An adaptive electric vehicle traction force control method and system based on maximum wheel end transmissible torque

Abstract

The application discloses a kind of self-adapting electric vehicle traction force control method and system based on maximum wheel end transmissible torque, it is related to electric vehicle dynamics control technical field;The system includes: signal acquisition and pretreatment module, road surface state discrimination module, adaptive scheduling module, NMETE calculation module, torque attenuation limiting module and motor controller;The method judges the road adhesion state by wheel-vehicle acceleration difference, then determines slip and online updates relaxation factor by torque ratio and acceleration difference, and then obtains the maximum effective torque estimation value;Under low adhesion road surface, according to the low adhesion duration Attenuation coefficient is constructed, the output torque is limited, and the final torque upper limit is obtained, so as to improve the traction utilization rate and driving stability under complex road surface and multiple working conditions.Compared with the conventional METE controller, the application can better adapt to complex road surface and multiple working conditions, improve the slip rate change, and improve the traction utilization rate and vehicle driving stability.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle dynamics control technology, and more specifically, to an adaptive electric vehicle traction control method and system based on the maximum wheel-end transmittable torque. This invention is applicable to traction control of electric vehicles under complex road conditions and can be applied to drive anti-skid control systems for passenger cars, commercial vehicles, and special vehicles. Background Technology

[0002] With the rapid development of electric vehicles, electric drive systems have been widely used due to their advantages such as fast response, flexible control, and compact structure. However, under low-traction road conditions, electric vehicles are prone to wheel slippage, leading to loss of traction, decreased vehicle stability, and even safety accidents.

[0003] Traditional traction control methods each have significant shortcomings in implementation. Slip ratio-based control methods aim to keep the wheel slip ratio near a target value, preventing the wheel from entering unstable regions. Typical examples include PID control and fuzzy control. These methods require precise slip ratio measurement or estimation and are sensitive to model parameters. Model predictive control (MPC) methods build a vehicle-tire system model to predict the system state at future moments and optimize control inputs. While theoretically superior, they suffer from high computational complexity and difficulty in meeting real-time requirements.

[0004] The method based on Maximum Effective Torque Estimation (METE) was first proposed by Dejun Yin et al. (see Yin D, Oh S, Hori Y. A Novel Traction Control for EV Based on Maximum Transmissible Torque Estimation[J]. IEEE Transactions on Industrial Electronics, 2009, 56(6):2086-2094.). It estimates the maximum available torque of the wheels and limits the driving torque to prevent wheel slippage. The core formula of the METE method is: ;in, The relaxation factor is typically set to a fixed value. This method is simple to calculate and easy to implement, and has been widely used in engineering. However, traditional METE controllers still have the following limitations: First, the relaxation factor... Traditional METE (Mean Interference-Assisted Traction) systems are typically set to fixed values, making it impossible to adjust in real time according to changes in road surface adhesion conditions and slippage. On high-adhesion surfaces, this can lead to excessive torque restriction, reducing traction utilization; on low-adhesion surfaces, it may result in insufficient restriction, leading to inadequate slippage suppression. Secondly, traditional METE lacks the ability to online discern road conditions, making it difficult to detect sudden changes in adhesion or shifts in operating conditions, resulting in a certain degree of lag in the control strategy. Thirdly, traditional METE has limited robustness under multiple operating conditions and varying road surfaces, and its adaptability to random driver inputs and complex road surface combinations is insufficient. Fourthly, while traditional METE can achieve basic anti-skid measures, its ability to suppress slippage trends is limited, easily leading to problems such as large fluctuations in slip ratio, slow recovery, and insufficient traction utilization, thus affecting the vehicle's longitudinal dynamics and driving stability. Therefore, there is an urgent need for an improved traction control method that can adapt to multiple operating conditions, perceive road surface changes in real time, adjust control parameters online, and effectively suppress slippage trends. Summary of the Invention

[0005] The present invention aims to overcome the shortcomings of the existing technology by proposing an adaptive electric vehicle traction control method and system based on the maximum wheel-end transmittable torque. The goal is to be able to determine the road surface adhesion state in real time, adaptively adjust control parameters, effectively suppress slippage tendency and improve traction utilization, thereby significantly improving the traction control performance and driving stability of electric vehicles under complex road conditions.

[0006] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: The adaptive electric vehicle traction control method based on the maximum wheel-end transmittable torque of the present invention is characterized by the following steps: Step 1, data collection The vehicle wheel-end dynamics signals at time points include: Wheel angular velocity at time , longitudinal acceleration of the vehicle body at any moment , The driver constantly requests reference torque from the drive system. as well as Wheel end torque estimate at time 1 ;right After differentiation, the longitudinal acceleration of the wheel at time t is obtained. , thus calculating Wheel-vehicle acceleration difference at time ; Step 2, for Wheel-vehicle acceleration difference at time After filtering, the result is obtained. Wheel-vehicle acceleration difference after time processing ; Step 3, according to ,right The road surface adhesion state at any given time is determined to obtain... Road surface adhesion status indicator at any time ,like express The road surface adhesion state at a given time is a high-adhesion road surface. express The road surface adhesion state at any given time is low adhesion. Step 4, during the preset duration Down: Always greater than the minimum preheating time threshold ,and Filtered wheel speed at time Always greater than the minimum effective wheel speed threshold ,and wheel-end acceleration estimate at time t The absolute value is always no greater than the upper limit of acceleration. ,and Maximum effective torque estimate at time 1 Always greater than the positive threshold ,make Otherwise, let Thus constructing Controller takeover permission flag at any time ; Step 5, based on ,judge Is the vehicle in dead time at any given moment? when At that time, the judgment The vehicle's state at any given moment is in a dead time, and causes... ,make ,in, for Measure the motor torque command at any given time and execute step 4; when If so, proceed to step 6; Step 6, if Road surface adhesion status indicator at any time and Then let ;in, This indicates the entry threshold, and ; Indicates the sampling period; like and ,make ;in, Indicates the exit threshold, and , ; In other cases, keep constant; Step 7, Construction Torque ratio indication at time ;in, Indicates that the drive motor is in The estimated maximum available torque at time [time]. This parameter is used to prevent the denominator from being zero. Step 8, based on and ,right Determine the vehicle's slippage status at any given moment: when or At that time, the judgment The vehicle skidded at that moment, causing Time-shifting markers Otherwise, determine The vehicle remained stationary at all times, and... ;in, The torque ratio threshold, The threshold value for wheel-vehicle acceleration difference; Step 9, Construction Slide duration timer for moments and Integral quantity of slip intensity at time step ; when At that time, Slide duration timer for moments and Integral quantity of slip intensity at time step Update and get Slide duration timer for moments and slip strength integral ,in, This represents the upper limit of the slip strength integral. Represents a saturation function; The sampling period; when At that time, the preset recovery strategy was applied to each case. and Attenuation; Step 10, according to and Construct separately Time adjustment factor and Integral adjustment factor at time step , thus obtaining Comprehensive adjustment factor at time , and then calculate relaxation factor at time ;in, This is the lower limit of the relaxation factor. This is the upper limit of the relaxation factor. For time adjustment coefficients, This is the coefficient for integral adjustment. This is the upper limit for integral adjustment; Step 11, Calculation Longitudinal driving force between the tire and the road surface at all times and calculate Maximum effective torque estimation coefficient at time t Thus obtain Maximum effective torque estimate at time 1 ;in, This represents the moment of inertia of the wheel. Indicates the equivalent vehicle mass. Indicates the rolling radius of the wheel. express The longitudinal driving force between the tire and the road surface at all times; Step 12, according to ,right By applying secondary restrictions, we obtain The final torque limit at any moment : when At that time, Low-level duration count value at time Set to zero, and let The final torque limit at any moment ; when At that time, calculate Low-level duration count value at time and obtain Logarithmic decay coefficient at time Thus obtain Decaying torque at time ; and thus obtain The final torque limit at any moment ,in, This indicates the attenuation rate adjustment parameter. This represents the steady-state lower limit of the attenuation coefficient. Indicates the absolute minimum torque threshold; Step 13, Motor command torque at any moment The output is sent to the motor controller to limit the actual drive torque in order to achieve the desired traction for the electric vehicle. Real-time adaptive anti-slip control.

