A fault-tolerant control method for failure of drive motor of distributed electric vehicles

By constructing dynamic models and sliding mode surface functions in distributed electric vehicles, combining fuzzy algorithms and quadratic planning algorithms, the problem of slow vibration and convergence speeds in sliding mode algorithms is solved, and a faster and more stable sliding mode is achieved Control improves the stability and responsiveness of the vehicle.

CN118677306BActive Publication Date: 2024-11-05SHANDONG UNIV OF SCI & TECH
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411156571.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-22
Publication Date
2024-11-05
Estimated Expiration
2044-08-22

AI Technical Summary

Technical Problem

The existing slip-mode algorithms have vibration phenomena in distributed electric vehicles, which affects system performance and cannot effectively adjust real-time state according to the dynamic characteristics of the system's moving points, resulting in slow convergence time and high-frequency vibration.

Method used

A fault-tolerant control method for failure of distributed electric vehicle drive motors was designed. By constructing a dynamic model and sliding mode surface function, combining fuzzy algorithms to adjust the sliding mode algorithm parameters in real time, and using a quadratic planning algorithm to redistribute torques to realize Stability and optimization of sliding mode control signals.

Benefits of technology

It effectively suppresses the vibration phenomenon in the sliding mode algorithm, improves the rate of movement of the system state to the sliding mode surface, reduces steady-state vibration, improves the yaw motion characteristics and response speed of the vehicle, and ensures that the system quickly reaches a stable state.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118677306B_ABST
    Figure CN118677306B_ABST
Patent Text Reader

Abstract

The invention discloses a fault-tolerant control method for failure of a drive motor of a distributed electric vehicle, belonging to the field of fault-tolerant control. The method combines a fuzzy algorithm on the basis of a conventional power reaching law of a sliding mode algorithm, dynamically adjusts the coefficient parameters and exponential parameters of the power reaching law, so that the movement rate of the system in the sliding motion stage is associated with its own state, and suppresses the vibration amplitude in the sliding mode stage. The improved control method effectively improves the response speed and accuracy of the system. At the same time, a drive motor failure diagnosis method is proposed, which combines a tire model and a wheel dynamics model to obtain the failure parameters of the drive motor, and uses the parameters for torque distribution, so that the distributed electric vehicle has better dynamic and static performance when the drive motor fails.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of fault-tolerant control, and in particular relates to a fault-tolerant control method for failure of a distributed electric vehicle drive motor. Background Art

[0002] In recent years, new energy vehicles have gradually become an important development direction in the automotive field due to their advantages of high efficiency, low noise, and ease of use. Among them, distributed electric vehicles have shown significant technical advantages in terms of efficient transmission, flexible control, and high utilization of vehicle space, and have been widely used in the industrial field. However, electromagnetic faults inside the hub motor can cause its drive failure, causing vehicle instability and even causing serious traffic accidents, which puts higher requirements on the reliability of the electric drive system. At present, some advanced algorithms have been used for fault-tolerant control of distributed electric vehicles. The sliding mode algorithm has become an important method to improve the fault-tolerant performance of distributed electric vehicles due to its strong robustness and rapid dynamic response.

[0003] However, due to the discontinuity of switch control in the sliding mode algorithm, there is a chattering phenomenon when the state trajectory reaches the sliding mode surface, which affects the system performance and may cause mechanical wear and increase energy consumption. The current conventional use of sliding mode convergence law to weaken chattering has not fully considered the dynamic characteristics of the sliding mode motion point, and cannot achieve dynamic adjustment of the sliding mode control gain according to the real-time state of the system motion point, resulting in a slow convergence time and a certain high-frequency chattering at the origin. Summary of the invention

[0004] In view of the above problems existing in the prior art, the present invention proposes a fault-tolerant control method for failure of a distributed electric vehicle drive motor, which has a reasonable design, solves the shortcomings of the prior art and has good effects.

