CN101683867B

CN101683867B - Electric power steering system with variable weighting coefficient linear combination controller

Info

Publication number: CN101683867B
Application number: CN2008102006312A
Authority: CN
Inventors: 刘庆; 陈慧; 杨磊
Original assignee: SHANGHAI LIANSHENG AUTOMOBILE ELECTRONIC CO Ltd
Current assignee: DIAS Automotive Electronic Systems Co Ltd
Priority date: 2008-09-27
Filing date: 2008-09-27
Publication date: 2012-01-11
Anticipated expiration: 2028-09-27
Also published as: CN101683867A

Abstract

The invention discloses an electric power steering system with a variable weighting coefficient linear combination controller, which is formed by adding a torque sensor, an electronic control unit, a diversion assisted motor and a retarding mechanism thereof on a traditional machinery steering system (mainly comprising a handwheel, a steering shaft, a diverter, a steering drag-link mechanism and a steering wheel). The electric power steering system designed by the invention can perform liner weighting to the controller by some parameter measured on line, thus ensuring that the electric power steering system obtains expectant optimal steering performance.

Description

Electric power steering system with variable weighting coefficient linear combination controller

Technical Field

The invention belongs to the technical field of automobiles, and relates to an Electric Power steering system, in particular to an Electric Power steering system (EPS for short) with a variable weighting coefficient linear combination controller.

Background

The electric power-assisted steering system is composed of a traditional mechanical steering system (mainly comprising a steering wheel, a steering shaft, a steering gear, a steering pull rod mechanism and a steering wheel) and additionally provided with a torque sensor, an electronic control unit, a power-assisted steering motor and a speed reducing mechanism thereof. Wherein the torque sensor is mounted on the steering shaft to detect a driver's steering torque; the electronic control unit calculates a driving current value required to be provided for the power-assisted motor according to the current vehicle running condition and the magnitude of the operating torque; the power-assisted motor can be arranged on a steering shaft or a steering gear, power-assisted torque is provided for a driver through a speed reducing mechanism, the power-assisted torque can be set by software, different power-assisted gains are set at different vehicle speeds, and ideal steering hand feeling can be obtained. Generally, the main control function modules of the electronic control unit include a torque control module (including basic power control, active return control, damping control, etc.), a motor current control module, and a fault diagnosis module.

The traditional electric power steering control system achieves the effect of inhibiting road surface interference by adding moment differential control in a basic power assisting module, the traditional differential control is used, the aim of inhibiting the road surface interference can be achieved by repeatedly debugging or switching differential gains under different vehicle speeds, the requirement of road sensing performance is difficult to ensure in the whole vehicle speed range, the process is very complicated, and a designer is required to have considerable debugging experience, so that the time is very long.

Disclosure of Invention

The invention aims to provide an electric power steering system with a variable weighting coefficient linear combination controller, which can effectively ensure the steering assistance performance of the electric power steering system at different vehicle speeds, in order to improve the road feel characteristic of the electric power steering system.

In order to achieve the above purpose, the technical scheme of the invention is as follows: an electric power-assisted steering system with a variable weighting coefficient linear combination controller is composed of a traditional mechanical steering system (mainly comprising a steering wheel, a steering shaft, a steering gear, a steering pull rod mechanism and a steering wheel) and a torque sensor, an electronic control unit, a power-assisted steering motor and a speed reducing mechanism thereof additionally arranged on the traditional mechanical steering system. The torque sensor is arranged on the steering shaft to detect the torque on the hand of a driver; the electronic control unit calculates a driving current value required to be provided for the power-assisted motor according to the current vehicle running condition and the hand torque; the power-assisted motor is generally arranged on a steering shaft or a steering gear and provides power-assisted torque for a driver through a speed reducing mechanism; generally, the main control function modules of the electronic control unit include a torque control module (including basic power control, active return control, damping control, etc.), a motor current control module, and a fault diagnosis module. The invention is characterized in that a basic power-assisted module is improved, a linear combination controller with a variable weighting coefficient is designed, and the linear combination controller enables an electric power-assisted steering system to obtain expected steering power-assisted performance under any working condition and can ensure the stability of the electric power-assisted steering system under any working condition.