[0007] The present invention provides an adaptive electric vehicle traction control system based on maximum wheel-end transmittable torque, characterized by comprising: a signal acquisition and preprocessing module, a road surface condition discrimination module, and... Adaptive scheduling module, NMETE calculation module, torque attenuation limiting module and motor controller; The signal acquisition and preprocessing module is used for acquisition. Vehicle wheel-end dynamics signals at any given time, including Wheel angular velocity at time , longitudinal acceleration of the vehicle body at any moment , The driver constantly requests reference torque from the drive system. as well as Wheel end torque estimate at time 1 ; and on Differentiation yields longitudinal acceleration of the wheel at any moment and calculate Wheel-vehicle acceleration difference at time Afterwards, The wheel-vehicle acceleration difference is obtained by filtering. ; The road surface condition determination module is based on and Controller takeover permission flag at any time ,right The road surface adhesion state at any given time is determined to obtain... Road surface adhesion status indicator at any time ,in Indicates a high-adhesion road surface. Indicates a low-adhesion road surface; The The adaptive scheduling module is used to construct Torque ratio indication at time And combined with the wheel-vehicle acceleration difference determination Whether the vehicle slipped at any given moment, thus obtaining... Time-shifting markers And when slippage is detected, update Slide duration timer for moments and Integral quantity of slip intensity at time step Thus update relaxation factor at time ; The NMETE calculation module is based on Wheel end torque estimate at time 1 , longitudinal acceleration of the wheel at any moment and relaxation factor at time ,calculate Longitudinal driving force between the tire and the road surface at all times , thus calculating Maximum effective torque estimate at time 1 ; The torque attenuation limiting module is based on ,right Apply secondary restrictions: based on Low-level duration count value at time ,calculate Decay coefficient at time Thus obtain The final torque limit at any moment ; The motor controller receives The final torque limit at any moment ,get Motor command torque at any moment It is used to limit the actual driving torque and achieve adaptive anti-slip control of electric vehicle traction.

[0008] Furthermore, the road surface condition determination module includes the following steps: Step a1, during the preset duration Down: Always greater than the minimum preheating time threshold ,and Filtered wheel speed at time Always greater than the minimum effective wheel speed threshold ,and wheel-end acceleration estimate at time t The absolute value is always no greater than the upper limit of acceleration. , Maximum effective torque estimate at time 1 Always greater than the positive threshold ,make Otherwise, let Thus constructing Controller takeover permission flag at any time ; Step a2, when At that time, the judgment The vehicle is in a dead zone at any given time, and the road surface is marked with status indicators. ,make ,in, for The motor command torque at any given moment; when At that time, if Road surface adhesion status indicator at the sampling time ,and Then let ;in, This indicates the entry threshold, and ; Indicates the sampling period; when At that time, if ,and Then let ;in, Indicates the exit threshold, and , In other cases, maintain constant.

[0009] Furthermore, the aforementioned The adaptive scheduling module includes the following steps: Step b1, Construction Torque ratio indication at time ,in, This represents the maximum available torque estimate from the previous sampling period. This parameter is used to prevent the denominator from being zero. Step b2, based on and ,right Determine the vehicle's slippage status at any given moment: when or At that time, the judgment The vehicle skidded at that moment, causing Otherwise, let ;in, The torque ratio threshold, The threshold value for wheel-vehicle acceleration difference; Step b3, when At that time, Slide duration timer for moments and moments Slip strength integral Update and get , ,in, This represents the upper limit of the slip strength integral. Represents a saturation function; when At that time, the preset recovery strategy is used for and Attenuation; Step b4, according to and Construct separately Time adjustment factor and Integral adjustment factor at time step , thus obtaining Comprehensive adjustment factor at time , and then calculate relaxation factor at time ;in, This is the lower limit of the relaxation factor. This is the upper limit of the relaxation factor. For time adjustment coefficients, This is the coefficient for integral adjustment. This is the upper limit for integral adjustment.

[0010] Furthermore, the NMETE calculation module includes: Step c1, according to Wheel end torque estimate at time 1 , longitudinal acceleration of the wheel at any moment Wheel rotational inertia and wheel rolling radius ,calculate Longitudinal driving force between the tire and the road surface at all times ; Step c2, according to relaxation factor at time Wheel rotational inertia Equivalent vehicle mass and wheel rolling radius ,calculate Maximum effective torque estimation coefficient at time t ; Step c3, Calculation Maximum effective torque estimate at time 1 .

[0011] Furthermore, the torque attenuation limiting module includes: Step d1, when At that time, Low-level duration count value at time Set to zero, and let ; when At that time, calculate Low-level duration count value at time ; and with Construct with the independent variable Decay coefficient at time ,in, This indicates the attenuation rate adjustment parameter. This represents the steady-state lower limit of the attenuation coefficient; Step d2, Calculation Decaying torque at time ; Step d3, Calculation The final torque limit at any moment ,in, This represents the absolute minimum torque threshold.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The system of this invention uses a road surface condition discrimination module, The synergistic effect of the adaptive scheduling module, NMETE calculation module, and torque attenuation limiting module enables adaptive multi-condition control by adapting to various working conditions such as single road surface, connected road surface, and random input. Compared with the traditional METE controller, this invention can suppress slippage trend more promptly, improve the utilization of road adhesion, and improve traction utilization and longitudinal dynamic performance while ensuring stability.

[0013] 2. The method of this invention adopts a road surface condition discrimination strategy of "no pulse entry, pulse exit", which can make timely discrimination even on pure low-adhesion road surfaces (without docking pulses), improving the robustness of the system and avoiding the missed discrimination problem of traditional pulse-dependent methods. At the same time, the introduction of filtering and dead-zone mechanisms improves the anti-interference ability.

[0014] 3. This invention allows for the calibration of the relaxation factor range, road condition discrimination threshold, and low adhesion torque attenuation parameter according to different vehicle models, different drive system parameters, and different driving needs, thereby balancing traction performance, stability, and engineering implementation flexibility.

[0015] 4. The method of the present invention is simple to calculate, has good real-time performance, is easy to implement in electric vehicle ECU, has low computational load, low hardware cost, and has good engineering application prospects. Attached Figure Description

[0016] Figure 1 This is an overall flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the module structure of the system of the present invention; Figure 3 This is a module logic diagram of the system of the present invention; Figure 4 This is a schematic diagram illustrating the working principle of the road surface condition determination module. Figure 5 for Working principle diagram of the adaptive scheduling module; Figure 6 Relaxation factor under typical continuous slip conditions The change curve; Figure 7 This is a schematic diagram of the working principle of the NMETE calculation module; Figure 8 A flowchart of the low-adhesion logarithmic torque attenuation limiting module; Figure 9 A comparison diagram of the torque response of the classic METE controller and the present invention under a single low-adhesion road surface condition; Figure 10 A comparison chart showing the slip ratio performance of the classic METE controller and the present invention under dual-interlocking road surface conditions; Figure 11 Performance verification diagram of the present invention under snow conditions and random driver input; Figure 12 The performance verification diagram of the present invention is shown under random road conditions and a driver's step input of 100 Nm. Detailed Implementation

