A multi-motor distributed drive energy management method and system

Abstract

This application discloses a multi-motor distributed drive energy management method and system, belonging to the field of electric vehicle drive control technology. This application aims to solve the problems of fixed power allocation and energy waste and uneven distribution caused by the lack of consideration for the actual power utilization differences of each wheel in existing technologies. The method includes: obtaining the total available power of the drive system, the real-time torque demand of each wheel, and the actual power utilization of each wheel motor; determining the torque allocation ratio of each wheel based on the real-time torque demand, and allocating basic available power accordingly; calculating the difference between the basic available power and the actual utilization power of each wheel as the remaining power; distributing the remaining power of each wheel evenly to the other wheels to adjust the final available power of each wheel; and controlling the output of the corresponding motor based on the final available power. This application can dynamically recover and redistribute underutilized power, improving the overall vehicle energy utilization efficiency.

Description

Technical Field

[0001] This application belongs to the field of electric vehicle drive control technology, specifically relating to a multi-motor distributed drive energy management method and system. Background Technology

[0002] Currently, multi-motor distributed drive electric vehicles have become an important direction for electric vehicle technology development due to their advantages such as independent controllability of each wheel, high transmission efficiency, and rapid response. In the drive control of this type of vehicle, the core task of energy management strategy is to rationally distribute the total available power of the drive system to each wheel motor. Existing technologies generally adopt a forward open-loop allocation method based on torque distribution coefficients. This method determines the torque distribution ratio of each wheel based on pre-calibrated rules or efficiency optimization algorithms (such as optimal torque distribution based on motor efficiency MAP), according to the driver's accelerator pedal opening, vehicle steering angle, vehicle speed, and other state information. Then, the total power is distributed to each motor according to a fixed ratio or a ratio calculated offline to achieve vehicle drive.

[0003] However, the aforementioned existing energy management methods have significant limitations in practical applications: due to the real-time changes in dynamic loads, road surface adhesion conditions, and the efficiency characteristics of the motors themselves during actual driving, there is often a discrepancy between the actual power utilized by each motor and its allocated power. When the actual power utilized by a wheel is lower than its allocated power, the unconsumed power cannot be recovered or reused by the system, resulting in energy waste. At the same time, other wheels may not be able to meet real-time torque requirements due to insufficient allocated power, leading to limited vehicle power response or reduced energy efficiency. This power management method based on static allocation or open-loop control fails to utilize real-time feedback information on the actual power consumption of each motor for dynamic adjustment, making it difficult to optimize the utilization of the total system power and hindering further improvement in the energy utilization efficiency of the entire vehicle. Summary of the Invention

[0004] The technical problem solved by this application is that the power distribution method of the existing multi-motor distributed drive system is fixed or based on open-loop control, and fails to make dynamic adjustments by utilizing the real-time feedback information of the actual power used by each motor. This results in energy waste when the power of a certain wheel is not fully utilized, and the power of other wheels cannot be replenished in time when it is insufficient, resulting in low energy utilization efficiency of the whole vehicle.

[0005] The purpose of this application is to overcome the shortcomings of fixed power allocation and low energy utilization efficiency in existing multi-motor distributed drive systems, and to provide an energy management method and system that can dynamically adjust power allocation according to the real-time torque demand and actual power consumption of each wheel. By introducing a surplus power redistribution mechanism, the system power is optimized, thereby improving the overall vehicle energy efficiency and dynamic performance.

[0006] To achieve the above objectives, this application provides the following technical solution: In a first aspect, this application provides a multi-motor distributed drive energy management method, applied to an electric vehicle with multiple wheels, each driven by an independent motor, comprising the following steps: The process includes: an acquisition step, acquiring the total available power of the drive system, the real-time torque requirement of each wheel, and the actual power utilized by the motor of each wheel; a first allocation step, determining the torque allocation ratio of each wheel based on the real-time torque requirement of each wheel, and allocating the total available power to each wheel based on the torque allocation ratio and the efficiency coefficient of each motor, as the base available power of each wheel; a residual power calculation step, for each wheel, calculating the difference between the base available power of that wheel and the actual power utilized by that wheel, as the residual power of that wheel; a second allocation step, dynamically redistributing the residual power of each wheel to other wheels to adjust the available power of other wheels, wherein the final available power of each wheel is equal to the base available power of that wheel plus the sum of the additional power allocated from each of the other wheels; and an execution step, controlling the output of the corresponding motor based on the final available power of each wheel.

[0007] Optionally, in the second allocation step, the remaining power of each wheel is evenly distributed to the other wheels, and the additional power of each wheel is equal to the sum of the remaining power of each of the other wheels divided by the number of the other wheels.

[0008] Optionally, the remaining power of each wheel is allocated to the other wheels according to a dynamic weight. The additional power of each wheel is equal to the sum of the remaining power of each other wheel multiplied by the corresponding allocation weight, wherein the allocation weight is dynamically calculated based on the real-time status of each wheel.

[0009] Optionally, the allocation weights are determined based on at least one of the following factors: the torque demand gap of each wheel, the current operating efficiency of the motors of each wheel, the road adhesion coefficient of each wheel, and the priority of each wheel in vehicle stability control.

[0010] Optionally, the allocation weights are determined according to the following principles: wheels with a larger torque demand gap receive a higher allocation weight; wheels corresponding to motors with higher current operating efficiency receive a higher allocation weight; wheels with a lower road surface adhesion coefficient receive a lower allocation weight; and wheels with higher priority in stability control receive a higher allocation weight.

[0011] Optionally, the allocation weights are calculated in real time by a pre-trained machine learning model, which takes the real-time state parameters of the vehicle as input and outputs the optimized allocation weights for each wheel.

[0012] Optionally, the machine learning model adopts an online learning approach, continuously updating the model parameters based on the deviation between historical allocation results and actual power consumption.

[0013] Optionally, the online learning uses a recursive least squares algorithm or an online gradient descent algorithm to update the model parameters.

[0014] Optionally, the torque distribution ratio is determined based on the proportion of the product of the real-time torque demand of each wheel and the correction factor to the sum of the products of all wheels.

