Optimal torque distribution method for dual-motor electric vehicle with wheel slip rate constraint

Abstract

The application discloses a kind of optimal torque distribution method of double-motor electric vehicle with wheel slip ratio as constraint, comprising: calculating the slip ratio of each wheel;According to the principle of high selection, front axle slip rate and rear axle slip rate are obtained;According to the front axle slip rate and rear axle slip rate, it is judged whether to start the drive anti-skid control strategy;Optimal torque distribution strategy based on particle swarm algorithm, the optimal torque distribution coefficient under different driving conditions is calculated;According to the optimal torque distribution coefficient and the demand torque of whole vehicle, the output torque of front motor and the output torque of rear motor after distribution are obtained.The application aims to distribute torque based on efficiency optimization principle when electric vehicle drives on high adhesion road, but when driving on low adhesion road, design drive anti-skid control strategy to modify front and rear motor torque, and this correction value is used as dynamic constraint condition into torque distribution strategy, so that the wheel slip ratio of driving wheel tends to the expected value, and the driving stability of whole vehicle is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of drive torque distribution control, in particular, an optimal torque distribution method of a dual-motor electric vehicle with wheel slip ratio as a constraint. BACKGROUND

[0002] Drive torque distribution control is to improve the efficiency of the motor and reduce the energy consumption of the vehicle. Zheng Q uses a nonlinear particle swarm optimization algorithm to design a torque optimal distribution control strategy, instantaneously searches the optimal torque distribution coefficient of the front and rear motors, and establishes a dual-motor efficiency optimal model. Finally, the test in the loop shows that the working efficiency of the drive motor is improved under the control strategy. Kim H proposes a torque distribution control strategy for a dual-motor drive system, which aims to minimize the loss power of the motor, combines the driving efficiencies of the two motors, finds the highest efficiency point at different speeds, and determines the torque distribution proportion coefficient based on this. The simulation results show that the proposed torque distribution control strategy can effectively reduce the energy consumption of the vehicle. Holdstock designs a dual-motor drive, clutchless, and power continuous transmission system to improve the economy of the vehicle, and formulates a torque distribution control strategy based on the maximum efficiency of the motor and power system. Finally, the energy consumption performance simulation comparison with the traditional single-motor drive system in Cruise verifies the effectiveness of the proposed control strategy. Wang Xiaowei proposes a two-layer torque distribution control strategy, the upper layer is the drive torque decision layer, and the lower layer is the drive torque distribution layer. A multi-objective optimization function of vehicle economy, stability, and motor failure is formulated, and the simulation verifies that the control strategy can reduce the energy consumption of the vehicle. Ren Zhiyong takes a dual-motor driven articulated vehicle as the research object and proposes a torque distribution control strategy based on threshold value. The control strategy is applied to real vehicle test, and the test results show that the proposed control strategy reduces the energy consumption of the vehicle by about 5% compared with the power distribution, improving the economy of the vehicle. Xu Shuoshuo adopts the simulated annealing particle swarm filter algorithm to establish a vehicle driving state prediction model in order to improve the economy and stability of the pure electric vehicle. According to the prediction algorithm, the corresponding torque distribution strategy is formulated for different vehicle driving states, and the effectiveness of the control strategy is verified through hardware-in-the-loop testing. Meng Lingju establishes a mathematical model of the dual-motor drive system based on the working characteristics of the motor to determine the optimal torque distribution coefficient and design a dual-motor optimal torque distribution control strategy. The Interface co-simulation method is used for simulation analysis, which verifies that the torque distribution control strategy can improve the economy of the vehicle. The current torque distribution strategies mainly focus on improving the overall utilization efficiency of the motor during the driving process of the electric vehicle, while ignoring the driving stability on low adhesion road surface. SUMMARY

[0003] The technical problem to be solved by the present application is that the current torque distribution strategy mainly focuses on improving the comprehensive utilization efficiency of the motor during the driving process of the electric vehicle, and ignores the deficiency of driving stability on low adhesion road surface, and the present application provides a double-motor electric vehicle optimal torque distribution method considering longitudinal stability and taking wheel slip ratio as a constraint.

[0004] To achieve the above technical purpose, the technical solution adopted by the present application is:

[0005] A double-motor electric vehicle optimal torque distribution method taking wheel slip ratio as a constraint, comprising:

[0006] Step 1, calculating the slip ratio of each wheel at the current time;

[0007] Step 2, obtaining the front axle slip ratio and the rear axle slip ratio according to the high selection principle;

[0008] Step 3, judging whether to start the drive anti-skid control strategy according to the front axle slip ratio and the rear axle slip ratio;

[0009] If the front axle needs to start the drive anti-skid control strategy, the front motor torque at the last sampling time is corrected to obtain the corrected front motor torque instruction value, otherwise, the front motor torque at the last sampling time is directly taken as the corrected front motor torque instruction value;

[0010] If the rear axle needs to start the drive anti-skid control strategy, the rear motor torque at the last sampling time is corrected to obtain the corrected rear motor torque instruction value, otherwise, the rear motor torque at the last sampling time is directly taken as the corrected rear motor torque instruction value;

[0011] Step 4, if the sum of the corrected front axle motor torque instruction value and the rear axle motor torque instruction value is less than or equal to the whole vehicle demand torque, taking the corrected front axle motor torque instruction value as the output torque of the front axle motor at the current time, and taking the corrected rear axle motor torque instruction value as the output torque of the rear axle motor at the current time, and the torque distribution process at the current time is ended;

[0012] If the sum of the corrected front motor torque instruction value and the rear motor torque instruction value is greater than the whole vehicle demand torque, defining the output torque value range of the front motor, the output torque value range of the rear motor and the torque distribution coefficient value range, and sequentially executing step 5 and step 6;

[0013] Step 5, optimal torque distribution strategy based on particle swarm algorithm, calculate the optimal torque distribution coefficient under different driving conditions;

[0014] Step 6, according to the optimal torque distribution coefficient and the demand torque of the whole vehicle, the output torque of the front motor and the output torque of the rear motor after distribution are obtained, and the torque distribution process at the current time is ended.

[0015] As a further improved technical solution of the application, the step 1 is:

[0016] Calculate the slip ratio of each wheel at the current time:

[0017]

[0018] Wherein, S i is the slip ratio of the i-th wheel of the automobile at the current time; v is the vehicle speed at the current time, r is the wheel radius of the automobile, ω i is the angular velocity of the i-th wheel of the automobile at the current time.