[0017] In this embodiment, an adaptive electric vehicle traction control method based on the maximum wheel-end transmittable torque is described, such as... Figure 1 As shown, the specific logic is as follows: Figure 3 This includes the following steps: Step 1: Acquire vehicle wheel-end dynamic signals, including: Wheel angular velocity at time , longitudinal acceleration of the vehicle body at any moment , The driver constantly requests reference torque from the drive system. as well as Wheel end torque estimate at time 1 ;make Differentiation yields longitudinal acceleration of the wheel at any moment Based on the longitudinal acceleration of the wheels and the longitudinal acceleration of the vehicle body, the difference in wheel acceleration is calculated using equation (1). : (1) This step is used to acquire and preprocess the basic signals required for subsequent control, and it is the input layer of the entire method. Unlike traditional anti-skid methods that focus on the target slip ratio, this invention does not use the slip ratio as a direct closed-loop input to the controller. Instead, it prioritizes utilizing measurable dynamic quantities at the wheel ends and on the vehicle body to construct the wheel-vehicle acceleration difference. And use it as an important basic quantity for road surface condition judgment and slippage trend perception; In engineering implementation, wheel angular velocity The longitudinal acceleration of the vehicle body can be obtained by converting the wheel speed sensor, shaft encoder, or motor-side speed signal through the transmission ratio. Acceleration estimates can be provided from the IMU, vehicle acceleration sensors, or existing acceleration estimates from the vehicle controller; the driver may request reference torque. This usually originates from the motor torque command requested by the driver's accelerator pedal; It should be noted that this is denoted as The longitudinal acceleration of the wheel is essentially the circumferential acceleration calculated by combining the rate of change of the wheel angular velocity with the wheel rolling radius. The advantages of this approach are twofold: firstly, it avoids the reliance on high-precision real-time estimation of the vehicle's longitudinal velocity; secondly, it directly reflects the transient response of the drive wheels to torque input, and is more conducive to characterizing the dynamic characteristics of the wheel speed being faster than the vehicle speed in the initial stage of slippage. In other embodiments, if the vehicle platform has a high-quality wheel-end acceleration measurement channel, the wheel-end acceleration can be directly used as... It can be used without further differentiation of wheel speed; however, the wheel-vehicle acceleration difference in its subsequent construction... The relevant decision-making logic remains unchanged. If necessary, it can also be based on the wheel radius. and vehicle longitudinal speed The slip ratio is calculated using equation (2). Used for operational status monitoring, simulation result analysis, or performance evaluation: (2) In equation (2), max[·] represents the function that takes the maximum value. This represents a parameter to prevent the denominator from being zero.

[0018] The vehicle parameters used in this embodiment are shown in Table 1: Table 1 Vehicle Parameters

[0019] Step 2: Since differentiating the wheel angular velocity can easily amplify high-frequency noise, and the vehicle's longitudinal acceleration also contains certain transient fluctuations during start-up, sudden road changes, or random driver input, the original value is used directly. Road surface condition determination can easily lead to false triggering, jittery switching, or premature controller takeover. Therefore, this invention introduces filtering processing before road surface condition determination. Wheel-vehicle acceleration difference at time A first-order discrete low-pass filter is used for low-pass filtering to eliminate measurement noise, thereby obtaining the result using equation (3). Wheel-vehicle acceleration difference after time processing : (3) In equation (3), These are the filter coefficients used to suppress high-frequency noise, while also striking a trade-off between noise suppression and maintaining dynamic response. This represents the wheel-vehicle acceleration difference after processing in the previous cycle. The purpose of this step is not to change the original dynamic relationship, but to improve the reliability of subsequent road surface condition determination. Especially when the present invention adopts the "no pulse entry, pulse exit" determination logic, if the input signal is not properly filtered, it is easy to misidentify initialization noise or measurement interference as a change in adhesion state. In other embodiments, equivalent discrimination quantities can be obtained by means of moving average filtering, band-limited differentiators, state observers or Kalman filtering. As long as the signal reliability can be improved without destroying the subsequent discrimination logic, it can be regarded as an equivalent implementation of the present invention.

[0020] Step 3, according to ,right The road surface adhesion state at any given time is determined to obtain... Road surface adhesion status indicator at any time ,like express The road surface adhesion state at a given time is a high-adhesion road surface. express The road surface adhesion at any given time is low adhesion. This step mainly provides a reference for the controller output and torque limiting section.

[0021] Step 4, Introduction of the present invention This serves as a pre-emergence enable management signal. Only when the preceding signal meets certain comprehensive conditions is the subsequent road condition determination and torque control logic truly allowed to take over. A controller takeover permission flag is constructed. And based on this, make a judgment Is the current time within the dead zone? Indicates the wheel speed after filtering. This represents the estimated wheel-end acceleration. This represents the estimated maximum effective torque. Let : System uptime Greater than the minimum preheating time threshold Wheel speed after filtering Greater than the minimum effective wheel speed threshold wheel-end acceleration estimate The absolute value is not greater than the upper limit of acceleration. Maximum effective torque estimate Greater than the positive threshold Furthermore, the duration for which the above conditions are met consecutively satisfies the preset duration. Otherwise, let The core purpose of this step is not to directly determine low adhesion, but to first determine whether the controller has met the conditions for safe takeover. This idea stems from considerations of METE engineering implementation issues: if the controller is immediately allowed to enter torque limiting logic during the stages of vehicle start-up, controller initialization, wheel speed derivative establishment, before the filter has entered a stable state, or when the maximum effective torque estimate is abnormal for a short time, it is very easy to cause a sudden drop in torque, false torque limiting, and unnecessary control oscillations.

[0022] Step 5, based on ,judge Is the vehicle in dead time at any given moment? when At that time, the judgment The vehicle's state at any given moment is in a dead time, and causes... ,make ,in, for Measure the motor torque command at any given time and execute step 4; when If so, proceed to step 6; Figure 4 The implementation logic for this step corresponds to this. Unlike simply setting a fixed dead time, this invention uses a more robust Ready determination method: when the input signal is not yet reliable, even if the dead time has passed, it will not rashly enter low-adhesion discrimination and torque limiting control; this can effectively suppress system malfunctions. In specific calibration, , , , and All of these thresholds can be calibrated based on vehicle model, wheel speed sensor mass, derivative bandwidth, and vehicle control system refresh cycle. This invention does not limit the specific values ​​of these thresholds, but requires that they collectively ensure that the controller is only allowed to take over when the input is valid, the estimate is reliable, and the system transient has essentially ended.

[0023] Step 6, if Road surface adhesion status indicator at any time and ,make ;in, Indicates the entry threshold; Indicates the sampling period; like and ,make ;in, Indicates the exit threshold; and This creates a hysteresis window to prevent state jitter; otherwise, it maintains... constant; The state transition equation is given by equation (4): (4) In equation (4), Indicates that, Indicates or; In this case, the parameter settings are: entry threshold. Relaxed conditions ensure timely entry and exit thresholds. Strict conditions ensure genuine exit. The design philosophy of this strategy is: lenient entry conditions to ensure timely detection even on low-friction surfaces (such as ice) without docking pulses; and strict exit conditions to prevent frequent switching caused by a drop in wheel acceleration due to torque reduction.

[0024] Step 7: Determine the vehicle slip state based on the torque ratio indication and wheel-vehicle acceleration difference, and update the relaxation factor accordingly. This step is used to distinguish whether the slip state that requires adaptive adjustment of the relaxation factor has been entered. Instead of using a single criterion, the relative relationship between torque demand and maximum available torque and the difference in wheel-vehicle dynamic response are combined to make the judgment. Figure 5 The first part corresponds to this step, and the output of this step is a binary glide flag. Its direct target is not torque limitation, but rather the subsequent... An adaptive scheduling module. In other words, this invention does not explicitly track a target slip ratio, but rather adjusts control parameters online by detecting slip trends; Construct using equation (5) Torque ratio indication at time : (5) In equation (5), This represents the estimated maximum available torque of the drive motor at the previous moment. This represents a parameter to prevent the denominator from being zero.

