Whole-vehicle energy management and endurance optimization method for low-speed four-wheel electric vehicle

Abstract

The invention discloses a whole-vehicle energy management and endurance optimization method for a low-speed four-wheel electric vehicle, particularly relates to the technical field of energy management of electric vehicles, and is used for solving the problems that an existing low-speed four-wheel electric vehicle is extensive in energy distribution and low in completion reliability of key task nodes. Task information including a stop point sequence and a task importance degree is acquired before a task starts, and a guarantee unit, a compressible unit and a sacrifice unit are divided in combination with historical task records; under the constraint of battery available energy and charge lower limit, energy budget is distributed according to traction and accessory energy characteristics of each unit, and an energy envelope monotonically decreasing along with task propulsion is constructed; therefore, the battery energy is finely planned around the task structure on the premise that the hardware cost is not increased, it is preferentially guaranteed that key stop points are completed according to plans, the risk of insufficient energy is exposed in advance, and the endurance stability and task completion power of the low-speed four-wheel electric vehicle in repeated scenes such as park commuting are improved.

Description

Technical Field

[0001] This invention relates to the field of energy management technology for electric vehicles, specifically a method for overall energy management and range optimization of a low-speed four-wheeled electric vehicle. Background Technology

[0002] Existing low-speed four-wheeled electric vehicles are widely used in scenarios such as park commuting, community transportation, and last-mile delivery. These vehicles often perform multiple starts and stops, passenger pick-up and drop-off, and loading and unloading tasks within fixed or semi-fixed routes. In existing technologies, vehicle energy management is usually controlled primarily based on the current state of charge of a single vehicle. Most of these solutions employ fixed lower charge thresholds, mileage estimation, and simple strategies such as "limiting power when the battery is low" and "finding the nearest charging point." A few solutions incorporate factors such as ambient temperature and average energy consumption to adjust the driving range. However, these solutions mostly estimate the total energy demand roughly from the perspective of the entire vehicle, without detailing the energy distribution between different stops and task segments, nor distinguishing the importance of tasks at each stop. This results in the vehicle being forced to conservatively reduce power or temporarily suspend tasks when energy is scarce, lacking precise and controllable guarantees that key stops can be completed as planned.

[0003] While some improvement solutions have begun to utilize historical operational data for statistical analysis of vehicle energy consumption, they largely remain at the level of "total energy consumption per mission" and "average energy consumption." They fail to incorporate factors such as the order of stops, differences in loading conditions, and mission completion records from the mission information to create a structured model of the mission process by stop point or mission unit. Furthermore, they typically do not construct a target energy trajectory that monotonically decreases as the mission progresses before the mission begins, and lack a systematic planning mechanism that separately constrains traction energy and accessory energy consumption. Consequently, when the battery's available energy approaches the critical level for mission completion, existing technologies struggle to promptly identify which stops are essential and which can be reduced or abandoned. They also find it difficult to quantify the risks of subsequent mission completion early on, relying solely on driver experience for on-the-spot judgment, resulting in significant uncertainty regarding range performance and mission completion.

[0004] Therefore, in the context of repetitive low-speed four-wheeled electric vehicle missions, there is still an urgent need for an energy management method that can, before the mission begins, combine mission information and historical mission records to classify and plan the traction energy and accessory energy requirements of different mission units, pre-allocate the energy budget of each mission unit under the constraints of available battery energy and lower charge limit, and form a clear target energy evolution trajectory, while simultaneously combining actual energy deviation and remaining energy during execution to quantitatively assess the risks of subsequent mission completion. This would solve the problems of disconnect between mission structure and energy planning, insufficient support capabilities for key docking points, and untimely exposure of energy shortage risks in existing technologies. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for energy management and range optimization of low-speed four-wheeled electric vehicles, in order to solve the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for energy management and range optimization of a low-speed four-wheeled electric vehicle, comprising: S1. Obtain mission information and historical data, divide into support units, compressible units and sacrificial units, and statistically analyze the traction energy and accessory energy characteristics of each mission unit. S2. Under the constraints of available battery energy and lower charge limit, allocate traction energy budget and accessory energy budget to each task unit according to energy characteristics and priority to form an accumulated energy envelope. S3. Based on the energy envelope, generate a set of control parameters for each task unit, consisting of the upper limit of traction power, the regenerative braking torque coefficient, and the power limit of accessories, and store them into the vehicle controller in sequence. S4. During the execution process, periodically collect the traction energy and accessory energy of each task unit, calculate the actual and budgeted deviations, and combine the remaining energy and the energy budget of subsequent task units to obtain the risk index distribution of subsequent task completion. S5. Based on the energy deviation and the risk index distribution of subsequent task completion, obtain the energy-saving adjustment coefficient, scale the upper limit of traction power, the speed limit and the acceleration limit according to the energy-saving adjustment coefficient, increase the regenerative braking torque coefficient and reduce the accessory power limit; S6. At the end of the task, update the energy characteristic table, risk calculation parameters and energy saving adjustment coefficient mapping table according to the actual energy of each task unit and the task completion status, so that they can be called when constructing the energy envelope and generating the control parameter set.

[0007] Furthermore, S1 includes: The vehicle controller retrieves task information and historical task records, aligns the time sequence of the measurement signals according to the sampling rhythm, and marks records that exceed the physical limit range as invalid measurement points; Based on the order of stops and the status of the doors, historical mission records are divided into mission units. Based on the mission importance and the amount of mission completion, the mission units are further divided into support units, compressible units, and sacrificial units. The traction energy, accessory energy, and corresponding power characteristics of various task units are statistically analyzed and written into the energy characteristic table according to task unit category and task unit order.

[0008] Furthermore, S2 includes: The vehicle controller determines the available energy of the battery based on the battery state of charge, battery nominal capacity, lower charge limit, and efficiency reduction factor. Read the average traction energy and average accessory energy from the energy characteristic table in the order of task units as the initial values ​​of traction energy budget and accessory energy budget, and calculate the initial total energy demand. When the initial total energy demand is greater than the available energy of the battery, the accessory energy compression coefficient and the traction energy compression coefficient are set for the compressible unit and the sacrificial unit. The compression coefficient is reduced according to the compression round. The final energy budget is obtained under the constraint that the traction energy budget and accessory energy budget of each task unit are not lower than their respective safe energy consumption lower bound. The vehicle controller constructs an energy envelope curve based on the final energy budget and stores the energy envelope curve and the task session identifier in the task parameter area.

[0009] Furthermore, S3 includes: The vehicle controller determines the expected travel time of each task unit based on the expected mileage in the task information and the statistical results of the vehicle speed of each task unit in the energy characteristic table. The average traction power boundary is obtained by dividing the traction energy budget of each task unit by the expected driving time. It is then compared with the vehicle's announced power, the motor's long-term allowable power, and the temperature rise constraint power one by one, and the minimum value is taken as the upper limit of the traction power of the task unit. Write the upper limit of traction power, as well as the speed limit and acceleration limit that match the upper limit of traction power, into the control parameter set in the task parameter area.

[0010] Furthermore, the vehicle controller determines the energy redundancy of each task unit based on the energy envelope value range, selects the regenerative braking torque coefficient by inputting the energy redundancy and the slope difficulty level into the mapping table, and corrects the regenerative braking torque coefficient under the constraint of the upper limit of battery charging current. The vehicle controller classifies on-board accessories into safety-related accessories and comfort-related accessories based on mission information. It determines the average total power limit of the accessory system based on accessory energy budget and expected duration, and determines the power limits of safety-related accessories and comfort-related accessories according to the priority principle of safety-related accessories. The upper limit of traction power, regenerative braking torque coefficient, and accessory power limit are associated with the task session identifier and parameter version identifier and then sent to the traction control module and accessory control module.

[0011] Furthermore, S4 includes: The vehicle controller acquires the cumulative traction energy and accessory energy values ​​according to the sampling period; The energy deviation ratio is calculated based on the traction energy budget and the accessory energy budget, and the remaining energy is obtained based on the battery state of charge and the battery available energy and compared with the energy envelope curve. The vehicle controller generates a risk index for subsequent task completion based on the energy deviation ratio, the degree of deviation between the remaining energy and the envelope curve, and the subsequent energy budget according to the risk mapping relationship. When it detects that the traction control module information and accessory control module information have not been updated, a conservative strategy is adopted to reduce the upper limit of traction power and the power limit of accessories.

[0012] Furthermore, S5 includes: At the end of the energy-saving adjustment cycle, the vehicle controller reads the traction energy deviation ratio, the accessory energy deviation ratio, and the maximum task completion risk index of the support unit. The basic energy-saving adjustment coefficient is obtained by looking up the energy-saving adjustment coefficient mapping table, and then the energy-saving adjustment coefficient is corrected according to the changing trend of the maximum task completion risk index.