[0005] A fault-tolerant control method for failure of a distributed electric vehicle drive motor comprises the following steps:

[0006] Step 1: Construct a distributed electric vehicle dynamics model in the geodetic coordinate system, a magic tire model, a wheel rotation dynamics model in the wheel coordinate system, and a linear two-degree-of-freedom reference model;

[0007] Step 2: Obtain the failure parameters of the drive motor through drive motor failure diagnosis;

[0008] Step 3: Define the state variables and sliding surface functions of the system and design the reaching law;

[0009] Step 4: Use fuzzy algorithm to adjust the reaching law parameters of sliding mode algorithm in real time to stabilize the control signal;

[0010] Step 5: Redistribute the torque of each tire to obtain a control signal of the torque of each tire.

[0011] Furthermore, in step 1, the dynamic model of the distributed electric vehicle in the geodetic coordinate system is:

[0012] ; (1)

[0013] in:

[0014] ; (2)

[0015] Where: , is the vehicle mass and the body moment of inertia; , , are the longitudinal speed, lateral speed and yaw rate respectively; , , They are the longitudinal force, lateral force and yaw moment of the vehicle respectively; is the front wheel turning angle; , , , They are the longitudinal forces of the left front, right front, left rear and right rear tires respectively. , , , They are the lateral forces of the left front, right front, left rear and right rear tires respectively; , , They are the wheelbase, the distance from the center of mass to the front axle, and the distance from the center of mass to the rear axle;

[0016] The Magic Tire model is:

[0017] ; (3)

[0018] in, is the peak factor, is the stiffness factor, is the curvature factor, is the shape factor, is the horizontal drift of the curve, is the vertical drift of the curve, is the slip rate, is the longitudinal force;

[0019] In the wheel coordinate system, the wheel rotation dynamics model is:

[0020] , ; (4)

[0021] Where: , are the wheel moment of inertia and wheel effective rolling radius respectively; is the rolling resistance of each tire; is the angular velocity of each wheel, and its acceleration is obtained by the corresponding sensor;

[0022] The linear two-degree-of-freedom reference model is:

[0023] ; (5)

[0024] Where: , are the cornering stiffness of the front and rear tires, respectively; is the sideslip angle of the center of mass.

[0025] Furthermore, the step 2 includes the following sub-steps:

[0026] Step 21: The vertical load of each tire is:

[0027] ; (6)

[0028] in, , , , They are the vertical loads of the left front, right front, left rear and right rear tires respectively. is the acceleration due to gravity, is the vehicle longitudinal acceleration, is the vehicle lateral acceleration, is the height of the vehicle's center of mass;

[0029] The longitudinal slip rate of each tire is:

[0030] ; (7)

[0031] in, is the tire slip rate, , is the longitudinal velocity of the wheel center;

[0032] According to the magic tire model, the longitudinal force of each tire is expressed as:

[0033] ; (8)

[0034] Step 22: According to the wheel rotation dynamics model, the motor torque corresponding to each tire It is expressed as:

[0035] , (9)

[0036] in:

[0037] ; (10)

[0038] Where: and Calculate the factor for rolling resistance;

[0039] Step 23: Failure parameters of the drive motor corresponding to each tire for:

[0040] , ; (11)

[0041] in, It is the control signal of each tire torque, which is obtained by the motor controller.

[0042] Furthermore, in step 3, the sliding surface function is defined for:

[0043] ; (12)

[0044] In the formula, , are the desired yaw rate and the desired sideslip angle, respectively. is a constant, and >0;

[0045] The design convergence law is as follows:

[0046] ; (13)

[0047] in, is the approach rate parameter.

