Electro-hydraulic coordination control method for braking mode switching process of electric vehicle

Abstract

The invention discloses an electro-hydraulic coordination control method for a braking mode switching process of an electric vehicle, which realizes optimal distribution of electro-hydraulic braking force by comprehensively considering braking stability and energy recovery rate through a multi-objective optimization algorithm, and meanwhile, controls a motor to compensate hydraulic torque through a seagull optimized model predictive control algorithm controller, so that the braking mode switching process of the electric vehicle is optimized. The difference of dynamic response characteristics of a motor braking system and a hydraulic braking system can be eliminated, and the impact degree and the impact duration in the electro-hydraulic switching process are reduced to the maximum extent.

Description

Technical Field

[0001] The invention relates to the technical field of automobile control, and in particular to an electro-hydraulic coordinated control method for a braking mode switching process of an electric vehicle. Background Art

[0002] With the current energy crisis and environmental pollution becoming increasingly serious, electric vehicles have received widespread attention due to their zero emissions, high efficiency, and independence from conventional energy. The electro-hydraulic composite braking system consists of a hydraulic braking system and an electric motor braking system. By adopting appropriate braking control strategies during the braking process, braking modes such as pure hydraulic braking, pure electric braking, and electro-hydraulic composite braking can be achieved. During the electro-hydraulic and hydraulic composite braking, the difference in dynamic response characteristics between the motor and the hydraulics will lead to changes in the braking intensity, which will cause torque fluctuations during the switching of the braking mode, thus causing a certain degree of impact on the system components and the vehicle, resulting in excessive braking impact and prolonged impact duration, thus affecting the stability and comfort of the vehicle's braking. Therefore, the goal of the electro-hydraulic composite braking control strategy is to coordinate hydraulic braking and electric motor braking, and ensure the stability of the vehicle during braking while achieving braking energy recovery.

[0003] Most of the current electro-hydraulic composite brake control systems do not consider the difference in time domain response characteristics between hydraulic braking and motor regenerative braking, resulting in a large impact when the electric motor force is added and exited. The Chinese invention patent "A coordinated control system and method for electric vehicle electro-hydraulic composite brake torque fluctuation" with application number CN202111352746.5 published on January 21, 2022 uses the sensor on the brake pedal to obtain the displacement information of the brake pedal, and distributes the appropriate motor and hydraulic braking torque through the braking force distribution unit, and finally uses the motor force to compensate for the difference between the required hydraulic braking force and the actual hydraulic braking force, reducing the torque fluctuation problem caused by the electro-hydraulic composite brake system. The Chinese invention patent "A coordinated control strategy for electro-hydraulic composite brake torque based on type II fuzzy logic" with application number CN202310659609.9 published on August 11, 2023 uses the braking intensity z and battery SOC as fuzzy logic inputs to distribute the regenerative braking torque and hydraulic torque, and uses the motor to complete the compensation of the hydraulic torque when the road adhesion coefficient changes suddenly. However, the above two inventions do not fully consider the driver's braking operation factors during the braking process. During the compound braking process, when the motor and hydraulic system modes are switched, the car is in a constant braking intensity. When the car is in a variable braking intensity, the braking impact intensity and impact duration will change to a certain extent. In particular, when the braking mode is switched, excessive braking impact and long impact duration will affect the stability and comfort of the compound braking process. Summary of the invention

[0004] The present invention aims to solve the problem of excessive impact and long duration of the whole vehicle during mode switching of the electric-hydraulic composite braking process of an electric vehicle, and provides an electro-hydraulic coordinated control method for the braking mode switching process of an electric vehicle.

[0005] To solve the above problems, the present invention is achieved through the following technical solutions:

[0006] An electro-hydraulic coordinated control method for a braking mode switching process of an electric vehicle comprises the following steps:

[0007] Step 1: Obtain the braking intensity z of the electric vehicle, and calculate the required total braking force F of the electric vehicle based on the braking intensity z. b ; The total braking force required is F b for:

[0008] F b =mgz-F w -F f -F i

[0009] Step 2: Establish a multi-objective optimization model with braking stability and energy recovery rate as the main objectives; the multi-objective optimization model is:

[0010]

[0011] Step 3: by giving the constraints of the braking force distribution coefficient β and the motor force proportional coefficient k to be solved in the multi-objective optimization model established in step 2, a multi-objective genetic algorithm is used to solve the multi-objective optimization model to obtain a solution set of the multi-objective optimization model, that is, a solution set of multiple sets of braking force distribution coefficients and motor force proportional coefficients {(β o ,k o )};

[0012] Step 4: Use the ideal decision method to calculate the solutions of each group (β o ,k o ) and the ideal solution (β ′ ,k ′ ), and the solution with the smallest relative Euclidean distance is selected as the final decision solution, thereby obtaining the braking force distribution coefficient β and the motor force proportional coefficient k; the relative Euclidean distance d is:

[0013]

[0014] Step 5: Calculate the expected motor braking torque T according to the braking force distribution coefficient β and the motor braking force proportional coefficient k obtained in step 4. m_req and the desired hydraulic braking torque T h_req ;in

[0015] Expected motor torque T m_req for:

[0016] T m_req =F b ×β×k×r

[0017] Expected hydraulic braking torque T h_req for:

[0018] T h_req =F b ×T m_req

[0019] Step 6: Set the expected hydraulic braking torque T h_req It is sent to the hydraulic predictive control algorithm based on fuzzy control to obtain the actual hydraulic braking torque T h ;