[0013] Furthermore, the vehicle controller reduces the upper limit of traction power, vehicle speed limit and acceleration limit according to the task unit category and energy saving adjustment coefficient by traction adjustment weight, amplifies the regenerative braking torque coefficient by regenerative adjustment weight and is constrained by the upper limit of battery charging current, and reduces the power limit of comfort-related accessories according to the accessory compression ratio. When recording excessively high risk markers, the upper limit of traction power and the power limit of comfort-related accessories will be reverted to the conservative upper limit of traction power and the conservative power limit of accessories.

[0014] Furthermore, S6 includes: After each task is completed, the vehicle controller updates the energy characteristic table and energy fluctuation range in the historical observation task window based on the task operation record; Adjust the safety factor in the risk calculation rules according to the incomplete task status, and interpolate and smooth the energy-saving adjustment coefficient mapping table according to the trajectory of the energy-saving adjustment coefficient as the input quantity changes; After each update, a rule version identifier and a threshold version identifier are generated. The new and old version identifiers and key task indicators are written to the log area using an append-only write method to achieve version locking and parameter traceability.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. By acquiring task information including stop order, trip attributes, and task importance before the task begins, and combining historical task records, the system automatically divides the task into support units, compressible units, and sacrificial units based on stop points and task completion status. Under the constraints of available battery energy and lower charge limit, the system allocates traction energy budget and accessory energy budget to each task unit based on traction energy characteristics and accessory energy characteristics. It also constructs an energy envelope that monotonically decreases as the task progresses and the energy envelope value range of each task unit. This allows for fine-grained planning of battery energy around the task structure without increasing the overall vehicle hardware cost. It prioritizes ensuring that key stops are completed as planned and exposes the risk of insufficient energy in advance, thereby improving the range stability and task completion power of low-speed four-wheeled electric vehicles in repetitive task scenarios such as park commuting, community transportation, and last-mile delivery.

[0016] 2. By pre-generating control parameter sets for each task unit based on energy envelope, including traction power upper limit, regenerative braking torque coefficient, and power limits for safety-related accessories and comfort-related accessories, the system periodically calculates the traction energy deviation ratio, accessory energy deviation ratio, and subsequent task completion risk index distribution during execution. Energy-saving adjustment coefficients are then obtained by looking up tables. Traction power upper limit and vehicle speed acceleration limit are tightened according to task unit category, regenerative braking torque coefficient is increased, and comfort-related accessory power is reduced. Simultaneously, after the task ends, the mapping relationship between energy characteristics, risk calculation parameters, and energy-saving adjustment coefficients is smoothly updated using historical task observation windows. Version locking and evidence retention are achieved through rule version identifiers and log areas. This ensures consistency and idempotency of energy-saving adjustment behavior even under changing operating conditions and communication or measurement anomalies, improving the adaptability, robustness, and traceability of vehicle energy management, and facilitating continuous optimization of energy-saving strategies and scheduling schemes by the fleet side. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the energy management and range optimization method for a low-speed four-wheeled electric vehicle according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example: Figure 1 A flowchart illustrating a method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to the present invention is provided. The method includes: S1. Obtain mission information and historical data, divide the mission into support units, compressible units, and sacrificial units, and statistically analyze the traction energy and accessory energy characteristics of each mission unit. The specific implementation is as follows: In repetitive scenarios such as commuting in industrial parks, community transportation, and last-mile delivery, before each vehicle starts and confirms entering a complete task, the vehicle controller first obtains the task information for this task through the onboard human-machine interface and remote dispatch terminal. The task information is a description of this trip entered or issued by the driver or dispatcher before the task begins. It includes at least the names of stops in sequence, the estimated mileage between adjacent stops, the cargo or passenger level markings planned to be loaded at each stop, and the importance markings of the corresponding tasks at each stop. The stop names can correspond to park patrol points, delivery stations, or community entrances and exits. The estimated mileage can be set to the planned route length or historical average driving distance. The loading level markings can be set to empty, partially loaded, and fully loaded levels. The task importance markings can be set to mandatory completion, compressible, and optional levels.

[0020] After receiving the task information, the vehicle controller selects several historical task records within a pre-set observation time window from the vehicle's local storage medium before the task begins. The observation time window preferably covers multiple tasks with the same route and the same start and end stops in the near future. Tasks of the same type are preferably limited to tasks with completely identical stop name sequences in the task information. When the stop name sequence changes, it is counted separately as a new task type. The length of the observation time window and the number of historical tasks extracted are preferably set to ensure that the number of historical samples of the same type of task is not less than the preset minimum sample size, such as covering several dozen of the same type of task in the recent weeks, so that the statistical results can reflect the current road conditions and weather conditions.

[0021] Each historical mission record continuously records vehicle speed, acceleration / deceleration, motor terminal voltage, motor phase current, accessory power estimation, ambient temperature, and mission completion status, all collected according to a uniform sampling rhythm. The uniform sampling rhythm is set to a fixed sampling period by the vehicle controller. At the end of each sampling period, all signals are assigned a monotonically increasing time sequence number and the corresponding value is recorded.

[0022] Vehicle speed is the instantaneous driving speed converted from the wheel rotation speed measured by the vehicle speed sensor in each sampling period through the transmission ratio, and recorded in units of length divided by time; acceleration / deceleration is the longitudinal acceleration calculated by subtracting the vehicle speed of the previous sampling period from the current sampling period's vehicle speed and then dividing by the sampling period length, and recorded in units of acceleration; motor terminal voltage is the periodic average of the motor DC bus voltage or equivalent terminal voltage read from the motor control module in each sampling period, and recorded in units of voltage; motor phase current is the periodic average of the effective values ​​of each phase current read from the motor control module in each sampling period, and recorded in units of current; accessory power estimation is calculated by the air conditioning controller, body control module, and multimedia control module in each sampling period based on their respective control parameters. The estimated power values ​​calculated based on the operating status of the air conditioning compressor, heater fan, windshield defroster, and advertising display screen are summed within the sampling period and recorded in power units. The ambient temperature is the temperature value measured by temperature sensors placed outside the vehicle or near the air intake within each sampling period and recorded in temperature units. The task completion status is a task result marker generated by the vehicle controller at the end of each historical task based on whether the vehicle stopped near each stop in the task information and whether the actual dwell time reached the set lower limit. At the end of the task, whether each stop was reached and whether the task was executed as planned are written into the recording area in the form of Boolean markers or multi-level markers. The lower limit of dwell time can be set to a time length greater than zero and sufficient to complete the loading / unloading or passenger loading / unloading operations.

[0023] After acquiring historical task records within the observation time window, the vehicle controller preferably aligns the time tags of each channel according to a unified sampling rhythm. Specifically, when reading historical records, the vehicle speed, acceleration / deceleration, motor terminal voltage, motor phase current, accessory power estimation, and ambient temperature records for each sampling period are arranged on the same time axis according to time sequence. When it is found that the time interval between two adjacent time sequences of a certain channel is greater than twice the length of the sampling period, that interval is identified as a recording gap and recorded as a possible communication interruption or measurement loss. Subsequently, each channel is checked one by one to see if the measured values ​​exceed the physical limits of the vehicle. The physical limits are preferably based on the maximum design speed, maximum allowable acceleration, motor rated voltage, motor rated current, and local historical highest and lowest temperatures given in the vehicle technical manual. These rated values ​​and historical extreme values ​​are extended upwards or downwards with a fixed margin. The margin range for motor current and voltage can be set as a certain multiple range near the rated values. The multiple range refers to the upper and lower limit range formed by multiplying the rated values ​​by a coefficient not less than one and a coefficient not greater than one.

[0024] For measurement points where the vehicle speed exceeds the maximum design speed margin or is less than zero, the acceleration / deceleration exceeds the maximum allowable acceleration margin or is less than the negative margin of that acceleration, the motor terminal voltage exceeds the motor rated voltage margin, the motor phase current exceeds the motor rated current margin, the accessory power estimate exceeds the upper limit margin of the total power of the vehicle accessories, or the ambient temperature is significantly higher than the margin of the local historical highest temperature or significantly lower than the margin of the local historical lowest temperature, the vehicle controller will mark them as invalid measurement points on the corresponding time sequence number and treat them as missing values ​​in subsequent statistics.

[0025] After aligning time tags and marking invalid measurement points, the vehicle controller fills in the missing short-time intervals marked in each historical task record using linear interpolation of adjacent valid measurement points. Specifically, within the missing interval, a straight line that changes linearly with time is constructed based on the adjacent valid measurement values ​​at the start and end of the missing interval, and the value corresponding to the straight line is taken as the fill value at the time sequence number of each sampling period within the interval. When the number of consecutive missing sampling points in a certain missing interval exceeds a preset upper limit, no interpolation is performed, and the interval is removed from subsequent energy statistics. This upper limit can preferably be set to several sampling periods, such as no less than several dozen sampling periods, to avoid the filling of long-term missing points affecting the accuracy of the statistical results.