[0019] As a further improved technical solution of the application, the step 2 is specifically:

[0020] Step 2.1, the maximum slip ratio of the two wheels on the same shaft is taken as the slip ratio of the shaft;

[0021] Step 2.2, the front axle slip ratio S f and the rear axle slip ratio S r are obtained respectively according to the method of step 2.1.

[0022] As a further improved technical solution of the application, the step 3 is specifically:

[0023] Step 3.1, the front axle slip ratio S f and the rear axle slip ratio S r are compared with the optimal slip ratio respectively:

[0024] ΔS f =S ep -S f ; ΔS r =S ep -S r ;

[0025] S ep represents the optimal slip ratio, which is 0.2;

[0026] Step 3.2, when ΔS f and ΔS r are greater than 0, the front axle and the rear axle need to start the drive anti-skid control strategy:

[0027] 3.2.1, Calculate the front motor torque correction amount AT f :

[0028]

[0029] Where k p is the proportional coefficient; k i is the integral coefficient; k d is the differential coefficient;

[0030] The corrected front motor torque command value is obtained by adding the correction amount AT f to the front motor torque command T f at the previous sampling time, and the corrected front motor torque command value T fl is expressed as:

[0031] T fl = T f + AT f ;

[0032] 3.2.2, Calculate the rear motor torque correction amount AT r :

[0033]

[0034] The corrected rear motor torque command value is obtained by adding the correction amount AT r to the rear motor torque command T r at the previous sampling time, and the corrected rear motor torque command value T rl is expressed as:

[0035] T rl = T r + AT r ;

[0036] Step 3.3, when AS f is greater than 0, and AS r is less than or equal to 0; the front axle needs to start the drive anti-slip control strategy, that is, the corrected front motor torque command value T fl is calculated according to the method of step 3.2.1; the rear motor torque command T r at the previous sampling time is directly used as the corrected front motor torque command value T rl ;

[0037] Step 3.4, when AS r is greater than 0, and AS f is less than or equal to 0; the rear axle needs to start the drive anti-slip control strategy, that is, the corrected rear motor torque command value T rl is calculated according to the method of step 3.2.2; the rear motor torque command T fThe rear motor torque instruction value T fl ;

[0038] Step 3.5, when ΔS f is less than or equal to 0, ΔS r is less than or equal to 0, the rear motor torque instruction value T r at the last sampling time is directly taken as the corrected front motor torque instruction value T rl ; the rear motor torque instruction value T f at the last sampling time is directly taken as the corrected front motor torque instruction value T fl .

[0039] As a further improved technical solution of the application, in step 4, the output torque value range of the front motor, the output torque value range of the rear motor and the torque distribution coefficient value range are defined, specifically:

[0040] The output torque T f of the front motor at the current time is in the range of T req -T rl <T f <T fl ;

[0041] The output torque T r of the rear motor at the current time is in the range of T req -T fl <T r <T rl ;

[0042] The torque distribution coefficient λ at the current time is in the range of

[0043] Wherein, T req represents the whole vehicle demand torque.

[0044] As a further improved technical solution of the application, step 5 is specifically:

[0045] Step 5.1, select an fitness function;

[0046] The fitness function f is:

[0047]

[0048] Wherein, η(n, λ·T a ) represents the efficiency when the front motor speed is n and the torque is λ·T a ; η(n, (1-λ)T a ) represents the efficiency when the rear motor speed is n and the torque is (1-λ)T a ; λ·T a =Tf (1-λ)T a = T r , T a = T req ;

[0049] Step 5.2, define the position and extreme value of the particle;

[0050] Step 5.3, establish a constraint condition;

[0051] The initial torque distribution coefficient is set as λ=0.5; and the constraint optimal torque distribution coefficient λ satisfies 0.5<λ<1;

[0052] Step 5.4, set the particle swarm algorithm parameters;

[0053] Step 5.5, initialize the particle swarm;

[0054] Calculate the fitness value of each particle;

[0055] Update the individual optimal value; update the global optimal value;

[0056] Judge whether the iteration reaches the maximum iteration number, if not, update the speed and position of the particle, if yes, the fitness maximum value at this time is the global optimal value, and the corresponding particle position is the optimal torque distribution coefficient.

[0057] As a further improved technical solution of the application, the step 6 is specifically:

[0058] The output torque of the front motor at the current moment is:

[0059] T f = λ·T a ;

[0060] The output torque of the rear motor at the current moment is:

[0061] T r =(1-λ)T a

[0062] Wherein T a =T req .

[0063] The application has the beneficial effects that:

[0064] The current torque distribution strategy mainly focuses on improving the overall utilization efficiency of the motor during the driving process of the electric vehicle, while ignoring the driving stability on low adhesion road surface. Based on this, this paper proposes an optimal torque distribution method for dual-motor electric vehicles with wheel slip ratio as the constraint. When the electric vehicle drives on high adhesion road surface, the front and rear motors adopt the torque distribution strategy based on the particle swarm algorithm with the goal of efficiency optimization. When driving on low adhesion road surface, the driving anti-skid control strategy is designed to modify the torque of the front and rear motors, and this modified value is used as the dynamic condition constraint for the torque distribution strategy with the goal of efficiency optimization, so that the driving wheel slip ratio tends to the expected value, and the driving stability of the vehicle is improved. BRIEF DESCRIPTION OF DRAWINGS

[0065] Figure 1 Figure 1 is a diagram of the dual-motor dual-axle drive configuration.

[0066] Figure 2 Figure 2 is a diagram of the wheel dynamics model.

[0067] Figure 3 Figure 3 is a diagram of the road surface utilization adhesion coefficient-slip ratio curve.

[0068] Figure 4 Figure 4 is a logic diagram of the driving anti-skid control strategy.

[0069] Figure 5 Figure 5 is a simulation result diagram of the uniform low adhesion road surface.

[0070] Figure 5 Figure 6(a) is a simulation result diagram of the slip ratio without control on the uniform low adhesion coefficient road surface.

[0071] Figure 5 Figure 6(b) is a simulation result diagram of the slip ratio with control on the uniform low adhesion coefficient road surface.

[0072] Figure 5 Figure 7(c) is a simulation result diagram of the motor torque without control on the uniform low adhesion coefficient road surface.

[0073] Figure 6 Figure 7(d) is a simulation result diagram of the motor torque with control on the uniform low adhesion coefficient road surface.

[0074] Figure 6 Figure 8 is a simulation result diagram of the low adhesion to high adhesion interface road surface.

