of the steer-by-wire system, the steering motor provides steering power for the steer-by-wire system. The dynamics equation of the steer-by-wire system is expressed as:

[0237]

[0238] In the formula, f a is the aperiodic torque disturbance; τ ripple is the periodic torque ripple.

[0239] As Figure 2 shown, it is the structural block diagram of the composite disturbance suppression strategy for the steer-by-wire system of the present invention. The difference between the desired steering angle and the actual steering angle is input into the nonsingular fast terminal sliding mode algorithm to output the current required by the steering motor; the linear extended state observer inputs the observed value of the steering angle, the observed value of the differential value of the steering angle, and the observer of the aperiodic disturbance into the nonsingular fast terminal sliding mode algorithm. The torque observer receives the q-axis current and speed of the steering motor and outputs the observed value of the torque. The observed value of the torque of the torque observer, the actual q-axis current i of the steering motor qThe output current of the non-singular fast terminal sliding mode angle tracking controller is substituted into the spatial domain iterative learning algorithm with a forgetting factor to obtain the compensation current for time-varying periodic torque pulsations. The compensation current is added to the output current of the non-singular fast terminal sliding mode angle tracking controller and the compensation current of the linear extended state observer to suppress aperiodic disturbances to obtain the actual input current of the steering motor. The current is input to the steering motor, which provides steering power to the steer-by-wire system.

[0240] like Figure 3 As shown, using a sinusoidal signal as the desired turning trajectory, and after compensation by a composite disturbance suppression method, the non-singular fast terminal sliding mode corner tracking controller achieves stable and accurate tracking of the sinusoidal signal.

[0241] like Figure 4 As shown, using the sweep frequency signal as the desired turning trajectory, and after compensation by the composite disturbance suppression method, the non-singular fast terminal sliding mode corner tracking controller achieves stable and accurate tracking of the sweep frequency signal.

[0242] like Figure 5 As shown, using the ramp signal as the desired turning trajectory, and after compensation by the composite disturbance suppression method, the non-singular fast terminal sliding mode corner tracking controller achieves stable and accurate tracking of the ramp signal.

[0243] This invention has many specific applications. The above description is only a preferred embodiment of this invention. It should be noted that for those skilled in the art, several improvements can be made without departing from the principle of this invention, and these improvements should also be considered within the scope of protection of this invention.

Claims

1. A method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system, based on the steer-by-wire system, characterized in that, The steer-by-wire system includes: an angle controller, an inverter, a steering motor, an angle sensor, a steering gear, a steering rack, a steering tie rod, and a front wheel; The steering angle controller is electrically connected to the inverter and sends control commands to the inverter; the inverter is electrically connected to the steering motor, which generates three-phase current to drive the steering motor, which is a permanent magnet synchronous motor; the output end of the steering motor is mechanically connected to the steering gear, with the connection method being a shaft and key fit; the steering gear meshes with the steering rack, and the steering rack is mechanically connected to one end of the steering tie rod, with the connection method being a threaded connection; the other end of the steering tie rod is mechanically connected to the front wheel, with the connection method being a threaded connection; the steering tie rod drives the front wheel to generate a steering angle; the steering angle sensor is installed between the output shaft of the steering motor and the steering gear, outputting the motor steering angle signal; the steering angle sensor is electrically connected to the steering angle controller and transmits the steering angle signal to the steering angle controller through a signal line; The method steps are as follows: 1) The steering angle controller receives the vehicle's desired steering angle from an external input, and subtracts the desired steering angle from the vehicle's actual steering angle to obtain the difference between the vehicle's desired steering angle and the vehicle's actual steering angle. 2) Establish an uncertainty model for the steer-by-wire system that includes non-periodic disturbances and time-varying periodic torque pulsations. Design an error dynamic equation based on the uncertainty model of the steer-by-wire system. Input the difference between the vehicle's desired steering angle and the vehicle's actual steering angle and its differential value into the error dynamic equation. Design a non-singular fast terminal sliding surface. Design a non-singular fast terminal sliding angle tracking controller based on the non-singular fast terminal sliding surface and the error dynamic equation. By adjusting the parameters of the non-singular fast terminal sliding angle tracking controller, ensure that the difference between the vehicle's desired steering angle and the vehicle's actual steering angle and its differential value converge to 0. The output current is calculated by the non-singular fast terminal sliding angle tracking controller. 3) Design a linear extended state observer to differentiate the actual steering angle of the vehicle collected by the steering angle sensor to obtain the differential value of the actual steering angle. Input the actual steering angle of the vehicle, the differential value of the actual steering angle, and the output current of the non-singular fast terminal sliding mode steering angle tracking controller to the linear extended state observer. After calculation, the estimated value of the steering angle, the estimated value of the differential value of the steering angle, and the estimated value of the non-periodic disturbance are obtained. Input the above estimated values ​​to the non-singular fast terminal sliding mode steering angle tracking controller to obtain the compensation current for the non-periodic disturbance. 4) Design a torque observer, input the steering motor speed and the output current of the non-singular fast terminal sliding mode angle tracking controller to the torque observer, and calculate the observed value of the steering motor output torque; 5) Combine the torque observation value from the torque observer with the actual q-axis current i of the steering motor. q The output current of the non-singular fast terminal sliding mode corner tracking controller is substituted into the spatial domain iterative learning algorithm with a forgetting factor to calculate the compensation current that suppresses time-varying periodic torque pulsation. 6) The output current of the non-singular fast terminal sliding mode angle tracking controller is added to the compensation current of non-periodic disturbance and time-varying periodic torque pulsation to obtain the steering motor input current. The steering motor is driven by the input current to provide the vehicle with an ideal steering angle.