[0015] Optionally, the correction factor is dynamically adjusted according to the vehicle status, which includes at least one of steering conditions, road surface adhesion coefficient, motor efficiency operating range, and vehicle load transfer.

[0016] Optionally, the adjustment methods for the correction coefficient include: reducing the correction coefficient of the inner wheel under steering conditions; adjusting the correction coefficient of the corresponding wheel under low-traction road conditions; prioritizing power allocation to the motor currently in the high-efficiency operating range, with a corresponding increase in the correction coefficient; increasing the correction coefficient of the rear wheel under acceleration conditions; and increasing the correction coefficient of the front wheel under braking conditions.

[0017] Optionally, the basic available power is calculated by multiplying the total available power by the torque distribution ratio of the corresponding wheel, and then multiplying it by the comprehensive efficiency coefficient of the motor and transmission system corresponding to that wheel.

[0018] Optionally, the remaining power is calculated by subtracting the actual utilized power from the basic available power. A positive difference indicates that the wheel has surplus power, while a negative difference indicates that the wheel has insufficient power.

[0019] Optionally, before distributing the remaining power equally to other wheels, the process further includes a step of smoothing the remaining power. The smoothing process uses a weighted sum of the remaining power at the current moment and the filtered remaining power at the previous moment, wherein the weighting coefficient of the remaining power at the current moment ranges from 0.6 to 0.9.

[0020] Optionally, there are a total of four wheels, and the final available power of each wheel is equal to the base available power of that wheel plus the sum of the remaining power of the other three wheels divided by three.

[0021] Optionally, the final available power satisfies at least one of the following constraints: single wheel power constraint, i.e., the final available power of each wheel does not exceed the maximum allowable power of the motor of that wheel; total system power constraint, i.e., the sum of the final available power of all wheels does not exceed a preset multiple of the total available power; power change rate constraint, i.e., the change rate of the final available power of each wheel does not exceed a preset maximum power change rate threshold.

[0022] Optionally, the preset multiplier is 1.05.

[0023] Optionally, the acquisition step, the first allocation step, the remaining power calculation step, the second allocation step, and the execution step are repeatedly executed at a frequency of 10 Hz to 100 Hz to achieve real-time dynamic power management.

[0024] Optionally, the actual power utilized is obtained through real-time feedback from the motor controller, and the total available power is dynamically determined based on the state of charge, temperature, and health status of the power battery.

[0025] Secondly, this application provides a multi-motor distributed drive energy management system for electric vehicles with multiple wheels, each driven by an independent motor. The system includes: a power battery system for providing drive energy; multiple independent motors and corresponding motor controllers, each motor independently driving one wheel; a sensor group for collecting vehicle status data; and a central processing unit communicatively connected to the power battery system, motor controllers, and sensor group. The central processing unit is configured to execute the multi-motor distributed drive energy management method described above.

[0026] Optionally, the central processing unit obtains the real-time torque demand of each wheel and the actual power utilized by each wheel motor through the vehicle's CAN bus, and determines the total available power based on the current state of the power battery.

[0027] Optionally, the central processing unit is configured to: when the remaining power is positive, transfer the surplus power corresponding to the positive value to other wheels by means of average distribution; when the remaining power is negative, make up for the power deficiency by receiving the average remaining power distributed from other wheels.

[0028] Optionally, the central processing unit is also configured to: limit the final available power to the maximum permissible power when the final available power exceeds the maximum permissible power; and proportionally reduce the final available power of each wheel when the sum of the final available powers exceeds 1.05 times the total available power.

[0029] Optionally, the central processing unit is also configured to convert the calculated final available power of each wheel into a corresponding motor torque limit and send it to the corresponding motor controller, which outputs the actual torque according to the motor torque limit and the driver's required torque, wherein the actual torque does not exceed the motor torque limit.

[0030] Optionally, the central processing unit integrates a lightweight machine learning inference engine to perform forward computation of the machine learning model, which is one of a neural network model, a decision tree model, or a linear regression model.

[0031] Optionally, the central processing unit is further configured to: collect the correspondence between actual power consumption data and allocation results during vehicle operation, periodically or in real time transmit the data back to the cloud server for model retraining, and send the updated model parameters to the vehicle.

[0032] Optionally, the central processing unit is further configured to convert the calculated final available power of each wheel into a corresponding motor torque limit and send it to the corresponding motor controller, wherein the motor controller outputs an actual torque based on the motor torque limit and the driver's required torque, wherein the actual torque does not exceed the motor torque limit.

[0033] Compared to existing technologies, this application offers at least the following advantages: By introducing the actual power utilized by each wheel motor as a closed-loop feedback quantity, this application calculates the difference between the basic available power and the actual utilized power of each wheel in real time—that is, the remaining power. This accurately identifies which wheels in the system have excess power and which have insufficient power. The excess power is then redistributed evenly to other wheels, allowing underutilized power to be recovered and reinjected into the system, avoiding energy waste caused by differences in actual consumption after power allocation in traditional methods. Simultaneously, this dynamic redistribution mechanism is simple and efficient, requiring no complex optimization iteration calculations. It can quickly respond to dynamic changes in vehicle driving conditions (such as rapid acceleration, steering, and changes in road surface adhesion), ensuring real-time balance in power distribution among wheels. Furthermore, by setting multiple constraints such as the maximum power of a single wheel, the total system power, and the rate of power change, it ensures that power distribution remains within a safe range, preventing motor overload or system power exceeding limits. This application significantly improves the overall vehicle energy utilization efficiency, enhances the system's dynamic adaptability and operational safety, and is primarily implemented based on software algorithms, resulting in low implementation costs and easy application on existing distributed drive platforms.

[0034] Furthermore, the average allocation scheme has the advantages of simple calculation, high real-time performance, and easy engineering implementation, making it suitable for scenarios with limited computing resources or extremely high requirements for response speed.

[0035] Furthermore, the dynamic weighted allocation scheme, based on average allocation, introduces a multi-dimensional weighted decision-making mechanism based on real-time status, which can more accurately allocate the remaining power to the wheels that need it most, are most effective, and are safest, thereby further improving energy utilization efficiency, power response performance, and driving safety.