[0020] Step 7: According to the expected motor torque T m_req , expected hydraulic braking torque T h_req and the actual hydraulic braking torque T h , calculate the target motor torque T′ m_req ; Where the target motor torque T′ m_req for:

[0021]

[0022] Step 8: Set the target motor torque T' m_req The actual motor torque T is obtained by inputting it into the motor predictive control algorithm based on model predictive control and Seagull optimization. m ;

[0023] Step 9: Based on the actual hydraulic braking torque T b and the actual motor torque T m Control the hydraulic braking system and motor braking system of electric vehicles;

[0024] In the formula, F b is the total braking force required by the electric vehicle, z is the braking strength of the electric vehicle, m is the mass of the electric vehicle, g is the acceleration of gravity, and F w is the air resistance of the electric vehicle, F g is the rolling resistance of electric vehicles, F i is the slope resistance of the electric vehicle; minf1 is the first optimization target, minf2 is the second optimization target, β is the braking force distribution coefficient of the electric vehicle, k is the electric braking force proportional coefficient of the electric vehicle, a is the distance from the center of mass of the electric vehicle to the center line of the front axle, b is the distance from the center of mass of the electric vehicle to the center line of the rear axle, L is the distance from the center line of the front axle to the center line of the rear axle of the electric vehicle, hg is the height of the center of mass of the electric vehicle, r is the motor power arm of the electric vehicle, η m is the motor power generation efficiency of the electric vehicle, n is the motor speed of the electric vehicle, θ is the motor power conversion coefficient of the electric vehicle; β o is the oth solution of the braking force distribution coefficient, k o is the oth solution of the motor force proportional coefficient, o = 1, 2, ... O, O is the number of solution groups; d is the relative Euclidean distance, β′ is the ideal solution of the braking force distribution coefficient, k′ is the ideal solution of the motor force proportional coefficient; T m_req is the expected motor torque, T h_req is the expected hydraulic braking torque; T h is the actual hydraulic braking torque, T′ m_req is the target motor torque.

[0025] The specific process of step 6 above is as follows:

[0026] Step 6.1, initialization, let the current iteration number t = 1, the last actual hydraulic braking torque T h (0) = 0;

[0027] Step 6.2: According to the current actual hydraulic braking torque T h (t) and the expected hydraulic braking torque T h_req , calculate the current hydraulic braking torque error E(t) and the current hydraulic braking torque error change rate ΔE(t); where

[0028] The current hydraulic braking torque error E(t) is:

[0029] E(t)=T h_req -T h (t-1)

[0030] The current hydraulic braking torque error change rate ΔE(t) is:

[0031]

[0032] Step 6.3: Send the current hydraulic braking torque error E(t) and the current hydraulic braking torque error change rate ΔE(t) to the fuzzy controller to obtain the current proportional coefficient k p (t), current integral coefficient k i (t) and the current differential coefficient k d (t);

[0033] Step 6.4: According to the current hydraulic braking torque error E(t) and the current proportional coefficient k p (t), current integral coefficient k i (t) and the current differential coefficient kd (t), calculate the current hydraulic pressure torque control signal u h (t); where the current hydraulic pressure torque control signal u h (t) is:

[0034]

[0035] Step 6.5: Based on the dynamic response of the hydraulic brake system, control the signal u according to the current hydraulic pressure torque h (t) Calculate and update the current actual hydraulic braking torque T h (t); where the current actual hydraulic braking torque T h (t) is:

[0036] T h (t) = u h (t)×e -5t

[0037] Step 6.6: Determine whether the iteration termination condition is reached, that is, the current actual hydraulic braking torque T h (t) is equal to the desired hydraulic braking torque T h_req Or the current actual hydraulic braking torque T h (t) No change after a certain preset number of times:

[0038] If yes, the current actual hydraulic braking torque T h (t) is the actual hydraulic braking torque T required h Output;

[0039] Otherwise, set the number of iterations to t+1 and return to step 6.2;

[0040] Where, E(t) is the current hydraulic braking torque error, T h_req is the expected hydraulic braking torque, T h (t-1) is the last actual hydraulic braking torque, ΔE(t) is the current hydraulic braking torque error change rate, T represents time; u h (t) is the current hydraulic pressure torque control signal, k p (t) is the current proportional coefficient, k i (t) is the current integral coefficient, k d (t) is the current differential coefficient; T h (t) is the current actual hydraulic braking torque, and t is the current iteration number.

[0041] In the above step 6.3, the fuzzy PID rule base in the fuzzy controller is:

[0042]

[0043] In the table, E(t) represents the current hydraulic braking torque error, ΔE(t) represents the current hydraulic braking torque error change rate, k p (t) represents the current proportional coefficient, k i (t) represents the current integral coefficient, k d (t) represents the current differential coefficient, NB represents large negative, NM represents medium negative, NS represents small negative, ZO represents zero, PS represents small positive, PM represents medium positive, and PB represents large positive.