[0026] After completing the unified rhythm alignment and missing completion, the vehicle controller divides each historical task record into several task units according to the stopping point sequence given in the task information. A task unit is a continuous driving segment with the boundary of the vehicle leaving a certain stopping point and stopping at the next stopping point. When dividing, the vehicle controller determines a parking event by detecting that the vehicle speed is lower than a set low speed threshold for several consecutive sampling periods and the door opening and closing status occurs during the time period. The low speed threshold can preferably be set to a vehicle uniform crawling speed slightly higher than zero, such as close to a certain number of kilometers per hour, to filter out the errors of vehicle start-up shaking and extremely low-speed vehicle maneuvering. The door opening and closing status is determined by the door lock switch signal provided by the body control module in each sampling period. When the door lock signal is detected to undergo a complete change process from closing to opening and then back to closing within a continuous sampling period, the time period is regarded as a passenger pick-up or drop-off or loading and unloading process.

[0027] During the time period for determining a parking event, the vehicle controller records the vehicle's parking position and matches it with the parking point position in the task information. A successful match is considered achieved when the distance between the parking position and the target parking point position is less than the allowable range of position error. The allowable range of position error can preferably be set according to the positioning accuracy of the on-board positioning module. For example, when using satellite positioning, the error radius can be set to a range of several meters. When the distance between the vehicle's parking position and the parking point position in the task information is less than this error radius, it is determined that a parking event has been completed at that parking point. In each historical task, the vehicle controller categorizes the sampling sequence from leaving the previous parking point event to arriving at the next parking point event into a task unit. Multiple task units are obtained sequentially in the same historical task, and the task units are arranged in the order of parking points in multiple historical tasks, so that task units in the same sequential position in different historical tasks can be compared laterally.

[0028] After obtaining a large number of task units, the vehicle controller categorizes each task unit based on the task importance markers and task completion status of each stop in the task information. When a stop is marked as mandatory in the task information and the historical task completion status within the observation time window shows that the proportion of planned task completion for that stop is not lower than a preset high percentage threshold (e.g., not lower than 80%) and the number of historical tasks is not lower than a preset minimum sample size, the corresponding task unit is classified as a guarantee unit. When a stop is marked as compressible or its completion method is adjustable in the task information, and the historical record shows that the proportion of occasional shortened stop times but complete skipping under energy constraints is not higher than a preset threshold... When a low percentage threshold is set, for example, no more than 20%, the corresponding task unit is classified as a compressible unit. When a stop is marked as abandonable in the task information and the percentage of stops skipped in the history under energy shortage is not less than a medium percentage threshold, for example, not less than 50%, the corresponding task unit is classified as a sacrificial unit. When the number of historical samples of a task unit is less than the preset minimum number of samples, the vehicle controller can temporarily treat the task unit as a guaranteed unit until the number of historical samples of the task unit meets the minimum number of samples requirement, and then reclassify it according to the above percentage threshold rules. This ensures that the classification results are stable and reliable and that the judgment logic has only one execution mode as understood by those skilled in the art.

[0029] After classification, the vehicle controller statistically analyzes traction energy and accessory energy characteristics within each task unit. Specifically, for each sampling period within each task unit, the traction power for that sampling period is calculated as the product of the motor terminal voltage and the motor phase current, multiplied by a symbolic value representing the operating state. When the vehicle speed is higher than a set low-speed threshold and the motor is in traction mode, the traction power is recorded as a positive value. When the vehicle decelerates and the motor is in generator mode, the recovered power is recorded separately and not included in the traction energy characteristics. Within that task unit, the traction power for all sampling periods is multiplied by the sampling period length and accumulated to obtain the traction energy for that task unit. The maximum and minimum values ​​of the traction power are recorded during the accumulation process. The accessory energy is calculated by directly multiplying the accessory power estimate by the sampling period length within each sampling period and then calculating it within the task unit. The accessory power estimates for all sampling periods are multiplied by the sampling period length and summed one by one to obtain the accessory energy of the task unit. At the same time, the average and maximum values ​​of the accessory power estimates within the task unit are recorded. Within the observation time window, for task units belonging to the same category and located in the same order, the vehicle controller calculates the average, maximum, and minimum values ​​of the traction energy. Preferably, the interval between the maximum and minimum values ​​is recorded as the traction energy fluctuation range. Similarly, the accessory energy fluctuation range is obtained. The maximum value of the traction power and the maximum value of the accessory power estimates are statistically analyzed to obtain the typical peak value of the traction power and the typical peak value of the accessory power. The above average value, fluctuation range, and typical peak value together constitute the traction energy characteristics and accessory energy characteristics of the task unit of this category and position. Each characteristic is recorded with a clear physical quantity name, unit, and typical value range.

[0030] Finally, the vehicle controller writes the traction energy characteristics and accessory energy characteristics of each category and each sequential position task unit into the energy characteristic table according to the order of the task units in the task information. The energy characteristic table is stored in the non-volatile storage area of ​​the vehicle controller in the form of structured records. Each record contains at least the task unit number, task unit category, average traction energy, traction energy fluctuation range, typical peak traction power, average accessory energy, accessory energy fluctuation range, typical peak accessory power, number of historical tasks involved in the statistics, and observation time window identifier. The energy characteristic table is directly called by the vehicle controller in the subsequent energy budget allocation and energy envelope construction steps of the task.

[0031] In a preferred embodiment, for example in an industrial park with a fixed patrol route, the sampling period can be set to once per second, the observation time window can be set to several dozen tasks completed on the same patrol route within the last few weeks, the reasonable range for vehicle speed can be set to zero to several tens of kilometers per hour, the reasonable range for acceleration / deceleration can be set to plus or minus several meters per square second, the reasonable range for motor terminal voltage can be set to a certain percentage above and below the nominal battery voltage, the reasonable range for motor phase current can be set to a range of several times above and below the rated motor current, the reasonable range for accessory power estimation can be set to a percentage range near the nominal total power of the vehicle accessories, and the reasonable range for ambient temperature can be set to the highest and lowest locally recorded temperatures. With a margin of several degrees added above and below, this configuration allows for relatively stable traction energy characteristics and accessory energy characteristics to be obtained after dozens of patrol missions, which can be used for energy planning in subsequent missions. In another embodiment, mission information can be obtained without an external scheduling platform. Instead, the vehicle controller can automatically identify recurring specific parking locations and connecting paths in a certain number of historical trajectories, summarize these locations and paths into standard mission information, and then generate an energy feature table according to the steps of unified sampling rhythm, time alignment, missing data completion, mission unit division, mission unit classification, and energy feature statistics. As long as the fields and construction rules of the energy feature table remain consistent, it can be regarded as an equivalent replacement for the above-described implementation method.

[0032] S2. Under the constraints of available battery energy and lower charge limit, allocate traction energy budget and accessory energy budget to each task unit according to energy characteristics and priority to form a cumulative energy envelope. The specific implementation is as follows: After completing the energy characteristic table statistics, the vehicle controller calculates the available battery energy for this task before the start of the task based on the current battery state of charge, the battery nominal capacity, and the lower limit of charge provided by the battery supplier. At the end of each sampling period, the battery management module calculates a dimensionless number between zero and one by using the average values ​​of signals such as battery terminal voltage, individual cell voltages in the battery pack, battery charging and discharging current, and temperature during that sampling period, and reports it to the vehicle controller. When reading the battery state of charge for the current sampling period, the vehicle controller compares it with the lower limit of charge provided by the battery supplier in the technical manual. The lower limit of charge can be set to be no less than the state of charge value corresponding to a fixed percentage of the battery nominal capacity to ensure that the battery still retains the necessary safety margin after the task ends. The nominal battery capacity is written into the configuration area of ​​the vehicle controller by the manufacturer in terms of electricity volume during the vehicle calibration phase.

[0033] After obtaining the current battery state of charge (SOC), nominal battery capacity, and lower charge limit, the vehicle controller preferably determines the available battery energy as follows: First, calculate the difference between the current SOC and the lower charge limit. Multiply this difference by the nominal battery capacity to obtain the theoretically releaseable charge under the current temperature conditions. Then, combine this with the efficiency reduction factor statistically obtained by the battery management module in recent tasks to correct the theoretical charge and form the available battery energy, which is recorded in the vehicle controller in energy units. The efficiency reduction factor can be determined by calculating the sum of traction energy and accessory energy in the current task at the end of each historical task, and then using the ratio of the difference between the battery SOC at the beginning and end of the current task multiplied by the nominal battery capacity. This ratio is used as the single efficiency coefficient for that historical task. Within the observation time window, the arithmetic mean or moving average of all single efficiency coefficients is taken, and the average result is used as the efficiency reduction factor for the current task, so as to gradually approximate the actual energy loss in subsequent tasks.