[0036] The two parallel schemes can be flexibly selected or switched according to vehicle configuration and driving mode, providing users with differentiated technical paths and meeting the needs of different levels of products from economical to high-end intelligent.

[0037] Furthermore, the machine learning self-learning scheme, as a further optimization of dynamic weighted allocation, enables the weight allocation strategy to learn and optimize adaptively without the need for manual calibration of complex weight rules. It has the ability to continuously evolve, significantly improving the system's intelligence level and adaptability to operating conditions. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the multi-motor distributed drive energy management method provided in this application; Figure 2 This is a schematic diagram of the multi-motor distributed drive energy management system provided in this application; Detailed Implementation The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application. The step numbers in the following embodiments are set only for ease of explanation, and there is no limitation on the order between the steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0039] The following are explanations of uncommon technical terms and English abbreviations used in this application, for reference when understanding the technical solutions.

[0040] 1. VCU (Vehicle Control Unit) The vehicle controller (VCU) is the core control unit of an electric vehicle, responsible for interpreting driver intentions (acceleration, braking, steering), coordinating the work of various subsystems (battery management system, motor controller, vehicle stability system, etc.), and executing the vehicle's energy management and torque distribution strategies. In this application, the VCU serves as the central processing unit, performing power distribution and redistribution calculations.

[0041] 2. BMS (Battery Management System) The battery management system (BMS) monitors the voltage, current, temperature, and other state parameters of the power battery, estimates the battery's state of charge and health, and calculates the battery's current maximum allowable charge and discharge power. In this application, the BMS provides a reference value of the total available power to the VCU via a CAN bus.

[0042] 3. CAN (Controller Area Network) A Controller Area Network (VCU) is a serial bus protocol for communication between various electronic control units within a vehicle, characterized by strong real-time performance and good anti-interference capabilities. In this application, the VCU obtains the actual power utilization of each motor, the torque requirements of each wheel, and battery status information via the CAN bus.

[0043] 4. SOC (State of Charge) State of charge (SOC) represents the percentage of the battery's current remaining charge compared to its total charge when fully charged, ranging from 0% to 100%. The Battery Management System (BMS) estimates the SOC value in real time, and the Vehicle Control Unit (VCU) dynamically adjusts the total available power based on the SOC value to prevent overcharging or over-discharging of the battery.

[0044] 5. SOH (State of Health) State of Health (SOH) characterizes the degree of performance degradation of a power battery relative to that of a new battery, typically expressed as capacity retention rate or internal resistance increase rate. As battery usage time increases, SOH gradually decreases, and the VCU adjusts the total available power based on the SOH.

[0045] 6. MCU (Motor Control Unit) The motor controller is the control unit that drives the motor. It receives torque commands from the VCU and controls the motor's current and voltage by adjusting the on-time of the power switching transistors in the inverter, thereby precisely controlling the motor's output torque. In this application, the MCU is also responsible for providing feedback on the actual power utilized by the motor.

[0046] 7. Torque Distribution Ratio Torque distribution ratio represents the proportion of driving torque that each wheel should bear under the current driving conditions to the total required torque. The sum of the torque distribution ratios of all wheels is 1. In this application, torque distribution ratio is a key parameter for calculating the basic available power of each wheel.

[0047] 8. Basic Available Power The basic available power is the initial power allocation value obtained by each wheel in the first allocation step. It is calculated by multiplying the total available power by the torque allocation ratio of the corresponding wheel, and then by the overall efficiency coefficient of the motor and transmission system.

[0048] 9. Actual Utilized Power The actual power consumed is calculated by the motor controller based on real-time acquisition of the motor's three-phase voltage and current. This parameter is a key feedback quantity that distinguishes this application from existing technologies, and is used to identify the deviation between power allocation and actual consumption.

[0049] 10. Remaining Power Residual power is the difference between the basic available power and the actual utilized power. When the residual power is positive, it means that the power allocated to that wheel is not being fully utilized, resulting in energy waste; when the residual power is negative, it means that the power allocated to that wheel is insufficient, requiring additional power support.

[0050] 11. Final Available Power Final available power is the actual power limit available to each wheel after the redistribution of surplus power. The final available power of each wheel is equal to the base available power of that wheel plus the sum of the surplus power distributed equally from the other wheels.

[0051] 12. Smoothing Filter Coefficient The smoothing filter coefficients, denoted by α, are weighting parameters in a first-order low-pass filter, ranging from 0.6 to 0.9. This coefficient controls the weight of the current sampled value in the filtering result. The closer α is to 1, the faster the filtering response but the weaker the smoothing effect; the closer α is to 0.6, the stronger the smoothing effect but the greater the response delay.

[0052] 13. Efficiency MAP The efficiency MAP is a two-dimensional table pre-calibrated using motor bench tests. It takes motor speed and torque as input and outputs the overall efficiency of the motor and transmission system at the current operating point. The VCU obtains the efficiency coefficient ηi through table lookup and interpolation.

[0053] 14. Power Constraint Conditions Power constraints are power limiting rules that ensure the safe operation of the system. They include the maximum power limit of a single wheel (not exceeding the maximum allowable power of the motor), the total power limit of the system (the sum of the power of each wheel does not exceed 1.05 times the total available power), and the power change rate limit (the power change between adjacent control cycles does not exceed a preset threshold).

[0054] 15. Average Redistribution Mechanism The average redistribution mechanism is the core algorithm of this application. It involves dividing the remaining power of each wheel by the number of other wheels and then distributing it equally among the other wheels. For a four-wheeled vehicle, the additional power of each wheel is the sum of the remaining power of the other three wheels divided by three. This mechanism is computationally simple, has low communication overhead, and can achieve power balancing without complex optimization.

[0055] In multi-motor distributed drive electric vehicles, each wheel is driven by an independent motor, making energy management a core issue. Current technologies typically rely on fixed ratios or simple algorithms for power allocation, failing to adequately consider variations in torque demand and actual power utilization at each wheel. This results in uneven energy distribution and low efficiency. For example, when the actual power utilization of a wheel is lower than its allocated power, the surplus power is wasted, while other wheels may suffer from insufficient power. Therefore, a dynamic energy management method is needed to adjust the available power of each wheel in real time to optimize overall energy efficiency.