[0044] The specific process of step 8 above is as follows:

[0045] Step 8.1, initialization, let the current iteration number t = 1, the last actual motor torque T m (0) = 0, the last motor torque control signal u m (0) = 0;

[0046] Step 8.2: Use the Seagull optimization algorithm to solve the weight matrix Q(t) of the output error and the weight matrix R(t) of the control input change by minimizing the fitness function J1; the fitness function J1 is:

[0047]

[0048] Step 8.3: Use the model predictive control algorithm to solve the current motor torque control signal u by minimizing the cost function J2 m (t); where the cost function J2 is:

[0049] J2=(T′ m_req -T m (t-1)) T ×Q(t)×(T′ m_req -T m (t-1))

[0050] +(u m (t)-u m (t-1)) T ×R(t)×(u m (t)-u m (t-1))

[0051] Step 8.4: Based on the speed response of the motor brake system, control the current motor torque signal u m (t) Calculate and update the current actual motor torque T m (t); where the current actual motor torque T m (t) is:

[0052] T m (t) = u m (t)×e-100t

[0053] Step 8.5: Determine whether the iteration termination condition is reached, that is, the current actual motor torque T m (t) is equal to the target motor torque T′ m_req Or the current actual motor torque T m (t) No change after a certain preset number of times:

[0054] If yes, then the current actual motor torque T m (t) is the actual motor torque T m Output;

[0055] Otherwise, set the number of iterations to t+1 and return to step 8.2;

[0056] Where v is the current speed of the electric vehicle, γ is the input matrix weight coefficient, α is the output matrix weight coefficient, Q(t) is the output error weight matrix, R(t) is the control input change weight matrix, and T represents time; u m (t) is the current motor torque control signal, u m (t-1) is the last motor torque control signal, T′ m_req is the target motor torque, T m (t-1) is the last actual motor torque, (■) T represents transpose; T m (t) is the current actual motor torque, and t is the current iteration number.

[0057] Compared with the prior art, the present invention has the following characteristics:

[0058] 1. The present invention uses a multi-objective optimization algorithm to comprehensively consider braking stability and energy recovery rate to achieve optimal distribution of electro-hydraulic braking force. Compared with the traditional rule-based electro-hydraulic braking force distribution method, the present invention can ensure braking safety while taking into account energy recovery rate and braking stability, achieve multi-objective optimization, and has stronger practicality.

[0059] 2. In the traditional electro-hydraulic brake system control method, the motor is added or withdrawn when the braking intensity is constant, and the situation of changing braking intensity is not considered. Based on the four working conditions where the braking intensity is constant or changing when the motor is added or withdrawn, the present invention designs an electro-hydraulic coordinated control strategy for the braking mode switching process of electric vehicles. The motor is controlled by the Seagull optimized MPC controller to compensate for the hydraulic torque, which can eliminate the difference in dynamic response characteristics between the motor brake system and the hydraulic brake system, and minimize the impact degree and impact duration of the electro-hydraulic switching process. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of the dynamic coordinated control principle of the present invention.

[0061] Figure 2 This is a vehicle braking force analysis diagram.

[0062] Figure 3 Graph for multi-objective optimization.

[0063] Figure 4 Relationship diagram of front and rear axle braking forces.

[0064] Figure 5 These are the membership function diagrams of the fuzzy controller, (a) the membership function diagram of the braking intensity z, (b) the membership function diagram of the vehicle speed v, (c) the membership function diagram of the battery SOC, and (d) the membership function diagram of the motor torque proportional coefficient k.

[0065] Figure 6 The result diagram of the motor exiting the constant braking intensity is as follows: (a) vehicle speed, (b) total torque on the front axle, and (c) impact degree.

[0066] Figure 7 The result diagram of motor exit under variable braking intensity, (a) vehicle speed, (b) total torque on the front axle, and (c) impact degree.

[0067] Figure 8 The result diagram of the motor under constant braking intensity is as follows: (a) vehicle speed, (b) total torque on the front axle, and (c) impact degree.

[0068] Fig. 9 The result diagram of the motor with variable braking intensity: (a) vehicle speed, (b) total torque on the front axle, and (c) impact degree. DETAILED DESCRIPTION

[0069] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in combination with specific examples and with reference to the accompanying drawings.

[0070] Considering that the current electro-hydraulic composite brake system considers the addition or withdrawal of the motor when the braking intensity is constant, and does not consider the situation where the braking intensity changes when the motor is added or withdrawn, the present invention proposes an electro-hydraulic coordinated control method for the braking mode switching process of an electric vehicle, such as Figure 1 As shown, the steps include:

[0071] Step 1: Obtain the braking intensity z of the electric vehicle, and calculate the required total braking force F of the electric vehicle based on the braking intensity z. b .

[0072] During the braking process of an electric vehicle, the driver applies the brake pedal to decelerate the electric vehicle. The deceleration resistance ∑F required by the electric vehicle is determined according to the braking intensity z during braking:

[0073] ∑F=mgz

[0074] Where ∑F is the total resistance required for deceleration, m is the mass of the electric vehicle, g is the acceleration of gravity, and z is the braking strength of the electric vehicle.

[0075] In the present invention, the electric motor of the electric vehicle acts on the front axle, and the hydraulic brake acts on the front axle and the rear axle at the same time, so the required total braking force F of the electric vehicle is b Equal to the front axle braking force (F reg +F hyb ) and rear axle braking force F br The sum of reg To power the electric vehicle, F hyb is the front axle hydraulic braking force of the electric vehicle, F br is the hydraulic braking force of the rear axle of the electric vehicle. At this time, according to the vehicle driving equation during the braking process, the total braking force required by the electric vehicle during the braking process can be obtained: b :

[0076] F b =(F reg +F hyb )+F br =∑FF w -F f -F i

[0077] In the formula, F b is the total braking force required by the electric vehicle, F w is the air resistance of the electric vehicle, F f is the rolling resistance of electric vehicles, F i is the slope resistance of the electric vehicle.