2. The method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system according to claim 1, characterized in that, Step 1) specifically includes: The desired steering angle of the vehicle is provided by the driver, and the actual steering angle is obtained by the steering angle sensor. The difference e1 between the desired steering angle and the actual steering angle is: In the formula, δ represents the desired steering angle. f This is the actual steering angle.

3. The method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system according to claim 1, characterized in that, The uncertainty model of the steer-by-wire system in step 2) includes aperiodic disturbances and time-varying periodic torque pulsations. The aperiodic disturbances, composed of unknown model parameter mismatch disturbances, unknown modeling dynamic disturbances, tire self-centering torque, and other external unknown disturbances, are modeled as bounded uncertainty disturbances. The time-varying periodic torque pulsations, composed of cogging torque, flux harmonic torque, inverter nonlinear torque, and current sampling error disturbances, are modeled as bounded time-varying periodic disturbances, thus obtaining the uncertainty model of the steer-by-wire system. Based on the uncertainty model of the steer-by-wire system, the error dynamic equation is obtained, and step 1) is... The difference between the desired steering angle and the actual steering angle of the vehicle is differentiated to obtain the differential value of the steering angle difference. The difference between the desired steering angle and the actual steering angle of the vehicle and its differential value are substituted into the error dynamic equation to design a non-singular fast terminal sliding surface. Based on the non-singular fast terminal sliding surface and the error dynamic equation, a non-singular fast terminal sliding angle controller is designed. By adjusting the parameters of the non-singular fast terminal sliding angle controller, the difference between the desired steering angle and the actual steering angle of the vehicle and its differential value are converged to 0. The input current required for the angle controller is calculated.