Due to the adoption of the technical scheme, the invention has the beneficial effects that: on the premise of not changing the hardware of the conventional electric power steering system, the steering power performance of the electric power steering system at different speeds is ensured by changing the control algorithm.

Drawings

Fig. 1 is a schematic diagram of an electric power steering system having a variable weighting factor linear combination controller according to the present invention.

Fig. 2 is a schematic diagram of a Linear parameter variation control method (LPV) employed by the electric power steering system with the variable weighting coefficient Linear combination controller according to the present invention.

Fig. 3 is a model schematic diagram of an electric power steering system with a variable weighting coefficient linear combination controller according to the present invention.

Fig. 4 is a schematic diagram of the design of an LPV controller of an electric power steering system with a variable weighting coefficient linear combination controller according to the present invention.

Fig. 5 is a bode plot of LPV controller design weighting functions for an electric power steering system with a variable weighting coefficient linear combination controller of the present invention. (10Hz)

Fig. 6 is a bode plot of LPV closed loop control for an electric power steering system with a variable weighting factor linear combination controller of the present invention. (10Hz)

Fig. 7 is a bode plot of LPV controller design weighting functions for an electric power steering system with a variable weighting coefficient linear combination controller of the present invention. (6Hz)

Fig. 8 is a bode plot of LPV closed loop control for an electric power steering system with a variable weighting factor linear combination controller of the present invention. (6Hz)

FIG. 9 is a schematic diagram of a robust controller design for an electric power steering system with a variable weighting factor linear combination controller of the present invention.

Fig. 10 is a robust control theory diagram.

FIG. 11 is a Bode diagram of the original EPS system and the robust controller added closed loop system of the electric power steering system with the variable weighting coefficient linear combination controller of the present invention.

Detailed Description

The invention will be further described with reference to examples of embodiments shown in the drawings.

An electric power-assisted steering system with a variable weighting coefficient linear combination controller is composed of a traditional mechanical steering system (mainly comprising a steering wheel, a steering shaft, a steering gear, a steering pull rod mechanism and a steering wheel) and a torque sensor, an electronic control unit, a power-assisted steering motor and a speed reducing mechanism thereof. The torque sensor is arranged on the steering shaft to detect the torque on the hand of a driver; the electronic control unit calculates a driving current value required to be provided for the power-assisted motor according to the current vehicle running condition and the hand torque; the power-assisted motor is generally arranged on a steering shaft or a steering gear and provides power-assisted torque for a driver through a speed reducing mechanism; generally, the main control function modules of the electronic control unit include a torque control module (including basic power control, active return control, damping control, etc.), a motor current control module, and a fault diagnosis module. The invention is characterized in that a basic power-assisted module is improved, a linear combination controller with a variable weighting coefficient is designed, and the linear combination controller enables an electric power-assisted steering system to obtain expected steering power-assisted performance under any working condition and can ensure the stability of the electric power-assisted steering system under any working condition.

The weighting mode of the linear combination controller can comprise the following two modes:

one is to change the characteristics of the transfer function by weighting the state matrix based on the weighting of the state space expression of the transfer function. Specifically, the following formula (I) represents:

(\begin{matrix} A & B \\ C & D \end{matrix}) = k (\begin{matrix} A_{1} & B_{1} \\ C_{1} & D_{1} \end{matrix}) + (1 - k) (\begin{matrix} A_{2} & B_{2} \\ C_{2} & D_{2} \end{matrix}) - - - (I)

another is the weighting of the output quantity based on the transfer function, which is expressed by the following equation (II):

K(s)＝k·K₁(s)+(1-k)·K₂(s) (II)

wherein, in the above formulas (I) and (II)

(\begin{matrix} A & B \\ C & D \end{matrix})

Is the state space matrix expression of the transfer function, K(s) is the Laplace transform expression of the transfer function, K is the variable weighting coefficient, in the patent, K can change on line along with the parameters of the vehicle speed, etc., wherein K is between 0 and 1, K is the variable weighting coefficient₁(s) and K₂(s) in the Laplace transform expression for the solved controllers with different performance.

The first type of the weighting mode of the Linear combination controller is the weighting of a state space expression based on a transfer function, and a solving method of the mode is to design a control system by using a Linear parameter variation control method (LPV for short).

