Regenerative Braking Control Method, Storage Medium and Vehicle Based on Parameter Estimation

Abstract

The present invention relates to a regenerative braking control method based on parameter estimation, which acquires the structural parameters of the whole vehicle and uses sensors to obtain the parameters during vehicle driving; constructs a vehicle model considering the estimated parameters; constructs a parameter estimator to provide an accurate parameter estimation value; constructs an energy recovery optimization model and a torque distribution model to obtain the calculation coefficients corresponding to the highest efficiency of the dual-motor device under different wheel speeds and different total regenerative braking torque requirements; calculates the braking safety area according to the structural parameters of the whole vehicle, the parameter estimation value, the ECE braking safety regulations, and the ideal braking distribution curve, and generates a regenerative braking strategy according to the calculation coefficients and the braking safety area to obtain a regenerative braking strategy with the highest regenerative braking energy recovery efficiency. Parameter estimation is introduced before braking force distribution to improve the accuracy of the parameters on which the braking distribution calculation depends, and to improve the control effect of the vehicle regenerative braking system to improve the energy recovery efficiency.

Description

Technical Field

[0001] The present invention relates to the field of regenerative braking, and more specifically, to a regenerative braking control method, a storage medium, and an automobile based on parameter estimation. Background Art

[0002] Regenerative braking, also known as feedback braking, is a braking technology used in electric vehicles. During braking, the kinetic energy of the vehicle is converted and stored, rather than being turned into useless heat. In the braking condition, the electric motor is switched to operate as a generator, and the inertia of the vehicle drives the rotor of the electric motor to rotate, generating a reverse torque, and converting a part of the kinetic energy or potential energy into electrical energy for storage or utilization. Therefore, this is a process of energy recovery. Regenerative braking is widely applied to pure electric vehicles, hybrid electric vehicles, and railway locomotives.

[0003] Among them, there is a regenerative braking control method for hybrid electric vehicles based on the DQN algorithm, which realizes multi-objective optimization of energy recovery and braking stability during the braking process of hybrid electric vehicles, so that the regenerative braking of hybrid electric vehicles can be reasonably distributed, thereby effectively improving the energy recovery efficiency and braking stability of hybrid electric vehicles.

[0004] However, regarding the regenerative braking of automobiles, in the above method, some parameters of the automobile are also required to be obtained. Since regenerative braking occurs during driving, if most of the parameters in the above method are only the parameters in the inherent state of the automobile, and the changes in the driving process of the automobile are rarely considered, the control effect of the vehicle regenerative braking control system that depends on the vehicle state will be insufficient. Summary of the Invention

[0005] In order to overcome the above problems in the prior art, the present invention provides a regenerative braking control method, a storage medium, and an automobile based on parameter estimation, which can obtain the information during the vehicle driving process in real time and improve the control effect of the vehicle regenerative braking system.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is: a regenerative braking control method based on parameter estimation, specifically including the following steps:

[0007] Step 1: Obtain the structural parameters of the whole vehicle and use sensors to obtain the parameters during the vehicle driving process;

[0008] Step 2: Construct a vehicle model considering the estimated parameters by obtaining the parameters during the vehicle driving process;

[0009] Step 3: Construct a parameter estimator to provide an accurate parameter estimation value;

[0010] Step 4: Construct an energy recovery optimization model and a torque distribution model to obtain the calculation coefficients corresponding to the highest efficiency of the dual-motor device under different wheel speeds and different total regenerative braking torque requirements,

[0011] Step 5: Calculate the braking safety area according to the structural parameters of the vehicle, the parameter estimation values, the ECE braking safety regulations, and the ideal braking distribution curve. Generate a regenerative braking strategy based on the calculation coefficients and the braking safety area to obtain a regenerative braking strategy with the highest regenerative braking energy recovery efficiency. The actual structural parameters are the parameters in the inherent state of the vehicle.

[0012] Vehicle driving parameters required in the braking force distribution calculation process, such as vehicle mass and road gradient, etc. These parameters change with the driving conditions. For example, the vehicle mass is affected by passengers and loaded goods, and different road gradients may also change. These parameters directly affect the calculation of braking force distribution. In this solution, parameter estimation is introduced before braking force distribution to improve the accuracy of the parameters on which the braking distribution calculation depends and improve the control effect of the vehicle regenerative braking system.