[0048] Further, in step 4, and As the input of the fuzzy algorithm, the coefficient parameter of the power approach law is and exponential parameters As the output, the input variables are quantized to the fuzzy domain interval [-2,2], and the fuzzy sets are set to {NB, NS, ZO, PS, PB}, where NB is negative, NS is negative, ZO is zero, PS is positive, and PB is positive. The output variables are quantized to the fuzzy domain interval [0,2], and the fuzzy sets are {ZO, PS, PB}, where ZO is zero, PS is positive, and PB is positive. The triangular membership function is used to fuzzify the input and output variables. and Design fuzzy control rules, use Mamdani algorithm for fuzzy reasoning, and use the center method to clarify the output quantity to obtain the specific values ​​of the power approach law parameters, as shown below:

[0049] ; (14)

[0050] ; (15)

[0051] in, , Respectively The central value of a domain, For the input variable The degree of membership of a domain.

[0052] Furthermore, in step 5, the equality constraint is established as:

[0053] ; (16)

[0054] in:

[0055] (17)

[0056] in, is the control signal of each tire torque after redistribution, ;

[0057] Define the utilization rate of each tire as:

[0058] , ; (18)

[0059] in, is the utilization rate of each tire, is the tire-road friction coefficient;

[0060] Design tire utilization sum function :

[0061] ; (19)

[0062] Design torque distribution error function :

[0063] ; (20)

[0064] in, is the weight coefficient matrix, is the expected yaw moment;

[0065] The objective function and constraints of the designed torque redistribution are:

[0066] ;(twenty one)

[0067] The quadratic programming algorithm is used to solve the distribution and obtain the control signal of each tire torque after redistribution. .

[0068] Beneficial technical effects brought by the present invention:

[0069] The present invention provides a fault-tolerant control method for failure of a distributed electric vehicle drive motor. In order to improve the driving stability of a distributed electric vehicle under failure of a drive motor and suppress the conventional power approach law chattering phenomenon in a sliding mode algorithm, a vehicle yaw torque fuzzy sliding mode fault-tolerant control system is constructed in combination with a fuzzy algorithm. Compared with the conventional power approach law, the improved power approach law has a higher rate of movement of the system state toward the sliding mode surface, and can suppress the chattering problem that occurs at the origin of the conventional power approach law. Moreover, the system state reaches a steady state at the origin, effectively weakening the steady-state chattering, so that the system has better steady-state performance. The distributed electric vehicle under the fuzzy sliding mode fault-tolerant control using the improved power approach law responds more quickly and has better yaw motion characteristics. Compared with the sliding mode control method of the traditional power approach law, the vehicle reaches a steady state faster and has no overshoot. After the failure of the drive motor, the system can be quickly and accurately adjusted to a stable state, with superior dynamic response and significant improvement in robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 The present invention is a flow chart of a fault-tolerant control method for failure of a distributed electric vehicle drive motor.

[0071] Figure 2 It is a comparison diagram of the vehicle yaw angular velocity under the double lane change condition in the present invention.

[0072] Figure 3 It is a comparison diagram of the vehicle center of mass side slip angle under the double lane change condition in the present invention. DETAILED DESCRIPTION

[0073] The specific implementation of the present invention is further described below in conjunction with specific embodiments:

[0074] A fault-tolerant control method for failure of a drive motor of a distributed electric vehicle adopts a hierarchical fault-tolerant control strategy. The upper layer adopts fuzzy sliding mode yaw torque control based on an improved power reaching law, and the lower layer adopts torque distribution control based on a quadratic programming algorithm. The upper layer of the fault-tolerant control uses a fuzzy algorithm to adjust the power reaching law parameters in real time, so that the approach rate of the system in the sliding motion stage is associated with its own state, and the vibration amplitude in the sliding mode stage is suppressed; when the system state moves far away from the sliding surface, the values ​​of the coefficient parameter and the exponential parameter are increased to make the approach rate larger, so as to ensure that the system state quickly approaches the sliding surface; when the system state is close to the sliding surface, the values ​​of the coefficient parameter and the exponential parameter are reduced, so that the system state repeatedly crosses both sides of the sliding surface while continuously decreasing, and gradually reaches stability at the origin; the lower layer of the fault-tolerant control uses a tire model and a wheel dynamics model to diagnose the failure of the drive motor, and uses the failure parameters for torque distribution;