[0025] Step 8, based on and ,right Determine the vehicle's slippage status at any given moment: when or At that time, the judgment When the vehicle begins to skid, the skid sign is activated. Otherwise, determine The vehicle did not slip at all times, making ;in, The torque ratio threshold, The threshold value for wheel-vehicle acceleration difference; Torque ratio indication The physical meaning of is: the relative tension between the torque currently requested by the driver and the upper limit of the torque that the system can transmit. When A large value indicates that the driving force request has approached or exceeded the level allowed by the current adhesion capability; wheel-vehicle acceleration difference. This directly reflects whether the wheel's response to driving torque is significantly faster than the vehicle body, and is a dynamic sign of slippage. (Using...) Exceeding the threshold or Parallel criteria for exceeding thresholds can improve the sensitivity and robustness of slip detection. It should be noted that the specific threshold value can be calibrated based on the characteristics of the vehicle's powertrain, tire radius, drive torque level, and signal noise level.

[0026] Step 9, Construction Slide duration timer for moments and Integral quantity of slip intensity at time step ; when At that time, according to the sampling period The slip duration timer and slip intensity integral are updated using equations (6) and (7): (6) (7) In equations (6) and (7), This represents the upper limit of the slip strength integral. Represents a saturation function; when At that time, stop The accumulation; when At that time, according to the preset recovery strategy, equations (8) and (9) are used to respectively... and Attenuation: (8) (9) In equations (8) and (9), The recovery coefficient of the slip duration timer. The coefficients of restitution for the integral of slip strength are both positive constants; when and At this point, the system can be considered to have returned to its initial state without slippage.

[0027] Step 10, according to and Construct separately Time adjustment factor and Integral adjustment factor at time step and obtain Comprehensive adjustment factor at time : (10) (11) (12) in, Used to indicate how long the slip lasted. Both are used to reflect the severity of slip. They characterize the slip state from the time and intensity dimensions respectively, providing a more comprehensive reflection of the current working condition than a single instantaneous criterion. Through analysis of... Discrete integration can distinguish between short-term spikes and sustained severe slip; further, the integral can be constrained by a saturation function. Within this framework, measures should be taken to prevent the integral from increasing indefinitely and causing parameters to become uncontrollable. Time adjustment factor The exponential approach has the advantage of rapidly increasing the parameter value when slippage first occurs, and gradually saturating after a prolonged slippage, thus avoiding excessive parameter variation; integral adjustment factor. This directly reflects the cumulative effect of slip strength. The sum of the two is then truncated to... The interval is used to obtain the comprehensive adjustment factor. Then mapped to Interval. Update again according to equation (13). relaxation factor at time : (13) in, This is the lower limit of the relaxation factor. The upper limit of the relaxation factor, This is the time adjustment factor. This is the integral adjustment coefficient. Adjust the upper limit for points; This step corresponds to The core of the adaptive scheduling module, such as Figure 5 As shown. Its basic idea is: in traditional METE... Most are fixed values, while fixed It is difficult to simultaneously achieve both high traction utilization under high adhesion conditions and anti-slip performance under low adhesion conditions. Therefore, this invention no longer... Instead of treating it as a constant, it is updated online based on the slip duration and slip intensity, making... Adaptive to changes in operating conditions Figure 6 The relaxation factor under typical continuous slip conditions is shown. The curve showing the change. It's also important to note that here... Indicates the duration of the slip, used for Adaptive scheduling; and in subsequent step 9 This indicates the duration of low adhesion, used for low adhesion torque decay. Although both are time-related quantities, their triggering conditions and physical meanings are different and should not be used interchangeably.

[0028] Step 11, as follows Figure 7 As shown, the longitudinal driving force between the tire and the road surface is estimated using the driving torque and the translational / rotational relationship. This formula essentially originates from wheel-end dynamic balance: a portion of the driving torque is used to overcome the acceleration requirement caused by the wheel-end rotational inertia, and the remainder is converted into the longitudinal driving force between the tire and the road surface. According to... The driving torque and wheel longitudinal acceleration at time t are calculated using equation (14). Longitudinal driving force between the tire and the road surface at all times: (14) Subsequently, based on Construct using equation (15) Maximum effective torque estimation coefficient at time t .because The adaptive update based on the current slip state has already been performed in step 6, therefore here... It is no longer a fixed parameter, but can change with the operating conditions: (15) Finally, using equation (16) we obtain Maximum effective torque estimate at time 1 : (16) In equation (16), express The driver requests a reference torque at any given moment. express The longitudinal acceleration of the wheel at any given moment, This represents the moment of inertia of the wheel. Indicates the equivalent vehicle mass. Indicates the rolling radius of the wheel. express The longitudinal driving force between the tire and the road surface at all times; This step corresponds to the NMETE calculation module, and its goal is to calculate the relaxation factor based on the classic METE, which has already been updated online. By incorporating the maximum effective torque estimation process, a torque upper limit estimate that is more adapted to the current operating conditions is obtained, compared with the traditional fixed estimate. Compared to METE, the improvement of this step is that the maximum effective torque estimate is no longer determined by a fixed relaxation factor, but is coupled with the degree of slip and duration, thereby achieving a more reasonable torque upper limit switching between high and low adhesion conditions.

[0029] Step 12: After the improved METE module calculation, this embodiment sets the parameters as follows: Figure 8 The low-adhesion logarithmic torque attenuation limiting module shown uses the duration of low adhesion as the independent variable, according to... right By applying secondary restrictions, we obtain The final torque limit at any moment ; when When the system determines that the current surface is a high-adhesion road surface, no additional attenuation is needed. Low-level duration count value at time Set to zero and use equation (17) to obtain The final torque limit at any moment : (17) when When the system is in a low-adhesion state, relying solely on the basic NMETE upper limit is sometimes insufficient to quickly suppress continuous slippage. Equation (18) is then used to calculate... Low-level duration count value at time : (18) The logarithmic decay coefficient at time t is obtained using equation (19). : (19) Therefore, the damping torque at time t can be obtained using equation (20). : (20) This means that the longer the low-adhesion state lasts, the more conservative the torque output strategy the controller tends to adopt. Finally, using equation (21), we obtain... The final torque limit at any moment : (twenty one) In equation (21), This indicates the attenuation rate adjustment parameter. This represents the steady-state lower limit of the attenuation coefficient. Indicates the sampling period. Indicates the absolute minimum torque threshold; With this design, the torque can still maintain good responsiveness when the vehicle first enters a low-traction surface; if the low-traction state continues, the torque will be further compressed into a safer range, thus taking into account starting responsiveness, traction utilization and continuous anti-skid capability.

[0030] Step 13: To balance the driver's need to follow the steering intentions on well-paved roads with the need for rapid intervention and control in the event of wheel slippage, a semi-open-loop control strategy is constructed. The driver requests a reference torque at any given moment. and The comparison will ultimately determine the command torque. The output is sent to the motor controller to limit the actual drive torque in order to achieve the desired traction for the electric vehicle. Real-time adaptive anti-slip control; This step is the execution output layer of the entire method, corresponding to... Figure 1 The "output torque limit value" and Figure 2 The motor controller 30 in the middle. In engineering implementation, the calculated... It can be sent to the motor controller via CAN bus, vehicle control network or shared variables inside the controller, and the motor controller will use it as the saturation upper limit or constraint boundary of the actual torque command; The actual drive torque ultimately executed by the motor controller is no longer solely determined by the driver's torque request, but is constrained by the final torque upper limit calculated by this invention. When there is high adhesion and no significant slippage, the system's constraint on the driver's request is weak, allowing the vehicle to fully release its traction performance; when there is low adhesion or a significant slippage tendency, the system will... The combined effect of adaptive scheduling, NMETE upper limit estimation, and low-adhesion torque decay gradually tightens the torque output, thereby suppressing slippage and improving longitudinal stability. Thus, this invention forms a complete closed loop in its control structure: the sensor module collects signals from the wheel ends and the vehicle body, the traction controller completes state discrimination, parameter adjustment, and torque upper limit calculation, the motor controller executes the final torque limit, and the drive motor and vehicle dynamics object then feed back the actual response to the sensor module. The simulation and comparison results in subsequent step 14 will further illustrate the effectiveness and superiority of steps 1 to 9 under different operating conditions.