[0078] Step 2: Establish a multi-objective optimization model with braking stability and energy recovery rate as the main objectives.

[0079] The braking force distribution strategy of the present invention mainly considers the two goals of braking stability and energy recovery rate, and reasonably distributes the front and rear axle braking forces and the motor hydraulic braking force according to the specific braking conditions, thereby achieving the purpose of braking stability and energy recovery. The braking force distribution coefficient β is the front axle braking force (F reg +F hyb ) and the required total braking force F bThe ratio of the motor force F to the front and rear axle brake force reflects the distribution of the front and rear axle brake force, and its magnitude has an important influence on the braking stability. reg and front axle braking force (F reg +F hyb ), which reflects the distribution of the motor and hydraulic braking forces, and its magnitude has an important impact on the recovery of braking energy. To this end, the present invention selects the braking force distribution coefficient β and the motor force proportional coefficient k as the control variables of the multi-objective optimization, and establishes a multi-objective function based on them.

[0080] Since the most ideal braking state of an electric vehicle is when the front and rear wheels are locked at the same time, the braking efficiency is the highest and the braking safety is the best. Therefore, the present invention sets the first optimization target to minimize the deviation between the front and rear axle adhesion coefficient and the braking strength. The expression for establishing the first optimization target minf1 is:

[0081]

[0082] Where β is the braking force distribution coefficient of the electric vehicle, z is the braking strength of the electric vehicle, a is the distance from the center of mass of the electric vehicle to the center line of the front axle, b is the distance from the center of mass of the electric vehicle to the center line of the rear axle, L is the distance from the center line of the front axle to the center line of the rear axle of the electric vehicle, and h is g is the center of mass height of the electric vehicle.

[0083] Since the braking energy recovery of electric vehicles is achieved through the generation of power by the drive motor, the effective power generation of the drive motor can characterize the adequacy of the braking energy recovery. Therefore, the present invention takes the power generation of the drive motor as the second optimization target. Considering that the higher the power generation, the more energy is recovered, but when establishing the objective function, the minimum value of the optimization target is generally taken as the optimal target. Therefore, the present invention takes a negative value for the power generation and establishes the expression of the second optimization target minf2 as follows:

[0084]

[0085] Where k is the electric vehicle's braking force proportional coefficient, β is the electric vehicle's braking force distribution coefficient, and F b is the total braking force required by the electric vehicle, r is the braking force arm of the electric vehicle, ηm is the motor power generation efficiency of the electric vehicle, n is the motor speed of the electric vehicle, and θ is the motor power conversion coefficient of the electric vehicle (generally θ=9549).

[0086] Accordingly, the multi-objective optimization model constructed by the present invention is:

[0087]

[0088] Where minf1 is the first optimization objective, minf2 is the second optimization objective, β is the braking force distribution coefficient of the electric vehicle to be solved, and k is the motor braking force proportional coefficient of the electric vehicle to be solved.

[0089] Step 3: By giving the constraints of the braking force distribution coefficient β and the motor force proportional coefficient k to be solved in the multi-objective optimization model, a multi-objective genetic algorithm is used to solve the multi-objective optimization model to obtain a solution set of the multi-objective optimization model.

[0090] The constraint condition of the braking force distribution coefficient β is set based on the braking stability and braking efficiency requirements of relevant regulations. In this embodiment, by analyzing the force of the electric vehicle braking process (such as Figure 2 As shown), combined with the United Nations Economic Commission for Europe automobile regulations (ECE), the braking force distribution coefficient β is obtained as:

[0091]

[0092] The constraint condition of the motor torque proportional coefficient k is set based on the motor's own external characteristics. In this embodiment, the motor's own external characteristics require the expected motor torque T m_req It should be smaller than the maximum torque T that the motor can output. regmax ,Right now:

[0093] T m_req =F b ×β×k×r<T regmax

[0094] After determining the constraints and the objective function, the present invention adopts the NSGA-II algorithm (multi-objective genetic algorithm) to solve the multi-objective optimization model and obtains the solution set of the multi-objective optimization model, that is, the solution of multiple groups of braking force distribution coefficients β and motor force proportional coefficients k {(β o ,k o )}, where o = 1, 2, ... O, O is the number of solution groups. In this implementation, the NSGA-II algorithm uses the fast non-dominated sorting genetic algorithm function gamultiobj with an elite strategy to solve the multi-objective optimization model, which has the characteristics of simple algorithm calculation and good convergence. Set the optimal front-end individual coefficient ParetoFraction = 0.3, population size PopulationSize = 100, maximum genetic generation number Generations = 200, and fitness function deviation TolFun = 0.01 to solve the multi-objective optimization model. Taking the braking condition of vehicle speed v = 50km / h and braking intensity z = 0.1 as an example, the results of multi-objective optimization are as follows Figure 3 shown.

[0095] Step 4: Use the ideal decision method to calculate the solutions of each group (βo ,k o ) and the ideal solution (β ′ ,k ′ ) and select the solution with the smallest relative Euclidean distance as the final decision solution, thereby obtaining the optimal braking force distribution coefficient β and the optimal motor force proportional coefficient k.