[0075] Figure 6 Figure 9(a) is a simulation result diagram of the slip ratio without control from the low adhesion coefficient road surface to the high adhesion coefficient road surface.

[0076] Figure 6 Figure 9(b) is a simulation result diagram of the slip ratio with control from the low adhesion coefficient road surface to the high adhesion coefficient road surface.

[0077] Figure 6 Fig. 3 is a simulation result diagram of the motor torque without control from the low adhesion coefficient road surface to the high adhesion coefficient road surface.

[0078] Figure 7 Fig. 4 is a simulation result diagram of the motor torque with control from the low adhesion coefficient road surface to the high adhesion coefficient road surface.

[0079] Figure 8 Fig. 5 is a flow chart of the particle swarm algorithm.

[0080] Figure 9 Fig. 6 is a flow chart of the driving torque distribution.

[0081] Figure 10 Fig. 7 is a diagram of the optimal torque distribution coefficient.

[0082] Figure 11 Fig. 8 is a vehicle speed following curve diagram of the whole vehicle model under the NEDC cycle condition.

[0083] Figure 12 Fig. 9 is a diagram of the front and rear motor torque distribution coefficient distribution under the cycle condition.

[0084] Figure 13 Fig. 10 is a diagram of the front and rear motor torque distribution result using the strategy A.

[0085] Figure 14 Fig. 11 is a diagram of the front and rear motor torque distribution result using the strategy B.

[0086] Figure 15 Fig. 12 is a diagram of the power battery SOC change under the three torque distribution strategies.

[0087] Figure 16 Fig. 13 is a diagram of the whole vehicle driving range and battery SOC change under the three torque distribution strategies.

[0088] Figure 1 Fig. 14 is a diagram of the whole vehicle total power consumption change under the three torque distribution strategies. DETAILED DESCRIPTION

[0089] The specific embodiments of the present application are further described below with reference to the accompanying drawings:

[0090] The current torque distribution strategy focuses on improving the overall utilization efficiency of the electric motor during the driving process of the electric vehicle, while ignoring the driving stability on low adhesion road surfaces. Based on this, an optimal torque distribution method for dual-motor electric vehicles with wheel slip ratio as a constraint is proposed. When the electric vehicle is driving on high adhesion road surfaces, the front and rear motors use an efficiency optimal torque distribution strategy based on the particle swarm algorithm. When driving on low adhesion road surfaces, a drive slip control strategy is designed to modify the front and rear motor torque, and this modified value is used as a dynamic condition constraint for the efficiency optimal torque distribution strategy, so that the drive wheel slip ratio tends to the desired value and improves the vehicle's driving stability.

[0091] The dual-motor drive system expands the working area of the motor in the high efficiency region through precise control of the motor. For the torque distribution problem of the front and rear axle motors of a dual-motor four-wheel drive electric vehicle, an optimal torque distribution strategy based on the particle swarm algorithm is proposed, and a forward simulation model of the dual-motor electric vehicle is established. Considering the influence of the drive wheel slip ratio on the stability of the vehicle, the front and rear motor torque command values are modified with the vehicle slip ratio as the control target. Considering the modified motor torque command values, the constraint conditions of the optimal torque distribution problem are designed, and the efficiency mathematical model of the dual-motor drive system is established based on the vehicle configuration. The highest overall utilization efficiency of the dual-motor drive system is used as the objective function, and the torque distribution coefficient is iteratively optimized within the constraint range by the particle swarm algorithm. The effectiveness of the proposed method is verified through single working condition points and NEDC cycle conditions. The simulation results show that the proposed optimal torque distribution strategy can achieve reasonable torque distribution to meet the vehicle's driving requirements. Under the NEDC cycle condition, the torque distribution strategy designed in this paper can effectively improve the battery SOC decline, prolong the vehicle's driving range, and significantly reduce the vehicle's energy consumption, which has certain practicality in improving the vehicle's economy.

[0092] 1. Model establishment:

[0093] 1.1. Vehicle configuration of dual-motor drive system:

[0094] The dual-motor dual-axle drive configuration is to mount a set of drive devices on the front and rear axles, as shown in Vehicle parameters . It can distribute the required torque to the two motors according to the control strategy, expand the motor's high-efficiency operating speed and torque range, and to some extent, increase the vehicle's driving range, effectively solving the problem that a single-motor drive configuration cannot simultaneously meet the vehicle's power and energy consumption requirements. Therefore, this drive configuration has gradually increased its market share. The pure electric vehicle studied in this paper adopts a front and rear axle dual-motor drive configuration, and its basic parameters are shown in Table 1.

[0095] Table 1. Basic parameters of pure electric vehicle

[0096] Symbol Value Unit Kerb mass kg m0 2150 Full load mass kg m1 2525 Wind area Tire radius A 2.7 m2 Rolling resistance coefficient r 0.359 m Air resistance coefficient f 0.012 Cd Mechanical system transmission efficiency 0.3 ηt Rotational mass conversion coefficient 0.9 δ Motor parameters 1.2

[0097] The front and rear shafts of the dual-motor drive system electric vehicle are configured with two permanent magnet synchronous motors of the same specification, and are respectively matched with fixed speed ratio reducers. The front and rear shaft motor speeds are calculated through the vehicle speed. The main parameters of the components of the dual-motor drive system are shown in Table 2.

[0098] Table 2, drive motor parameter matching results

[0099] Value Unit Rated voltage Peak power 320 V kW 80 Rated power kW 32 Peak rotational speed r / min 12000 Rated rotational speed r / min 4000 Peak torque N·m 191 Rated torque N·m 76 Figure 2

[0100] 1.2, efficiency model of the dual-motor drive system:

[0101] The relationship between the motor torque and the motor efficiency is nonlinear, and therefore the torque distribution strategy aims to reasonably distribute the torque instructions of the front and rear drive motors, reduce the input power of the front and rear motors, improve the comprehensive efficiency of the front and rear motors, and improve the economic performance of the vehicle. Under the same total demand torque, the control strategy changes the torque instruction distribution ratio of the front and rear motors, reduces the sum of the loss power of the front and rear motors, and the loss power of the motor is as follows:

[0102]

[0103] wherein, P lossi is the loss power of a single motor; η i (n, T i ) is the efficiency of a single motor with a speed of n and a torque of T i ; P elei is the electrical power input to a single motor by the motor controller; and P mechi is the mechanical power output by a single motor.