The Linear parameter variation control method (LPV) is a system in which some parameters of an actual controlled object are varied with time and the parameters can be measured online, such as some physical parameters of a missile control system varied with flight height and speed, and a power-assisted gain varied with vehicle speed in an electric power steering system to which the patent is directed.

Here, consider first an LPV system:

z＝C₁(θ(t))x+D₁₁(θ(t))w+D₁₂(θ(t))u

y＝C₂(θ(t))x+D₂₁(θ(t))w+D₂₂(θ(t))u (1)

where w, u in equation (1) are the external disturbance input signal and the control input signal, z, y are the control output signal and the measurement output signal, and x is the state variable.

Next, the system object is described in the form of a multicellular system, as in formula (2):

<math> <mrow> <mrow> <mfenced open='(' close=')'> <mtable> <mtr> <mtd> <mi>A</mi> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> <mtd> <msub> <mi>B</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> <mtd> <msub> <mi>B</mi> <mn>2</mn> </msub> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <msub> <mi>C</mi> <mn>1</mn> </msub> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> <mtd> <msub> <mi>D</mi> <mn>11</mn> </msub> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> <mtd> <msub> <mi>D</mi> <mn>12</mn> </msub> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <msub> <mi>C</mi> <mn>2</mn> </msub> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> <mtd> <msub> <mi>D</mi> <mn>21</mn> </msub> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> <mtd> <msub> <mi>D</mi> <mn>22</mn> </msub> <mrow> <mo>(</mo> <mi>θ</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> </mtd> </mtr> </mtable> </mfenced> </mrow> <mo>&Element;</mo> <mi>ρ</mi> <mo>:</mo> <mo>=</mo> <mi>Co</mi> <mrow> <mo>{</mo> <mrow> <mo>(</mo> <mrow> <mfenced open='(' close=')'> <mtable> <mtr> <mtd> <msub> <mi>A</mi> <mi>i</mi> </msub> </mtd> <mtd> <msub> <mi>B</mi> <mrow> <mn>1</mn> <mi>i</mi> </mrow> </msub> </mtd> <mtd> <msub> <mi>B</mi> <mrow> <mn>2</mn> <mi>i</mi> </mrow> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>C</mi> <mrow> <mn>1</mn> <mi>i</mi> </mrow> </msub> </mtd> <mtd> <msub> <mi>D</mi> <mrow> <mn>11</mn> <mi>i</mi> </mrow> </msub> </mtd> <mtd> <msub> <mi>D</mi> <mrow> <mn>12</mn> <mi>i</mi> </mrow> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>C</mi> <mrow> <mn>2</mn> <mi>i</mi> </mrow> </msub> </mtd> <mtd> <msub> <mi>D</mi> <mrow> <mn>21</mn> <mi>i</mi> </mrow> </msub> </mtd> <mtd> <msub> <mi>D</mi> <mrow> <mn>22</mn> <mi>i</mi> </mrow> </msub> </mtd> </mtr> </mtable> </mfenced> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mi>r</mi> <mo>=</mo> <mn>1,2</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mi>r</mi> <mo>}</mo> </mrow> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow></math>

then look for an LPV controller of the form as equation (3):

u＝C_k(θ(t))x_k+D_k(θ(t))y (3)

and ensures that secondary H is achieved for a closed loop system as shown in figure 2_∞The property γ.

Wherein

w is an externally perturbed input signal;

u is a control input signal;

z is a control output signal;

y is the measurement output signal;

p (t) is a parameter that can be measured online.

Road feel may be characterized by the transfer of a restoring torque to the driver's hand. In order to research road feel, an electric power steering system model is established, and a rack force loading mechanism simulates road resistance; then fixing the steering wheel, wherein the situation is consistent with the situation that a driver firmly stabilizes the steering wheel in the driving process of the vehicle; through model simplification, a dynamic equation with rack force as input and a torque sensor as output can be established, and fig. 3 is a simplified electric power steering system model.