[0075] like Figure 1 As shown, the method specifically comprises the following steps:

[0076] Step 1: Construct a distributed electric vehicle dynamics model in the geodetic coordinate system, a magic tire model, a wheel rotation dynamics model in the wheel coordinate system, and a linear two-degree-of-freedom reference model;

[0077] Step 2: Obtain the failure parameters of the drive motor through drive motor failure diagnosis;

[0078] Step 3: Define the state variables and sliding surface functions of the system and design the reaching law;

[0079] Step 4: Use fuzzy algorithm to adjust the reaching law parameters of sliding mode algorithm in real time to stabilize the control signal;

[0080] Step 5: Redistribute the torque of each tire to obtain a control signal of the torque of each tire.

[0081] Specifically, in step 1, the distributed electric vehicle dynamics model in the geodetic coordinate system is:

[0082] ; (1)

[0083] in:

[0084] ; (2)

[0085] Where: , is the vehicle mass and the moment of inertia of the vehicle body; , , are the longitudinal speed, lateral speed and yaw rate respectively; , , They are the longitudinal force, lateral force and yaw moment of the vehicle respectively; is the front wheel turning angle; , , , They are the longitudinal forces of the left front, right front, left rear and right rear tires respectively. , , , They are the lateral forces of the left front, right front, left rear and right rear tires respectively; , , They are the wheelbase, the distance from the center of mass to the front axle, and the distance from the center of mass to the rear axle;

[0086] The Magic Tire model is:

[0087] ; (3)

[0088] in, is the peak factor, is the stiffness factor, is the curvature factor, is the shape factor, is the horizontal drift of the curve, is the vertical drift of the curve, is the slip rate, is the longitudinal force;

[0089] In the wheel coordinate system, the wheel rotation dynamics model is:

[0090] , ; (4)

[0091] Where: , are the wheel moment of inertia and wheel effective rolling radius respectively; is the rolling resistance of each tire; is the angular velocity of each wheel, and its acceleration is obtained by the corresponding sensor;

[0092] The linear two-degree-of-freedom reference model is:

[0093] ; (5)

[0094] Where: , are the cornering stiffness of the front and rear tires, respectively; is the sideslip angle of the center of mass.

[0095] Specifically, step 2 includes the following sub-steps:

[0096] Step 21: The vertical load of each tire is:

[0097] ; (6)

[0098] in, , , , They are the vertical loads of the left front, right front, left rear and right rear tires respectively. is the acceleration due to gravity, is the vehicle longitudinal acceleration, is the vehicle lateral acceleration, is the height of the vehicle's center of mass;

[0099] The longitudinal slip rate of each tire is:

[0100] ; (7)

[0101] in, is the slip rate of each tire, , is the longitudinal speed of the wheel center;

[0102] According to the magic tire model, the longitudinal force of each tire is expressed as:

[0103] ; (8)

[0104] Step 22: According to the wheel rotation dynamics model, the motor torque corresponding to each tire It is expressed as:

[0105] , (9)

[0106] in:

[0107] ; (10)

[0108] Where: and Calculate the factor for rolling resistance;

[0109] Step 23: Failure parameters of the drive motor corresponding to each tire for:

[0110] , ; (11)

[0111] in, It is the control signal of each tire torque, which is obtained by the motor controller.

[0112] Specifically, in step 3, the sliding surface function is defined as for:

[0113] ; (12)

[0114] In the formula, , are the desired yaw rate and the desired sideslip angle, respectively. is a constant, and >0;

[0115] The design convergence law is as follows:

[0116] ; (13)

[0117] in, is the approach rate parameter.