[0013] Preferably, Step 4 is specifically as follows:

[0014] S4.1: The torque distribution model includes a dual-motor efficiency optimization model and a dual-axis drive dual-motor loss model; establish a dual-motor efficiency optimization model and give the variables to be optimized;

[0015] S4.2: Combine the variables to be optimized with the loss model of a single motor to establish a dual-axis drive dual-motor loss model;

[0016] S4.3: The efficiency of the motor in the power generation state. For a given rotational speed ω and Te, the power generation efficiency of the motor is linearly related to the loss. By reducing the loss power of the motor, the power generation efficiency is improved, thereby improving the energy recovery efficiency. Take the derivative of the total loss of the dual motors with respect to the torque distribution coefficient of the front-axis motor. When the derivative is zero to obtain the extreme point, that is, dP L / d α = 0, the torque distribution coefficient at the minimum loss of the dual-motor system can be obtained, and the torque distribution coefficient is the calculation coefficient.

[0017] Preferably, the energy recovery optimization model and the dual-motor efficiency optimization model are specifically as follows:

[0018]

[0019] In the formula, ω1 and ω2 are the electrical angular velocities of the front and rear motors respectively; p is the number of pole pairs of the front and rear motors; Te1 and Te2 are the braking torques of the front and rear axle motors respectively; ɑ is the torque distribution coefficient of the front-axis motor; P L1 、P L2are the front and rear motor loss powers respectively; Maxη sys is the maximum value of the double-motor utilization efficiency and also the maximum value of the braking energy recovery; Te is the total regenerative braking torque;

[0020] The double-axis drive double-motor loss model is specifically as follows:

[0021]

[0022] In the formula, P L is the total loss of the double motor; R a1 and R a2 are the stator winding phase resistances of the front and rear motors respectively; ψ f1 and ψ f2 are the magnetic fluxes generated by the permanent magnets of the front and rear motors respectively; R c1 and R c2 are the equivalent iron resistances of the front and rear motors respectively; L1 and L2 are the inductances of the front and rear motors respectively; i wd1 and i wd2 are the active component values of the stator d-axis of the front and rear motors respectively; K f1 is the friction resistance coefficient of the rear motor; K f2 is the friction resistance coefficient of the rear motor.

[0023] Preferably, step four is specifically:

[0024] Define the double-motor utilization efficiency as:

[0025]

[0026] In the formula, η sys is the double-motor utilization efficiency; T e1 and T e2 are the torques of the front and rear motors respectively; n1 and n2 are the rotational speeds of the front and rear motors respectively; η e1 is the efficiency of the front motor at torque T e1 , rotational speed n1; η e2 is the efficiency of the rear motor at torque T e2 , rotational speed n2;

[0027] Establish the energy recovery optimization model and torque distribution model as follows, where the energy recovery optimization model is the optimal braking energy recovery model for double-axis drive

[0028]

[0029] In the formula: α is the front axle motor torque distribution coefficient; T e is the total regenerative braking torque; T e1max (n1) is the maximum torque that the front motor can output at a rotational speed of n1; T e2max(n2) is the maximum torque that the rear motor can output when the rotational speed is n2; α L is the lower bound of the value range of α, and α U is the upper bound of the value range of α;

[0030] Calculate the proportion of the front-wheel regenerative braking torque in the total regenerative braking torque corresponding to different rotational speeds on the wheels and different regenerative braking torques required on the wheels according to the map diagrams of the front and rear motors, and the proportionality coefficient makes the utilization efficiency of the front and rear motor systems the highest.

[0031] Preferably, in step five, during the calculation of the braking force, the vehicle satisfies the longitudinal dynamics model:

[0032]

[0033] In the formula, Fw is the air resistance, C D is the air resistance coefficient; A is the frontal area, v is the vehicle speed; T1 and T2 are the rolling resistance couples of the front and rear wheels; m is the vehicle mass; δ is the conversion coefficient of the rotating mass of the vehicle, θ is the inclination angle of the ramp; r is the rolling radius of the wheel; F xb1 is the front axle braking force, and F xb2 is the rear axle braking force;

[0034] Specifically, when the ideal braking force curve is that the braking force of the vehicle is distributed along the I curve, the relationship between the front-wheel braking force and the rear-wheel braking force is as follows:

[0035]

[0036] In the formula, F xb1 is the front axle braking force, F xb2 is the rear axle braking force, F xb1 and F xb2 are obtained by substituting the vehicle mass as the parameter estimated value into the longitudinal dynamics model; L is the wheelbase; hg is the height of the center of mass; b is the distance from the center of mass to the rear axle; g is the acceleration due to gravity;

[0037] The ECE braking safety regulations limit the range of the braking force intensity threshold and obtain the relationship between the front and rear braking forces:

[0038]

[0039] The range formed by the ECE braking safety regulations and the ideal braking force curve is the safe braking area.