From fig. 2, the dynamic equations of the electric power steering system can be written

Similarly, neglecting the output shaft rigidity of the motor, the motor rotation angle and the pinion rotation angle have a proportional relation,

θ_m＝G*θ_p (6)

the rack displacement and the pinion rotation angle are also

θ_p＝X_r/r_p (7)

The magnitude of the motor assist is also determined by the assist gain and the torque sensor output, and since the steering wheel is fixed, θ c is 0, that is:

T_com＝K_a*T_s＝K_a*K_s*(-θ_p) (8)

wherein:

jm is the moment of inertia of the motor,

mr is the mass of the rack,

theta c is the steering column angle

θ m is motor rotation angle

Xr is the displacement of the rack,

ks is torsion bar rigidity of torque sensor

Cm is the damping of the motor,

cr is the damping of the rack,

rp is the radius of the pinion gear and,

g is the transmission ratio of the worm gear and the worm,

ta is the actual moment of assistance of the motor

The Tcom target commands a power-assisted torque,

θ p is the pinion rotational angle.

An LPV controller schematic diagram designed to improve road feel using the LPV control method is shown in figure 4,

wherein,

ka is the power-assisted gain and,

ks is the torque sensor torsion bar stiffness,

ta is the actual torque assistance of the motor,

the Tcom target commands a power-assisted torque,

θ p is the pinion rotational angle,

fr is the rack force exerted by the EPS hardware on the ring simulation platform,

ts is the output of the torque sensor and,

the LPV Controller is a robust Controller we want to design.

Based on the design goals, we choose the control goals as follows,

D = F_{r} - \frac{1 + K_{a}}{r_{p}} * T_{s} - - - (9)

namely, the difference value of the rack force Fr and the force converted from the hand torque and the motor assisting torque to the rack is taken as a control target, if D approaches to 0, the rack force Fr is completely transmitted to the steering wheel, and if D approaches to Fr, the rack force Fr is completely attenuated by the system and is not transmitted to the steering wheel. Therefore, by defining the control target D by a weighting function w(s), it is possible to control the transfer characteristic of the rack force Fr to the steering wheel torque, thereby changing the road feel of the electric power steering system.

It can be seen that the expression of the control target D contains a variable power assist gain Ka, which is the variable online measurable parameter p (t) shown in fig. 2 when establishing an electric power steering system augmented LPV system.

It is assumed that the chosen weighting function w(s) has the form,

since the weighting function is a transfer function and also has dynamic characteristics, the system needs to add a state variable w1, so the evaluation output Z ═ D × w(s) can be written in the following form:

an LPV augmentation system of an electric power steering system can be obtained:

z＝C₁(K_a(t))x+D₁₁(K_a(t))w+D₁₂(K_a(t))u

y＝C₂(K_a(t))x+D₂₁(K_a(t))w+D₂₂(KX(t))u (13)

wherein

w＝F_r

u＝T_com

z＝Z

y＝T_s

D_{11} (K_{a} (t)) = \frac{bc}{a}

D₁₂(K_a(t))＝0

C₂(K_a(t))＝[0 -K_s 0]

D₂₁(K_a(t))＝0

D₂₂(K_a(t))＝0

It can be seen that in the LPV augmented system matrix, only A (K) is present_a(t)) and C₁(K_a(t)) is related to the assist gain Ka, since we generally choose the maximum assist gain Ka of 20 in the process of designing the electric power steering system, it is not assumed that Ka is in the interval [0, 20 ]]So that the LPV system can be written as a convex combination of two multi-cell vertices,

wherein,

after selecting the appropriate weighting function w(s), the LPV controller is solved using the hinfgs command in the LMI toolbox in MATLAB/SIMULINK.

Selecting different weighting functions may result in LPV controllers with different control performance.

The weighting function w(s) is used to define the control target D, and if it is desired that the rack force Fr of a certain frequency range can be completely transmitted to the steering wheel torque, that is, D approaches 0 in the frequency range, the gain of the weighting function w(s) in the frequency range can be set to a relatively large positive value; if it is desired that a certain frequency range of rack force Fr cannot be transmitted to the steering wheel torque, i.e., D approaches Fr in this frequency range, the gain of the weighting function w(s) in this frequency range may be set to 0 or a positive value very close to 0.

For information from a road surface, a frequency band range within 10Hz is only related to the motion state of an automobile, most road surface information with frequency higher than 10Hz belongs to useless information or interference information, a weighting function W(s) is selected as a high-pass function, the gain of the weighting function W(s) is higher before a turning frequency set value 1, and the gain of W(s) approaches to 0 in a section higher than the turning frequency set value 1.