[0104] In the optimal torque distribution strategy, the torque distribution coefficient is defined as follows:

[0105]

[0106] wherein, λ is the torque distribution coefficient; T f and T r are the front shaft motor torque instruction and the rear shaft motor torque instruction, respectively.

[0107] The total output torque of the drive motor is the sum of the front and rear motor torque instructions, which can be expressed as:

[0108] T a = T f + T r (3);

[0109] wherein, T a is the total output torque of the motor.

[0110] According to the analysis of the relationship between the input power and the output power of the motor according to formula (1), the working efficiency η of the motor m may be expressed as:

[0111]

[0112] Therefore, the comprehensive utilization efficiency η of the dual-motor drive system com may be expressed as:

[0113]

[0114] The optimal torque distribution problem can be described as finding a front and rear axle motor torque distribution coefficient for the electric vehicle at any operating point, so that the motor comprehensive efficiency is the highest. Taking the torque distribution coefficient as the control variable and the motor comprehensive efficiency as the objective function, the motor comprehensive efficiency calculation formula can be transformed as:

[0115]

[0116] In the formula, η(n, T f ) is the efficiency of the front axle motor when the speed is n and the torque is T f ; η(n, T r ) is the efficiency of the rear axle motor when the speed is n and the torque is T r .

[0117] Substituting formula (2) and (3) into (6), the efficiency model of the dual-motor drive system is:

[0118]

[0119] 2. Drive anti-skid control strategy:

[0120] When the pure electric vehicle accelerates or starts on low adhesion road surface such as wet or icy road surface, due to the limited ground adhesion, the ground tangential reaction force is lower than the vehicle driving force, and the greater the driving force, the more likely it is to occur. The stability of the pure electric vehicle is damaged. In order to prevent the excessive slip of the electric vehicle from affecting the dynamic performance and handling stability of the vehicle, the output torque of the drive motor needs to be adjusted when the torque distribution strategy is formulated, so that the slip rate reaches the expected value.

[0121] 2.1. Wheel dynamics model:

[0122] The selected pure electric vehicle in this paper is four-wheel drive with front and rear axle motors. The driving torque of the driving wheels can be output to the wheel model through the motor model and the transmission system model. Since the electric vehicle is a highly complex system, in order to study the excessive slip of the vehicle on the low adhesion road, the wheel dynamics model needs to be simplified. This paper adopts a single wheel model, which only needs to consider the key factors of vehicle longitudinal driving. Assuming that the four wheels are of the same specification, the wheel dynamics model is shown in FIG. 8. In the wheel coordinate system, the dynamics equations of the four wheels are shown in equation (8): Figure 3

[0123]

[0124] In the equation, ω i is the driving wheel angular velocity; J is the wheel rotational inertia; T f / r is the front and rear motor output torque respectively; F xi is the wheel longitudinal force; T bi is the wheel braking torque; i = 1, 2, 3, 4, representing the left front wheel, right front wheel, left rear wheel and right rear wheel respectively.

[0125] When the braking condition is not considered, the driving wheel angular velocity can be expressed as:

[0126]

[0127] In the equation, T di is the driving wheel driving torque. Generally, when the vehicle is in the driving state, the wheel slip ratio can be written as:

[0128]

[0129] In the equation, S i is the wheel slip ratio; v is the vehicle speed.

[0130] 2.2, Optimal slip ratio:

[0131] This paper uses the road adhesion coefficient-driving wheel slip ratio model proposed by Burckhardt to describe the relationship between the driving wheel slip ratio S and the road adhesion coefficient μ of the pure electric vehicle in the acceleration or braking condition. This model can be expressed as:

[0132]

[0133] In the equation, C1, C2 and C3 are the fitting coefficients of different roads.

[0134] In the μ-S model, the optimal slip ratio and the maximum utilization adhesion coefficient of the road can be expressed as:

[0135] ​

[0136] The common μ-S model road surface is used to apply the relationship curve between the adhesion coefficient and the slip ratio, such as Figure 3 As shown, the optimal slip ratio is assumed to correspond to the peak road surface adhesion coefficient of the five curves. Figure 4 It is known that when the slip ratio is less than the optimal slip ratio, the road surface adhesion coefficient increases with the increase of the slip ratio. At this time, if the wheel driving force is greater than the adhesion force, the wheel slip ratio increases, and the adhesion coefficient also increases accordingly. The increased wheel adhesion force actually reduces the trend of increasing wheel slip ratio, resulting in stable vehicle driving. When the slip ratio is greater than the optimal slip ratio, the road surface adhesion coefficient decreases with the increase of the slip ratio. At this time, the wheel slip ratio will continuously increase over time until the wheel slips, affecting vehicle driving stability. This paper refers to the typical tire slip ratio and road surface adhesion coefficient characteristics. The optimal slip ratio is generally between 15% and 20%. To ensure vehicle power and maximize driving force, a slip ratio greater than 20% is taken as the control standard for the drive anti-slip control strategy. That is, when the wheel slip ratio is greater than 20%, the drive anti-slip control strategy intervenes in the torque control strategy and controls the wheel slip ratio to around 20%.

[0137] 2.3 Design of Drive Anti-Slip Control Strategy:

[0138] Currently, the main control algorithms for drive anti-slip control strategies include fuzzy control, sliding mode control, PID control, and gate control. Considering the slow response of fuzzy control, the need for on-site testing data for gate control, and the high-frequency jitter characteristics of sliding mode control which are detrimental to the torque output of the drive motor, this paper adopts PID control to control the slip ratio and limit the driving torque of the electric vehicle on low-traction surfaces. The control strategy logic is as follows: Figure 4 As shown.

[0139] Depend on Figure 5 It can be seen that the front and rear axle slip ratio S when the electric vehicle is in motion is... f / r And the optimal slip ratio S on the current road surface ep The difference ΔS is used as the criterion for judging whether the drive wheel is slipping. Since the driving force of the left and right wheels on the same axle comes from the same drive motor, the driving torque of the left and right wheels cannot be controlled independently. This paper adopts the high selection principle. When the wheel with the maximum slip ratio does not slip, the other wheel on the same axle with a smaller slip ratio will also not slip. Therefore, the wheel with the larger slip ratio among the left and right wheels is used as the benchmark value for motor torque correction. The slip ratios of the front and rear axles are S1, S2, and S3, respectively. f S r This is compared with the optimal slip ratio to ensure the vehicle slip ratio is within the range of maximum road adhesion coefficient. ΔS can be expressed as:

[0140] ΔS=S ep -S f / r (14);

[0141] When the front-rear axle slip ratio is greater than the optimal slip ratio, it is considered that the drive has a tendency to slip, and the drive motor torque command needs to be adjusted by PID control to obtain the correction amount ΔT f / ΔT r The front axle motor correction amount calculation formula is as follows:

[0142]

[0143] In the formula, k p is the proportional coefficient; k i is the integral coefficient; and k d is the differential coefficient.