[0034] Once the available battery energy is determined, the vehicle controller sequentially calls the traction energy characteristics and accessory energy characteristics corresponding to the task unit number and category in the energy characteristic table, according to the task unit's order in this task. The traction energy characteristics include the average traction energy and traction energy fluctuation range of this type of task unit within the historical observation time window. The accessory energy characteristics include the average accessory energy and accessory energy fluctuation range. When reading the traction energy characteristics and accessory energy characteristics, the vehicle controller regards the average traction energy as the baseline traction energy required for the task unit to complete the task under standard operating conditions, the average accessory energy as the baseline accessory energy required for the task unit to complete the task under standard operating conditions, and the traction energy fluctuation range and accessory energy fluctuation range as the reasonable upper and lower limits of the energy demand variation of the task unit under different weather, load, and driving habits.

[0035] The vehicle controller then assigns priority weights to each task unit according to the task importance given in the task information. These priority weights can be set to three discrete levels, corresponding to support units, compressible units, and sacrificial units, with support units having the highest priority weight, compressible units having a medium priority weight, and sacrificial units having a low priority weight. In the initial allocation phase, the vehicle controller sets the initial traction energy budget for each task unit to the average traction energy of that task unit, and the initial accessory energy budget to the average accessory energy of that task unit. The initial sum of the initial traction energy budgets and accessory energy budgets for all task units is accumulated sequentially to obtain the initial total energy demand. The vehicle controller compares this initial total energy demand with the available battery energy. If the initial total energy demand is not greater than the available battery energy, the vehicle controller directly uses this initial allocation scheme as the traction energy budget and accessory energy budget for this task. If the initial total energy demand is greater than the available battery energy, the vehicle controller initiates the energy compression process.

[0036] During each round of compression, the energy compression process must ensure that the traction energy budget and accessory energy budget of each task unit are not lower than the lower limit of the safe energy consumption corresponding to their respective energy fluctuation range. During the energy compression process, the vehicle controller configures the accessory energy compression coefficient and traction energy compression coefficient for each compressible unit and sacrificial unit in memory. The value range of the accessory energy compression coefficient and traction energy compression coefficient is limited to between zero and one. At the initial moment, the accessory energy compression coefficient of the compressible unit can be set to a value slightly less than one, the accessory energy compression coefficient of the sacrificial unit can be set to a smaller value, and the traction energy compression coefficient of all task units can be initially set to one.

[0037] In the first round of compression, the vehicle controller maintains the traction energy compression coefficient of all task units at one, thereby ensuring that the traction energy budget of the unit, compressible unit, and sacrificial unit is equal to the average traction energy. Only the accessory energy budget of the compressible unit and the sacrificial unit is compressed, that is, the initial value of the accessory energy budget of these task units is multiplied by the corresponding accessory energy compression coefficient to obtain a new accessory energy budget. The sum of the traction energy budget and accessory energy budget of all task units is then accumulated in the order of task units to obtain the total energy demand after the first round of compression. This is compared with the available energy of the battery. When the total energy demand after the first round of compression is still greater than the available energy of the battery, the vehicle controller enters the second round of compression.

[0038] In the second round of compression, the vehicle controller reduces the accessory energy compression coefficient of the compressible unit and the sacrificial unit according to a preset step size. The step size can be set to a fixed compression step size, and the compressible unit and the sacrificial unit can use different compression step sizes. For example, the accessory energy compression step size of the sacrificial unit can be set to be larger than that of the compressible unit, so that the accessory energy budget of the sacrificial unit decreases faster in the same round. At the same time, in the second round and subsequent rounds, the vehicle controller begins to moderately compress the traction energy budget of the sacrificial unit, that is, multiply the initial value of the traction energy budget of the sacrificial unit by the traction energy compression coefficient to obtain a new traction energy budget, and reduces the traction energy compression coefficient by a preset step size in each round, while keeping the traction energy compression coefficient of the guarantee unit at one. Only when the accessory energy budget has been compressed to near the lower limit of the accessory energy fluctuation range and still cannot meet the total energy constraint, is the traction energy compression coefficient of the compressible unit slightly reduced, thereby generating the traction energy budget and accessory energy budget of each task unit after the second round of compression.

[0039] After the second compression cycle, the vehicle controller sums the traction energy budgets and accessory energy budgets of all task units to obtain a new total energy demand. This new total energy demand is compared with the available battery energy. If the total energy demand is still greater than the available battery energy, the next compression cycle begins with the same compression step size, repeating the cycle of "updating the compression coefficient, recalculating the budget, and comparing with the available battery energy." When, after a compression cycle, the traction energy budgets of all task units are not lower than the lower limit of the corresponding average traction energy minus the traction energy fluctuation range, the accessory energy budgets of all task units are not lower than the lower limit of the corresponding average accessory energy minus the accessory energy fluctuation range, and the total energy demand is less than or equal to the available battery energy, the vehicle controller uses the traction energy budget and accessory energy budget obtained from that compression cycle as the final allocation result. If, during a compression cycle, it is found that the traction energy budgets of some sacrificial units have already been reduced... If, after compression to near the average traction energy minus the lower limit of the traction energy fluctuation range, the total energy demand still exceeds the available battery energy after further compression of the auxiliary energy budgets of these sacrificial units, the vehicle controller marks these sacrificial units as candidate cancellation units. During subsequent operation or task assignment phases, the controller will issue a prompt to the driver or dispatcher through the human-machine interface or scheduling system, suggesting that these candidate cancellation units be skipped during task execution to reduce the overall energy demand. If, after a round of compression, it is determined that to ensure the total energy demand does not exceed the available battery energy, the traction energy budget of certain backup units must be compressed to below the average traction energy minus the lower limit of the traction energy fluctuation range of that backup unit, the vehicle controller will generate an alarm message indicating that the task cannot be completed as planned under the current available battery energy conditions. This alarm message will be reported to the upper-level dispatch platform via the vehicle communication network so that the task can be manually adjusted or mid-journey charging can be arranged.

[0040] After the traction energy budget and accessory energy budget are allocated, the vehicle controller accumulates the traction energy budget and accessory energy budget for each task unit according to the order of the task units in this task. The available battery energy is subtracted from the current accumulated value to obtain the expected remaining energy value at the end of each task unit. These expected remaining energy values ​​are connected in the order of the task units to form a curve that monotonically decreases as the task progresses. This curve is recorded in the vehicle controller as the energy envelope curve. The energy envelope curve shows the expected change trajectory of the remaining battery energy when each task unit consumes energy according to the budget in the task sequence dimension.

[0041] When recording the energy envelope curve, the vehicle controller simultaneously determines the energy envelope value range for each task unit based on the traction energy fluctuation range and accessory energy fluctuation range of each task unit in the energy characteristic table. Specifically, based on the expected remaining energy value at the end of the task unit, the traction energy fluctuation range and accessory energy fluctuation range are summed to obtain a total fluctuation. This total fluctuation is multiplied by a preset amplification factor as a redundancy boundary. The preset amplification factor can be set to a constant greater than zero and less than one. Then, the redundancy boundary is added to the expected remaining energy value to obtain the upper bound of the energy envelope value range of the task unit. The lower bound of the energy envelope value range for each task unit is obtained by subtracting the expected remaining energy value, thus forming the upper and lower allowable ranges of the battery's remaining energy at the end of each task unit. In addition, the vehicle controller can calculate the budget redundancy for the entire task. The budget redundancy is a portion of the energy difference reserved between the battery's available energy and the sum of the traction energy budget and accessory energy budget of all task units. It is used to absorb the additional consumption caused by environmental changes and differences in driving behavior during operation. The budget redundancy can be set as a small proportion of the battery's available energy and is reflected in the energy envelope curve as the portion of the expected remaining energy that is greater than zero at the end of the last task unit.

[0042] Inside the vehicle controller, the energy envelope curve, the energy envelope value range of each task unit, and the budget redundancy are stored in a structured form along with the task session identifier in the task parameter area. During subsequent operation, the vehicle controller will calculate the actual traction energy and accessory energy in the current task unit based on the cumulative results of the actual traction power and accessory power estimates in each sampling cycle, and compare them with the corresponding traction energy budget and accessory energy budget for that task unit. At the same time, the actual remaining energy at the current progress of the task is calculated by converting the current battery state of charge obtained from the battery management module. The actual remaining energy is compared with the expected remaining energy at the corresponding position of the energy envelope curve and the upper and lower bounds of the energy envelope value range at that position to determine whether the current task execution is within or deviates from the budget range, providing a basis for subsequent energy-saving adjustments and task completion risk assessment.