[0031] Step 14: The improved maximum effective torque estimation algorithm is denoted as NMETE. To verify the effectiveness of the invention, a quarter vehicle model was built in the MATLAB / Simulink environment, and the NMETE controller and the traditional METE controller were compared and tested. Test Condition 1: Driver Input: Inclined slope with a gradient of 100, stabilizing after 1 second; Road Surface Condition: Low adhesion (snow accumulation) road surface; Test Duration: 8 seconds. Test results are shown in Table 2. Table 2 Comparison of performance of single low-adhesion pavement

[0032] Figure 9 The torque variation curves under these test conditions are shown. It can be seen that the classic METE performs rather conservatively, failing to respond promptly to the driver's torque demands, resulting in a slower acceleration performance. The improved NMETE, on the other hand, follows the driver's intentions better, responding quickly when acceleration is needed, while also appropriately limiting torque at suitable times and maintaining output through fine adjustments, thus balancing acceleration and driving safety. Test Condition 2: Driver Input: 100 Nm step input; Road Condition: Dry asphalt → Snow (2-second switch); Test Duration: 8 seconds. Test results are shown in Table 3. Table 3 Comparison of the performance of the connecting road surface

[0033] Test condition 3: Driver input: 200 Nm step input; Road surface condition: dry asphalt → snow (1 second switching); Test duration: 7 seconds. Test results are shown in Table 4. Table 4 Comparison of performance of double-jointed pavement

[0034] Figure 10 The performance of different controllers under this operating condition was demonstrated. The classic METE controller showed weak adaptability to sudden changes in adhesion under low-adhesion conditions, with a slow and persistently high slip ratio, resulting in insufficient slip suppression. In contrast, the NMETE controller was able to reduce excess drive torque more quickly and continuously after the peak, allowing the slip ratio to fall along a steeper trajectory and gradually approach the reasonable operating range for low-adhesion surfaces, thus achieving better anti-skid performance and traction utilization on snow.

[0035] Test Condition 4: Multi-condition robustness verification, using snow + random driver input ( Figure 11 ) and random road conditions + driver 100Nm step input ( Figure 12The test results verified the multi-condition adaptability of the NMETE controller. The test results show that the NMETE controller can maintain stable control performance under different operating conditions, a capability lacking in the overly conservative classic METE controller.

[0036] In summary, compared to the traditional METE controller, this invention improves traction control performance. On low-adhesion surfaces (μ=0.3), the maximum average acceleration increases from 0.6804 m / s² to 2.5910 m / s²; the maximum traction utilization rate increases from 0.2312 to 0.8830; and the root mean square error of the slip ratio decreases from 0.137597 to 0.073483. The data demonstrate that this invention significantly improves traction utilization and acceleration performance while maintaining stability.

[0037] In this embodiment, an adaptive electric vehicle traction control system based on the maximum wheel-end transmittable torque is described below. Figure 2 and Figure 3 The system includes: a signal acquisition and preprocessing module, a road surface condition determination module, and a slippage determination and... The system includes an adaptive scheduling module, an NMETE calculation module, a torque attenuation limiting module, and a motor controller. Figure 2 A schematic diagram of the module structure of the system of the present invention is provided. Figure 3 A module logic diagram of the system of the present invention is given. Figure 1 Corresponding to the overall process of the method shown, this embodiment explains the inputs, outputs, functional divisions, and interactions of each module from a system perspective. (Subsequent...) Figures 4-8 Corresponding to road surface condition identification, Adaptive scheduling Key modules include trend analysis, NMETE calculation, and low-adhesion logarithmic torque attenuation limitation. Figures 9-12 This is used to illustrate the control effect of the system of the present invention under different operating conditions. The entire control process follows this sequence: signal acquisition and preprocessing → road surface condition determination → slip determination and... Update → Maximum effective torque estimation → Low adhesion torque decay limit → Output torque constraint are gradually developed.

[0038] The purpose of this embodiment is not to simply repeat the method claims, but to illustrate from a system implementation perspective: what tasks each functional module is responsible for, how signals are transmitted between modules, why this module division is adopted, and how it is implemented in the vehicle controller. Compared with the method embodiment, which only focuses on algorithm steps, the system embodiment emphasizes the correspondence between "module—signal—output—object of action". The system embodiment and the method embodiment together constitute a complete disclosure of the technical solution of the present invention, wherein the method embodiment focuses on step-by-step execution logic, and the system embodiment focuses on modular deployment logic.

[0039] like Figure 2 As shown, the system of this invention adopts a layered and modular structural organization. The signal acquisition and preprocessing module is located at the front end of the system and is used to complete the acquisition, conversion, and basic processing of wheel-end and vehicle body dynamic signals; the road surface condition discrimination module and the slip determination and... The adaptive scheduling module, together with other modules, forms the parameter adaptive layer, used to identify the current operating condition and adjust control parameters online; the NMETE calculation module provides an estimate of the maximum effective torque under the current operating condition; the torque attenuation limiting module further corrects the upper limit of the basic torque under low-adhesion conditions; and the motor controller, as the execution layer, imposes a limit on the actual drive torque based on the final upper limit of torque. This structure gives the system good hierarchy, maintainability, and portability in engineering implementation.

[0040] Figure 3 The signal transmission paths between the modules are further illustrated from a logical perspective. The input signals on the left include the wheel angular velocity. longitudinal acceleration of the vehicle body Driver requests reference torque Wheel end torque estimation value and system uptime These signals first enter the signal acquisition and preprocessing module to generate the wheel-vehicle acceleration difference. and its filtering quantity Subsequently, the Ready state is determined and the road surface condition is updated, outputting the road surface adhesion status. Slip determination and The adaptive scheduling module according to , , The relaxation factor is calculated using the maximum available torque estimate from the previous sampling period. The NMETE calculation module is based on , and Calculate the estimated maximum effective torque. Finally, the torque attenuation limiting module according to and Output final torque limit This is executed by the motor controller.

[0041] The advantages of this modular approach are twofold: firstly, the controller's front-end identification logic, parameter adaptation logic, and execution logic are separated, facilitating calibration and maintenance; secondly, the modules are connected through clear state variables and intermediate variables, avoiding coupling all control functions into a single black-box algorithm, thereby improving the system's deployability across different controller hardware platforms. For electric vehicle control systems, this modular structure is particularly suitable for implementation on vehicle controllers, drive controllers, or independent control computing units.