[0096] The ideal decision method requires an ideal solution to be given in advance. In the present invention, the ideal solution β of the braking force distribution coefficient β ′ and the ideal solution k of the motor force proportionality coefficient k ′ The results are summarized through a large number of experiments and analyses in the early stage. For example:

[0097] In determining the ideal solution of the braking force distribution coefficient β ′ When the braking safety ECE regulations, the ideal braking force curve I and the front wheel anti-lock curve f are combined, the following is obtained: Figure 4 The front and rear axle braking force relationship diagram, where point O is the origin of the coordinate system, point A is the intersection of the ECE regulation and the horizontal axis, point B is the tangent point of the β line and the ECE curve, point C is the intersection of the curve and the I curve, and point D is a point on the I curve. While ensuring braking safety, as much energy recovery as possible is performed, and as much torque as possible is distributed to the front axle. The β value should be as large as possible, so as to perform as much motor braking as possible. Therefore, the distribution ratio is dynamically optimized along OA, AB, BC, and CD, that is, the ideal solution of the braking force distribution coefficient β is β ′ for:

[0098]

[0099] In the process of energy recovery, since the energy recovery efficiency is affected by multiple factors such as braking intensity z, vehicle speed v, battery SOC, and the participation of motor braking is uncertain, in order to determine the ideal solution of the motor braking force proportional coefficient under different braking states, the present invention uses braking intensity z, vehicle speed v, and battery SOC as the three input variables of the fuzzy controller, and the motor braking force proportional coefficient k ′ As the output variable of the fuzzy controller. The fuzzy controller constructs a membership function for these three input variables: for the braking intensity z, it is divided into different intervals of light braking, medium braking, and heavy braking, and a corresponding membership function is determined for each interval to reflect the degree of influence on k under different braking intensities. For the vehicle speed v, it is also divided into low speed, medium speed, and high speed intervals, and the corresponding membership function is designed to reflect the effect on k under different vehicle speeds. For the battery SOC, it is divided into low power, medium power, and high power intervals, and their respective membership functions are determined, considering the restriction or promotion of the battery power state on the motor driving force proportional coefficient. In this embodiment, the three-input and one-output membership function is as follows Figure 5As shown in the figure, (a) the membership function diagram of braking intensity z, (b) the membership function diagram of vehicle speed v, (c) the membership function diagram of battery SOC, and (d) the membership function diagram of motor braking torque proportional coefficient k. The fuzzy controller designs fuzzy rules based on different input combinations. For example, when the braking intensity is light, the vehicle speed is medium, and the battery SOC is low, the motor braking force proportional coefficient can be set higher; when the braking intensity is heavy, the vehicle speed is high, and the battery SOC is low, the motor braking force proportional coefficient should be lower. By formulating comprehensive fuzzy rules to ensure that all possible input combinations can be covered, the fuzzy controller can accurately determine the ideal solution k′ of the motor braking force proportional coefficient k under various braking conditions.

[0100] When the ideal solution β′ of the braking force distribution coefficient β and the ideal solution k′ of the motor force proportional coefficient k are determined, the solutions (β o , k o ) and the ideal solution (β′, k′), and sort the relative Euclidean distances of each group of solutions. Therefore, the set with the smallest relative Euclidean distance has the smallest difference with the current actual state and is more in line with actual needs and constraints. Therefore, the solution with the smallest relative Euclidean distance is the decision solution, where the relative Euclidean distance d is:

[0101]

[0102] Step 5: Calculate the expected motor braking torque T according to the determined braking force distribution coefficient β and motor braking force proportional coefficient k. m_req and the desired hydraulic braking torque T h_req .

[0103] Expected motor torque T m_req for:

[0104] T m_req =F b ×β×k×r

[0105] Expected hydraulic braking torque T h_req for:

[0106] T h_req =F b ×T m_req

[0107] Step 6: Set the expected hydraulic braking torque T h_req It is sent to the hydraulic predictive control algorithm based on fuzzy control to obtain the actual hydraulic braking torque T h .

[0108] Step 6.1, initialization, let the current iteration number t = 1, the last actual hydraulic braking torque Th (0)=0.

[0109] Step 6.2: According to the current actual hydraulic braking torque T h (t) and the expected hydraulic braking torque T h_req , calculate the current hydraulic braking torque error E(t) and the current hydraulic braking torque error change rate ΔE(t).

[0110] The current hydraulic braking torque error E(t) is:

[0111] E(t)=T h_req -T h (t-1)

[0112] The current hydraulic braking torque error change rate ΔE(t) is:

[0113]

[0114] Step 6.3: Send the current hydraulic braking torque error E(t) and the current hydraulic braking torque error change rate ΔE(t) to the fuzzy controller to obtain the current proportional coefficient k p (t), current integral coefficient k i (t) and the current differential coefficient k d (t).

[0115] In the fuzzy optimization PID control process, the fuzzy rule base performs online self-tuning on the proportional coefficient, integral coefficient and differential coefficient of the PID controller according to different torque errors E(t) and torque error change rates ΔE(t).

[0116] If the error E(t) and the error change rate ΔE(t) are large, the fuzzy rule base will adjust the proportional coefficient kp to a larger value to increase the response speed, but it should be noted that it may cause a large overshoot of the system; the fuzzy rule base will adjust the integral coefficient k accordingly i Take a smaller value, because at this time, the larger the error and the error change rate, the larger k i It may lead to integral saturation and cause the system to overshoot; the fuzzy rule base has an influence on the differential coefficient k d It needs to be adjusted according to the actual situation. d The value may make the system regulation advance, and the smaller k d Values ​​may cause system regulation lag.