[0040] Preferably, a first braking force intensity threshold and a second braking force intensity threshold are set, and the regenerative braking strategy is specifically as follows:

[0041] When the actual braking force intensity is less than or equal to the first braking force intensity threshold, the distribution relationship of the front and rear wheel braking forces is as follows:

[0042]

[0043] Wherein, F bf1 and F bf2 are the hydraulic braking force of the front axle and the hydraulic braking force of the rear axle respectively;

[0044] When the actual braking force intensity is greater than the first braking force threshold and less than the second braking force threshold, the braking force distribution relationship is as follows:

[0045]

[0046] Wherein, F FI is the front axle braking force distributed according to the I line; F RI is the rear axle braking force distributed according to the I line; i is the transmission ratio; (F e1 *η e1 +F e2 *η e2 )*η1*η3 / i ≤ T bat , T bat is the maximum charging torque allowed by the battery pack, and the torque provided for charging the battery during energy recovery should be less than the torque that the battery can bear; η1 is the mechanical transmission efficiency; η3 is the battery charging efficiency;

[0047] When the actual braking force intensity is greater than or equal to the second braking force threshold, the braking force distribution relationship is as follows:

[0048]

[0049] Wherein, z is the braking force intensity h g is the height of the center of mass; L is the wheelbase.

[0050] Preferably, in step two, a corresponding vehicle longitudinal kinematic model or longitudinal dynamic model is established according to the parameters to be estimated.

[0051] Preferably, in step three, the parameter estimator can be based on a trained neural network, an extended Kalman filter or an adaptive Kalman filter.

[0052] A storage medium stores a computer program, and when the computer program is read and run by a processor, the above-mentioned parameter-estimation-based regenerative braking control method is implemented.

[0053] A vehicle includes a braking system, and when the braking system performs regenerative braking, it executes according to the above-mentioned parameter-estimation-based regenerative braking control method.

[0054] Compared with the prior art, the beneficial effects of the present invention are as follows: Before the braking force distribution, parameter estimation is introduced in the present invention to improve the accuracy of the parameters on which the braking distribution calculation depends, and to improve the control effect of the vehicle regenerative braking system so as to improve the energy recovery efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a flowchart of the regenerative braking control method based on parameter estimation of the present invention;

[0056] Figure 2 is the braking force distribution diagram of an ECE regulation car;

[0057] Figure 3 is the braking safety area diagram. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0058] The drawings are only for illustrative purposes and should not be construed as limitations of this patent; for a better illustration of this embodiment.

[0059] The technical solutions of the present invention will be further specifically described below through specific embodiments and in conjunction with the drawings:

[0060] Embodiment 1

[0061] As Figure 1 shown in the embodiment of the regenerative braking control method based on parameter estimation, the method specifically includes the following steps:

[0062] Step 1: Obtain the structural parameters of the whole vehicle and use sensors to obtain the parameters during the vehicle driving process;

[0063] Step 2: Construct a vehicle model considering the estimated parameters by obtaining the parameters during the vehicle driving process;

[0064] Step 3: Construct a parameter estimator to provide an accurate parameter estimation value;

[0065] Step 4: Construct an energy recovery optimization model and a torque distribution model to obtain the torque distribution coefficient of the dual-motor regenerative braking corresponding to the highest efficiency of the dual-motor device under different rotational speeds of the wheels and different total regenerative braking torque requirements; the specific process is as follows:

[0066] S4.1: The torque distribution model includes a dual-motor efficiency optimization model and a dual-axis drive dual-motor loss model; establish a dual-motor efficiency optimization model and give the variables to be optimized;

[0067] S4.2: Combine the variables to be optimized with the loss model of a single motor to establish a dual-axis drive dual-motor loss model;

[0068] S4.3: The efficiency of the motor in the power generation state. For a given rotational speed ω and Te, the power generation efficiency of the motor is linearly correlated with the losses. By reducing the loss power of the motor, the power generation efficiency can be improved, thereby enhancing the energy recovery efficiency. Let dP L2 / d α = 0, and the torque distribution coefficient at the minimum loss of the dual-motor system can be obtained. This torque distribution coefficient is the calculation coefficient.