[0042] In this embodiment, the signal acquisition and preprocessing module corresponds to the system's front-end input layer, used to transform raw physical quantities into state and discrimination quantities that can be directly used by subsequent control algorithms. Specifically, the wheel angular velocity... Reflecting the rotational state of the wheel ends, the longitudinal acceleration of the vehicle body Reflecting the vehicle's longitudinal response, the driver requests a reference torque. Wheel-end torque estimate reflects the driver's power demand. This is used in the NMETE calculation module to estimate the longitudinal driving force between the tire and the road surface; the signal acquisition and preprocessing module operates according to the following sub-steps: Step 1: Collect the wheel angular velocity of the vehicle. longitudinal acceleration of the vehicle body , The driver constantly requests reference torque from the drive system. and wheel end torque estimation value ; Step 2, Measure the wheel angular velocity Differentiating the longitudinal acceleration of the wheel yields the following results. ; Step 3, according to Calculate the wheel-vehicle acceleration difference ; This is the most critical physical quantity in this module, and its significance lies in the following: When the drive wheels and the vehicle body are in a state of good adhesion, the longitudinal acceleration of the wheels and the longitudinal acceleration of the vehicle body usually maintain a strong consistency; when low adhesion or slippage begins to occur, the wheels tend to respond to the drive torque more quickly, while the vehicle body, due to the limitations of adhesion conditions, experiences a relatively slower increase in longitudinal acceleration. It will increase. Therefore, It can serve as an important basic quantity for subsequent road surface condition assessment and slippage trend detection; Step 4: Wheel-vehicle acceleration difference The wheel-vehicle acceleration difference is obtained by filtering. ; in, express The wheel angular velocity at any given moment, express The longitudinal acceleration of the vehicle body at any given moment, express The driver requests a reference torque at any given moment. express The estimated wheel-end torque at time [time]. express The longitudinal acceleration of the wheel at any given moment; Figure 3 The sub-units "derivative of wheel angular velocity," "calculation of wheel-vehicle acceleration difference," and "low-pass filtering" correspond to the sub-steps of this module. This processing distinguishes raw signals such as wheel angular velocity and vehicle acceleration from the discriminant quantities used in subsequent control; it also makes the system more adaptable to different types of sensor configurations. For example, in one implementation, wheel longitudinal acceleration... This can be obtained by differentiating the wheel speed signal; in another embodiment, if the vehicle platform has a reliable wheel-end acceleration measurement channel, this quantity can also be directly collected, but subsequent... and The usage method remains unchanged; In the specific implementation, Filtering is necessary. This is because differentiating the wheel speed signal amplifies measurement noise; directly applying the original signal would amplify the noise. Used for state determination, it is prone to misjudgment. The filtered result... It is more suitable for input into the subsequent road surface condition determination module. The form of the filter is not limited to a single implementation method. In addition to the first-order low-pass filter, equivalent replacements can also be made by moving average filtering, band-limited differentiators, or state observation methods.

[0043] In this embodiment, the road surface condition discrimination module is the pre-discrimination layer of the system of the present invention, and its core functions are twofold: first, to determine whether the current system has met the conditions for safe takeover; second, based on the processed wheel-vehicle acceleration difference... The road surface adhesion status is determined. Figure 4 The working principle diagram of this module is given, and Figure 3 The circular Ready state diagram further illustrates the state transition of the Ready takeover decision from a system logic perspective; The road surface condition determination module operates according to the following sub-steps: Step 1: Construct the controller takeover permission flag If and only if the system runtime Greater than the minimum preheating time threshold Filtered wheel speed Greater than the minimum effective wheel speed threshold Wheel-end acceleration estimates The absolute value is not greater than the upper limit of acceleration. Maximum effective torque estimate Greater than the positive threshold Furthermore, the duration for which the above conditions are met consecutively satisfies the preset duration. season Otherwise, Compared to traditional methods that rely solely on a fixed dead time, this invention introduces a controller takeover permission flag. The design philosophy is as follows: when the vehicle starts, the controller initializes, the filter and derivative are not yet stable, or the maximum effective torque estimate is temporarily distorted, although the system has already started running, it is not appropriate for the controller to directly intervene in torque limiting at this time. Abruptly taking over could easily cause a sudden drop in torque, jitter, or malfunction. Therefore, this invention integrates minimum warm-up time, wheel speed effectiveness, reasonable wheel-end acceleration range, maximum effective torque estimate effectiveness, and duration confirmation mechanisms to determine whether the controller has met the reasonable conditions for takeover. Step 2, when At that time, the judgment The vehicle is in a dead zone at any given time, and the road surface is marked with status indicators. ; Step 3, when At that time, if the road surface adhesion status indicator at the previous sampling time ,and Then let ; Step 4, when At that time, if the road surface adhesion status indicator at the previous sampling time ,and Then let ; Step 5, in all other cases, maintain )constant; in, This indicates the entry threshold, and ; Indicates the exit threshold, and and satisfy ; Indicates the sampling period; when In this invention, the road surface condition is directly set to a high-adhesion state. Essentially, this is a safety strategy before the controller intervenes. At this stage, even if there are short-term fluctuations in the original signal, it will not trigger subsequent road surface detection and torque limiting, thus avoiding erroneous torque limiting during the initialization phase. Only when... Only then will the system allow it according to The state at the previous sampling time Enter the formal road surface status update logic; When the system is currently in a high-attachment state, as long as Exceeding a smaller positive threshold This allows for a more sensitive identification of low adhesion trends; however, when the system is already in a low adhesion state, it can only be detected when... Decrease to a more stringent negative threshold Only after this is exiting the low-adhesion state allowed. The advantage of this design is that it ensures timely low-adhesion recognition while avoiding delays caused by the controller already limiting torque. The drop in adhesion caused it to prematurely exit the low-adhesion state. from Figure 4 As can be seen, the Ready determination and the road surface state update are not two isolated processes, but rather sequential: the Ready takeover determination first decides whether the controller can intervene, and the state update then determines whether the current condition is high or low adhesion. This two-layer logical structure improves the system's robustness and engineering practicality.

[0044] In this embodiment, the slip determination and The adaptive scheduling module is the online parameter adjustment layer in the system of this invention. Unlike traditional METE controllers that use a fixed relaxation factor... The methods are different; this invention will Designed as an adaptive variable that can dynamically change with operating conditions, its update is not based on the target slip ratio or the optimal slip ratio, but on the torque ratio indication. Wheel-vehicle acceleration difference Slip duration and the integral of slip strength . Figure 5 The working principle diagram of this module is given. Figure 6 The typical working condition of continuous slip is given. The curve of change; Ratio Indicator Its function is to measure the relative relationship between the driver's current torque demand and the maximum available torque that the system can provide. A larger ratio indicates that the vehicle is under high driving force demand, and the drive system is more likely to be close to the road's traction limit; wheel-vehicle acceleration difference. This directly reflects whether the wheel response is too fast relative to the vehicle body response, and is an important dynamic sign of slippage. Combining these two types of information for slippage determination can improve the reliability of detection; Slip determination and The adaptive scheduling module operates according to the following sub-steps: Step 1, Construction Torque ratio indication at time ,in, This represents the maximum available torque estimate from the previous sampling period. This parameter is used to prevent the denominator from being zero. Step 2, when or At that time, the judgment The vehicle skidded at that moment, causing Otherwise, ;in, The torque ratio threshold, The threshold value for wheel-vehicle acceleration difference; Step 3, when At that time, according to the sampling period Slide duration timer based on the previous sampling time and slip strength integral Update and get , ,in, This represents the upper limit of the slip strength integral. The saturation function is represented; instead of simply adjusting the instantaneous over-threshold parameter, this invention further constructs a sliding duration timer. and slip strength integral These two quantities reflect "how long the slip lasted" and "how severe the slip was," respectively. This avoids misjudging short-term spikes as sustained severe slip, and also allows parameter adjustments to have a certain degree of memory, rather than relying entirely on single-sampling point judgments. Step 4, when At that time, the preset recovery strategy was applied to each case. and according to and Attenuation is performed; among which, The recovery coefficient of the slip duration timer. The coefficients of restitution for the integral of slip strength are both positive constants; when and At this point, the system can be considered to have returned to its initial state without slippage; The advantage of this design is that the parameter adjustment is both responsive and does not remain at an overly conservative setting for an extended period after the operating conditions return to normal. Step 5, according to and The time-based adjustment factor and the integral adjustment factor are constructed separately to obtain the comprehensive adjustment factor. and calculate ,in, This is the lower limit of the relaxation factor. This represents the upper limit of the relaxation factor. Figure 6 The relaxation factor under continuous slip condition is given. The change curve can intuitively reflect the design concept of this module: the longer the slip duration, the greater the slip intensity. The closer to the upper limit, the more likely it is that there will be no slippage for a long time. It will gradually fall back to a lower range. In this way, the present invention achieves better parameter matching under both high and low adhesion conditions; In other implementations, the specific functional forms of the time adjustment factor and the integral adjustment factor can also be adjusted according to the vehicle model and controller implementation method. For example, piecewise linear functions, hyperbolic functions, or lookup table methods can be used, as long as their essence remains "based on slip duration and slip intensity". Online updates are all equivalent implementations of this invention.