When the weighting function w(s) is selected as shown in fig. 5, the turning frequency set value 1 is set to 10Hz, and the closed loop bode plot from the rack force Fr to the torque sensor output Ts under the action of the LPV controller is shown in fig. 6, it can be seen that when the power-assisted gain Ka changes from 0 to 20, the closed loop system attenuates the road information above 10Hz, but retains the road information within 10Hz related to the vehicle motion state, and the EPS is still working to maintain road feel even when Ka is 0, that is, the electric power steering system does not provide the steering power; moreover, as the assist gain Ka becomes larger, the closed loop system is always kept stable.

When the weighting function w(s) is selected and the turning frequency set value 1 is set to 6Hz as shown in fig. 7, and the closed loop bode plot from the rack force Fr to the torque sensor output Ts under the action of the LPV controller is shown in fig. 8, it can be seen that when the assist gain Ka is changed from 0 to 20, the closed loop system attenuates the road information above 6Hz, and when the assist gain Ka becomes large, the system remains stable.

Since setting the turning frequency setting value 1 to 10Hz is the best choice, we choose the weighting function with the turning frequency of 6Hz to prove that LPV controllers with different performances can be obtained by changing the weighting function w(s).

Another weighting method of the linear combination controller is based on the weighting of output quantities of transfer functions, and a solving method of the method can be that two controllers with different performances are respectively solved, the output quantities of the two controllers are weighted, and the total output quantity obtained by weighting is used as the output quantity of the whole control system.

This patent utilizes robust control theory to go to solve two controllers of different performance respectively, to restraining the motor output pulsating robust controller and to the robust controller of road feel, and the weight function of the linear combination controller of design is the function of speed of a motor vehicle, along with the increase of the speed of a motor vehicle promptly, the weight of the robust controller of road feel increases gradually, and the weight of the robust controller of restraining the pulsation reduces gradually.

The following is a design process for a robust controller for suppressing motor output ripple:

the frequency range of the output torque pulsation of the EPS power-assisted motor is very wide, and the frequency range is based on H_∞The robust control method of the norm is used for designing a robust controller capable of suppressing motor pulsation so as to eliminate the influence on the maneuverability of a driver. A schematic diagram of a robust controller for suppressing motor ripple is shown in fig. 9.

Wherein Ta is the actual assistance torque of the motor, Tcom target instruction assistance torque, Tmd is the torque pulsation of the motor, Ts is the output of the torque sensor, W1(s) is a weighting function, Z is the evaluation output, and the Robust Controller is the Robust Controller to be designed.

According to based on H_∞The norm robust control method converts the control schematic diagram into a standard H shown in FIG. 10_∞And (4) controlling the problem. Then H according to the standard_∞Solving the control problem, obtaining a robust controller K(s) which can stabilize the closed-loop system internally and input H of the closed-loop transfer function from Tmd to Z from the outside_∞The minimum or sub-optimal solution is satisfied, i.e., less than a constant value γ.

Where g(s) is an amplification matrix of the system, w is an external input, here, torque ripple of the motor, Z is an evaluation output, here, Z ═ T_sW1(s), y is the system output, here the product of the torque sensor output Ts and the assist gain Ka, u is the control input, here the target commanded assist torque Tcom.

It is assumed that the chosen weighting function W1(s) has the following form,

since the weighting function is a transfer function and also has dynamic characteristics, the system needs to add a state variable n1, and an amplification matrix of the electric power steering system can be obtained as shown in equation (15). After selecting the appropriate weighting function W1(s), i.e. determining the parameters a, b, c, the desired robust controller can be found using the robust control toolkit in MATLAB.

z＝C₁x+D₁₁w+D₁₂u

y＝C₂x+D₂₁w+D₂₂u (15)

The choice of W1(s) has a direct relationship to the performance of a robust controller, and if W1(s) is a high-pass transfer function, the amount of disturbance in the frequency range above its cutoff frequency is greatly attenuated and not transferred to the steering wheel. In general driving, the steering wheel input of the driver is below 1Hz, and the frequency of the motor pulsation is generally higher than 1Hz, so the turning frequency set value 2 of W1(s) is set to be a high-pass transfer function of about 1Hz in the patent. After the weighting function is selected, the robust controller obtained by the matlab solution is a transfer function K1(s) of the third order.