[0117] When the error E(t) and the error change rate ΔE(t) are small, the fuzzy rule base will adjust the proportional coefficient k accordingly. p Take a smaller value to ensure system stability and avoid excessive overshoot; the fuzzy rule base will adjust the integral coefficient k accordingly i Take a larger value to eliminate the system steady-state error; the fuzzy rule base will adjust the differential coefficient kd Make fine adjustments to eliminate system oscillations.

[0118] According to the above adjustment experience, in this embodiment, the fuzzy PID rule base in the fuzzy controller is shown in Table 1.

[0119] Table 1 Fuzzy PID rule base

[0120]

[0121] In the table: the input is E(t) and ΔE(t), and the output is k p , k i and k d The seven language variables are NB (negative large), NM (negative medium), NS (negative small), ZO (zero), PS (positive small), PM (positive medium), and PB (positive large).

[0122] Step 6.4: According to the current hydraulic braking torque error E(t) and the current proportional coefficient k p (t), current integral coefficient k i (t) and the current differential coefficient k d (t), calculate the current hydraulic pressure torque control signal u h (t). Perform PID control on the current hydraulic braking torque error E(t) to obtain the current hydraulic pressure torque control signal u h (t), and applies the control signal to the hydraulic brake system.

[0123] Current hydraulic pressure torque control signal u h (t) is:

[0124]

[0125] Wherein, T represents time.

[0126] Step 6.5: Based on the dynamic response of the hydraulic brake system, control the signal u according to the current hydraulic pressure torque h (t) Calculate and update the current actual hydraulic braking torque T h (t).

[0127] Dynamic response of hydraulic brake system using first-order inertia link Simulation, where T B is the time constant of the hydraulic braking system, which is taken as 0.2, and s is the complex variable in the Laplace transform. Then, the actual hydraulic braking torque T can be obtained by converting it into the time domain expression. h (t) is:

[0128] T h (t) = u h (t)×e -5t

[0129] Where t is the current iteration number.

[0130] Step 6.6: Determine whether the iteration termination condition is reached, that is, the current actual hydraulic braking torque T h (t) is equal to the desired hydraulic braking torque T h_req Or the current actual hydraulic braking torque T h (t) No change after a certain preset number of times:

[0131] If yes, the current actual hydraulic braking torque T h (t) is the actual hydraulic braking torque T required h Output;

[0132] Otherwise, set the number of iterations to t+1 and return to step 6.2.

[0133] Step 7: Based on the expected motor torque T m_req , expected hydraulic braking torque T h_req and the actual hydraulic braking torque T h , calculate the target motor torque T′ m_req .

[0134] Target motor torque T′ m_req for:

[0135] T′ m_req =T m_req +T h_req -T h

[0136] Step 8: Set the target motor torque T' m_req The actual motor torque T is obtained by inputting it into the motor predictive control algorithm based on model predictive control and Seagull optimization. m .

[0137] Step 8.1, initialization, let the current iteration number t = 1, the last actual motor torque T m (0) = 0, the last motor torque control signal u m (0)=0.

[0138] Step 8.2: Use the Seagull optimization algorithm to optimize the output error weight matrix Q(t) and the control input change weight matrix R(t).

[0139] The present invention uses the weight matrix Q of the output error to measure the importance of the deviation between the output and the expected output, and the weight matrix R of the control input change to measure the cost of the control input change. Its purpose is to accurately adjust the control strategy of the system during the motor control process so that the actual output can be as close to the expected output as possible. In order to obtain the optimal weight matrix Q of the output error and the weight matrix R of the control input change, the seagull optimization algorithm is used to achieve it.

[0140] The Seagull Optimization Algorithm (SOA) is a bionic optimization algorithm that simulates the migration and attack behavior of seagulls. It continuously adjusts individual positions and conducts local searches, and searches for the optimal solution in the solution space by minimizing the fitness function J1. It has strong global and local search capabilities.

[0141] The fitness function J1 of the Seagull optimization algorithm is:

[0142]

[0143] Among them, v is the vehicle speed, γ and α are the set weight coefficients used to balance the importance of different parts, Q(t) is the weight matrix of the output error, R(t) is the weight matrix of the control input change, and T represents time.

[0144] Step 8.3: Use the model predictive control algorithm to optimize and solve the current motor torque control signal u m (t).

[0145] Model predictive control algorithm (MPC) is an advanced control algorithm based on model. It uses system model to predict future state and determines the optimal control strategy by minimizing cost function J2. It can effectively handle multivariable constraint problems.

[0146] The cost function J2 of the model predictive control algorithm is:

[0147] J2=(T′ m_req -T m (t-1)) T ×Q(t)×(T′ m_req -T m (t-1))

[0148] +(u m (t)-u m (t-1)) T ×R(t)×(u m (t)-u m (t-1))

[0149] In the formula, u m (t) is the current motor torque control signal, u m(t-1) is the last motor torque control signal, T′ m_req is the target motor torque, T m (t-1) is the last actual motor torque, (■) T Indicates transpose.

[0150] Step 8.4: Based on the speed response of the motor brake system, control the current motor torque signal u m (t) Calculate and update the current actual motor torque T m (t).