[0118] Specifically, in step 4, and As the input of the fuzzy algorithm, the coefficient parameter of the power approach law is and exponential parameters As the output, the input variables are quantized to the fuzzy domain interval [-2,2], and the fuzzy sets are set to {NB, NS, ZO, PS, PB}, where NB is negative, NS is negative, ZO is zero, PS is positive, and PB is positive. The output variables are quantized to the fuzzy domain interval [0,2], and the fuzzy sets are {ZO, PS, PB}, where ZO is zero, PS is positive, and PB is positive. The triangular membership function is used to fuzzify the input and output variables. and Design fuzzy control rules, as shown in Table 1 and Table 2:

[0119] Table 1 Parameters Fuzzy rule table

[0120] ;

[0121] Table 2 Parameters Fuzzy rule table

[0122] ;

[0123] Coefficient parameters The fuzzy rules are designed as follows:

[0124] (1) If For NB, NB, at this time, the system moving point is far away from the sliding surface and moves away from the sliding surface at a large speed. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0125] (2) If For NB, NS, at this time, the system moving point is far away from the sliding surface and moves away from the sliding surface at a small approaching rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0126] (3) If For NB, is ZO, at this time the system motion point is almost stationary far away from the sliding surface, design The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0127] (4) If For NB, PS, at this time, the system moving point is far away from the sliding surface and approaches the sliding surface at a smaller rate. The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0128] (5) If For NB, PB, at this time, the system moving point is far away from the sliding surface and approaches the sliding surface at a larger rate. The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger approaching rate;

[0129] (6) If For NS, NB, at this time, the system moving point is close to the sliding surface and moves away from the sliding surface at a larger rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0130] (7) If For NS, NS, at this time, the system moving point is close to the sliding surface and moves away from the sliding surface at a smaller rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0131] (8) If For NS, is ZO, at which point the system motion point is almost stationary near the sliding surface. The fuzzy rule is PS, which makes the system moving point approach the sliding mode surface and smoothly enter the sliding mode stage;

[0132] (9) If For NS, PS, at this time, the system moving point is close to the sliding surface and approaches the sliding surface at a smaller rate. The fuzzy rule is PS, reducing the parameter The value of makes the system decelerate continuously in the process of approaching the sliding surface and smoothly enter the sliding mode stage;

[0133] (10) If For NS, PB, at this time, the system moving point is close to the sliding surface and approaches the sliding surface at a larger rate. The fuzzy rule is PS, reducing the parameter The value of makes the system decelerate continuously in the process of approaching the sliding surface and smoothly enter the sliding mode stage;

[0134] (11) If is ZO, at this time the system motion point is near the sliding surface, and the approach rate should be as close to 0 as possible. As a coefficient parameter, the smaller its value is, the smaller the absolute value of the power approaching law is, so the design The fuzzy rule is ZO, which makes the system perform sliding mode motion while reducing the vibration amplitude;

[0135] (12) If For PS, NB, at this time, the system moving point is close to the sliding surface and approaches the sliding surface at a larger rate. The fuzzy rule is PS, reducing the parameter The value of makes the system decelerate continuously in the process of approaching the sliding surface and smoothly enter the sliding mode stage;

[0136] (13) If For PS, NS, at this time, the system moving point is close to the sliding surface and approaches the sliding surface at a smaller rate. The fuzzy rule is PS, reducing the parameter The value of makes the system decelerate continuously in the process of approaching the sliding surface and smoothly enter the sliding mode stage;

[0137] (14) If For PS, is ZO, at which point the system motion point is almost stationary near the sliding surface. The fuzzy rule is PS, which makes the system moving point approach the sliding mode surface and smoothly enter the sliding mode stage;