[0043] In a preferred embodiment, for example, in scenarios where low-speed four-wheeled electric vehicles are used for patrolling industrial parks and short-haul transportation within factories, the sampling period for the battery state of charge (SCC) can be set to once per second. The lower limit of SCC is set to the SCC corresponding to a certain percentage of the battery's nominal capacity. The budget redundancy is set to a certain percentage of the battery's available energy. The compression step size for accessory energy and traction energy is set to a fixed value not greater than one. When performing a patrol mission comprising several dozen task units, the vehicle controller calls and allocates the energy characteristic table in the above manner. After several rounds of compression, an energy envelope curve is obtained where the expected remaining energy at the endpoint is significantly greater than zero. In subsequent patrol missions, it was found that after adopting this energy envelope allocation strategy, the corresponding stopping points of the guarantee unit were completed as planned in the vast majority of missions, and the actual remaining energy at the end of the mission fell within the energy envelope value range, only approaching the energy envelope value under extreme temperature conditions and abnormal driving behaviors such as prolonged rapid acceleration by the driver. The lower bound of the energy envelope and the budget redundancy boundary demonstrate that the above allocation and compression strategies are feasible and stable under typical operating conditions. In another embodiment, for delivery scenarios with significant changes in task structure, the initial values ​​of the traction energy budget and the accessory energy budget can be set as the weighted result of the average traction energy value and the average accessory energy value and the corresponding upper bound of the energy fluctuation range. The redundancy boundary amplification factor can be appropriately increased to improve adaptability to sudden high-load road sections. The upper and lower bounds of the energy fluctuation range are preferably based on the historical minimum and maximum energy values, or on this basis, a preset percentage margin can be added or subtracted. As long as the calculation order of the battery available energy, the priority order of the energy compression process, and the construction rules of the energy envelope curve and the energy envelope value range remain unchanged, the same calculation process can still be reproduced and similar technical effects can be achieved by selecting specific compression step size and ratio parameters according to the above disclosure in different scenarios. This is an equivalent implementation of the method.

[0044] S3. Based on the energy envelope, generate a set of control parameters for each task unit, consisting of the upper limit of traction power, the regenerative braking torque coefficient, and the power limit of accessories, and store them into the vehicle controller in sequence. Specifically, this is implemented as follows: After completing the calculation of available battery energy, the allocation of traction energy budget and accessory energy budget for each task unit, and the generation of energy envelope curves, the vehicle controller generates a set of control parameters for each task unit based on the above results before the start of this task. The set of control parameters includes at least the upper limit of traction power, the regenerative braking torque coefficient, and the accessory power limit, and is stored in the parameter area corresponding to this task in the order of the task units.

[0045] Specifically, the vehicle controller first determines the expected travel time of the task unit based on the estimated mileage in the task information and the statistical results of the vehicle speed within the task unit in the historical task records. The expected travel time can be calculated synchronously during the energy characteristic table statistics. That is, the vehicle speed of all sampling periods in the task units of the same location and category in the historical tasks is averaged, and the ratio of the average vehicle speed to the estimated mileage of the task unit in the task information is calculated to obtain the historical average travel time. Then, the average of multiple historical average travel times is taken within the observation time window to obtain the expected travel time of the task unit. In this task, the vehicle controller directly calls the expected travel time when constructing the control parameter set.

[0046] For each task unit, the vehicle controller divides the traction energy budget by the expected travel time to obtain the average traction power boundary required for the task unit to complete the task within the expected time scale. This average traction power boundary is then compared one by one with the thermal performance constraints comprised of the vehicle's advertised power, the motor's long-term allowable power, and the motor's temperature rise limit. The vehicle's advertised power is the maximum continuous output power given in the vehicle's advertised parameters. The long-term allowable power is calculated based on the motor's thermal balance during the vehicle design phase. The motor's temperature rise limit can be obtained by statistically analyzing the relationship between motor winding temperature and motor output power in historical tasks to obtain the upper limit power curve. During actual calculations, the vehicle controller stores a set of safe power upper limit tables corresponding to motor temperature segments in memory. It selects the appropriate temperature rise constraint by comparing the current ambient temperature and the typical traction power peak of the task unit in the energy characteristic table with the safe power upper limit table. The power value is then used as the minimum value among the average traction power boundary, the vehicle's announced power, the long-term allowable power, and the temperature rise constraint power as the upper limit of the traction power for this task unit. During subsequent operation, the traction power upper limit is used by the traction control module as the target upper bound of the motor controller. In each sampling period, the actual traction power is constrained below this upper limit by limiting the motor current command. At the same time, based on the determined traction power upper limit, the vehicle controller can combine the vehicle's announced maximum speed, acceleration and deceleration performance indicators, and the expected travel time of the task unit to generate speed limits and acceleration limits that match the traction power upper limit for the task unit. The speed limit is preferably not higher than the vehicle's announced maximum speed and not lower than the minimum speed required to ensure on-time arrival of the task. The acceleration limit is preferably not higher than the maximum allowable acceleration and deceleration of the vehicle to ensure that the longitudinal dynamics of the vehicle still meet the task requirements under the condition of limited traction power.

[0047] When determining the regenerative braking torque coefficient, the vehicle controller first calculates the energy redundancy of several subsequent task units based on the energy envelope curve and energy envelope value range obtained in step two. Before the start of the current task, the vehicle controller calculates an energy redundancy for each task unit in the task parameter region. The energy redundancy can be defined as the difference between the upper and lower bounds of the energy envelope value range at the end of the task unit, or as the difference between the lower bound of the energy envelope value range at the end of the task unit and zero energy. The vehicle controller adopts the former definition when generating the regenerative braking torque coefficient, that is, when constructing the control parameter set, it reads the energy redundancy of the current task unit and several subsequent task units, and sums or performs a weighted summation of these energy redundancies to obtain a... An energy redundancy index reflects the overall energy margin of subsequent tasks. The energy redundancy index is a dimensionless or energy-dimension comprehensive index that characterizes the overall energy margin of subsequent tasks. It can be a weighted sum or equivalent conversion value of energy redundancy. At the same time, the slope difficulty level is estimated based on the terrain information or the statistical results of longitudinal acceleration and deceleration within the task unit in the historical task records. The slope difficulty level can be calculated by reading the average slope on the path of the task unit from the map data, or by analyzing the distribution of vehicle speed decrease and traction power within the task unit in the historical tasks to deduce the three typical working conditions of uphill, flat road and downhill. Before the start of this task, the vehicle controller preferably uses a combination of map data and historical statistics to write the slope difficulty level in the parameter record of each task unit.

[0048] When calculating the regenerative braking torque coefficient, the vehicle controller uses energy redundancy index and gradient difficulty level as two independent variables. On the one hand, when the energy redundancy index is low, meaning the energy margin for subsequent tasks is tight, it is necessary to maximize the energy recovery ratio during deceleration. On the other hand, when the gradient difficulty level is downhill or long downhill, a larger proportion of electric motor braking is allowed to participate in deceleration to reduce the burden on mechanical braking. The vehicle controller retrieves a basic regenerative braking torque coefficient from a mapping table with energy redundancy index and gradient difficulty level as inputs, according to a pre-calibrated two-dimensional mapping relationship. This basic coefficient is then compared with the torque boundary corresponding to the upper limit of the battery charging current. The upper limit of the charging current can be provided by the battery management module through a lookup table under the current temperature and current state of charge. The vehicle controller compares the regenerative torque corresponding to the basic regenerative braking torque coefficient with the torque boundary. When the charging current caused by the torque corresponding to the basic coefficient does not exceed the upper limit of the charging current given by the battery management module, the basic regenerative braking torque coefficient is directly used as the regenerative braking torque coefficient of the task unit. When the torque corresponding to the basic coefficient may cause the charging current to exceed the upper limit, the vehicle controller reduces the regenerative braking torque coefficient proportionally to achieve a balance between the regenerative braking force generated based on the coefficient and the maximum deceleration requirement of the vehicle and the safe range of battery charging.

[0049] In determining the power limits of accessories, the vehicle controller first divides the vehicle accessories into two main categories: safety-related accessories and comfort-related accessories, based on the task attributes of each stop point in the task information and the distribution of accessory power estimates within each task unit in the energy characteristic table. Safety-related accessories include front and rear lights, brake lights, turn signals, windshield defroster components, and anti-fog devices, while comfort-related accessories include air conditioning compressors, heater fans, and in-vehicle advertising displays. In the classification process, the vehicle controller uses the function that affects driving safety as the standard. That is, if turning off an accessory in rainy or foggy weather or at night would significantly reduce the driver's ability to perceive the road conditions ahead or reduce the ability of the vehicle to be perceived by surrounding road users, then that accessory is classified as a safety-related accessory.

[0050] After obtaining the accessory energy budget and expected duration of the task unit, the vehicle controller divides the accessory energy budget by the expected duration to obtain the upper bound of the average total power that the accessory system of the task unit can consume. Then, according to the principle of prioritizing safety-related accessories and compressing comfort-related accessories, the upper bound of the average total power is decomposed into power limits for safety-related accessories and power limits for comfort-related accessories. Preferably, the power limit for safety-related accessories is not lower than the average value of the power estimates of safety-related accessories obtained from the energy characteristic table of the task unit. The power limit for comfort-related accessories is adjusted according to the remaining power space and energy redundancy index. When the energy redundancy index is high, the power limit for comfort-related accessories can be appropriately increased. When the energy redundancy index is low, the power limit for comfort-related accessories is reduced, or even the comfort-related accessories are limited to a very low level in the sacrificial unit.