[0151] Dynamic response of hydraulic brake system using first-order inertia link Simulation, where T c is the time constant of the motor braking system, which is taken as 0.01, and s is the complex variable in the Laplace transform, which is converted into the time domain expression. The current actual motor torque T m (t) is:

[0152] T m (t) = u m (t)×e -100t

[0153] Where t is the current iteration number.

[0154] Step 8.5: Determine whether the iteration termination condition is reached, that is, the current actual motor torque T m (t) is equal to the target motor torque T′ m_req Or the current actual motor torque T m (t) No change after a certain preset number of times:

[0155] If yes, then the current actual motor torque T m (t) is the actual motor torque T m Output; otherwise, set the number of iterations to t+1 and return to step 8.2.

[0156] Step 9: Based on the actual hydraulic braking torque T h and the actual motor torque T m The hydraulic brake system and motor brake system of the electric vehicle are controlled to achieve electro-hydraulic coordinated control of the electric vehicle.

[0157] The electro-hydraulic coordinated control method for the electric vehicle braking mode switching process implemented by the present invention can realize electro-hydraulic coordinated control under the condition that the braking intensity is constant or changes when the motor is added or removed, thereby reducing the impact of the whole vehicle during the control process. When the electro-hydraulic coordinated control strategy is tested for four working conditions where the braking intensity is constant or changes when the motor is added or removed, the test results are as follows: Figures 6 to 9 As shown, Figure 6 This is the result diagram of the motor exiting the constant braking intensity. Figure 7 This is the result diagram of the motor exiting the variable braking intensity. Figure 8 The result diagram of the motor with constant braking intensity is shown below. Fig. 9 The results of adding variable braking intensity to the motor show that the maximum impact reduction is 5.8m / s compared with the unoptimized coordinated control strategy. 3 , minimum reduction of 5m / s 3 , the impact duration is reduced by at least 0.05s. Compared with the uncoordinated control strategy, the maximum impact degree is reduced by 18.4m / s 3 , the minimum reduction is 13.9m / s 3 , the impact duration is reduced by at least 0.15s. Based on the test results, the designed control parameters can be further optimized to improve the performance of the control system.

[0158] It should be noted that although the embodiments of the present invention described above are illustrative, they are not intended to limit the present invention, and therefore the present invention is not limited to the above specific embodiments. Without departing from the principles of the present invention, any other embodiments obtained by those skilled in the art under the guidance of the present invention are deemed to be within the protection of the present invention.

Claims

1. An electro-hydraulic coordinated control method for the braking mode switching process of an electric vehicle, characterized in that: The steps include: Step 1: Obtain the braking intensity z of the electric vehicle, and calculate the required total braking force F of the electric vehicle based on the braking intensity z. b ; The total braking force required is F b for: F b =mgz-F w -F f -F i Step 2: Establish a multi-objective optimization model with braking stability and energy recovery rate as the main objectives; the multi-objective optimization model is: Step 3: by giving the constraints of the braking force distribution coefficient β and the motor force proportional coefficient k to be solved in the multi-objective optimization model established in step 2, a multi-objective genetic algorithm is used to solve the multi-objective optimization model to obtain a solution set of the multi-objective optimization model, that is, a solution set of multiple sets of braking force distribution coefficients and motor force proportional coefficients {(β o ,k o )}; Step 4: Use the ideal decision method to calculate the solutions of each group (β o ,k o ) and the ideal solution (β ′ ,k ′ ), and the solution with the smallest relative Euclidean distance is selected as the final decision solution, thereby obtaining the braking force distribution coefficient β and the motor force proportional coefficient k; the relative Euclidean distance d is: Step 5: Calculate the expected motor braking torque T according to the braking force distribution coefficient β and the motor braking force proportional coefficient k obtained in step 4. m_req and the desired hydraulic braking torque T h_req ;in Expected motor torque T m_req for: T m_req =F b ×β×k×r Expected hydraulic braking torque T h_req for: T h_req =F b ×r-T m_req Step 6: Set the expected hydraulic braking torque T h_req It is sent to the hydraulic predictive control algorithm based on fuzzy control to obtain the actual hydraulic braking torque T h ; Step 7: According to the expected motor torque T m_req , expected hydraulic braking torque T h_req and the actual hydraulic braking torque T h , calculate the target motor torque T′ m_req ; Where the target motor torque T′ m_req for: Step 8: Set the target motor torque T' m_req The actual motor torque T is obtained by inputting it into the motor predictive control algorithm based on model predictive control and Seagull optimization. m ; Step 9: Based on the actual hydraulic braking torque T h and the actual motor torque T m Control the hydraulic braking system and motor braking system of electric vehicles; In the formula, F b is the total braking force required by the electric vehicle, z is the braking strength of the electric vehicle, m is the mass of the electric vehicle, g is the acceleration of gravity, and F w is the air resistance of the electric vehicle, F f is the rolling resistance of electric vehicles, F i is the slope resistance of the electric vehicle; minf1 is the first optimization target, minf2 is the second optimization target, β is the braking force distribution coefficient of the electric vehicle, k is the electric braking force proportional coefficient of the electric vehicle, a is the distance from the center of mass of the electric vehicle to the center line of the front axle, b is the distance from the center of mass of the electric vehicle to the center line of the rear axle, L is the distance from the center line of the front axle to the center line of the rear axle of the electric vehicle, h g is the height of the center of mass of the electric vehicle, r is the motor power arm of the electric vehicle, η m is the motor power generation efficiency of the electric vehicle, n is the motor speed of the electric vehicle, θ is the motor power conversion coefficient of the electric vehicle; β o is the oth solution of the braking force distribution coefficient, k o is the oth solution of the motor force proportional coefficient, o = 1, 2, ... O, O is the number of solution groups; d is the relative Euclidean distance, β′ is the ideal solution of the braking force distribution coefficient, k ′ is the ideal solution of the motor force proportionality coefficient; T m_req is the expected motor torque, T h_req is the expected hydraulic braking torque; T h is the actual hydraulic braking torque, T′ m_req is the target motor torque.