The robust controller design process for road feel is as follows:

because the road surface input and the motor output pulsation are system inputs with the same transfer characteristic, similar to the design of a robust controller for restraining the pulsation, a similar augmentation system matrix is obtained firstly, and then a weighting function W2(S) meeting the road feel requirement is selected, so that the expected robust controller can be obtained.

The frequency range of the road surface input information is wide, but only the road surface information within about 10Hz reflects the motion state of the vehicle, and is the road feel required for safe driving of the driver. If W2(s) is a high-pass function, road information in a frequency range above the W2(s) cutoff frequency is greatly attenuated and not transmitted to the steering wheel.

The method selects W2(s) as a high-pass function, sets the turning frequency set value 3 as 10Hz, and filters out road surface information with the frequency range larger than about 10 Hz. The corresponding robust controller K2(s) can be solved using the robust control toolkit in MATLAB.

The following is the selection process of the controller weight function:

when the vehicle is running, the demand for road feel is gradually increased with the increase of the vehicle speed, and the demand for suppressing motor pulsation is gradually decreased. It is therefore possible to design the weight function of the controller as a function of vehicle speed, i.e. as the vehicle speed increases, the weight Q1 of the road feel controller increases gradually and the weight Q2 of the ripple suppression controller decreases gradually.

The weight function selected by the invention is:

Q1＝V/40

Q2＝1-V/40

where V is the vehicle speed. The designed robust controller is connected into the EPS simulation system and compared with the original system without the controller. Fig. 11 shows a comparison of Bode plots with a controller and an original system at different vehicle speeds, wherein motor pulsation and road surface information are used as system inputs, and a torque sensor Ts is used as a system output.

From the figure, when the vehicle speed is 0Km/h, the input information above 1Hz is completely transmitted to the steering wheel for the original system, and a resonance point is arranged at about 12 Hz. Thus, the output torque ripple of the motor cannot be suppressed, which affects the steering comfort. When the controller is added to the original system, the robust controller takes the suppression of the output pulsation of the motor as a main control target, the suppression pulsation controller designed according to the cut-off frequency of 1Hz can attenuate the pulsation input of the motor above 1Hz, and the resonance point existing in the original system is well suppressed.

When the vehicle speed is 40Km/h, road surface disturbance above 10Hz is transmitted to the steering wheel for the original system, and a resonance point exists around 11Hz, road surface useless information of the frequency can be amplified and transmitted to the steering wheel, so that the steering comfort is greatly influenced; after the controller is added, the robust controller takes the acquisition of road feel as a main control target, and the road feel controller designed according to the cut-off frequency of 10Hz effectively attenuates road surface disturbance above 10Hz and well inhibits a resonance point;

when the vehicle speed is between 0 and 40Km/h, the robust controller balances the two control requirements of road feel and pulsation suppression. Taking the operating mode of the vehicle speed of 20Km/h as an example, as can be seen from fig. 11, the cutoff frequency is about 2Hz, the motor pulsation suppression effect is reduced compared with the case of the vehicle speed of 0, and more road feel information can be fed back to the driver.

The foregoing description of the two embodiments is provided to facilitate the understanding and application of the present invention to those of ordinary skill in the art. It will be readily apparent to those skilled in the art that various modifications to these embodiments may be made, and the generic principles described herein may be applied to other embodiments without the use of the inventive faculty. Therefore, the present invention is not limited to the embodiments described herein, and those skilled in the art should make improvements and modifications to the present invention based on the disclosure of the present invention within the protection scope of the present invention.