[0138] (15) If For PS, PS, at this time, the system moving point is close to the sliding surface and moves away from the sliding surface at a small rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0139] (16) If For PS, PB, at this time, the system moving point is close to the sliding surface and moves away from the sliding surface at a greater rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0140] (17) If For PB, NB, at this time, the system moving point is far away from the sliding surface and approaches the sliding surface at a larger rate. The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0141] (18) If For PB, NS, at this time, the system moving point is far away from the sliding surface and approaches the sliding surface at a smaller rate. The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0142] (19) If For PB, is ZO, at this time the system motion point is almost stationary far away from the sliding surface, design The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0143] (20) If is PB, ds / dt is PS, at this time, the system moving point is far away from the sliding surface and moves away from the sliding surface at a small rate, design The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0144] (21) If For PB, PB, at this time, the system moving point is far away from the sliding surface and moves away from the sliding surface at a greater rate. The fuzzy rule is PB, increasing the parameter The value of makes the system movement direction change rapidly.

[0145] Index parameters The fuzzy rules are designed as follows:

[0146] (1) If For NB, NB, at this time, the system moving point is far away from the sliding surface and moves away from the sliding surface at a large speed. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0147] (2) If For NB, NS, at this time, the system moving point is far away from the sliding surface and moves away from the sliding surface at a small approaching rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0148] (3) If For NB, is ZO, at this time the system motion point is almost stationary far away from the sliding surface, design The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0149] (4) If For NB, PS, at this time, the system moving point is far away from the sliding surface and approaches the sliding surface at a smaller rate. The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0150] (5) If For NB, PB, at this time, the system moving point is far away from the sliding surface and approaches the sliding surface at a larger rate. The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger approaching rate;

[0151] (6) If For NS, NB, at this time, the system moving point is close to the sliding surface and moves away from the sliding surface at a larger rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0152] (7) If For NS, NS, at this time, the system moving point is close to the sliding surface and moves away from the sliding surface at a smaller rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0153] (8) If For NS, is ZO, at which point the system motion point is almost stationary near the sliding surface. The fuzzy rule is PS, which makes the system moving point approach the sliding mode surface and smoothly enter the sliding mode stage;

[0154] (9) If For NS, PS, at this time, the system moving point is close to the sliding surface and approaches the sliding surface at a smaller rate. The fuzzy rule is PS, reducing the parameter The value of makes the system decelerate continuously in the process of approaching the sliding surface and smoothly enter the sliding mode stage;

[0155] (10) If For NS, PB, at this time, the system moving point is close to the sliding surface and approaches the sliding surface at a larger rate. The fuzzy rule is ZO, reducing the parameter The value of makes the system decelerate continuously in the process of approaching the sliding surface and smoothly enter the sliding mode stage;

[0156] (11) If ZO, at this time the system motion point is near the sliding surface, the approach rate should be as close to 0 as possible. <1, As a power parameter, the larger its value, the smaller the absolute value of the power approaching law, so the design The fuzzy rule is PS, which makes the system perform sliding mode motion while reducing the vibration amplitude;

[0157] (12) If For PS, NB, at this time, the system moving point is close to the sliding surface and approaches the sliding surface at a larger rate. The fuzzy rule is ZO, reducing the parameter The value of makes the system decelerate continuously in the process of approaching the sliding surface and smoothly enter the sliding mode stage;

[0158] (13) If For PS, NS, at this time, the system moving point is close to the sliding surface and approaches the sliding surface at a smaller rate. The fuzzy rule is PS, reducing the parameter The value of makes the system decelerate continuously in the process of approaching the sliding surface and smoothly enter the sliding mode stage;

[0159] (14) If For PS, is ZO, at which point the system motion point is almost stationary near the sliding surface. The fuzzy rule is PS, which makes the system moving point approach the sliding mode surface and smoothly enter the sliding mode stage;