[0069] Preferably, the energy recovery optimization model and the dual-motor efficiency optimization model are specifically as follows:

[0070]

[0071]

[0072] In the formula, ω1 and ω2 are the electrical angular velocities of the front and rear motors respectively; p is the number of pole pairs of the front and rear motors; Te1 and Te2 are the braking torques of the front and rear axle motors respectively; ɑ is the torque distribution coefficient of the front axle motor; P L1 、P L2 are the loss powers of the front and rear motors respectively; Maxη sys is the maximum value of the utilization efficiency of the dual-motor, and also the maximum value of the braking energy recovery; Te is the total regenerative braking torque;

[0073] The loss model of the shaft-driven dual-motor is specifically as follows:

[0074]

[0075] In the formula, P L is the total loss of the dual-motor; R a1 、R a2 are the stator winding phase resistances of the front and rear motors respectively; ψ f1 、ψ f2 are the magnetic fluxes generated by the permanent magnets of the front and rear motors respectively; R c1 、R c2 are the equivalent iron resistances of the front and rear motors respectively; L1 and L2 are the inductances of the front and rear motors respectively; i wd1 、i wd2 are the active component values of the stator d-axis of the front and rear motors respectively; K f1 is the friction resistance coefficient of the rear motor; K f2 is the friction resistance coefficient of the rear motor.

[0076] Step Five: Calculate the braking safety area based on the structural parameters of the vehicle, the parameter estimation values, the ECE braking safety regulations, and the ideal braking distribution curve. Generate a regenerative braking strategy based on the calculation coefficient and the braking safety area to obtain a regenerative braking strategy with the highest regenerative braking energy recovery efficiency. During the calculation of the braking force, the vehicle satisfies the longitudinal dynamics model:

[0077]

[0078] Wherein, Fw is the air resistance, C D is the air resistance coefficient; A is the frontal area, v is the vehicle speed; T1 and T2 are the rolling resistance couple moments of the front and rear wheels; m is the vehicle mass; δ is the conversion coefficient of the rotating mass of the vehicle, θ is the inclination angle of the ramp; r is the rolling radius of the wheel; F xb1 is the braking force of the front axle, F xb2 is the braking force of the rear axle.

[0079] Specifically, when the braking force of the vehicle is distributed along the I curve, the braking force of the front wheel and the braking force of the rear wheel of the ideal braking force curve satisfy the following relationship:

[0080]

[0081] Wherein, F xb1 is the braking force of the front axle, F xb2 is the braking force of the rear axle, F xb1 and F xb2 are obtained by substituting the vehicle mass as the parameter estimated value into the longitudinal dynamics model; L is the wheelbase; hg is the height of the center of mass; b is the distance from the center of mass to the rear axle; g is the acceleration due to gravity;

[0082] To ensure the braking direction stability and braking efficiency. The ECE regulations for car braking are as Figure 2 shown. For various vehicles between z = 0.2 and 0.8, the utilization coefficient of adhesion should satisfy φ ≤ (z + 0.07) / 0.85. When the braking force intensity z is between 0.3 and 0.4 and the utilization coefficient curve of the rear axle does not exceed the straight line φ = z + 0.05, the utilization coefficient of the rear axle can be higher than that of the front axle.

[0083] According to the braking force distribution regulations for cars in the ECE regulations, when z = 0.2 to 0.8, the relationship between the front and rear braking forces can be obtained:

[0084]

[0085] When z ≤ 0.2, the regulations do not make strict provisions. And in the actual driving process, the road surface with a coefficient of adhesion above 0.5 is relatively common. Therefore, when the braking force intensity is 0.2, the possibility of wheel lock-up is relatively small. So in this embodiment, when the braking force intensity does not exceed 0.2, it is allowed that the utilization coefficient of the rear axle is higher than that of the front axle. To sum up, the obtained safe braking area is Figure 3 the area OABCDEFO enclosed by the thick black line in. Points A, B, E, and F are the intersections of the equal braking force distribution line of z = 0.2 with the y-axis, the I line, the lower boundary line of the ECE regulations, and the x-axis respectively. Points C and D are the intersections of the f line of φ = 0.8 with the I curve and the lower boundary line of the ECE regulations respectively.

[0086] The calculation of the safety braking area is consistent with that of the prior art. The difference is that the front axle braking force and the rear axle braking force in the prior art are default known values, while the front axle braking force and the rear axle braking force in this embodiment are calculated values obtained by substituting the vehicle mass as a parameter estimate into the longitudinal dynamics model. Since the default known values are not the actual values of the vehicle, and the calculated values are estimated values calculated according to the current situation of the vehicle, they are more accurate than the default known values.