[0051] The vehicle controller constructs a control parameter record for each task unit in the task parameter area. This record includes at least the task unit number, traction power limit, vehicle speed limit, acceleration limit, regenerative braking torque coefficient, power limit of safety-related accessories, power limit of comfort-related accessories, and the corresponding task session identifier and parameter version identifier. The task session identifier is generated by the vehicle controller according to the current date and time and task number each time a new task is confirmed, and is used to uniquely identify a complete task. The parameter version identifier is incremented once each time a complete set of control parameters is constructed, and is used to distinguish different construction rounds.

[0052] Before the mission begins, the vehicle controller sends the control parameter set to the traction control module and accessory control module via the vehicle's internal communication network according to the mission unit sequence number. During the sending, the message log simultaneously carries the mission session identifier, mission unit sequence number, and parameter version identifier. Upon receiving the message carrying the mission session identifier and parameter version identifier, the traction control module first compares the parameter version identifier in the message with its own currently effective parameter version identifier. If the parameter version identifier in the message is higher than the currently effective version, the traction power limit and regenerative braking torque coefficient from the message are loaded into the internal control parameter buffer, and the currently effective version is updated to the parameter version identifier in the message. If the parameter version identifier in the message is lower than the currently effective version, the traction control module records a parameter version mismatch error code in its internal status log and discards the parameter update in that message, thereby avoiding... The old version of control parameters is not overridden by the new version. The attachment control module adopts the same version comparison strategy for the reception and activation of attachment power limits. That is, each time an attachment power limit message carrying a task session identifier and parameter version identifier is received, the parameter version identifier in the message is compared with the current effective version. When the version identifier in the message is behind the current version, the message is recorded as a version lag event and the existing attachment power limit remains unchanged. The internal attachment power limit is only updated when the version identifier in the message is equal to or higher than the current version. This ensures that the traction control module and the attachment control module always work together around the same task session identifier and parameter version identifier throughout the entire task execution process, reducing the confusion of control parameters under network transmission delay and repeated issuance. This ensures that this step is still stable and feasible even under complex field conditions and occasional packet loss in communication.

[0053] S4. During execution, periodically collect the traction energy and auxiliary energy of each task unit, calculate the actual and budgeted deviations, and combine the remaining energy and the energy budget of subsequent task units to obtain the risk index distribution for the completion of subsequent tasks. The specific implementation is as follows: After the vehicle enters the actual task execution phase, the vehicle controller continuously runs an online monitoring and risk assessment logic according to the same rhythm as the aforementioned sampling cycle. At the end of each sampling cycle, it collects the cumulative traction energy and accessory energy values ​​within the current task unit from the traction control module and accessory control module. The cumulative traction energy value is obtained by resetting the accumulated amount to zero at the beginning of the task unit, multiplying the motor terminal voltage and motor phase current in each sampling cycle to obtain the traction power of the current cycle, multiplying the traction power by the sampling cycle length to obtain the traction energy increment of the current cycle, and then summing the traction energy increments of all completed sampling cycles within the task unit to obtain the current cumulative traction energy value. The cumulative accessory energy value is obtained by resetting the cumulative accessory energy value to zero at the beginning of the task unit, reading the accessory power estimate reported by the accessory control module in each sampling cycle, multiplying the accessory power estimate by the sampling cycle length to obtain the accessory energy increment of the current cycle, and then summing the accessory energy increments in each cycle within the task unit to obtain the current cumulative accessory energy value.

[0054] At the end of each sampling cycle, the vehicle controller compares the accumulated traction energy consumed by the current task unit with the traction energy budget determined in step two. The remaining traction energy is obtained by subtracting the accumulated traction energy from the budget. The remaining traction energy is then divided by the budget to obtain a dimensionless traction energy deviation ratio. When the traction energy deviation ratio is positive and has a large absolute value, it indicates that the traction energy consumption is significantly lower than the budget. When the traction energy deviation ratio is negative and has a large absolute value, it indicates that the traction energy consumption significantly exceeds the budget. At the same time, the accumulated accessory energy is compared with the accessory energy budget corresponding to the task unit. The remaining accessory energy is obtained by subtracting the accumulated accessory energy from the budget. The accessory energy deviation ratio is calculated in the same way. The traction energy deviation ratio and accessory energy deviation ratio are stored in the vehicle controller to characterize the degree of deviation of the current task unit's energy usage from the budget.

[0055] At the end of each sampling period, the vehicle controller also reads the battery state of charge (SOC) at the end of the current sampling period from the battery management module. It converts the SOC into the current actual remaining energy in a way consistent with the calculation method of the battery's available energy in step two. That is, it multiplies the difference between the current SOC and the lower limit of charge by the battery's nominal capacity and then by the efficiency reduction factor to obtain the remaining energy that the battery can still use for traction and accessories at this moment under the condition that it is not lower than the lower limit of charge. The vehicle controller, in conjunction with the current task unit number and the progress of this task, finds the expected remaining energy value at the corresponding position on the energy envelope curve, and subtracts the expected remaining energy from the actual remaining energy to obtain the remaining energy deviation along the task sequence. Then, it divides the remaining energy deviation by the difference between the upper and lower limits of the energy envelope value range to obtain the remaining energy deviation degree. When the remaining energy deviation degree is close to zero, it indicates that the actual remaining energy is highly consistent with the energy envelope curve. When the deviation degree is negative and the absolute value is close to one, it indicates that the actual remaining energy is close to the lower limit of the energy envelope or even close to the budget redundancy boundary.

[0056] After obtaining the traction energy deviation ratio, accessory energy deviation ratio, and remaining energy deviation within the current task unit, the vehicle controller, in conjunction with the traction energy budget, accessory energy budget, and the traction energy fluctuation range and accessory energy fluctuation range recorded in the energy characteristic table for all subsequent task units, calculates the task completion risk index for each subsequent task unit according to the pre-set risk calculation rules. The task completion risk index is preferably represented by a real number between zero and one. A value near zero indicates that the subsequent task unit can basically complete the task according to the budget under the current remaining energy conditions, while a value near one indicates that the task unit is likely unable to simultaneously meet the traction energy budget and the minimum accessory requirements.

[0057] The risk calculation rule can be set as follows: First, in the vehicle controller, simulate the remaining energy changes from the current task unit to the end of each subsequent task unit in the order of task units. Take the current actual remaining energy as the starting point, subtract the sum of the traction energy budget and accessory energy budget required from the unexecuted part of the current task unit to the end of the subsequent task unit, and subtract a certain safety factor corresponding to the energy fluctuation range to obtain the expected remaining energy to reach the end of the subsequent task unit under the pessimistic assumption. Then, compare the expected remaining energy with zero energy and the minimum accessory energy requirement acceptable to the task unit in the energy characteristic table. The minimum accessory energy requirement is preferably defined as the lower limit of accessory energy consumption in the task unit when only safety-related accessories are guaranteed to work continuously and comfort-related accessories are kept in the minimum working mode. It can be calculated based on the average power estimate of safety-related accessories in the energy characteristic table and the expected duration. When the expected remaining energy is significantly higher than the traction of the task unit, the risk calculation rule is applied. When the sum of the energy budget and the minimum accessory requirement is reached, the task completion risk index for this task unit is set to a small value close to zero. When the expected remaining energy is close to the sum of the traction energy budget and the minimum accessory requirement, the task completion risk index is set to a medium value. When the expected remaining energy is lower than the sum of the traction energy budget and the minimum accessory requirement, the task completion risk index is set to a large value close to one. Specifically, this can be achieved by storing a one-dimensional mapping table in the vehicle controller, with the difference between the expected remaining energy and the total task requirement as input and the task completion risk index as output. During each risk calculation, this difference is used as the independent variable to look up the task completion risk index in the table, thus making the calculation rule of the risk index clear and stable in implementation and avoiding ambiguity in interpretation. The one-dimensional mapping table can be implemented by dividing the difference between the expected remaining energy and the total task requirement into several intervals and presetting the corresponding task completion risk index for each interval. Preferably, a piecewise linear interpolation method is used to generate the risk index of the intermediate point.

[0058] The vehicle controller organizes the task completion risk index of all subsequent task units into a subsequent task completion risk index distribution according to the task unit order, and writes it into the vehicle controller's operating status area at the end of each sampling period along with the current task session identifier and parameter version identifier. The operating status area retains the risk index distribution records of the most recent sampling periods so that they can be reviewed and analyzed after the task is completed, and provide a basis for subsequent energy-saving adjustment strategies and parameter version adjustments.