[0160] (15) If For PS, PS, at this time, the system moving point is close to the sliding surface and moves away from the sliding surface at a small rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0161] (16) If For PS, PB, at this time, the system moving point is close to the sliding surface and moves away from the sliding surface at a greater rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0162] (17) If For PB, NB, at this time, the system moving point is far away from the sliding surface and approaches the sliding surface at a larger rate. The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0163] (18) If For PB, NS, at this time, the system moving point is far away from the sliding surface and approaches the sliding surface at a smaller rate. The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0164] (19) If For PB, is ZO, at this time the system motion point is almost stationary far away from the sliding surface, design The fuzzy rule is PB, increasing the parameter The value of makes the system approach the sliding surface at a larger rate;

[0165] (20) If For PB, PS, at this time, the system moving point is far away from the sliding surface and moves away from the sliding surface at a smaller rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0166] (21) If For PB, PB, at this time, the system moving point is far away from the sliding surface and moves away from the sliding surface at a greater rate. The fuzzy rule is PB, increasing the parameter The value of causes the system's motion direction to change rapidly;

[0167] The Mamdani algorithm is used for fuzzy reasoning, and the output is clarified using the center method to obtain the specific values ​​of the power approach law parameters, as shown below:

[0168] ; (14)

[0169] ; (15)

[0170] in, , Respectively The central value of a domain, For the input variable The degree of membership of a domain.

[0171] Specifically, in step 5, the equality constraint is established as:

[0172] ; (16)

[0173] in:

[0174] ; (17)

[0175] in, is the control signal of each tire torque after redistribution, ;

[0176] Define the utilization rate of each tire as:

[0177] , ; (18)

[0178] in, is the utilization rate of each tire, is the tire-road friction coefficient;

[0179] Design tire utilization sum function :

[0180] ; (19)

[0181] Design torque distribution error function :

[0182] ; (20)

[0183] in, is the weight coefficient matrix, is the expected yaw moment;

[0184] The objective function and constraints of the designed torque redistribution are:

[0185] ;(twenty one)

[0186] The quadratic programming algorithm is used to solve the distribution and obtain the control signal of each tire torque after redistribution. .

[0187] Figure 2 , Figure 3 The figures are respectively the vehicle yaw rate and center of mass sideslip angle in single wheel failure mode after adopting fault-tolerant control. It can be seen that the vehicle yaw stability is improved after adopting fault-tolerant control. Compared with the normal situation, the peak value of the vehicle yaw rate increased by 12.9% and the peak value of the center of mass sideslip angle increased by 6.5% when the sliding mode fault-tolerant control method was adopted. For the fuzzy sliding mode algorithm, the peak value of the vehicle yaw rate increased by 7.6% and the peak value of the center of mass sideslip angle increased by 1.1%, which shows that the fuzzy sliding mode algorithm can further improve the stability of vehicle driving.

[0188] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A fault-tolerant control method for failure of a distributed electric vehicle drive motor, characterized in that: The following steps are involved: Step 1: Construct a distributed electric vehicle dynamics model in the geodetic coordinate system, a magic tire model, a wheel rotation dynamics model in the wheel coordinate system, and a linear two-degree-of-freedom reference model; Step 2: Obtain the failure parameters of the drive motor through drive motor failure diagnosis; Step 3: Define the state variables and sliding surface functions of the system and design the reaching law; Step 4: Use fuzzy algorithm to adjust the reaching law parameters of sliding mode algorithm in real time to stabilize the control signal; Step 5: Redistribute the torque of each tire to obtain a control signal of the torque of each tire; In step 3, define the sliding surface function for: ; In the formula, , are the desired yaw rate and the desired sideslip angle, is a constant, and >0; The design convergence law is as follows: ; in, is the approach rate parameter; In step 4, and As the input of the fuzzy algorithm, the coefficient parameter of the power approach law is and exponential parameters As the output, the input variables are quantized to the fuzzy domain interval [-2,2], and the fuzzy sets are set to {NB, NS, ZO, PS, PB}, where NB is negative, NS is negative, ZO is zero, PS is positive, and PB is positive. The output variables are quantized to the fuzzy domain interval [0,2], and the fuzzy sets are {ZO, PS, PB}, where ZO is zero, PS is positive, and PB is positive. The triangular membership function is used to fuzzify the input and output variables. and Design fuzzy control rules, use Mamdani algorithm for fuzzy reasoning, and use the center method to clarify the output quantity to obtain the specific values ​​of the power approach law parameters, as shown below: ; ; in, , Respectively The central value of a domain, For the input variable The degree of membership of a domain.