[0056] The core of this application lies in establishing a multi-level power distribution mechanism based on torque distribution ratio and real-time power feedback. This mechanism first allocates basic power according to the torque demand of each wheel, and then dynamically redistributes the underutilized power (residual power) to other wheels that need power by monitoring the actual power usage of each motor, forming a closed-loop power management system.

[0057] In existing technologies, multi-motor distributed drive systems typically employ a fixed ratio or a power allocation method based on offline optimization lookup tables. This fails to utilize real-time feedback information on the actual power consumption of each motor for dynamic adjustment, resulting in energy waste when the power allocated to a certain wheel is not fully utilized, while other wheels cannot receive timely replenishment when their power is insufficient.

[0058] Regarding this issue, firstly, such as Figure 1 As shown, this embodiment provides a multi-motor distributed drive energy management method, which is applied to electric vehicles with multiple wheels, each driven by an independent motor, such as four-wheel independently driven electric passenger vehicles or commercial vehicles. The method specifically includes the following steps.

[0059] Acquisition Steps: The central processing unit (e.g., the vehicle control unit, VCU, or a separate domain controller) collects the total available power of the drive system, the real-time torque demand of each wheel, and the actual power utilized by each wheel motor in real time via the vehicle's CAN bus. The total available power is dynamically calculated by the battery management system based on the current state of charge, temperature, and health of the power battery and sent to the CAN bus. This total available power can also be further adjusted to account for the power consumption of other electrical accessories in the vehicle. The real-time torque demand of each wheel is calculated by the upper-level controller based on the driver's accelerator pedal opening, brake pedal opening, steering wheel angle, and vehicle stability control requirements (such as yaw moment control and traction control). The actual power utilized by each wheel motor is calculated in real time based on the three-phase current, voltage, and power factor of the motor fed back from each motor controller. This actual power utilized reflects the actual electrical power consumed by the motor at the current operating point.

[0060] The first allocation step: The central processing unit determines the torque allocation ratio for each wheel based on the real-time torque demand of each wheel. Then, based on this torque allocation ratio and the overall efficiency coefficient of each motor and transmission system, it allocates the total available power to each wheel as the base available power for each wheel. Specifically, the torque allocation ratio reflects the proportion of driving tasks that each wheel should undertake under the current driving conditions. The central processing unit calculates the torque allocation ratio for each wheel based on the real-time torque demand of each wheel and the vehicle's dynamic state (such as steering angle, road adhesion coefficient, load transfer, etc.). Subsequently, the total available power is multiplied by the torque allocation ratio of the corresponding wheel, and then multiplied by the overall efficiency coefficient of the corresponding motor and transmission system to obtain the base available power for that wheel. The overall efficiency coefficient is obtained in real-time by looking up an efficiency MAP table obtained from pre-testing on a motor test bench. This coefficient integrates the efficiency of the motor itself, the inverter efficiency, and the mechanical transmission efficiency of the reducer.

[0061] Remaining power calculation steps: For each wheel, the central processing unit calculates the difference between the wheel's base available power and its actual utilized power, and uses this difference as the wheel's remaining power. A positive remaining power value indicates that the wheel's currently allocated base available power is not being fully utilized, resulting in a power surplus; a negative remaining power value indicates that the wheel's current actual utilized power exceeds its allocated base available power, resulting in a power shortage.

[0062] The second allocation step: The central processing unit distributes the surplus power of each wheel equally among the other wheels to adjust the available power of the other wheels. Specifically, for any given wheel, the average surplus power allocated from each of the other wheels is equal to the surplus power of that wheel divided by the number of the other wheels. The final available power of that wheel equals its base available power plus the sum of the average surplus power allocated from each of the other wheels. Through this averaging mechanism, the positive surplus power of power-surplus wheels is transferred to power-deficient wheels, while the negative surplus power of power-deficient wheels is compensated from other wheels, achieving dynamic power balance in the system.

[0063] Execution steps: The central processing unit sends torque commands to each motor controller via the CAN bus based on the final available power calculated for each wheel. Each motor controller controls the output torque of the corresponding motor according to the received torque commands, so that the actual power consumed by the motor does not exceed the limit of the final available power.

[0064] This embodiment introduces actual utilized power as a closed-loop feedback quantity to identify wheels with excess power and insufficient power in real time, and realizes dynamic redistribution of power through an average distribution mechanism, which effectively reduces energy waste and improves the overall vehicle energy efficiency.

[0065] This embodiment further defines the calculation method for the torque distribution ratio. In this embodiment, the torque distribution ratio is determined based on the proportion of the product of the real-time torque demand of each wheel and the correction coefficient to the sum of the products of all wheels. Specifically, the central processing unit first obtains the real-time torque demand T of each wheel. i And determine the correction factor C for each wheel based on the current vehicle condition. i This correction factor is used to weight and adjust the original torque demand. This calculation method ensures that the sum of the torque distribution proportions of all wheels is 1, satisfying the power conservation constraint.

[0066] Specifically, the torque distribution ratio is calculated as follows: Torque distribution ratio K i It represents the proportion of driving tasks that each wheel should undertake under the current driving conditions, and its calculation comprehensively considers various vehicle dynamics factors:

[0067] Where Ci is a correction factor based on vehicle state, with a value ranging from 0.9 to 1.1, and its calculation takes into account: Steering condition correction: When steering, the proportion of the inner wheel distribution should be appropriately reduced; Road surface adhesion coefficient correction: Adjustment of the wheel distribution ratio on roads with low adhesion; Motor efficiency MAP correction: Prioritize power allocation to motors operating in the high-efficiency region; Vehicle load transfer correction: The proportion of load on the rear wheels increases during acceleration, and the proportion on the front wheels increases during braking.