[0045] In this embodiment, the NMETE calculation module corresponds to the maximum effective torque estimation layer in the system of the present invention. Its function is to estimate the torque based on the current wheel-end dynamic state and relaxation factor. Give the upper limit of the basic torque under the current operating conditions. . Figure 7 A schematic diagram illustrating the working principle of the NMETE calculation module is provided. The NMETE calculation module operates according to the following sub-steps: Step 1: Based on the estimated wheel end torque value longitudinal acceleration of wheels Wheel rotational inertia and wheel rolling radius ,calculate Longitudinal driving force between the tire and the road surface at all times ,in, express The longitudinal driving force between the tire and the road surface at all times; Part of the wheel-end torque reflects the effective driving force that actually acts on the tire-road contact interface. When the longitudinal acceleration of the wheel is large, a portion of the wheel-end torque is used to overcome the wheel's rotational inertia, thus reducing the portion that can actually generate longitudinal driving force. Step 2: Based on the relaxation factor Wheel rotational inertia Equivalent vehicle mass and wheel rolling radius ,calculate Maximum effective torque estimation coefficient at time t ; Due to the introduction here The previous module has already updated the output online based on the current slip state, therefore the final output of this module is... It also has the ability to adapt to different operating conditions; Step 3, according to get Maximum effective torque estimate at time 1 .

[0046] Compared to traditional METE, the NMETE module of this invention does not simply change a formula, but organically combines adaptive parameter scheduling and maximum effective torque estimation. The advantage of this is that when road surface adhesion is good and slippage is minimal, At lower levels, the system can release traction performance to the maximum extent; when low adhesion and slippage persist, Being lifted, thus More conservative, which is more conducive to preventing slippage; In one specific embodiment, Figure 9 and Figure 10 The simulation results shown all demonstrate that, compared to the classic METE controller, the NMETE calculation module of this invention can provide a more reasonable upper limit of torque under single low adhesion and dual docking road conditions, enabling the system to improve traction utilization while maintaining stability.

[0047] In this embodiment, the torque attenuation limiting module corresponds to the subsequent protective layer in the system of the present invention. Its function is to adjust the torque based on the current road surface adhesion state, provided that the NMETE calculation module has already given the basic torque upper limit. A decision will be made on whether to further revise the basic upper limit. Figure 8 A flowchart of the low-adhesion logarithmic torque attenuation limiting module is provided; The torque attenuation limiting module operates according to the following sub-steps: Step 1, when At this time, the system determines that the current road surface is of high adhesion, and therefore no additional torque attenuation is required. Low-level duration count value at time Set to zero, and let ; Step 2, when When this occurs, it indicates that the current system is in a low-attachment state. At this point, relying solely on the basic NMETE upper limit is sometimes insufficient to quickly suppress continued slippage. Calculation Low-level duration count value at time ; Step 3, with Construct logarithmic decay coefficients for the independent variable ,in, This indicates the attenuation rate adjustment parameter. This represents the steady-state lower limit of the attenuation coefficient; Attenuation coefficient The introduction of this feature allows the system to gradually adopt a more conservative torque limiting strategy as low-adhesion conditions persist, rather than immediately and drastically reducing the torque to a fixed value upon entering a low-adhesion state. This approach balances two requirements: ensuring a certain level of dynamic response during the initial stages of low-adhesion detection, and gradually strengthening protection as low-adhesion persists. This is achieved by further superimposing an absolute minimum torque threshold. and "not exceeding" By applying the upper limit constraint, the final upper limit of torque can be obtained. ; Step 4: Calculate the damping torque ; Step 5: Combine with the absolute minimum torque threshold and "not exceeding" The upper limit constraint is used to calculate the final torque upper limit. ,in, Indicates the absolute minimum torque threshold; Step 6, finally The driver requests a reference torque at any given moment. and The comparison will ultimately determine the command torque. Output to motor controller; Figure 11 and Figure 12 System verification results are presented under two conditions: snowy conditions with random driver input and random road surface conditions with a 100 Nm step input. Combined with... Figure 8 As can be seen from the torque attenuation limiting logic shown, this module can provide additional safety margin for the system under complex operating conditions, enabling the present invention to still have good robustness under random driver input and complex road surface changes.

[0048] In this embodiment, the motor controller is the execution layer of the system of the present invention. Its main function is not to participate in the front-end road condition judgment or parameter update, but to receive the final torque limit. Then, the upper limit of torque is compared with the reference torque requested by the driver. The process involves comprehensive analysis to generate the final motor command torque output to the drive system. The motor controller is used to receive the final torque limit. And impose limits on the actual driving torque to achieve the traction of the electric vehicle at Real-time adaptive anti-slip control; The motor controller can be based on The final motor command torque is obtained. The significance of this design is that when the driver requests torque below the system's safety limit, the system does not actively intervene in normal driving requests; only when the driver requests torque exceeding the safety limit allowed under the current operating conditions does the system impose constraints. Therefore, this invention can ensure traction control safety while preserving the driver's normal acceleration intentions as much as possible. It should be noted that the specific hardware form of the motor controller does not constitute a limitation of this invention. It can be a control unit inside the drive motor inverter, or it can be the execution result of a torque-limiting command sent to the drive controller by the vehicle controller or an independent control computing platform. As long as it can receive the final torque limit of this invention and impose a limit on the actual drive torque, it falls within the protection scope of this invention.

[0049] In summary, the system embodiments of the present invention utilize a signal acquisition and preprocessing module, a road surface condition discrimination module, and a slippage determination and... The hierarchical collaboration of the adaptive scheduling module, NMETE calculation module, torque attenuation limiting module, and motor controller realizes a complete closed loop from front-end signal acquisition, road surface adhesion state discrimination, slippage trend detection, adaptive parameter adjustment, maximum effective torque estimation, secondary torque limitation under low adhesion conditions to final execution control. Figure 1 The overall process of the method is given. Figure 2 The system structure is given. Figure 3 The system module logic is given. Figures 4-8 The implementation principles of the key sub-modules are explained respectively; Figures 9-12 This further verifies the effectiveness and superiority of the system of the present invention under typical working conditions such as single low-adhesion road surface, double-connected road surface, random snow input, and random road surface step input. This demonstrates that the present invention not only possesses a clear modular hierarchy in its theoretical structure, but also exhibits excellent control performance and promising engineering application prospects under actual working conditions.