Claims

1. An electric power steering system with a variable weighting coefficient linear combination controller comprises a mechanical steering system, a torque sensor, an electronic control unit, a power steering motor and a speed reducing mechanism thereof; wherein the torque sensor is mounted on a steering shaft of the mechanical steering system to detect a driver manipulation torque; the electronic control unit calculates a driving current value required to be provided for the power steering motor according to the current vehicle running condition and the magnitude of the operation torque; the steering power-assisted motor can be arranged on a mechanical steering shaft or a steering gear and provides power-assisted torque for a driver through a speed reducing mechanism; the electronic control unit comprises a torque control module, a motor current control module and a fault diagnosis module; the moment control module comprises a basic power control module, an active return control module and a damping control module, and is characterized in that the basic power control module comprises a linear combination controller with a variable weighting coefficient;

the weighting mode of the linear combination controller with the variable weighting coefficients is realized by adopting one of the following modes:

(1) changing the characteristics of the transfer function by weighting the state matrix based on the weighting of the state space expression of the transfer function; is specifically expressed by the following formula

(\begin{matrix} A & B \\ C & D \end{matrix}) = k (\begin{matrix} A_{1} & B_{1} \\ C_{1} & D_{1} \end{matrix}) + (1 - k) (\begin{matrix} A_{2} & B_{2} \\ C_{2} & D_{2} \end{matrix}) - - - (I);

(2) The weighting of the output quantities based on the transfer function is expressed by the following formula

K(s)＝k·K₁(s)+(1-k)·K₂(s) (II)

Wherein,

(\begin{matrix} A & B \\ C & D \end{matrix})

is a state space matrix expression of the transfer function, k(s) is a laplace transform expression of the transfer function, k is a variable weighting coefficient, where k varies between 0-1.

2. An electric power steering system having a variable weighting factor linear combination controller according to claim 1, characterized in that the solving method of the weighting of the state space expression based on the transfer function is calculated using a linear parameter variation control method.

3. The electric power steering system having the variable weighting factor linear combination controller according to claim 2, wherein the calculation using the linear parameter variation control method is to build an electric power steering system model using a transmission characteristic of a rack force to an output of the torque sensor by fixing a steering wheel angle, and to reduce the electric power steering system to a single-input single-output model.

4. The electric power steering system having a variable weighting factor linear combination controller according to claim 3, wherein the electric power steering system model is established by first selecting a control target as

D = F_{r} - \frac{1 + K_{a}}{r_{p}} * T_{s}

I.e. rack force F_rThe difference value of the steering torque and the force converted from the motor assisting torque to the rack is used as a control target, and a weighting function W(s) is used for limiting a control target D so as to control the rack force F_rTransfer characteristics to the steering wheel torque, thereby changing the road feel of the electric power steering system, wherein F_rRack force, r, applied to the EPS hardware on the ring simulation platform_pIs the pinion radius, K_aIs the boost gain, T_sIs the torque sensor output.

5. The electric power steering system of claim 4, wherein the weighting function W(s) is selected as a transfer function having a gain of 40dB-50dB before the turning frequency and a gain approaching 0 above the turning frequency, and the turning frequency is set to a set value of 1 according to the road feel requirement to suppress the interference information from the road surface with a frequency greater than the set value of 1.

6. The electric power steering system having a variable weighting factor linear combination controller according to claim 5, wherein the solving method for the weighting of the output quantities based on the transfer function is to solve two controllers with different performances, respectively, to weight the output quantities of the two controllers, and the total output quantity obtained by weighting is used as the output quantity of the linear combination controller having the variable weighting factor.

7. The electric power steering system having a variable weighting factor linear combination controller of claim 6, wherein the controller solving for two different performances is a robust controller suppressing motor output pulsation and a robust controller for road feel; and a robust control theory is adopted, aiming at a robust controller for inhibiting the output pulsation of the motor and a robust controller for road feel, the weight function of the designed linear combination controller is a function of the vehicle speed, namely the weight of the robust controller for road feel is gradually increased along with the increase of the vehicle speed, and the weight of the robust controller for inhibiting the pulsation is gradually reduced.

8. The electric power steering system of claim 7, wherein the weighting function selected for the robust controller for suppressing the output ripple of the motor is a high-pass transfer function with a cutoff frequency of a set value of 2, so that the robust controller for suppressing the output ripple of the motor suppresses the ripple interference from the motor with a frequency higher than the set value of 2.

9. The electric power steering system having a variable weighting factor linear combination controller according to claim 7, wherein the weighting function selected by the robust controller for road feel is a high-pass transfer function with a cutoff frequency of a set value of 3, so that the robust controller for road feel can suppress interference from a road surface with a frequency higher than the set value of 3.