[0068] This embodiment further defines the basis for adjusting the correction coefficient. In this embodiment, the correction coefficient is dynamically adjusted according to the vehicle state, which includes at least one of steering conditions, road surface adhesion coefficient, motor efficiency operating range, and vehicle load transfer. The central processing unit obtains steering angle information through the steering wheel angle sensor, estimates the road surface adhesion coefficient through wheel speed and acceleration sensors, obtains the current efficiency range by querying the motor efficiency MAP table based on the motor speed and torque fed back by the motor controller, and estimates the vehicle load transfer amount through the vehicle body acceleration sensor and suspension height sensor. The central processing unit dynamically calculates the correction coefficient for each wheel based on the above information, so that the torque distribution ratio can adapt to the real-time changes in the vehicle's driving state.

[0069] This embodiment further defines the specific adjustment method of the correction coefficient. In this embodiment, the adjustment method of the correction coefficient includes the following scenarios: During steering, the central processing unit reduces the correction coefficient of the inner wheel. For example, when the vehicle turns left, the left wheel is the inner wheel, and its correction coefficient is reduced to 0.92, while the right wheel is the outer wheel, and its correction coefficient is increased to 1.08. This reduces the torque distribution ratio of the inner wheel and increases the torque distribution ratio of the outer wheel, improving steering maneuverability. On low-traction surfaces (such as icy or slippery surfaces), the central processing unit reduces the correction coefficient of the wheel with the lower coefficient of adhesion, reducing the torque distribution of that wheel and preventing it from slipping due to excessive torque. Regarding motor efficiency optimization, the central processing unit queries the efficiency MAP table for each motor. For motors with higher efficiency at the current operating point, its correction coefficient is increased, for example, from 1.0 to 1.05, thereby prioritizing power allocation to motors operating in the high-efficiency zone and improving the overall system efficiency. Regarding vehicle load transfer, the central processing unit increases the correction factor for the rear wheels during acceleration (e.g., from 1.0 to 1.06) and the correction factor for the front wheels during braking (e.g., from 1.0 to 1.05) to accommodate the impact of axle load transfer on wheel adhesion.

[0070] This embodiment further defines the calculation method for basic available power. In this embodiment, the basic available power is calculated by multiplying the total available power by the torque distribution ratio of the corresponding wheel, and then multiplying by the comprehensive efficiency coefficient of the motor and transmission system corresponding to that wheel. The comprehensive efficiency coefficient ηi is obtained through a pre-calibrated efficiency MAP table, which takes motor speed and motor torque as input and outputs the comprehensive efficiency of the motor and transmission system. In actual operation, the central processing unit calculates the expected torque based on the current motor speed and torque distribution ratio, queries the efficiency MAP table to obtain the comprehensive efficiency coefficient ηi, and then calculates the basic available power according to the basic available power calculation method. This calculation method considers the actual efficiency loss of the motor and transmission system, making the basic available power more accurately reflect the actual available electrical power at the motor input.

[0071] Specifically, the basic available power calculation method is as follows: The basic available power of each wheel is initially allocated from the total available power according to its torque distribution ratio:

[0072] Where, η i The overall efficiency coefficient of each motor and transmission system can be obtained in real time through the pre-tested efficiency MAP table.

[0073] This embodiment further defines the calculation method and physical meaning of surplus power. In this embodiment, the surplus power is calculated by subtracting the actual utilized power from the base available power. When the calculation result is positive, it indicates that the wheel has surplus power, meaning that the actual power consumed is less than the allocated power, and the surplus power is not utilized, constituting a potential source of waste. When the calculation result is negative, it indicates that the wheel has insufficient power, meaning that the actual power consumed is greater than the allocated power, and the wheel needs additional power support to meet real-time torque requirements. By calculating the surplus power of each wheel, the central processing unit quantitatively identifies which wheels in the system have energy waste and which wheels have power deficits, providing an accurate data basis for subsequent redistribution.

[0074] The specific solution is as follows: residual power calculation and dynamic redistribution, including the following detailed steps: 1) Remaining power identification

[0075] When Premain_i > 0, it means that the wheel has excess power; when Premain_i < 0, it means that the wheel has insufficient power.

[0076] 2) Residual power normalization processing To avoid system oscillations caused by sudden changes in power distribution, the remaining power is smoothed out:

[0077] Where α is the filtering coefficient, with a value ranging from 0.6 to 0.9.

[0078] 3) Residual power redistribution The filtered remaining power of each wheel is evenly distributed to the other three wheels:

[0079] 4) Power redistribution constraints To prevent excessive power allocation, the following constraints are set: Single wheel maximum power limit: P avail_i ≤ P motor_max ; System total power limit: ΣP avail_i ≤ P total × 1.05 (allowing a 5% fluctuation margin); Power change rate limit: |P avail_i (t) - P avail_i (t-1)| ≤ ΔP max .

[0080] This embodiment further defines the smoothing filtering process. In this embodiment, before evenly distributing the remaining power to other wheels, a step of smoothing the remaining power is included. Specifically, the central processing unit uses a first-order low-pass filter to filter the remaining power. The filtering formula is: the filtered remaining power at the current moment equals the original remaining power at the current moment multiplied by the filtering coefficient α, plus the filtered remaining power at the previous moment multiplied by (1-α). The filtering coefficient α ranges from 0.6 to 0.9; in a specific implementation of this embodiment, α is set to 0.8. This filtering process can suppress instantaneous fluctuations in remaining power, avoid system oscillations and torque shocks caused by sudden changes in power distribution, and improve the stability and driving smoothness of the energy management system.

[0081] This embodiment further defines the specific calculation method in a four-wheel scenario. In this embodiment, there are a total of four wheels, meaning the vehicle has a four-wheel independent drive configuration. The final available power of each wheel is equal to the basic available power of that wheel plus the sum of the remaining power of the other three wheels, divided by three. This average allocation mechanism has the advantages of simple calculation and low communication overhead, and can achieve dynamic power balance without complex optimization algorithms.

[0082] In this application, the second allocation step provides two parallel implementation schemes, which can be selected according to vehicle configuration and control requirements or switched between under different operating conditions.

[0083] The first implementation scheme: average distribution; in this scheme, the central processing unit distributes the remaining power of each wheel equally among the other wheels. For any wheel i, its final available power is calculated using the following formula:

[0084] Where m is the total number of wheels. When m=4, the final available power of each wheel equals the base available power plus the sum of the remaining power of the other three wheels, divided by three. This scheme is simple to calculate, has low communication overhead, and good real-time performance, making it suitable for embedded controllers with limited computing resources.