[0087] Set the first braking force intensity threshold and the second braking force intensity threshold. In this embodiment, the first braking force intensity threshold and the second braking force intensity threshold are 0.2 and 0.5 respectively. When the actual braking force intensity is less than or equal to the first braking force intensity threshold, the braking force distribution relationship between the front and rear wheels is as follows:

[0088]

[0089] In the formula, F bf1 and F bf2 are the front axle hydraulic braking force and the rear axle hydraulic braking force respectively;

[0090] When the actual braking force intensity is greater than the first braking force threshold and less than the second braking force threshold, the braking force distribution relationship is as follows:

[0091]

[0092] In the formula, F FI is the front axle braking force distributed according to the I line; F RI is the rear axle braking force distributed according to the I line; i is the transmission ratio; (F e1 *η e1 +F e2 *η e2 )*η1*η3 / i ≤ T bat , T bat is the maximum charging torque allowed by the battery pack;

[0093] When the actual braking force intensity is greater than or equal to the second braking force threshold, the braking force distribution relationship is as follows:

[0094]

[0095] In the formula, z is the braking force intensity; h g is the height of the center of mass; L is the wheelbase.

[0096] Working principle or working process of this embodiment: Vehicle driving parameters required in the braking force distribution calculation process, such as vehicle mass and road slope, etc. These parameters change with the change of driving conditions. For example, the vehicle mass is affected by passengers and loaded goods, and different road slopes may also change. These parameters directly affect the calculation of braking force distribution. This solution introduces parameter estimation before performing braking force distribution to improve the accuracy of the parameters on which the braking distribution calculation depends and improve the control effect of the vehicle regenerative braking system.

[0097] Beneficial effects of this embodiment: The present invention introduces parameter estimation before performing braking force distribution to improve the accuracy of the parameters on which the braking distribution calculation depends and improve the control effect of the vehicle regenerative braking system to improve the energy recovery efficiency.

[0098] Embodiment 2

[0099] As Figure 1 shown in Embodiment 2 of the regenerative braking control method based on parameter estimation, specifically including the following steps:

[0100] Step 1: Obtain the structural parameters of the whole vehicle and use sensors to obtain the parameters during vehicle driving;

[0101] Step 2: Construct a vehicle model considering estimated parameters by obtaining the parameters during vehicle driving;

[0102] Step 3: Construct a parameter estimator to provide an accurate parameter estimation value;

[0103] Step 4: Construct an energy recovery optimization model and a torque distribution model to obtain the proportional coefficient of the front-wheel regenerative braking torque to the total regenerative braking torque corresponding to the highest efficiency of the dual-motor device under different wheel speeds and different total regenerative braking torque requirements on the wheels; The specific process is as follows:

[0104] Define the utilization efficiency of the dual motors as:

[0105]

[0106] In the formula, η sys is the utilization efficiency of the dual motors; T e1 and T e2 are the torques of the front and rear motors respectively; n1 and n2 are the rotational speeds of the front and rear motors respectively; η e1 is the efficiency of the front motor at torque T e1 , rotational speed n1; η e2 is the efficiency of the rear motor at torque T e2 , rotational speed n2;

[0107] The energy recovery optimization model and torque distribution model are established as follows. The energy recovery optimization model is the optimal braking energy recovery model for a two-axis drive.

[0108]

[0109] In the formula: α is the torque distribution coefficient of the front axle motor; T e is the total regenerative braking torque; T e1max (n1) is the maximum torque that the front motor can output at a speed of n1; T e2max (n2) is the maximum torque that the rear motor can output at a speed of n2; α L is the lower bound of the value of α, and α U is the upper bound of the value of α;

[0110] According to the map diagrams of the front and rear motors, calculate the proportional coefficient of the front-wheel regenerative braking torque to the total regenerative braking torque corresponding to different wheel speeds and different wheel regenerative braking torques required. This proportional coefficient maximizes the utilization efficiency of the front and rear motor systems.

[0111] Step Five: Calculate the braking safety area based on the structural parameters of the vehicle, the parameter estimation values, the ECE braking safety regulations, and the ideal braking distribution curve. The braking safety area is the same as that in Embodiment 1. Generate a regenerative braking strategy based on the calculated coefficient and the braking safety area to obtain a regenerative braking strategy with the highest regenerative braking energy recovery efficiency. Set the first braking force intensity threshold and the second braking force intensity threshold. In this embodiment, the regenerative braking strategy is specifically as follows:

[0112] The first braking force intensity threshold and the second braking force intensity threshold are 0.2 and 0.5 respectively. When the actual braking force intensity is less than or equal to the first braking force intensity threshold, the braking force distribution relationship between the front and rear wheels is as follows:

[0113]

[0114] In the formula, F bf1 and F bf2 are the front axle hydraulic braking force and the rear axle hydraulic braking force respectively;

[0115] When the actual braking force intensity is greater than the first braking force threshold and less than the second braking force threshold, the braking force distribution relationship is as follows:

[0116]

[0117] In the formula, F FI is the front axle braking force distributed according to the I-line; F RI is the rear axle braking force distributed according to the I-line; i is the transmission ratio; (F e1 *η e1 +F e2 *ηe2 ) * η1 * η3 / i ≤ T bat , T bat is the maximum charging torque allowed for the battery pack;

[0118] When the actual braking force intensity is greater than or equal to the second braking force threshold, the braking force distribution relationship is as follows:

[0119]

[0120] In the formula, z is the braking force intensity; h g is the height of the center of mass; L is the wheelbase.

[0121] Preferably, in step two, a corresponding vehicle longitudinal kinematic model or longitudinal dynamic model is established according to the parameters to be estimated.

[0122] Preferably, in step three, the parameter estimator can be a trained neural network, an extended Kalman filter, or an adaptive Kalman filter.

[0123] Embodiment 3

[0124] Embodiment 3 of the regenerative braking control method based on parameter estimation, based on the method of Embodiment 1 or 2, is different in that step two is further limited.

[0125] In step two, a vehicle model considering the estimated parameters is constructed. The parameter estimated in this embodiment is the vehicle mass, and a longitudinal dynamic model considering the vehicle mass and road slope is constructed as follows:

[0126]

[0127] In the formula: T tq is the driving torque, i g is the transmission ratio of the transmission, i0 is the reduction ratio of the final drive, η T is the mechanical efficiency of the transmission chain, r is the rolling radius of the wheel, C d is the air resistance coefficient, A f is the frontal area, ρ is the air density, v x is the vehicle speed, m is the vehicle mass, g is the acceleration due to gravity, y is the slope angle, f is the rolling resistance coefficient, a x is the vehicle acceleration.

[0128] Embodiment 4

[0129] Embodiment 4 of the regenerative braking control method based on parameter estimation, based on the method of Embodiment 3, is different in that step three is further limited.

[0130] In step three, a neural network-based vehicle mass estimator is designed, and the specific process is as follows:

[0131] S31: Select parameters such as vehicle speed and acceleration as the input feature A of the neural network according to the kinetic model t , and use the vehicle mass as the output. Train the neural network model using the vehicle parameter data collected by the sensors.

[0132] S32: The neural network selects a multi-layer perceptron. Hyperparameters such as the number of network layers and learning rate are selected according to data tuning. There is no restriction on the deep learning framework on which this neural network is based, including but not limited to Pytorch, Tensorflow, PaddlePaddle, MXNet, etc.

[0133] S33; Deploy the trained network model to the vehicle, and based on the real-time data of the on-vehicle sensors, calculate the accurate estimated value m of the vehicle mass in real time nn .

[0134] Example 5

[0135] Example 5 of the regenerative braking control method based on parameter estimation, based on the method of Example 3, the difference lies in further limitation of Step 3.

[0136] The parameter estimator in Step 3 is based on the vehicle mass estimation method of adaptive Kalman filtering, specifically:

[0137] S31: Design a weight regulator

[0138] Although the estimation result of the neural network estimator is accurate, it is unstable. This instability is harmful to the control system. To avoid the damage caused by this oscillation to the system, a weight regulator is introduced to calculate the weight factor and design an adaptive law. The weight factor can be used to adjust the influence of the estimation results of the neural network estimator and the extended Kalman filter on the final estimation result.

[0139] S32: Embed the pre-estimated value provided by the neural network estimator into the extended Kalman filter, and introduce an adaptive law to design an adaptive extended Kalman filter. The steps are as follows:

[0140]

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147]

[0148] Among them, H changes from [1 0 0] to [1 1 0],

[0149] The observed quantity is then extended to:

[0150]

[0151] τ k =(1 - β k ) / β k ∈[0, +∞). When the new input data of the neural network is far from the original data set, β k →0, τ k increases, and the posterior estimation depends more on the matching degree between the model and the actual working conditions. As the similarity increases, β k →1, τ k decreases and approaches 0. The observation covariance part of the neural network is equal to zero, and the final estimation result depends more on the neural network.

[0152] Example 6

[0153] Example 6 of the regenerative braking control method based on parameter estimation, based on the method of Example 1 or 2, is different in that steps two and three are further defined.