2. The fault-tolerant control method for failure of a distributed electric vehicle drive motor according to claim 1, characterized in that: In step 1, the dynamic model of distributed electric vehicles in the geodetic coordinate system is: ; in: ; Where: , is the vehicle mass and the body moment of inertia; , , are the longitudinal speed, lateral speed and yaw rate respectively; , , They are the longitudinal force, lateral force and yaw moment of the vehicle respectively; is the front wheel turning angle; , , , They are the longitudinal forces of the left front, right front, left rear and right rear tires respectively. , , , They are the lateral forces of the left front, right front, left rear and right rear tires respectively; , , They are the wheelbase, the distance from the center of mass to the front axle, and the distance from the center of mass to the rear axle; The Magic Tire model is: ; in, is the peak factor, is the stiffness factor, is the curvature factor, is the shape factor, is the horizontal drift of the curve, is the vertical drift of the curve, is the slip rate, is the longitudinal force; In the wheel coordinate system, the wheel rotation dynamics model is: , ; Where: , are the wheel moment of inertia and wheel effective rolling radius respectively; is the rolling resistance of each tire; is the angular velocity of each wheel, and its acceleration is obtained by the corresponding sensor; The linear two-degree-of-freedom reference model is: ; Where: , are the cornering stiffness of the front and rear tires, respectively; is the sideslip angle of the center of mass.

3. The fault-tolerant control method for failure of a distributed electric vehicle drive motor according to claim 2, characterized in that: The step 2 includes the following sub-steps: Step 21: The vertical load of each tire is: ; in, , , , They are the vertical loads of the left front, right front, left rear and right rear tires respectively. is the acceleration due to gravity, is the vehicle longitudinal acceleration, is the vehicle lateral acceleration, is the height of the vehicle's center of mass; The longitudinal slip rate of each tire is: ; in, is the slip rate of each tire, , is the longitudinal speed of the wheel center; According to the magic tire model, the longitudinal force of each tire is expressed as: ; Step 22: According to the wheel rotation dynamics model, the motor torque corresponding to each tire It is expressed as: , ; in: ; Where: and Calculate the factor for rolling resistance; Step 23: Failure parameters of the drive motor corresponding to each tire for: , ; in, It is the control signal of each tire torque, which is obtained by the motor controller.

4. The fault-tolerant control method for failure of a distributed electric vehicle drive motor according to claim 3 is characterized in that: In step 5, the equality constraint is established as: ; in: ; in, is the control signal of each tire torque after redistribution, ; Define the utilization rate of each tire as: , ; in, is the utilization rate of each tire, is the tire-road friction coefficient; Design tire utilization sum function : ; Design torque distribution error function : ; in, is the weight coefficient matrix, is the expected yaw moment; The objective function and constraints of the designed torque redistribution are: ; The quadratic programming algorithm is used to solve the distribution and obtain the control signal of each tire torque after redistribution. .

Citation Information

Patent Citations

  • Sliding-mode control method for parameter-free driving-insufficient UUV (Unmanned Underwater Vehicle) vertical plane route tracking

    CN106444794A

  • Permanent magnet synchronous motor fuzzy sliding mode control system and method with improved exponential approach law

    CN109951122A

  • Permanent magnet synchronous motor control method and system based on speed change approach rate

    CN110061671A

  • Permanent magnet synchronous motor control system based on fuzzy sliding membrane variable structure

    CN110266227A

  • Dead-beat direct torque control and implementation method based on sliding mode strategy

    CN113258833A