[0144] The actual motor torque command value is obtained by adding the correction amount to the motor torque command T f / T r at the previous sampling time, and the front axle motor torque actual command value T f l can be expressed as:

[0145] T fl =T f +ΔT f (16);

[0146] Where ΔT f is the front motor torque correction value. The calculation formula of the rear axle motor actual torque command value T rl is the same as that of the front motor.

[0147] 2.4, Verification of the drive anti-slip control strategy:

[0148] In order to verify the effectiveness of the drive anti-slip control strategy designed in this paper, a drive anti-slip control strategy model is built in MATLAB / Simulink, and simulation analysis is carried out on uniform low adhesion coefficient road and low adhesion to high adhesion coefficient road respectively, and the simulation time is set to 5s. At the same time, in order to better analyze the control effect of the drive anti-slip control strategy in this paper, the torque correction is applied to the motor and the torque correction is not applied to the motor respectively for comparison test.

[0149] (1) Uniform low adhesion coefficient road simulation:

[0150] When simulating on a uniform low adhesion coefficient road, the adhesion coefficient of the road is set to 0.2 in this paper. Since the road driving conditions of the same shaft left and right wheels are consistent, the slip ratio changes of the same shaft left and right wheels should also be consistent, so only the left front wheel and the left rear wheel are considered when analyzing the simulation results, and the simulation results are shown in Figure 5 .

[0151] Figure 5(a), (c) in Fig. 6 are simulation results of the double-motor four-wheel drive electric vehicle driving on the uniform low adhesion road surface without the driving anti-slip control. Due to the poor adhesion condition of the road surface, the slip rates of the front and rear drive wheels are far greater than the expected slip rate, and the phenomenon of excessive slip occurs, and the motor output torque is not adjusted accordingly. When the driving anti-slip control exists, the simulation results of the whole vehicle are shown in (b), (d) in Fig. 6. When the slip rates of the front and rear drive wheels are greater than the expected slip rate, the driving anti-slip control is started, the front and rear motor output torques are adjusted, and the slip rates of the drive wheels are controlled at about 0.2. Figure 6

[0152] (2) Simulation of low adhesion to high adhesion coefficient road surface:

[0153] For the simulation from the low adhesion coefficient road surface to the high adhesion coefficient road surface, the vehicle model is set to drive from the road surface with an adhesion coefficient of 0.3 to the road surface with an adhesion coefficient of 0.8, and only the left front wheel and the left rear wheel are considered in the simulation analysis. The simulation results are shown in (a), (c) in Fig. 7. Figure 6

[0154] Figure 6 (a), (c) in Fig. 7 are simulation results of the double-motor four-wheel drive electric vehicle driving on the low adhesion to high adhesion coefficient road surface without the driving anti-slip control. When the electric vehicle drives on the low adhesion road surface, due to the poor adhesion condition of the road surface, the slip rate of the drive wheel is large, and the phenomenon of excessive slip occurs. When the electric vehicle drives on the high adhesion road surface, the slip rate of the drive wheel is rapidly reduced, and the whole vehicle recovers stable driving. The front and rear motor torques are normally output without corresponding correction in the simulation process. When the driving anti-slip control exists, the simulation results of the whole vehicle are shown in (b), (d) in Fig. 7. During the driving process on the low adhesion road surface, when the slip rate of the drive wheel is greater than the expected slip rate, the driving anti-slip control rapidly adjusts the motor output torque to control the slip rate of each wheel at about 0.2. When driving on the high adhesion road surface, the slip rate of the drive wheel is less than the expected slip rate, and the driving anti-slip control does not control the motor output torque. Figure 7

[0155] 3, Optimal torque distribution strategy design:

[0156] 3.1, Constraint condition:

[0157] The torque distribution strategy based on the optimal efficiency needs to set the constraint condition. As can be seen from equation (7), the comprehensive efficiency of the motor is related to the front and rear axle motor torque distribution coefficient λ and the total output torque T a ​​​Therefore, the optimal torque distribution problem can be converted into the determination and constraint of the above two parameters. In the anti-slip control strategy of the pure electric vehicle drive, the front and rear axle motor torque instructions are modified to improve the vehicle driving stability. The sum of the modified front and rear axle motor torque instructions should be the current motor total output torque. Therefore, this paper determines the torque distribution coefficient and the value range of the total output torque by taking the modified front and rear axle motor torque instructions as dynamic constraint conditions.

[0158] When the sum of the modified front and rear axle motor torque instructions is less than or equal to the vehicle demand torque, i.e., T fl +T r ≤T req , the total output torque cannot meet the demand torque. In order to make the total output torque meet the vehicle power demand as much as possible, the modified front and rear axle motor torque instructions are directly taken as the output, i.e., T f =T fl , T r =T rl . The torque distribution coefficient λ = T fl / (T fl +T rl ), and the motor total output torque T a =T fl +T rl .

[0159] When the sum of the modified front and rear axle motor torque instructions is greater than the vehicle demand torque, i.e., T fl +T rl >T req , the total output torque can meet the vehicle power demand, and the motor total output torque T a satisfies T a =T req . The front and rear motor torque value ranges are respectively: T req -T rl <T f <T fl , T req -T fl <T r <T rl . Therefore, the torque distribution coefficient λ satisfies (T req -T rl ) / T req <λ<T fl / T req .

[0160] In summary, the constraint conditions of the torque distribution control strategy optimization problem are as follows:

[0161]

[0162] wherein, T mM represents the motor external characteristic torque, n m M represents the maximum speed.

[0163] 3.2, particle swarm algorithm flow:

[0164] In the torque distribution strategy designed in this paper involves multiple dimensions, which makes the time required for calculation in complex conditions greatly increased. Considering the above problems, this paper presents the optimal torque distribution strategy based on particle swarm optimization algorithm, the optimal torque distribution coefficient under different driving conditions.