[0154] In step two, the estimated parameter is the slope. A longitudinal kinematic model considering the road slope is constructed as follows:

[0155]

[0156] In the formula: a x Acceleration value measured by the longitudinal acceleration sensor.

[0157] The state quantity is selected as:

[0158] x2 = [v i] T

[0159] It can be further expressed as:

[0160]

[0161] In the formula: Δt is the sampling time, and W2 is the process noise.

[0162] The observed quantity is selected as:

[0163] y2 = [v]

[0164] The observation equation is established as follows:

[0165]

[0166] Where: Δt is the sampling time, and V2 is the process noise.

[0167] Finally, the state - space expression of the longitudinal kinematics is obtained as follows:

[0168]

[0169] Where: h2 is the observation matrix.

[0170] In step three, based on the constructed vehicle longitudinal kinematic model, a slope estimator is designed, and an extended Kalman filter (EKF) based on longitudinal kinematics is established.

[0171] 1) Prediction

[0172]

[0173] 2) Correction

[0174]

[0175] Where, x s2 = [v i] T , P s2 is the covariance matrix, Q s2 is the process - noise covariance matrix, R s2 is the measurement - noise covariance matrix, H s2 = h2 = [1 0].

[0176] Example 7

[0177] A storage medium stores a computer program, which, when read and run by a processor, implements the regenerative braking control method based on parameter estimation in any of the above - mentioned embodiments.

[0178] Example 8

[0179] A vehicle includes a braking system, which, when performing regenerative braking, executes according to the regenerative braking control method based on parameter estimation in any of the above - mentioned embodiments.

[0180] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, rather than limitations on the implementation manners of the present invention. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation manners here. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the claims of the present invention.

Claims

1. A regenerative braking control method based on parameter estimation, characterized in that, Specifically, it includes the following steps: Step 1: Obtain the structural parameters of the whole vehicle and use sensors to obtain the parameters during vehicle driving; Step 2: Construct a vehicle model considering the estimated parameters by obtaining the parameters during vehicle driving; Step 3: Construct a parameter estimator to provide a parameter estimation value according to the vehicle model considering the estimated parameters; Step 4: Construct an energy recovery optimization model and a torque distribution model to obtain the calculation coefficients corresponding to the highest efficiency of the dual-motor device of the vehicle under different wheel speeds and different total regenerative braking torque requirements on the wheels; S4.1: The torque distribution model includes a dual-motor efficiency optimization model and a dual-axis drive dual-motor loss model; establish a dual-motor efficiency optimization model and give the variables to be optimized; specifically: S4.2: Combine the variables to be optimized with the loss model of a single motor to establish a dual-axis drive dual-motor loss model; S4.3: The efficiency of the motor in the power generation state. For the given rotational speed ω and Te, the power generation efficiency of the motor is linearly related to the loss. Take the derivative of the total motor loss with respect to the torque distribution coefficient of the front-axis motor. When the derivative is zero to obtain the extreme point, the torque distribution coefficient at the minimum loss of the dual-motor system can be obtained, and the torque distribution coefficient is the calculation coefficient; The energy recovery optimization model and the dual-motor efficiency optimization model are specifically as follows: where ω1 and ω2 are the electrical angular velocities of the front and rear motors respectively; p is the number of pole pairs of the front and rear motors; T e1 and T e2 are the braking torques of the front and rear axle motors respectively; ɑ is the torque distribution coefficient of the front axle motor; P L1 and P L2 are the loss powers of the front and rear motors respectively; is the maximum value of the double-motor utilization efficiency and also the maximum value of the braking energy recovery; T e is the total regenerative braking torque; The dual-axis drive dual-motor loss model is specifically as follows: Where, P L is the total loss of the dual motors; R a1 , R a2 are the phase resistances of the stator windings of the front and rear motors respectively; ψ f1 , ψ f2 are the magnetic fluxes generated by the permanent magnets of the front and rear motors respectively; R c1 , R c2 are the equivalent iron resistances of the front and rear motors respectively; L1 and L2 are the inductances of the front and rear motors respectively; i wd1 , i wd2 are the active component values of the d-axis of the stators of the front and rear motors respectively; K f1 is the friction resistance coefficient of the rear motor; K f2 is the friction resistance coefficient of the rear motor; Step 5: Calculate the braking safety area according to the structural parameters of the whole vehicle, the parameter estimation value, the ECE braking safety regulations, and the ideal braking distribution curve. Generate a regenerative braking strategy according to the calculation coefficient and the braking safety area to obtain a regenerative braking strategy with the highest regenerative braking energy recovery efficiency; During the calculation of the braking force, the vehicle satisfies the longitudinal dynamics model: where \(F_w\) is the air resistance, \(v\) is the vehicle speed; \(T_1\) and \(T_2\) are the rolling resistance couple moments of the front and rear wheels; \(m\) is the vehicle mass; \(\delta\) is the conversion coefficient of the rotating mass of the vehicle, is the inclination angle of the ramp; \(r\) is the rolling radius of the wheel; \(F\) xb1 front axle braking force, \(F\) xb2 is the rear axle braking force; The ideal braking force curve is specifically that when the braking force of the vehicle is distributed along the I curve, the braking force of the front wheel and the braking force of the rear wheel satisfy the following relationship: Where, F xb1 is the front axle braking force, F xb2 is the rear axle braking force, F xb1 and F xb2 are obtained by substituting the vehicle mass as the parameter estimate value into the longitudinal dynamics model; L is the wheelbase; hg is the center of mass height; b is the distance from the center of mass to the rear axle; g is the acceleration due to gravity; The ECE braking safety regulations limit the range of the braking force intensity threshold and obtain the relationship between the front and rear braking forces: The range formed by the ECE braking safety regulations and the ideal braking force curve is the safety braking area.