4. The method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system according to claim 3, characterized in that, Step 2) specifically includes: 21) The non-periodic disturbance composed of unknown model parameter mismatch disturbance, unknown modeling dynamic disturbance, tire self-aligning torque and other external unknown disturbances is modeled as a bounded uncertainty disturbance, as follows: In the formula, t is time, and δ f τ is the front wheel steering angle. f Let τ be the Coulomb friction torque. a τ is the tire return torque. d For non-periodic torques generated by uncertain model parameters, unmodeled dynamics, and other external random disturbances, F f Let t be the Coulomb friction coefficient. p For the aerodynamic trajectory, t m For mechanical trajectory, C αf β is the front wheel slip stiffness coefficient, α is the slip angle, a is the distance from the front axle to the center of gravity, r is the yaw rate at the center, V is the longitudinal velocity component of the center of gravity, and F is the front wheel slip stiffness coefficient. d It is a non-periodic perturbation function; 22) The time-varying periodic torque pulsation, composed of cogging torque, flux harmonic torque, inverter nonlinear torque, and current sampling error disturbance, is modeled as a bounded time-varying periodic disturbance, as follows: In the formula, T cog For cogging torque; i smag T represents the stator current amplitude. n N represents the Fourier coefficients; s θ is the least common multiple of the number of tooth indentations and the number of pole pairs; n is a positive integer; θ m T is the rotor angle; ψf T is the magnetic flux harmonic torque. 6k ω represents the harmonic coefficients of the magnetic flux for each order, k = 1, 2, 3, ..., n; e Electric angular velocity; Δu α and Δu β θ represents the inverter nonlinear voltages along the α and β axes, respectively; Δu is the inverter voltage value; θ e ΔT is the electrical angle of the motor; offset For current bias error disturbance; ΔT scalig For current scaling error perturbation; K t This is the current error coefficient; For the offset phase; Δi as and Δi bs These are the measured current offsets for phases a and b, respectively; the measured current scaling factors for phases a and b are respectively expressed as K. a and K b t represents time; 23) The uncertainty model of the steer-by-wire system is described as follows: In the formula, J is the moment of inertia of the steering system; B is the damping coefficient of the steering system; τ m κ represents the output torque of the steering motor; κ is the transmission coefficient of the steering system. make For non-periodic torque disturbances, we get: In the formula, f a This is a non-periodic total disturbance; The output torque τ of the steering motor m Represented as: In the formula, P n ψ is the number of pole pairs of the steering motor; f i is the flux linkage value of the steering motor. q τ is the q-axis current of the steering motor. ripple It is a periodic torque pulsation; The electromagnetic output torque containing periodic torque pulsations is expressed as: In the formula, ψ dk Here are the flux linkage values ​​for each order; Δi offset , Δi scaling , Δi inverter These are the current errors caused by the offset and scaling errors of the current sensor and the nonlinearity of the inverter, respectively. The output torque of the steering motor consists of a DC component and harmonic components of the 1st, 2nd, 6th, 12th, and multiples of 6th orders. The output torque of the steering motor can be simplified as follows: In the formula, τ ek and φ k These are the amplitude and phase angle of the k-th harmonic component, respectively; 24) Establish a steer-by-wire angle tracking controller; Due to the compensation effect of the linearly extended state observer, the error between the vehicle's expected steering angle and its actual steering angle is expressed as: In the formula, δ represents the vehicle's desired steering angle. f This refers to the vehicle's actual steering angle. These are the observed values ​​of the steering angle; The observed value is the time derivative of the steering angle; The error dynamic equation is obtained as follows: In the formula, t1 = [e1, e2] Τ ; Linear extended state observer observes disturbances D represents the maximum disturbance value; u represents the control input. Let be the expected value of the second time derivative of the steering angle; The non-singular fast terminal sliding surface is designed as follows: In the formula, γ, p, and q are the coefficients of the non-singular fast terminal sliding mode controller; γ > 0; p and q are positive odd numbers, and p > q; The output current u of the non-singular fast terminal sliding mode angle controller is: In the formula, is the second derivative of the desired steering angle; 1 < p / q < 2; D is the controller parameter; η is the coefficient of the non-singular fast terminal sliding mode controller, η > 0; sgn is the sign function.

5. The method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system according to claim 1, characterized in that, In step 3), the steering angle sensor measures the actual steering angle of the vehicle, and differentiates it to obtain the differential value of the actual steering angle. A linear extended state observer is designed, and the actual steering angle of the vehicle, the differential value of the actual steering angle, and the output current of the non-singular fast terminal sliding mode steering angle tracking controller are used as inputs to the linear extended state observer. By adjusting the observer bandwidth, the observation error is ensured to converge to 0. The linear extended state observer outputs the estimated value of the steering angle, the estimated value of the differential value of the steering angle, and the estimated value of the non-periodic disturbance. The steering angle, the differential value of the steering angle, and the non-periodic disturbance are input into the error dynamic equation. The compensation current for the non-periodic disturbance is obtained by using the non-singular fast terminal sliding mode steering angle controller.