[0165] The optimization process based on PSO algorithm is shown in Population number The specific optimization process is:

[0166] (1) select fitness function:

[0167] In particle swarm optimization algorithm, fitness function is a key indicator to measure the optimization effect of the algorithm. In each iteration process, the current position of each particle needs to be calculated by fitness function. Therefore, the more concise the fitness function is designed, the higher the iteration efficiency of the algorithm is. According to formula (7), this paper takes the comprehensive efficiency of the driving motor as the optimization target, and the fitness function f can be represented as formula (18):

[0168]

[0169] (2) define the position and extreme value of particle:

[0170] The particle position in particle swarm optimization algorithm represents the torque distribution coefficient. Assuming that the algorithm optimization is in a D-dimensional search space, then the population X composed of n particles can be represented as:

[0171] X = (X1, X2,..., X n )(19);

[0172] The torque distribution coefficient X i of the i-th particle in the D-dimensional search space is:

[0173] X i = (X i1 ,X i2 ,...,X iD ) T (20);

[0174] The speed of the particle represents the change of the torque distribution coefficient between two iterations in the algorithm. The speed V i of the i-th particle in the D-dimensional search space is:

[0175] V i = (V i1 ,V i2..., V iD ) T (21);

[0176] The individual extreme value refers to the individual historical optimal fitness until the current iteration, and the position P i can be expressed as:

[0177] P i = (P i1 , P i2 ,..., P iD ) T (22);

[0178] The global extreme value refers to the global historical optimal fitness until the current iteration, and the position P g can be expressed as:

[0179] P g = (P g1 , P g2 ,..., P gD ) T (23).

[0180] (3) Update the particle position and velocity:

[0181] According to the definition of the position and extreme value of the particle, at each iteration of the algorithm, the particle updates its position and velocity through the individual extreme value and the global extreme value:

[0182]

[0183] where ω is the inertia weight; i is the particle, i = 1, 2,..., n; d is the spatial dimension, d = 1, 2,..., D; k is the current iteration number; is the position of the i-th particle in the d-th dimension at the k-th iteration; is the velocity of the i-th particle in the d-th dimension at the k-th iteration; c1 and c2 are learning factors, generally taking the same value; r1 and r2 are random numbers between 0 and 1. represents the position of the individual extreme value of the i-th particle in the d-th dimension, represents the position of the global extreme value of the i-th particle in the d-th dimension.

[0184] (4) Establish the constraint condition:

[0185] In this paper, the initial torque distribution coefficient is set to λ = 0.5. At the same time, considering that the pure electric vehicle studied in this paper mainly relies on the front axle motor for driving, when encountering different torque distribution coefficients leading to the same comprehensive efficiency, the front axle motor output scheme should be preferred, that is, the constraint torque distribution coefficient λ satisfies 0.5 < λ < 1.

[0186] (5) Set the particle swarm algorithm parameters:

[0187] The particle swarm algorithm parameters affect the stable operation of the system, and improper settings can lead to local optimization and slow convergence speed, resulting in unnecessary increase in the number of iterations. The specific parameter settings of the particle swarm algorithm are shown in Table 3.

[0188] Table 3, particle swarm algorithm parameters

[0189] Maximum iteration number Learning factor c1 Learning factor c2 Inertia weight Figure 8 100 30 1.5 1.5 1

[0190] Update the iteration to the maximum number of iterations, at which time the maximum fitness value is the global extreme value, and the corresponding particle position is the optimal torque distribution coefficient. Combined with the vehicle demand torque calculated by the driving control strategy described above, the output torque of the front and rear motors after distribution can be obtained according to the optimal torque distribution coefficient and the vehicle demand torque, and the torque distribution process is shown in Figure 9 .

[0191] 4、Torque distribution strategy simulation analysis:

[0192] 4.1、Based on single working condition point torque distribution result analysis:

[0193] To speed up the operation, the motor speed and output torque are discretized with a precision of 500 r / min and 16 Nm, the torque distribution result corresponding to the optimal comprehensive efficiency of each working condition point is calculated, and the distribution result is stored in the form of a table, and the full working condition point torque distribution result is shown in Figure 9 . During vehicle driving, the vehicle controller can control the front and rear motor output torque by using the table lookup method according to the vehicle demand torque.

[0194] As can be seen from Figure 10 , the front and rear motor torque distribution coefficient is mainly affected by the vehicle demand torque. When the total demand torque is low, the optimal torque distribution coefficient is 1 to avoid the driving motor operating in the low torque low efficiency region, and the front motor works alone; as the vehicle demand torque increases, the optimal torque distribution coefficient decreases and is between 0.5 and 1, at which time the front and rear motors provide part of the torque at the same time, making the dual motor drive system have the highest comprehensive efficiency; when the total demand torque is large, the torque distribution coefficient is 0.5, at which time the torque tends to be evenly distributed, and the vehicle demand torque is evenly distributed to the front and rear two motors, avoiding the motor working in the high torque low efficiency region, so that the driving motor runs in the high efficiency region as much as possible, improving the comprehensive utilization efficiency of the drive system.

[0195] To analyze the effectiveness of the optimal torque distribution strategy based on the particle swarm algorithm designed in this paper, a vehicle model is established in MATLAB / Simulink to drive under NEDC cycle conditions, Figure 12The vehicle model in this paper is shown in the NEDC cycle speed following curve, where the red implementation represents the target speed, and the blue dashed line represents the actual speed. From the speed tracking curve, it can be seen that the actual speed curve and the target speed curve are basically in the state of coincidence, the speed error is within the allowable range of the condition, the speed following condition is good, and the rationality of the vehicle driving control strategy designed in this paper is verified.

[0196] To analyze the effectiveness of the torque distribution strategy based on particle swarm algorithm designed in this paper, the vehicle model is set to drive under NEDC cycle conditions, and the torque distribution results of front and rear motors under the torque average distribution strategy and the torque distribution strategy based on particle swarm algorithm are simulated. The torque average distribution strategy is called strategy A, and the torque distribution strategy based on particle swarm algorithm in this paper is called strategy B. The driving motor torque distribution of the two strategies in the cycle condition is investigated, as shown in Figure 11 and 13 .

[0197] Figure 12 The distribution of front and rear motor torque distribution coefficients under the cycle condition is shown in the figure. The vehicle controller controls the output torque of the front and rear motors according to the current optimal torque coefficient. Figure 13 The torque distribution of the front and rear motors using strategy A is shown in the figure. The front and rear axle motors always output the same torque, and the pure electric vehicle driving system follows the four-wheel drive driving mode at every moment under the cycle condition. Figure 14 The torque distribution of the front and rear motors using strategy B is shown in the figure. When the vehicle is in the low and medium speed area, the acceleration and deceleration process is frequent, and the vehicle demand torque is large. Single motor cannot meet the demand of vehicle torque, and the torque distribution value of strategy B is close to that of strategy A. When the vehicle is in the high speed area, the vehicle demand torque is small, and strategy B tends to use the front motor alone, and the torque distribution coefficient is small, which is conducive to improving the comprehensive utilization efficiency of the driving system.