[0059] In actual operation, when the vehicle controller detects that power or energy acquisition information from the traction control module or accessory control module has not been updated for several consecutive sampling cycles, or when signals such as motor terminal voltage, motor phase current, accessory power estimation, and ambient temperature briefly exceed the aforementioned physically reasonable range, this time range is marked as a period of unexpected communication loss or short-term deviation of sensor readings from the physically reasonable range. The threshold for the number of consecutive sampling cycles without updates can be set to no more than a certain number of sampling cycles. During the marked time period, the vehicle controller stops including the traction energy increment and accessory energy increment within that time period in the energy deviation ratio. Instead of calculating the task completion risk index, a conservative strategy is used to limit the upper limit of traction power and the power limit of accessories for the current task unit. The conservative strategy can be set to temporarily reduce the upper limit of traction power to a certain percentage of the original upper limit of traction power and temporarily reduce the power limit of accessories to a certain percentage of the original power limit of accessories. The activation and deactivation time of the conservative strategy, the corresponding task unit number, and the marker of operation using the conservative upper limit of traction power and the power limit of accessories are written into the operation status area. In the subsequent on-site robustness review of this method, this record can be used to determine whether the task completion risk index is still kept within an acceptable range and the vehicle is still in a safe operating state under communication and measurement abnormalities.

[0060] In a preferred embodiment, for example in a routine patrol scenario in an industrial park, the online monitoring rhythm can be set to once per second. The threshold for triggering the energy-saving adjustment mechanism when the traction energy deviation ratio and the accessory energy deviation ratio exceed preset thresholds can be set to a certain percentage. The inflection point of the task completion risk index, corresponding to the difference between the expected remaining energy and the total energy required for the task, in the task completion risk mapping table can be set at a position slightly higher than the zero energy safety margin. The task completion risk mapping table is preferably constructed as a one-dimensional lookup table, with the difference between the expected remaining energy and the total energy required for the task as the independent variable and the task completion risk index as the output. Piecewise linear interpolation is used to obtain the risk index at the intermediate point. Statistical analysis during multiple missions revealed that when the actual remaining energy deviates from the energy envelope curve by no more than half of the energy envelope value range, the risk index for most mission completion remains at a low to medium level, and the vehicle can complete the mission without significantly compressing comfort-related accessories. However, in extreme temperature or long-distance uphill missions, when the remaining energy deviation approaches the lower limit of the energy envelope, the risk index for subsequent sacrificial units rapidly increases. The vehicle controller records the risk index distribution in the operating state area while triggering a conservative strategy and limiting comfort-related accessories, thereby verifying the feasibility and repeatability of the online monitoring and risk calculation steps in field conditions.

[0061] S5. Based on the energy deviation and the risk index distribution of subsequent task completion, obtain the energy-saving adjustment coefficient. Scale the traction power limit, vehicle speed limit, and acceleration limit according to the energy-saving adjustment coefficient, increase the regenerative braking torque coefficient, and decrease the accessory power limit. Specifically, the implementation is as follows: The energy-saving adjustment coefficient is a dimensionless coefficient between zero and one, used to uniformly characterize the overall tightening or relaxation of the traction power limit, regenerative braking torque coefficient, and accessory power limit under the current energy deviation and task completion risk conditions. During the vehicle's task execution, based on the aforementioned online monitoring and risk assessment logic, at the end of each energy-saving adjustment cycle that is an integer multiple of the sampling cycle, the vehicle controller extracts the maximum value from the internally stored current task unit's traction energy deviation ratio, accessory energy deviation ratio, and the task completion risk index corresponding to all guarantee units in the distribution of subsequent task completion risk indices in the most recent sampling cycles, and calculates the time-varying trend of this maximum risk index. These three quantities are used as inputs for energy-saving adjustment decisions. An energy-saving adjustment coefficient mapping table is pre-stored in the vehicle controller. This table takes the traction energy deviation ratio, accessory energy deviation ratio, and the maximum task completion risk index of the guarantee unit as inputs. The energy-saving adjustment coefficient mapping table can pre-quantify the traction energy deviation ratio, accessory energy deviation ratio, and the maximum task completion risk index of the guarantee unit into several levels, forming... A three-dimensional grid table is used, and intermediate values ​​are obtained through linear interpolation during operation. The independent variables correspond to the deviation between the actual energy consumption and the budget in the current task unit and the energy sufficiency in subsequent support unit tasks. Each input combination in the table corresponds to an energy-saving adjustment coefficient output between zero and one. The closer the energy-saving adjustment coefficient is to zero, the weaker the energy-saving intervention intensity; the closer it is to one, the stronger the energy-saving intervention intensity. At the end of each energy-saving adjustment cycle, the vehicle controller substitutes the current traction energy deviation ratio, accessory energy deviation ratio, and the maximum task completion risk index of the support unit into the energy-saving adjustment coefficient mapping table to find the basic energy-saving adjustment coefficient. The basic energy-saving adjustment coefficient is then fine-tuned according to the changing trend of the maximum risk index. When the maximum task completion risk index shows a continuous upward trend in the recent energy-saving adjustment cycles, the vehicle controller adds a fixed adjustment amount to the basic coefficient. When the maximum task completion risk index shows a downward trend, it reduces a fixed adjustment amount. The fine-tuned result is truncated between zero and one as the energy-saving adjustment coefficient at the current moment, so that the energy-saving adjustment coefficient takes into account both the current energy deviation and the risk exposure of future support tasks.

[0062] After obtaining the current energy-saving adjustment coefficient, the vehicle controller performs hierarchical adjustments on the control parameter set pre-generated for the current task unit and several subsequent task units in step three. In the traction section, the vehicle controller reads the upper limit of traction power of the current task unit and the speed limit and acceleration limit related to vehicle speed acceleration and deceleration control from the task parameter area according to the current task unit category. It obtains a traction tightening ratio by multiplying the energy-saving adjustment coefficient by the traction adjustment weight of the category to which the task unit belongs. The upper limit of traction power is reduced by a ratio of one minus the traction tightening ratio, and the speed limit and acceleration limit effective in the traction control module of the task unit are reduced by the same ratio. The traction adjustment weight of the guaranteed unit is preferably set to a small value, the compressible unit is set to a medium value, and the sacrificial unit is set to a large value, so that the traction capacity of the sacrificial unit is reduced to the maximum under the same energy-saving adjustment coefficient.

[0063] In the regenerative braking section, the vehicle controller reads the regenerative braking torque coefficient of the task unit, multiplies the energy-saving adjustment coefficient by the regenerative adjustment weight of the task unit's category to obtain the regenerative enhancement ratio, and amplifies the regenerative braking torque coefficient by adding the regenerative enhancement ratio. Simultaneously, it calculates the maximum permissible regenerative braking torque using the current charging current limit provided by the battery management module and the current motor speed. If the regenerative torque corresponding to the amplified regenerative braking torque coefficient might cause the charging current to exceed the limit, the vehicle controller proportionally reduces the regenerative braking torque coefficient until the corresponding regenerative torque matches the permissible regenerative torque, ensuring that the energy recovery ratio is increased without exceeding the battery's permissible charging rate. In the accessory section, the vehicle controller reads the power limits of safety-related accessories and comfort-related accessories for the task unit, and adjusts them according to energy saving... The adjustment coefficient and task unit category determine the accessory compression ratio. The power limit of comfort-related accessories is reduced by the ratio of one minus the accessory compression ratio. When the accessory compression ratio is high and the energy redundancy index is low, the working mode of comfort-related accessories can be further switched from comfort priority to energy priority working mode. The energy priority working mode can be set to prioritize ensuring that the power limit of safety-related accessories is maintained at the statistically obtained average level, while applying duty cycle restrictions or periodic shutdown strategies to comfort-related accessories such as air conditioning compressors, heater fans and advertising screens, and prompting the driver on the vehicle human-machine interface that the current energy-saving operation stage is in progress. The energy redundancy index preferably adopts the energy redundancy of the current task unit calculated based on the energy envelope curve in step three and stored in the task parameter area, or the weighted sum of the energy redundancy of the current task unit and several subsequent task units.

[0064] To ensure the idempotency and version consistency of energy-saving adjustment actions, the vehicle controller only accepts energy-saving adjustment calculation requests carrying the current task session identifier and parameter version identifier not lower than the currently effective version within each energy-saving adjustment cycle. The energy-saving adjustment cycle number is combined with the task session identifier and parameter version identifier to form an energy-saving adjustment record key. When multiple energy-saving adjustment requests from different modules are received within the same energy-saving adjustment cycle, the vehicle controller first checks whether an energy-saving adjustment coefficient has already been generated for that cycle based on the record key. If the record key already exists, the vehicle controller discards subsequent requests and retains only the earliest generated energy-saving adjustment coefficient, thus ensuring that the actual energy-saving adjustment coefficient used within the same rhythm cycle is unique. When the parameter version identifier is lower than the currently effective version, a version mismatch error is recorded in the running status area without any adjustment to the control parameter set. Through the aforementioned energy-saving adjustment record key, the idempotency of energy-saving adjustment actions within the same energy-saving adjustment cycle is achieved, avoiding inconsistencies in adjustment results caused by concurrent requests from multiple modules.