2. The electro-hydraulic coordinated control method for the braking mode switching process of an electric vehicle according to claim 1, characterized in that: The specific process of step 6 is as follows: Step 6.1, initialization, let the current iteration number t = 1, the last actual hydraulic braking torque T h (0) = 0; Step 6.2: According to the current actual hydraulic braking torque T h (t) and the expected hydraulic braking torque T h_req , calculate the current hydraulic braking torque error E(t) and the current hydraulic braking torque error change rate ΔE(t); where The current hydraulic braking torque error E(t) is: E(t)=T h_req -T h (t-1) The current hydraulic braking torque error change rate ΔE(t) is: Step 6.3: Send the current hydraulic braking torque error E(t) and the current hydraulic braking torque error change rate ΔE(t) to the fuzzy controller to obtain the current proportional coefficient k p (t), current integral coefficient k i (t) and the current differential coefficient k d (t); Step 6.4: According to the current hydraulic braking torque error E(t) and the current proportional coefficient k p (t), current integral coefficient k i (t) and the current differential coefficient k d (t), calculate the current hydraulic pressure torque control signal u h (t); where the current hydraulic pressure torque control signal u h (t) is: Step 6.5: Based on the dynamic response of the hydraulic brake system, control the signal u according to the current hydraulic pressure torque h (t) Calculate and update the current actual hydraulic braking torque T h (t); where the current actual hydraulic braking torque T h (t) is: T h (t)=u h (t)×e -5t Step 6.6: Determine whether the iteration termination condition is reached, that is, the current actual hydraulic braking torque T h (t) is equal to the desired hydraulic braking torque T h_req Or the current actual hydraulic braking torque T h (t) No change after a certain preset number of times: If yes, the current actual hydraulic braking torque T h (t) is the actual hydraulic braking torque T required h Output; Otherwise, set the number of iterations to t+1 and return to step 6.2; Where, E(t) is the current hydraulic braking torque error, T h_req is the expected hydraulic braking torque, T h (t-1) is the last actual hydraulic braking torque, ΔE(t) is the current hydraulic braking torque error change rate, T represents time; u h (t) is the current hydraulic pressure torque control signal, k p (t) is the current proportional coefficient, k i (t) is the current integral coefficient, k d (t) is the current differential coefficient; T h (t) is the current actual hydraulic braking torque, and t is the current iteration number.

3. The electro-hydraulic coordinated control method for the braking mode switching process of an electric vehicle according to claim 2, characterized in that: In step 6.3, the fuzzy PID rule base in the fuzzy controller is: In the table, E(t) represents the current hydraulic braking torque error, ΔE(t) represents the current hydraulic braking torque error change rate, k p (t) represents the current proportional coefficient, k i (t) represents the current integral coefficient, k d (t) represents the current differential coefficient, NB represents large negative, NM represents medium negative, NS represents small negative, ZO represents zero, PS represents small positive, PM represents medium positive, and PB represents large positive.

4. The electro-hydraulic coordinated control method for the braking mode switching process of an electric vehicle according to claim 1, characterized in that: The specific process of step 8 is as follows: Step 8.1, initialization, let the current iteration number t = 1, the last actual motor torque T m (0) = 0, the last motor torque control signal u m (0) = 0; Step 8.2: Use the Seagull optimization algorithm to solve the weight matrix Q(t) of the output error and the weight matrix R(t) of the control input change by minimizing the fitness function J1; the fitness function J1 is: Step 8.3: Use the model predictive control algorithm to solve the current motor torque control signal u by minimizing the cost function J2 m (t); where the cost function J2 is: J2=(T′ m_req -T m (t-1)) T ×Q(t)×(T′ m_req -T m (t-1))+(u m (t)-u m (t-1)) T ×R(t)×(u m (t)-u m (t-1)) Step 8.4: Based on the speed response of the motor brake system, control the current motor torque signal u m (t) Calculate and update the current actual motor torque T m (t); where Current actual motor torque T m (t) is: T m (t)=u m (t)×e -100t Step 8.5: Determine whether the iteration termination condition is reached, that is, the current actual motor torque T m (t) is equal to the target motor torque T′ m_req Or the current actual motor torque T m (t) No change after a certain preset number of times: If yes, then the current actual motor torque T m (t) is the actual motor torque T m Output; Otherwise, set the number of iterations to t+1 and return to step 8.2; Where v is the current speed of the electric vehicle, γ is the input matrix weight coefficient, α is the output matrix weight coefficient, Q(t) is the weight matrix of the output error, R(t) is the weight matrix of the control input change, and T represents time; u m (t) is the current motor torque control signal, u m (t-1) is the last motor torque control signal, T′ m_req is the target motor torque, T m (t-1) is the last actual motor torque, (■) T represents transpose; T m (t) is the current actual motor torque, and t is the current iteration number.

Images