2. The regenerative braking control method based on parameter estimation according to claim 1, wherein Step 4 is specifically: Define the utilization efficiency of the dual motor as: Where, η sys is the utilization efficiency of the dual motors; T e1 and T e2 are the torques of the front and rear motors respectively; n1 and n2 are the rotational speeds of the front and rear motors respectively. η e1 is the efficiency of the front motor at torque T e1 , speed n1; η e2 is the efficiency of the rear motor at torque T e2 , speed n2; Establish an energy recovery optimization model and a torque distribution model as follows, where the energy recovery optimization model is the optimal braking energy recovery model for dual-axis drive, Where: α is the front axle motor torque distribution coefficient; T e is the total regenerative braking torque; T e1max (n1) is the maximum torque that the front motor can output at a speed of n1; T e2max (n2) is the maximum torque that the rear motor can output at a speed of n2; α L is the lower bound of the value of α, α U is the upper bound of the value of α; Calculate the proportional coefficient of the front-wheel regenerative braking torque to the total regenerative braking torque corresponding to different wheel speeds and different wheel regenerative braking torques required according to the map of the front and rear motors. The proportional coefficient makes the utilization efficiency of the front and rear motor systems the highest, and the proportional coefficient is the calculation coefficient.

3. The regenerative braking control method based on parameter estimation according to claim 1, wherein Set the first braking force intensity threshold and the second braking force intensity threshold. The regenerative braking strategy is specifically as follows: When the actual braking force intensity is less than or equal to the first braking force intensity threshold, the braking force distribution relationship between the front and rear wheels is as follows: where F bf1 and F bf2 are the hydraulic braking force of the front axle and the hydraulic braking force of the rear axle, respectively; When the actual braking force intensity is greater than the first braking force threshold and less than the second braking force threshold, the braking force distribution relationship is as follows: Where F FI is the braking force of the front axle distributed according to the I-line; F RI is the braking force of the rear axle distributed according to the I-line; i is the transmission ratio; (F e1 *η e1 +F e2 *η e2 )*η 1* η3 / i*r ≤ T bat , T bat is the maximum charging torque allowed by the battery pack, and the torque provided for charging the battery during energy recovery should be less than the torque that the battery can bear; η1 is the mechanical transmission efficiency; η3 is the battery charging efficiency; When the actual braking force intensity is greater than or equal to the second braking force threshold, the braking force distribution relationship is as follows: where z is the braking force intensity; h g is the height of the center of mass; L is the wheelbase.

4. The regenerative braking control method based on parameter estimation according to claim 1, wherein In Step 2, establish a corresponding vehicle longitudinal kinematic model or longitudinal dynamics model according to the parameters to be estimated.

5. The regenerative braking control method based on parameter estimation according to claim 1, wherein In step three, the parameter estimator is based on a trained neural network, an extended Kalman filter, or an adaptive Kalman filter.

6. A storage medium, characterized in that, A computer program is stored, and when the computer program is read and run by a processor, it implements the regenerative braking control method based on parameter estimation according to any one of claims 1-5.

7. A vehicle, comprising a braking system, characterized in that, When the braking system performs regenerative braking, it executes according to the regenerative braking control method based on parameter estimation according to any one of claims 1-5.

Images