Claims

1. An adaptive traction control method for electric vehicles based on the maximum transmittable torque at the wheel end, characterized in that, Includes the following steps: Step 1, data collection The vehicle wheel-end dynamics signals at time points include: Wheel angular velocity at time , longitudinal acceleration of the vehicle body at any moment , The driver constantly requests reference torque from the drive system. as well as Wheel end torque estimate at time 1 ;right After differentiation, we get longitudinal acceleration of the wheel at any moment , thus calculating Wheel-vehicle acceleration difference at time ; Step 2, for Wheel-vehicle acceleration difference at time After filtering, the result is obtained. Wheel-vehicle acceleration difference after time processing ; Step 3, according to ,right The road surface adhesion state at any given time is determined to obtain... Road surface adhesion status indicator at any time ,like express The road surface adhesion state at a given time is a high-adhesion road surface. express The road surface adhesion state at any given time is low adhesion. Step 4, during the preset duration Down: Always greater than the minimum preheating time threshold ,and Filtered wheel speed at time Always greater than the minimum effective wheel speed threshold ,and wheel-end acceleration estimate at time t The absolute value is always no greater than the upper limit of acceleration. ,and Maximum effective torque estimate at time 1 Always greater than the positive threshold ,make Otherwise, let Thus constructing Controller takeover permission flag at any time ; Step 5, based on ,judge Is the vehicle in dead time at any given moment? when At that time, the judgment The vehicle's state at any given moment is in a dead time, and causes... ,make ,in, for Measure the motor torque command at any given time and execute step 4; when If so, proceed to step 6; Step 6, if Road surface adhesion status indicator at any time and Then let ;in, This indicates the entry threshold, and ; Indicates the sampling period; like and ,make ;in, Indicates the exit threshold, and , ; In other cases, keep constant; Step 7, Construction Torque ratio indication at time ;in, Indicates that the drive motor is in The estimated maximum available torque at time [time]. This parameter is used to prevent the denominator from being zero. Step 8, based on and ,right Determine the vehicle's slippage status at any given moment: when or At that time, the judgment The vehicle skidded at that moment, causing Time-shifting markers Otherwise, determine The vehicle remained stationary at all times, and... ;in, The torque ratio threshold, The threshold value for wheel-vehicle acceleration difference; Step 9, Construction Slide duration timer for moments and Integral quantity of slip intensity at time step ; when At that time, Slide duration timer for moments and Integral quantity of slip intensity at time step Update and get Slide duration timer for moments and slip strength integral ,in, This represents the upper limit of the slip strength integral. Represents a saturation function; The sampling period; when At that time, the preset recovery strategy was applied to each case. and Attenuation; Step 10, according to and Construct separately Time adjustment factor and Integral adjustment factor at time step , thus obtaining Comprehensive adjustment factor at time , and then calculate relaxation factor at time ;in, This is the lower limit of the relaxation factor. This is the upper limit of the relaxation factor. For time adjustment coefficients, This is the coefficient for integral adjustment. This is the upper limit for integral adjustment; Step 11, Calculation Longitudinal driving force between the tire and the road surface at all times and calculate Maximum effective torque estimation coefficient at time t Thus obtain Maximum effective torque estimate at time 1 ;in, This represents the moment of inertia of the wheel. Indicates the equivalent vehicle mass. Indicates the rolling radius of the wheel. express The longitudinal driving force between the tire and the road surface at all times; Step 12, according to ,right By applying secondary restrictions, we obtain The final torque limit at any moment : when At that time, Low-level duration count value at time Set to zero, and let The final torque limit at any moment ; when At that time, calculate Low-level duration count value at time and obtain Logarithmic decay coefficient at time Thus obtain Decaying torque at time ; and thus obtain The final torque limit at any moment ,in, This indicates the attenuation rate adjustment parameter. This represents the steady-state lower limit of the attenuation coefficient. Indicates the absolute minimum torque threshold; Step 13, Motor command torque at any moment The output is sent to the motor controller to limit the actual drive torque in order to achieve the desired traction for the electric vehicle. Real-time adaptive anti-slip control.

2. An adaptive electric vehicle traction control system based on the maximum transmittable torque at the wheel end, characterized in that, include: Signal acquisition and preprocessing module, road surface condition determination module, Adaptive scheduling module, NMETE calculation module, torque attenuation limiting module and motor controller; The signal acquisition and preprocessing module is used for acquisition. Vehicle wheel-end dynamics signals at any given time, including Wheel angular velocity at time , longitudinal acceleration of the vehicle body at any moment , The driver constantly requests reference torque from the drive system. as well as Wheel end torque estimate at time 1 ; and on Differentiation yields longitudinal acceleration of the wheel at any moment and calculate Wheel-vehicle acceleration difference at time Afterwards, The wheel-vehicle acceleration difference is obtained by filtering. ; The road surface condition determination module is based on and Controller takeover permission flag at any time ,right The road surface adhesion state at any given time is determined to obtain... Road surface adhesion status indicator at any time ,in Indicates a high-adhesion road surface. Indicates a low-adhesion road surface; The The adaptive scheduling module is used to construct Torque ratio indication at time And combined with the wheel-vehicle acceleration difference determination Whether the vehicle slipped at any given moment, thus obtaining... Time-shifting markers ; and when slippage is detected, update Slide duration timer for moments and Integral quantity of slip intensity at time step Thus update relaxation factor at time ; The NMETE calculation module is based on Wheel end torque estimate at time 1 , longitudinal acceleration of the wheel at any moment and relaxation factor at time ,calculate Longitudinal driving force between the tire and the road surface at all times , thus calculating Maximum effective torque estimate at time 1 ; The torque attenuation limiting module is based on ,right Apply secondary restrictions: based on Low-level duration count value at time ,calculate Decay coefficient at time Thus obtain The final torque limit at any moment ; The motor controller receives The final torque limit at any moment ,get Motor command torque at any moment It is used to limit the actual driving torque and achieve adaptive anti-slip control of electric vehicle traction.

3. The adaptive electric vehicle traction control system based on maximum wheel-end transmittable torque according to claim 2, characterized in that, The road surface condition determination module includes the following steps: Step a1, during the preset duration Down: Always greater than the minimum preheating time threshold ,and Filtered wheel speed at time Always greater than the minimum effective wheel speed threshold ,and wheel-end acceleration estimate at time t The absolute value is always no greater than the upper limit of acceleration. ,and Maximum effective torque estimate at time 1 Always greater than the positive threshold ,make Otherwise, let Thus constructing Controller takeover permission flag at any time ; Step a2, when At that time, the judgment The vehicle is in a dead zone at any given time, and the road surface is marked with status indicators. ,make ,in, for The motor command torque at any given moment; when At that time, if Road surface adhesion status indicator at the sampling time ,and Then let ;in, This indicates the entry threshold, and ; Indicates the sampling period; when At that time, if ,and Then let ;in, Indicates the exit threshold, and , In other cases, maintain constant.

4. The adaptive electric vehicle traction control system based on maximum wheel-end transmittable torque according to claim 2, characterized in that, The The adaptive scheduling module includes the following steps: Step b1, Construction Torque ratio indication at time ,in, This represents the maximum available torque estimate from the previous sampling period. This parameter is used to prevent the denominator from being zero. Step b2, based on and ,right Determine the vehicle's slippage status at any given moment: when or At that time, the judgment The vehicle skidded at that moment, causing Otherwise, let ;in, The torque ratio threshold, The threshold value for wheel-vehicle acceleration difference; Step b3, when At that time, Slide duration timer for moments and moments Slip strength integral Update and get , ,in, This represents the upper limit of the slip strength integral. Represents a saturation function; when At that time, the preset recovery strategy is used for and Attenuation; Step b4, according to and Construct separately Time adjustment factor and Integral adjustment factor at time step , thus obtaining Comprehensive adjustment factor at time , and then calculate relaxation factor at time ;in, This is the lower limit of the relaxation factor. This is the upper limit of the relaxation factor. For time adjustment coefficients, This is the coefficient for integral adjustment. This is the upper limit for integral adjustment.

5. An adaptive electric vehicle traction control system based on maximum wheel-end transmittable torque according to claim 2, characterized in that, The NMETE calculation module includes: Step c1, according to Wheel end torque estimate at time 1 , longitudinal acceleration of the wheel at any moment Wheel rotational inertia and wheel rolling radius ,calculate Longitudinal driving force between the tire and the road surface at all times ; Step c2, according to relaxation factor at time Wheel rotational inertia Equivalent vehicle mass and wheel rolling radius ,calculate Maximum effective torque estimation coefficient at time t ; Step c3, Calculation Maximum effective torque estimate at time 1 .

6. The adaptive electric vehicle traction control system based on maximum wheel-end transmittable torque according to claim 2, characterized in that, The torque attenuation limiting module includes: Step d1, when At that time, Low-level duration count value at time Set to zero, and let ; when At that time, calculate Low-level duration count value at time ; and with Construct with the independent variable Decay coefficient at time ,in, This indicates the attenuation rate adjustment parameter. This represents the steady-state lower limit of the attenuation coefficient; Step d2, Calculation Decaying torque at time ; Step d3, Calculation The final torque limit at any moment ,in, This represents the absolute minimum torque threshold.

Images