[0085] The total number of wheels is m. The final usable power of each wheel is equal to the basic usable power of that wheel plus the sum of the remaining power of the other n wheels, divided by n; m = n + 1. m is usually 4, but can also be other multi-wheeled vehicles.

[0086] Second implementation plan: Dynamic weighted allocation: In this scheme, the central processing unit distributes the remaining power of each wheel to the other wheels according to dynamic weights, rather than distributing it equally. For any wheel i, the final available power is calculated using the following formula:

[0087] Where, λ ij Let the weighting coefficients for allocating the remaining power of wheel j to wheel i satisfy the condition that for each wheel j, the sum of the weights allocated to other wheels is 1. = 1.

[0088] Weighting coefficient λ ij Based on the real-time dynamic calculations of each wheel, one or more of the following factors are considered: the torque demand gap of each wheel (i.e., the absolute value of the difference between the actual utilized power and the basic available power), the current operating efficiency of the motors of each wheel, the road surface adhesion coefficient of each wheel, and the priority of each wheel in vehicle stability control. The basic principle of weight allocation is: wheels with larger power gaps receive higher allocation weights, motors with higher efficiency receive higher allocation weights, wheels on low-adhesion surfaces receive lower allocation weights to avoid further slippage, and stability-critical wheels (such as steering wheels) receive higher allocation weights.

[0089] Furthermore, in the further optimization scheme of dynamic weighted allocation, the weight λ is allocated. ij The data is obtained in real time through a pre-trained machine learning model. This machine learning model takes real-time vehicle state parameters (including torque demand gaps for each wheel, motor speed, motor efficiency, road adhesion coefficient, yaw rate, steering wheel angle, etc.) as input and outputs an optimized weight vector for each wheel.

[0090] The machine learning model is continuously updated using online learning. The central processing unit (CPU) collects deviation data between historical power allocation results and actual power consumption, and updates model parameters in real time using recursive least squares or online gradient descent algorithms. This allows the weight allocation strategy to adaptively learn the driver's driving habits and the vehicle's power response characteristics. The CPU integrates a lightweight machine learning inference engine capable of performing forward computations for neural network models, decision tree models, or linear regression models. During vehicle operation, the CPU can also transmit collected data back to a cloud server for model retraining and distribute the updated model parameters back to the vehicle, enabling continuous model optimization.

[0091] This embodiment further defines the constraints that the final available power must meet. In this embodiment, the final available power must meet at least one of the following three constraints: Single wheel power constraint: The final available power of each wheel does not exceed the maximum allowable power of the motor of that wheel. This maximum allowable power is calculated and reported in real time by the motor controller based on the current temperature, voltage, and current limits of the motor. Total system power constraint: The sum of the final available power of all wheels does not exceed a preset multiple of the total available power, for example, 1.05 times, allowing a short-term power fluctuation margin of 5% to adapt to transient conditions. Power change rate constraint: The change rate of the final available power of each wheel does not exceed a preset maximum power change rate threshold, for example, 10kW / s, to prevent torque shocks caused by sudden changes in power commands. After calculating the final available power, the central processing unit checks the above constraints one by one, and performs limiting processing for cases that exceed the constraints.

[0092] This embodiment further defines a preset multiple for the total system power constraint. In this embodiment, the preset multiple for the total system power constraint is 1.05. That is, the sum of the final available power of all wheels is allowed to float up by 5% based on the total available power. This design takes into account the short-term overload capability of the power battery under transient conditions and the dynamic response characteristics of the motor controller, avoiding loss of power performance due to excessive limiting while ensuring system safety.

[0093] The execution frequency of the method is further defined based on Embodiment 1. In this embodiment, the acquisition step, the first allocation step, the remaining power calculation step, the second allocation step, and the execution step are repeatedly executed at a frequency of 10 Hz to 100 Hz. In a specific implementation, the execution frequency is set to 50 Hz, that is, a complete energy management calculation is completed and the torque commands of each motor are updated every 20 milliseconds. This execution frequency can match the vehicle's dynamic response speed and the CAN bus communication cycle, ensuring that power management can track changes in driver operation and vehicle status in real time, achieving real-time dynamic power management. In higher-performance controllers, an execution frequency of 100 Hz can be used to obtain a faster response speed.

[0094] Building upon Example 1, the data sources for actual utilized power and total available power are further defined. In this example, actual utilized power is obtained through real-time feedback from the motor controller. Each motor controller integrates voltage and current sensors to collect the three-phase voltage and current of the motor in real time. Combined with the power factor, the actual input power of the motor is calculated and periodically (e.g., every 10ms) transmitted to the central processing unit via the CAN bus. Total available power is dynamically determined based on the state of charge, temperature, and health status of the power battery. The battery management system monitors the voltage, current, and temperature of individual battery cells in real time, estimates the maximum allowable discharge power of the battery based on the battery equivalent circuit model, and, after considering health status corrections due to battery aging, transmits the data to the central processing unit via the CAN bus.

[0095] The process of the above method in this application includes: Step S101: Obtain the total available power P_total of the drive system. The total available power can be the total drive power provided by the battery system, or the available power calculated based on vehicle conditions (such as acceleration and vehicle speed).

[0096] Step S102: Calculate the torque distribution ratio K_i for each wheel. Here, i represents the wheel designation (e.g., left front wheel LF, right front wheel RF, left rear wheel LR, right rear wheel RR). The torque distribution ratio K_i can be calculated based on the vehicle dynamics model, such as parameters based on steering angle, acceleration, and road conditions, to ensure vehicle stability and efficiency. For example, K_i can be calculated using the following formula:

[0097] Where T_i is the torque requirement of wheel i.

[0098] Step S103: Calculate the base available power P_base_i for each wheel. For example, the base available power for the left front wheel is:

[0099] The same applies to the other wheels.

[0100] Step S104: Monitor the actual power used by each wheel, P_used_i. The actual power used can be obtained in real time through the motor controller or sensors, representing the power actually consumed by the wheel.