[0198] 4.2, based on the analysis of torque distribution results under the cycle condition:

[0199] The torque distribution strategy based on particle swarm optimization algorithm is to improve the total efficiency of the power system as the goal, the total demand torque and speed under a certain working condition is simplified as the working point, the torque distribution coefficient is optimized offline, and the economy of the vehicle is improved. Based on the torque distribution strategy based on particle swarm optimization algorithm, considering the vehicle driving on low adhesion road, the wheel slip ratio is taken as the control target, the output torque of the front and rear motors is corrected, and the corrected output torque of the motor is taken as the constraint to limit the value range of the torque distribution coefficient, so as to control the slip ratio to the expected value. This paper will take the comprehensive torque distribution strategy with slip ratio as the control target as strategy C. In order to intuitively verify the effect of the formulated torque distribution strategy in improving the economy of the vehicle, the SOC of the power battery, the endurance mileage and the power consumption per 100 kilometers are selected as the economy indicators to investigate the economy performance of the pure electric vehicle model under the cycle working condition.

[0200] The size of the power battery SOC value directly reflects the current storage capacity of the battery. Taking the power consumption of the vehicle model under the single NEDC cycle working condition as the standard, the initial SOC value of the battery is set to 90%, the same driving mode is adopted, and the simulation of strategy A, B and C is carried out respectively. The SOC change curves of the two strategies are shown in Figure 14

[0201] By comparing the SOC changes of the three torque distribution strategies in Figure 15 , it can be seen that when the vehicle drives in the urban cycle working condition with low speed, the optimal torque distribution coefficient tends to 0.5, and the SOC curves of the three torque distribution strategies decrease slowly. In the suburban cycle working condition with high speed, the SOC curve of strategy B decreases obviously slower than that of strategy A, and the SOC values of strategy B, strategy C and strategy A when completing a single cycle working condition are 86.85%, 86.65% and 86.54% respectively. The above results show that compared with the traditional torque average distribution strategy, when the vehicle drives in the working condition with large speed fluctuation range and high total demand torque, the reasonable distribution of the output torque of the front and rear motors by using the torque distribution strategy based on particle swarm optimization algorithm and the comprehensive torque distribution strategy can effectively reduce the energy consumption of the electric vehicle and improve the economy of the vehicle.

[0202] In order to further analyze the performance of the torque distribution strategy in improving the economy of the vehicle, the endurance mileage of the vehicle model is simulated. The influence of battery SOC on the maximum discharge voltage is not considered in the single cycle working condition, so the lower limit of SOC needs to be set as the simulation end condition when the repeated cycle working condition simulation is carried out. The initial power is set to 90%, and the endurance capability is taken as the evaluation index when the battery discharge depth is 80%, that is, the simulation is stopped when the vehicle SOC decreases to 10%. Under the NEDC cycle working condition, the endurance mileage and the battery SOC change of the vehicle under the three torque distribution strategies are shown in Figure 15 ​​

[0203] From Figure 16 It can be seen that, in the NEDC operating condition, when strategy A is adopted, the vehicle cruising range reaches 248.55km, and the driving time is 27011.67s; when strategy B is adopted, the vehicle cruising range can reach 271.86km, and the driving time is 29424.16s, compared with the torque average distribution strategy, the torque distribution strategy based on the particle swarm algorithm makes the vehicle cruising range increase by 9.37%, and the driving time increases by 8.93%; when strategy C is adopted, the vehicle cruising range can reach 256.7km, and the driving time is 28045.43s, compared with the torque average distribution strategy, the comprehensive torque distribution strategy makes the vehicle cruising range increase by 3.29%, and the driving time increases by 3.83%.

[0204] The 100km electric consumption under different cycle conditions is one of the evaluation indexes of the economy of pure electric vehicles. In order to further analyze the performance of the torque distribution strategy in improving the economy of the vehicle, the power consumption of the vehicle model under a single NEDC cycle condition is simulated, and the changes of the power consumption of the vehicle under three torque distribution strategies are as shown in Figure 16 .

[0205] From Index It can be seen that, in the NEDC operating condition, when strategy A is adopted, the vehicle cruising range reaches 248.55km, and the driving time is 27011.67s; when strategy B is adopted, the vehicle cruising range can reach 271.86km, and the driving time is 29424.16s, compared with the torque average distribution strategy, the torque distribution strategy based on the particle swarm algorithm makes the vehicle cruising range increase by 9.37%, and the driving time increases by 8.93%; when strategy C is adopted, the vehicle cruising range can reach 256.7km, and the driving time is 28045.43s, compared with the torque average distribution strategy, the comprehensive torque distribution strategy makes the vehicle cruising range increase by 3.29%, and the driving time increases by 3.83%.

[0206] Table 4, comparison of economy of three strategies under cycle conditions

[0207] Strategy A Strategy B Strategy C Final time battery SOC (%) SOC change amount (%) 86.54 86.85 86.65 Range (km) 3.46 3.15 3.35 Energy consumption (KJ) 248.55 271.86 256.7 100 km electricity consumption (kWh) 5076 4626 4914 ​ 12.809 11.673 12.4

[0208] In summary, according to the different quantitative indexes of the economy of the pure electric vehicle under the cycle condition, the simulation results are shown in Table 4. By comparing the economy indexes of the three strategies in Table 4, it can be seen that the torque distribution strategy based on the particle swarm algorithm designed in this paper can effectively improve the battery SOC drop, prolong the vehicle cruising range, and significantly reduce the vehicle energy consumption while being constrained by the wheel slip control, and has certain practicability in improving the economy of the vehicle.

[0209] The protection scope of the present application includes but is not limited to the above embodiments, the protection scope of the present application is subject to the claims, any replacement, deformation, improvement of the present technology that is easily thought of by the skilled in the art falls within the protection scope of the present application.

