Research on control strategy of electric power ste

Research on the control strategy of automotive electric power steering system

Abstract: Taking the control logic of electric power steering system as the research object, aiming at the nonlinear and time-varying parameters of automotive power steering system, the fuzzy control strategy with good dynamic response and strong adaptability is applied to control the power of steering, and the control rules of power assisted characteristics are constructed by using the fuzzy logic toolbox of MATLAB. At the same time, aiming at the problem that the vehicle's homing performance becomes worse after being equipped with electric power steering system, fuzzy control and PD control subsection control methods are used to improve the vehicle's homing performance. Taking a Chery car as an example, the control model of electric power steering system is built and simulated by Adams and MATLAB. The simulation results show that the fuzzy control can track the characteristic curve of the power steering well, and can obtain an ideal effect of power steering. At the same time, the fuzzy control and PD control subsection control methods also significantly improve the vehicle's return performance

key words: electric power steering fuzzy control return control

1 preface

electric power steering system (EPS) eradicates the inherent defects of hydraulic power steering system in system layout, installation tightness, energy loss, operating sensitivity and noise, and has many advantages that hydraulic power steering system cannot compare

in recent years, the rapid development of power electronic technology and the progress of control theory and information technology have provided high-quality and low-cost electronic devices and practical control strategies for the realization of electric power steering. At present, electric power steering system has been applied to many vehicles at home and abroad, and has a very broad application prospect

the electric power steering system uses the motor as the actuator, and uses the electronic control device to generate auxiliary power of appropriate size and direction to drive the steering system according to the driving state of the vehicle, so as to achieve the purpose of light and sensitive steering. Fuzzy control is a kind of computer intelligent control based on fuzzy set theory, fuzzy language variables and fuzzy logic reasoning. As a nonlinear control algorithm, it shows the advantages of good robustness and fast rise time for the control of nonlinear and complex objects. This paper uses fuzzy control theory to design a suitable control strategy of power steering

with the introduction of the motor, the inertia, damping and friction of the motor have a negative impact on the return performance of the steering system, and the return performance of the vehicle decreases. Therefore, using the respective advantages of fuzzy control and PD control, this paper uses the method of fuzzy control and PD control subsection control to improve the correction performance of the system, and has produced good results

2 establishment of steering power control strategy

eps in the process of steering, in order to reduce the control force of the steering wheel, the power assisted torque of the motor is applied to the steering shaft through the worm gear reduction mechanism, so that the steering control is light and sensitive. Boost control is the basic control strategy of EPS, which determines whether EPS has appropriate boost performance if arbitrary cross-section is selected

power assist characteristic refers to the law that power assist changes with the change of vehicle motion and force. The ideal power assisted characteristics should be able to fully coordinate the steering portability and road feel. The report believes that the British building code is "actually not applicable". The commonly used assistance features are linear, broken line and curve. The determination of linear shape is simple, which is convenient for the design of control system and adjustment. However, for the high and low areas of input torque, there is no difference and correspondence, and the road feeling is single, so the relationship between road feeling and portability cannot be well coordinated; Each curve of the curve shape itself can change according to the high and low torque input areas, but in the determination process, a large amount of and dense information about the torque characteristics of the ideal steering wheel is required, so it is not easy to determine and adjust; The advantages and disadvantages of broken line lie between the two [1]

because the fuzzy control adopts the control method that a pair of inputs and outputs correspond to a control rule, adjusting the control rule for different inputs can realize the curvilinear assistance characteristics, and it is very convenient to design and adjust

2.1 fuzzy controller design of steering power

2.1.1 establish the fuzzy space of input and output variables

this paper selects the steering wheel torque input TD and vehicle speed signal V as the input of the fuzzy controller, and the power moment I generated by the motor as the output of the single variable two-dimensional fuzzy controller structure. The widely used Mamdani type reasoning method is adopted in the fuzzy reasoning method

first set the basic universe, quantification factors and fuzzy sets of TD, V and I. The fuzzy sets of TD and V are [very slow, very slow, relatively slow, relatively fast, very fast, very fast] and [very small, very small, relatively small, relatively large, very large, very large]. They are represented by [t1, T2, T3, T4, T5, t6] and [v1, V2, V3, V4, V5, v6] respectively. Because the power moment I changes greatly in the size area of the steering wheel input torque, in order to control the appropriate output and ensure the sensitive and comfortable handling performance, the fuzzy set of the power moment is divided into nine fuzzy sets, which are [zero, very small, very small, small, relatively large, large, very large], represented by [i1, I2, I3, I4, i5, I6, i7, I8, i9] [2]

in order to achieve the purpose of rapid response, the input and output variables are set to obey the membership function of triangular curve

2.1.2 establishment of fuzzy control rules

fuzzy control rules describe the relationship between input and output, and express the discrimination process when people perform control on the controlled object. The design of power moment should follow the following basic requirements:

1) when the steering wheel torque is less than 1n m, the power moment is zero, so that the driver can maintain a better road feel and save electric energy

2) in the area where the input torque of the steering wheel is small, in order to maintain a better road feel, the power output should be small. In the area where the input torque of the steering wheel is large, in order to make the steering light, the power output should be large