[0101] Step S105: Calculate the remaining power P_remain_i of each wheel. For example, the remaining power of the left front wheel is:

[0102] If P_remain_i is negative, it means that the wheel has insufficient power. However, in this application, the remaining power is mainly used for redistribution. Therefore, a negative value may indicate that power needs to be obtained from other wheels (see the following description for details).

[0103] Step S106: Divide the remaining power P_remain_i of each wheel by 3 to obtain the average remaining power ΔP_i. For example, the average remaining power of the left front wheel is:

[0104] Step S107: Calculate the final available power P_avail_i for each wheel. For each wheel, its final available power equals its base available power plus the sum of the average remaining power of the other three wheels. For example, the final available power of the left front wheel is:

[0105] Similarly, the final available power of the other wheels is calculated in a similar way.

[0106] In practical applications, the redistribution of surplus power is carried out in real time to ensure dynamic power adjustment. If the surplus power of a certain wheel is negative (i.e., the actual power used is greater than the basic available power), then its average surplus power ΔP_i is negative, which means that the available power of other wheels will be reduced, thereby achieving power balance.

[0107] Secondly, such as Figure 2 As shown, this embodiment provides a multi-motor distributed drive energy management system applied to electric vehicles with multiple wheels, each driven by an independent motor. The system includes: a power battery system, multiple independent motors and corresponding motor controllers, a sensor array, and a central processing unit. The power battery system provides drive energy and typically uses lithium-ion battery packs with a rated voltage range of 300V to 800V. Each of the multiple independent motors and their corresponding motor controllers independently drives one wheel. The motor type can be a permanent magnet synchronous motor or an induction motor, and the power range is determined according to the vehicle type; for passenger vehicles, it is typically 50kW to 150kW per wheel. The sensor array is used to collect vehicle status data, including but not limited to wheel speed sensors, acceleration sensors, steering wheel angle sensors, and suspension height sensors. The central processing unit (such as a vehicle control unit (VCU) or an autonomous driving domain controller) communicates with the power battery system, motor controllers, and sensor array via a CAN bus or in-vehicle Ethernet. The central processing unit is configured to execute a multi-motor distributed drive energy management method to achieve closed-loop dynamic power management.

[0108] This embodiment further defines the data acquisition method of the central processing unit. In this embodiment, the central processing unit acquires the real-time torque requirements of each wheel and the actual power utilization of each wheel motor through the vehicle's CAN bus. The CAN bus adopts the ISO 11898 standard, and the communication rate is typically 500kbps or 1Mbps. Each motor controller encapsulates information such as the actual power utilization of the motor, current speed, and current torque into CAN messages and sends them periodically according to the agreed ID. The upper-level controller (such as the autonomous driving controller or the driver intent parsing module) also sends the calculated real-time torque requirements of each wheel through the CAN bus. The central processing unit, as a CAN node, receives the above messages and parses them to obtain the required data. At the same time, the central processing unit communicates with the battery management system through the CAN bus to obtain the current status of the power battery, including the state of charge, maximum allowable discharge power, temperature, etc., thereby determining the total available power.

[0109] This embodiment further defines how the central processing unit handles positive and negative surplus power. In this embodiment, the central processing unit is configured to: when the surplus power is positive, transfer the surplus power corresponding to the positive value to other wheels through an average distribution. For example, when the surplus power of the left front wheel is +2kW, the central processing unit distributes this 2kW of surplus power equally to the right front wheel, left rear wheel, and right rear wheel, with each wheel receiving approximately 0.67kW of additional power. When the surplus power is negative, for example, the surplus power of the left front wheel is -3kW, indicating that the wheel's power is less than 3kW, the central processing unit compensates for this power deficiency by receiving the average surplus power distributed from the other three wheels. When the sum of the surplus power of the other three wheels is greater than or equal to 3kW, the power deficiency of the left front wheel can be fully compensated; when the sum of the surplus power is less than 3kW, the final usable power of the left front wheel will still be less than its actual demand, at which point the central processing unit can further trigger vehicle speed limiting or power downgrading strategies.

[0110] This embodiment further defines the specific formula for the smoothing filtering process performed by the central processing unit. In this embodiment, the central processing unit performs smoothing filtering according to the following formula:

[0111] Where α is the filter coefficient, ranging from 0.6 to 0.9, t represents the current time, and t-1 represents the previous time. This represents the remaining power after filtering. The closer the filter coefficient α is to 1, the faster the filtering response, but the weaker the smoothing effect; the closer α is to 0.6, the stronger the smoothing effect, but the greater the response delay. In a specific implementation of this embodiment, α is set to 0.75, achieving a balance between response speed and smoothing effect. This first-order low-pass filter is simple to calculate; in an embedded controller, it only needs to store the filter value from the previous moment, resulting in minimal computational overhead.

[0112] This embodiment further defines the power limiting protection function of the central processing unit. In this embodiment, the central processing unit is also configured to perform the following power protection operations. When the final available power of a wheel exceeds the maximum allowable power of the wheel's motor, the central processing unit limits the final available power of that wheel to the maximum allowable power, discards the excess power, and sends a power-limited warning flag to the motor controller. When the sum of the final available power of all wheels exceeds 1.05 times the total available power, the central processing unit proportionally reduces the final available power of each wheel. Specifically, it calculates the excess power difference and then reduces it proportionally according to the current final available power of each wheel until the total power meets the constraint conditions. This power limiting protection mechanism ensures that the system always operates within the safe operating range of the power battery and motor.

[0113] This embodiment further defines the torque conversion and execution functions. In this embodiment, the central processing unit is also configured to convert the calculated final available power of each wheel into a corresponding motor torque limit and send it to the corresponding motor controller. The power-to-torque conversion formula is: T max_i = (P avail_i (× 9550) / n i T max_i P represents the maximum permissible torque (in Nm) of the motor corresponding to wheel i. avail_i n represents the final available power (in kW). i This represents the current motor speed (in rpm), with 9550 as the unit conversion factor. After receiving this torque limit, the motor controller combines it with the driver's torque demand (derived from the accelerator pedal opening) to output the actual torque, ensuring that the actual torque does not exceed the limit. For example, if the driver's torque demand is 200 Nm and the torque limit is 150 Nm, the motor controller outputs 150 Nm; if the driver's torque demand is 120 Nm, the motor controller outputs 120 Nm. This mechanism ensures that the actual power consumed by the motor is always limited to the final usable power range.