6. The method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system according to claim 1, characterized in that, Step 3) specifically includes: A linear extended state observer is designed to compensate for non-periodic disturbances in the uncertainty model of the steer-by-wire system; the design of the linear extended state observer specifically involves: Let the non-periodic torque disturbance f a Continuous and differential, linearly extended state observer extends the aperiodic torque disturbance f a As the new state variable x3; let For different state variables, b = κr / J; Described as: The linearly extended state observer is designed as follows: In the formula, For state variable x i The estimated values, i = 1, 2, 3; β 01 ,β 02 and β 03 These are the parameters to be designed; make This represents the estimation error of the state variables, combined with the formula with formula get: In the formula, Let ω0 be the bandwidth of the linearly extended state observer, and choose the characteristic polynomial (s+ω0). 3 =s 3 +β 01 s 2 +β 02 s+β 03 , represented as β 01 =3ω0, Ensure that matrix A is a Hurwitz matrix; For the estimated value of non-periodic torque disturbance Compensation current for non-singular fast terminal sliding mode for:

7. The method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system according to claim 1, characterized in that, Step 4) specifically includes: Design a torque observer to obtain torque observation values; construct the state equations of the permanent magnet synchronous motor: In the formula, B m =[1 / J m 0] Τ C m =[1 0];x′=[ω m T e ] Τ For state variables; y′=ω m T represents the motor speed. e J is the output torque of the motor. m b is the moment of inertia of the motor. m A is the motor damping coefficient; m B m With C m represents the coefficients of the state equation; u′ represents the observer input; The torque observer is represented as: In the formula, the desired state variable is The actual control output is y′=C m x′; the desired control output is K1 = [k1 k2] Τ And K2 = [k3 k4] Τ Here, k is the feedback matrix; k1, k2, k3, and k4 are the torque observer coefficients; Depend on and We can obtain: In the formula, This is the observation error; The differential value of the observation error; I is the identity matrix; s is the differential operator; The characteristic equation of the observer is expressed as: In the formula, λ is the independent variable of the characteristic equation; det represents the product of the eigenvalues ​​of the matrix corresponding to the linear transformation of the square matrix. Based on the desired pole σ, and the expected observer expression Then we can obtain: Assume b m =0, and design k1=k3=0, then the state feedback coefficient can be obtained as: The observed torque value of the permanent magnet synchronous motor can be obtained as follows: Received That is, the torque observation value T m .

8. The method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system according to claim 1, characterized in that, In step 5), the uncertainty model of the steer-by-wire system is simplified to an approximate linear integral model, and the state space of the approximate linear integral model is obtained. The state space of the approximate linear integral model is rewritten as an iterative learning state space. Based on the iterative learning state space, a spatial domain iterative learning algorithm with a forgetting factor is designed. The input of the algorithm includes the output of the previous iteration and the iterative learning tracking error of the previous iteration. The iterative learning tracking error is the difference between the actual torque value and the observed torque value. The learning law of the spatial domain iterative learning algorithm with a forgetting factor is designed. The learning law is the compensation current of the permanent magnet synchronous motor for periodic torque pulsation.

9. The method for suppressing composite disturbances in a permanent magnet synchronous motor steer-by-wire system according to claim 1, characterized in that, In step 6), the ideal input current of the steering motor is the sum of the control output current of the non-singular fast terminal sliding mode angle controller, the compensation current output by the spatial domain iterative learning algorithm with forgetting factor, and the compensation current for aperiodic disturbances. The compensation current output by the spatial domain iterative learning algorithm with forgetting factor includes the compensation current for periodic torque pulsations. The sum of the compensation current for aperiodic disturbances and the compensation current from the spatial domain iterative learning algorithm with forgetting factor is the compensation for the composite disturbance of aperiodic disturbances and periodic torque pulsations. At this time, the steering motor converts the ideal input current into electromagnetic torque, providing steering power for the steer-by-wire system. The rotation of the steering motor drives the steering gear, which in turn causes the steering rack to move laterally, and the steering tie rod to move accordingly, thereby driving the front wheels to perform steering.

Images