Claims

1. A method for optimal torque distribution of a dual-motor electric vehicle with wheel slip ratio as constraint, characterized in that, The method comprises the following steps: Step 1, calculating the slip ratio of each wheel; Step 2, obtaining the front axle slip ratio and the rear axle slip ratio according to the high selection principle; Step 3, judging whether to start the drive anti-skid control strategy according to the front axle slip ratio and the rear axle slip ratio; If the front axle needs to start the drive anti-skid control strategy, the front motor torque at the last sampling time is corrected to obtain the corrected front motor torque instruction value, otherwise, the front motor torque at the last sampling time is directly taken as the corrected front motor torque instruction value; If the rear axle needs to start the drive anti-skid control strategy, the rear motor torque at the last sampling time is corrected to obtain the corrected rear motor torque instruction value, otherwise, the rear motor torque at the last sampling time is directly taken as the corrected rear motor torque instruction value; Step 4, if the sum of the corrected front axle motor torque instruction value and the rear axle motor torque instruction value is less than or equal to the whole vehicle demand torque, the corrected front axle motor torque instruction value is taken as the output torque of the front axle motor, and the corrected rear axle motor torque instruction value is taken as the output torque of the rear axle motor; If the sum of the corrected front motor torque instruction value and the rear motor torque instruction value is greater than the whole vehicle demand torque, the output torque value range of the front motor, the output torque value range of the rear motor and the torque distribution coefficient value range are defined, and steps 5 and 6 are sequentially executed; Step 5, the optimal torque distribution strategy based on the particle swarm algorithm is used to calculate the optimal torque distribution coefficient under different driving conditions; Step 6, the output torque of the front motor and the output torque of the rear motor after distribution are obtained according to the optimal torque distribution coefficient and the whole vehicle demand torque.

2. The optimal torque distribution method for dual-motor electric vehicle with wheel slip ratio constraint according to claim 1, characterized in that, The step 1 is: The slip ratio of each wheel at the current time is calculated: Wherein, S i is the current time the car i wheel slip rate; v is the current time the car speed, r is the car wheel radius, ω i is the current time the car i wheel angular velocity.

3. The optimal torque distribution method for dual-motor electric vehicle with wheel slip ratio constraint according to claim 1, characterized in that, The step 2 is specifically: Step 2.1, the maximum slip ratio of the two wheels on the same axle is taken as the axle slip ratio; Step 2.

2. The front and rear slip ratios Sf and Sr are obtained respectively according to the method of Step 2.

1. f and the rear slip ratios S r .

4. The optimal torque distribution method for dual-motor electric vehicle with wheel slip ratio constraint according to claim 1, characterized in that, The step 3 is specifically: Step 3.1, the front axle slip ratio S f and the rear axle slip ratio S r are compared to the optimal slip ratio, respectively: ΔS f = S ep - S f ; ΔS r = S ep - S r ; S ep represents the optimal slip rate; Step 3.2, when ΔS f and ΔS r are both greater than 0, both front and rear axles need to initiate drive slip control strategy: 3.2.1, Calculate motor torque correction amount ΔT before f : In the formula, k p k is the proportionality coefficient. i k is the integral coefficient; d These are the differential coefficients; The corrected front motor torque command value is obtained by adding the correction amount ΔT to the front motor torque command T f The corrected front motor torque command value T is obtained by adding the correction amount ΔT to the front motor torque command T f The corrected front motor torque command value T is obtained by adding the correction amount ΔT to the front motor torque command T fl is expressed as T fl = T f + ΔT f ; 3.2.

2. Calculate the motor torque correction amount ΔT r : The corrected rear motor torque command value is obtained by adding the correction amount ΔT to the rear motor torque command T r The corrected rear motor torque command value T is obtained by adding the correction amount ΔT to the rear motor torque command T r The corrected rear motor torque command value T is obtained by adding the correction amount ΔT to the rear motor torque command T rl is expressed as: T rl = T r + ΔT r ; Step 3.3, when ΔS f is greater than 0, ΔS r is less than or equal to 0; then the front axle needs to start the drive slip control strategy, that is, the corrected front motor torque command value T fl is calculated according to the method of step 3.2.1; the rear motor torque command T r at the last sampling time is directly taken as the corrected front motor torque command value T rl ; Step 3.4, when ΔS r is greater than 0, ΔS f is less than or equal to 0, the rear axle needs to start the drive anti-slip control strategy, that is, the corrected rear motor torque command value T rl is calculated according to the method of step 3.2.2; the rear motor torque command T f of the last sampling time is directly taken as the corrected front motor torque command value T fl ; Step 3.5, when ΔS f is less than or equal to 0, ΔS r is less than or equal to 0, the rear motor torque command T r is directly used as the corrected front motor torque command value T rl ; the rear motor torque command T f is directly used as the corrected front motor torque command value T fl .

5. The optimal torque distribution method for dual-motor electric vehicles with wheel slip ratio constraints according to claim 1, characterized in that, In the step 4, the output torque value range of the front motor, the output torque value range of the rear motor and the torque distribution coefficient value range are defined, and the step 5 is specifically: Output torque T of the front motor at the current time f T req -T rl <T f <T fl ; Output torque T of the rear motor at the current time r T req -T fl <T r <T rl ; The torque distribution coefficient λ at the current time is in the range where T req represents the total vehicle demand torque.

6. The optimal torque distribution method for dual-motor electric vehicles with wheel slip ratio constraints according to claim 1, characterized in that, The step 5 is specifically: Step 5.1, selecting an adaptive function; The adaptive function f is: wherein η(n, λ · T a ) represents the efficiency when the front motor rotates at n and outputs λ · T a ; η(n, (1 - λ)T a ) represents the efficiency when the rear motor rotates at n and outputs (1 - λ)T a ; λ · T a = T f , (1 - λ)T a = T r , T a = T req ; Step 5.2, defining the position and extreme value of the particle; Step 5.3, establishing a constraint condition; The initial torque distribution coefficient is set to λ=0.5; and the optimal torque distribution coefficient λ satisfies 0.5<λ<1; Step 5.4, setting the particle swarm algorithm parameters; Step 5.5, initializing the particle swarm; The fitness value of each particle is calculated; The individual optimal value is updated; the global optimal value is updated; Whether the iteration reaches the maximum iteration number is judged, if not, the speed and position of the particle are updated, if yes, the current fitness maximum value is the global optimal value, and the corresponding particle position is the optimal torque distribution coefficient.

7. The optimal torque distribution method for dual-motor electric vehicles with wheel slip ratio constraints according to claim 1, characterized in that, The step 6 is specifically: The output torque of the front motor at the current time is: T f = λ · T a ; The output torque of the rear motor at the current time is: T r = (1 - λ)T a where T a = T req .

Images