[0114] A third objective of this application is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the aforementioned multi-motor distributed drive energy management method. The device also includes a communication interface and a bus.

[0115] The fourth objective of this application is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned multi-motor distributed drive energy management method.

[0116] A fifth objective of this application is to provide a computer program product comprising computer instructions that instruct a computer to execute the above-described multi-motor distributed drive energy management method.

[0117] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0118] These computer program instructions can also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0119] This application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, readable storage media, optical storage, etc.) containing computer-usable program code.

[0120] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0121] Obviously, the described embodiments are only some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort should fall within the scope of protection of this application.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and not to limit them. Although this application has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation methods of this application. Any modifications or equivalent substitutions that do not depart from the spirit and scope of this application should be covered within the protection scope of this application.

Claims

1. A multi-motor distributed drive energy management method, characterized in that, include: Obtain the total available power of the drive system, the real-time torque requirement of each wheel, and the actual power utilized by the motor of each wheel; The torque distribution ratio of each wheel is determined based on the real-time torque demand of each wheel, and the total available power is distributed to each wheel based on the torque distribution ratio and the efficiency coefficient of each motor, serving as the base available power for each wheel. For each wheel, the difference between the wheel's base available power and the wheel's actual utilized power is calculated as the wheel's remaining power. The remaining power of each wheel is dynamically redistributed to the other wheels to adjust the available power of the other wheels, wherein the final available power of each wheel is equal to the base available power of that wheel plus the sum of the additional power allocated from each of the other wheels; The output of the corresponding motor is controlled according to the final available power of each wheel.

2. The method according to claim 1, characterized in that, The remaining power of each wheel is evenly distributed to the other wheels, and the additional power of each wheel is equal to the sum of the remaining power of each of the other wheels divided by the number of the other wheels.

3. The method according to claim 1, characterized in that, The remaining power of each wheel is distributed to the other wheels according to a dynamic weight. The additional power of each wheel is equal to the sum of the remaining power of each other wheel multiplied by the corresponding distribution weight. The distribution weight is dynamically calculated based on the real-time status of each wheel. Preferably, the allocation weight is determined based on at least one of the following factors: the torque demand gap of each wheel, the current operating efficiency of the motor of each wheel, the road adhesion coefficient of each wheel, and the priority of each wheel in vehicle stability control.

4. The method according to claim 3, characterized in that, The allocation weights are determined according to the following principles: wheels with a larger torque demand gap receive a higher allocation weight; wheels corresponding to motors with higher current operating efficiency receive a higher allocation weight. Wheels with lower road adhesion coefficients receive lower allocation weights; in stability control, wheels with higher priority receive higher allocation weights.

5. The method according to claim 3, characterized in that, The allocation weights are calculated in real time by a pre-trained machine learning model, which takes the real-time state parameters of the vehicle as input and outputs the optimized allocation weights for each wheel. Preferably, the machine learning model adopts an online learning method, continuously updating the model parameters based on the deviation between historical allocation results and actual power consumption; the online learning uses a recursive least squares algorithm or an online gradient descent algorithm to update the model parameters.

6. The method according to claim 1, characterized in that, The torque distribution ratio is determined based on the proportion of the product of the real-time torque demand of each wheel and the correction coefficient to the sum of the products of all wheels. The correction coefficient is dynamically adjusted according to the vehicle status, which includes at least one of steering conditions, road surface adhesion coefficient, motor efficiency operating range, and vehicle load transfer.

7. The method according to claim 1, characterized in that, Before allocating the remaining power to other wheels, the process includes a step of smoothing and filtering the remaining power, wherein the smoothing and filtering process is a weighted sum of the remaining power at the current moment and the filtered remaining power at the previous moment.

8. The method according to claim 1, characterized in that, The final available power satisfies at least one of the following constraints: single wheel power constraint, i.e., the final available power of each wheel does not exceed the maximum allowable power of the motor of that wheel; total system power constraint, i.e., the sum of the final available power of all wheels does not exceed a preset multiple of the total available power; power change rate constraint, i.e., the change rate of the final available power of each wheel does not exceed a preset maximum power change rate threshold.

9. A multi-motor distributed drive energy management system, applied to an electric vehicle with multiple wheels, each wheel driven by an independent motor, characterized in that, include: The power battery system is used to provide driving energy; Multiple independent motors and their corresponding motor controllers, with each motor independently driving one wheel; Sensor array, used to collect vehicle status data; and, The central processing unit is communicatively connected to the power battery system, the motor controller, and the sensor group, and is configured to execute the multi-motor distributed drive energy management method according to any one of claims 1 to 8.

10. The system according to claim 9, characterized in that, The central processing unit is configured to: when the remaining power is positive, transfer the surplus power corresponding to the positive value to other wheels through redistribution; when the remaining power is negative, compensate for the power deficiency by receiving additional power allocated from other wheels. Preferably, the central processing unit is configured to: when using dynamic weighted allocation, collect in real time the torque demand gap, motor efficiency, road surface adhesion coefficient and stability priority of each wheel, calculate the allocation weight according to the preset weighting rules, or call the machine learning model to output the allocation weight; Preferably, the central processing unit integrates a lightweight machine learning inference engine for performing forward computation of the machine learning model, wherein the machine learning model is one of a neural network model, a decision tree model, or a linear regression model. Preferably, the central processing unit is further configured to: collect the correspondence between actual power consumption data and allocation results during vehicle operation, periodically or in real time transmit the data back to the cloud server for model retraining, and send the updated model parameters to the vehicle; Preferably, the central processing unit is further configured to: convert the calculated final available power of each wheel into a corresponding motor torque limit and send it to the corresponding motor controller, wherein the motor controller outputs an actual torque according to the motor torque limit and the driver's required torque, wherein the actual torque does not exceed the motor torque limit; Preferably, the multi-motor distributed drive energy management method is repeatedly executed at a frequency of 10 Hz to 100 Hz to achieve real-time dynamic power management.

Images