3) when the power moment increases to a certain value, it should be kept constant, so that the driver can obviously feel the increase of the road reaction force, guide safe driving, and avoid the motor failure due to excessive load

4) with the increase of vehicle speed, the applied steering power should be gradually reduced, so that the driver can get a good road feel and ensure driving safety

5) the transition between the size of assistance should be smooth to avoid the jumping feeling of steering force

according to the above principles, fuzzy control rules are established. When applying the control rules, it is also necessary to carry out specific design and commissioning according to the requirements of specific vehicle models with reference to the ideal power moment [3]. Figure 1 shows the pulse spectrum of the power curve composed of fuzzy rules. Figure 1 shows the setting of positive auxiliary torque, and the setting of negative auxiliary torque is symmetrical with this figure

Figure 1 pulse spectrum of power curve

2.1.3 defuzzification

the output of power moment obtained by fuzzy rules is a fuzzy quantity, which needs to be obtained through the defuzzification process. In this paper, the widely used weighted average method is used. The result of this method is easier to get a smooth output surface and reduce the fluctuation of the output power moment

2.2 simulation system model establishment

this paper takes a micro car of Chery company as an example, establishes the dynamic model of the whole vehicle in ADAMS, and carries out joint simulation with matlab/simulink, and constructs the steering power fuzzy control block diagram shown in Figure 2

Figure 2 steering power control block diagram

3 active return control strategy design

3.1 active return control strategy design therefore

because the electric power steering system uses the motor as the actuator, and the inertia and damping of the motor have a relatively large impact on the return performance of the system, therefore, it is necessary to introduce other control strategies on the basis of basic power control to improve the return performance of the system. Active return mainly includes two aspects: one is return control. When the vehicle is running at low speed, it relies on the motor to apply external torque, so that the steering wheel can quickly and accurately return to the middle position. The other is active damping control. In order to prevent overshoot, the damping effect of the motor torque on the system is used to make the steering wheel return to the middle position to avoid overshoot when the high-speed is returned [4,5]

in this paper, fuzzy control and PD control are used to control the correction performance of the whole vehicle. When the vehicle enters the alignment process, if the steering wheel angle is greater than a certain threshold, the fuzzy control with fast response and easy adjustment is used for control. When the steering wheel angle is less than a certain threshold, PD control with high control accuracy and good dynamic performance is used. The principle of active return control is shown in Figure 3

Figure 3 sectional control diagram of return control

3.2 design of controller

return control takes the steering wheel angle as the object of attention, so the steering wheel angle and its change rate are selected as the input and the auxiliary torque as the output. The controller also adopts Mamdani reasoning method. Considering the requirements of fast and accurate return performance, the steering wheel angle of 20 degrees is taken as the dividing point between fuzzy control and PD control

the triangular function form is still used here as the membership function of input and output variables. As for the steering wheel angle, under normal driving conditions, the steering wheel angle generally moves within 1 turn left and right, so it is necessary to design a narrower and thinner triangular function in this interval to make the variable response here more sensitive. As for the change of steering wheel angle, we are more concerned about the impact of the steering wheel angle return speed on the vehicle return overshoot performance. Therefore, it is necessary to narrow the membership function in the area with large steering wheel angle change rate. The membership function of the output power moment adopts the isosceles triangle function which is uniformly distributed and finer on both sides in the middle region

formulate a fuzzy rule table according to the steering wheel return rule

for the PD control part, the control coefficient needs to be continuously adjusted according to the vehicle model. It is necessary to ensure the requirements of vehicle homing performance and avoid the jumping change of steering force during the whole homing process

4 verification of control strategy results

4.1 analysis and verification of steering power control strategy

based on the joint simulation model of steering power, sine wave angle input is carried out on the whole vehicle to simulate the operation of vehicle changing line. The magnitude of steering wheel torque at 21km/h, 40km/h, 65km/h, 80km/h and 97km/h when the lateral acceleration of the vehicle reaches 0.3g after the application of steering power is analyzed respectively. This paper takes the ideal steering wheel torque of 0.3g tested by General Motors in 1999 as the preliminary reference target. Figure 3 shows the comparison between the steering torque of vehicles with power assisted control and the target value. It can be seen from Figure 3 that there is a small gap between the steering wheel torque value and the ideal torque after adding power assist control. Figure 4 shows the comparison of the steering wheel torque response of the vehicle to the sinusoidal steering angle input with and without power assistance when the vehicle speed is 65km/h. It can be seen from Figure 4 that the power steering system effectively reduces the steering hand force when the driver changes lanes or avoids obstacles, and ensures that the road feeling within the small torque range is not affected (for reasons of confidentiality, this paper processes some of the result data for reference only)

Figure 3 Comparison between power assist control and target value figure 4

Figure 4 Comparison of steering wheel torque at 65km/h

4.2 analysis and verification of active return control strategy

in order to evaluate the return effect after control, simulate the vehicle return with and without return control at low speed (v=20km/h) and high speed (v=80km/h) respectively. Make the car drive along a certain arc path at low speed and high speed respectively, and simulate the change of steering wheel angle in the process of car return