[0065] Within the protection unit, the vehicle controller preferably sets the traction adjustment weight to a small value, the regenerative adjustment weight to a medium value, and the upper limit of the accessory compression ratio to a low value. This allows for the prioritization of increasing regenerative braking force and slightly reducing the power of comfort-related accessories when the energy-saving adjustment coefficient increases, thus maintaining smooth changes in the upper limit of traction power and vehicle speed acceleration limits to avoid affecting the on-time arrival of critical tasks. Within the compressible unit, the vehicle controller sets the traction adjustment weight and the upper limit of the accessory compression ratio to a medium-high level. When the energy-saving adjustment coefficient increases, the upper limit of traction power and the power limit of comfort-related accessories are reduced simultaneously to compress the energy consumption of such tasks without interrupting the task. Within the sacrificial unit, the vehicle controller allows the traction adjustment weight and the upper limit of the accessory compression ratio to reach a large value. When the energy-saving adjustment coefficient approaches one, the upper limit of traction power and the power limit of comfort-related accessories for this task unit can be tightened to near the lower limit of the traction energy fluctuation range and the minimum accessory energy demand level. The controller then displays a prompt message to the driver through the human-machine interface suggesting that the task unit be reduced or skipped.

[0066] If, within a certain time period, the vehicle controller detects that the task completion risk index corresponding to the protection unit exceeds the preset high-risk threshold based on the subsequent task completion risk index distribution, and the energy-saving adjustment coefficient obtained from the energy-saving adjustment coefficient mapping table and trend correction has been close to one for several consecutive energy-saving adjustment cycles, then the vehicle controller determines that the current task is on the edge of high-risk operation. It records a high-risk flag in the operating status area. This flag includes the trigger time, task session identifier, current task unit number, current energy-saving adjustment coefficient, and maximum task completion risk index value. The traction control module and accessory control module read the high-risk flag according to the agreed interface fields in their respective cycle tasks. When the high-risk flag is detected to be valid, the local control parameters are immediately reverted to the aforementioned conservative traction power upper limit and conservative accessory power limit. A status prompt is displayed on the vehicle's human-machine interface, prompting the driver to find a safe parking location or return to the charging area as soon as possible. This prevents the energy-saving adjustment strategy from continuing to tighten the traction and accessory parameters and exceeding the safety and compliance boundaries under extreme energy shortage or prediction model mismatch conditions.

[0067] In preferred embodiments, for example, in a park commuting scenario with fixed patrol routes and several support units, the energy-saving adjustment cycle can be set to an integer multiple of several sampling cycles. The areas in the energy-saving adjustment coefficient mapping table corresponding to "the traction energy deviation ratio is negative and the absolute value is large" and "the maximum task completion risk index of the support unit is close to one" can be set as high energy-saving intensity areas. This allows the vehicle to increase the energy-saving adjustment coefficient at a faster speed when the vehicle experiences significant overconsumption in several consecutive task units and the risk of subsequent support units increases rapidly, thereby compressing the energy consumption of compressible and sacrificial units. In other delivery scenarios with significant route changes, the step size of the energy-saving adjustment coefficient can be appropriately reduced and the high-risk threshold can be increased to reduce frequent energy-saving adjustment oscillations caused by task structure fluctuations. As long as the calculation order of the energy-saving adjustment coefficient, the parameter adjustment rules for different types of task units, and the fallback logic after the high-risk flag is triggered remain consistent, the same energy-saving adjustment process can be reproduced on different vehicle models and control platforms based on the above-disclosed content to obtain similar range optimization effects.

[0068] S6. At the end of the task, update the energy characteristic table, risk calculation parameters, and energy-saving adjustment coefficient mapping table based on the actual energy of each task unit and the task completion status, so that they can be called during subsequent energy envelope construction and control parameter set generation. The specific implementation is as follows: After all task units of this mission have been completed, the vehicle controller enters the closed-loop update phase of this method. In this phase, the operation records of this mission are first summarized in the mission parameter area and the operation status area. These records include the actual traction energy, actual accessory energy, whether the task unit was completed as planned, the sequence number of the last task unit actually executed in this mission, the trajectory of the energy-saving adjustment coefficient used in each energy-saving adjustment cycle during the entire mission, and the high-risk markers and markers of operation using conservative traction power upper limit and accessory power limit values ​​recorded during the mission execution.

[0069] The actual traction energy and actual accessory energy are obtained by multiplying the estimated instantaneous traction power and accessory power by the sampling period length during task execution and then accumulating them within the task unit. The task completion mark is generated by the vehicle controller at the end of the task by comparing the predetermined stop sequence in the task information with the actual stop event sequence, and combining the value of the task completion status in step one in the current task record. If the stop corresponding to a certain task unit is identified within the preset time window, and the actual remaining energy at the end of the task unit is not less than the sum of the traction energy budget and the minimum accessory energy requirement of the task unit, then the task unit is marked as completed; otherwise, it is marked as incomplete.

[0070] After the vehicle is put into operation, a historical observation task window is configured for each type of task. The historical observation task window can be set as a set of tasks with completely consistent task information and similar environmental conditions in the most recent few tasks. The similar environmental conditions are preferably set as the ambient temperature falling within the same temperature range, the task occurring in the same shift or the same season label. In actual implementation, the vehicle controller can filter historical tasks by adding ambient temperature range identifiers and shift identifiers to the task records to form the historical observation task window corresponding to the task type.

[0071] After each task is completed, the vehicle controller writes the operation record of the current task to the historical observation task window corresponding to the task type. When the number of tasks in the historical observation task window exceeds the set limit, the earliest task record is removed in chronological order, and only the latest few tasks are retained as the basis for subsequent feature updates.

[0072] When updating the energy characteristic table, the vehicle controller adopts a smoothing update strategy for the average traction energy and accessory energy of the same category of task units at the same location in the historical observation task window. Specifically, the average value stored in the existing energy characteristic table is weighted with the actual traction energy and accessory energy of the current task according to a preset smoothing factor. That is, the new average value is equal to the smoothing factor multiplied by the old average value plus the supplementary factor multiplied by the actual energy of the current task. The sum of the smoothing factor and the supplementary factor is one. The smoothing factor is preferably greater than the supplementary factor, so that the energy characteristic table is gradually adjusted with the changes in actual working conditions without drastic fluctuations due to individual abnormal tasks. When implementing this, the vehicle controller can set a set of smoothing factor and supplementary factor values ​​for each task type in the configuration area and uniformly use this set of parameters for all task units.

[0073] When updating the energy fluctuation range, the vehicle controller recalculates the maximum and minimum values ​​of traction energy and accessory energy of each task unit under the task type in the historical observation task window, and makes limited adjustments to the upper and lower limits of the original fluctuation range based on the current statistical results. Preferably, the new upper and lower limits are limited to the range in which the original upper and lower limits expand outward or contract inward by no more than a preset percentage, so as to avoid excessive impact of a single task on the long-term fluctuation range.

[0074] When a task unit in the historical observation task window fails to complete multiple times, and the actual remaining energy at each failure is significantly lower than the energy safety boundary predicted by the corresponding risk level in the risk calculation rule, the vehicle controller records this phenomenon as a risk rule deviation. "Significantly lower" can be defined as the actual remaining energy being at least a preset percentage below the lower limit of the safety boundary, for example, 10% to 20% below the lower limit. The specific percentage can be determined by the manufacturer through testing and calibration for different vehicle models during the vehicle debugging phase. Furthermore, by statistically correlating the task completion risk index with actual incomplete events, the safety factor in the risk calculation rule is slightly adjusted to ensure that the new risk calculation rule... In this case, the actual probability of failure corresponding to the same risk level falls more stably within the target range in the observation window. The target range is preferably set to keep the proportion of failures within a fixed range in the high-risk level range, such as between 20% and 40%. When implementing the vehicle controller, the target proportion range for different risk levels can be set in the configuration area. During each window update, the safety factor is adjusted by comparing the difference between the actual statistical value and the upper and lower boundaries of the target range. The adjustment step size of the safety factor is preferably set to a fixed positive number less than 0.1 to ensure the smoothness and stability of the risk mapping relationship during multiple updates.

[0075] After the task is completed, the vehicle controller also makes a small-scale adjustment to the energy-saving adjustment coefficient mapping relationship. It extracts the energy-saving adjustment coefficient sequence for each energy-saving adjustment cycle in this task from the operating status area, as well as the corresponding traction energy deviation ratio, accessory energy deviation ratio, and task completion risk index sequence of the support unit. It checks whether there are sudden jumps in the energy-saving adjustment coefficient near certain input combinations, that is, the change amplitude of the input quantity in adjacent energy-saving adjustment cycles is lower than the set input change threshold while the change amplitude of the energy-saving adjustment coefficient is higher than the set output change threshold. When this situation is detected, the vehicle controller interpolates and smooths the output value of the corresponding area in the energy-saving adjustment coefficient mapping table. The interpolation and smoothing process preferably adopts the method of first linearly interpolating the output value of adjacent input points in this area according to the input variable, and then performing a moving average along the input axis direction on the interpolation result, so as to ensure that the mapping relationship maintains monotonicity and continuity in the input space, and avoids sudden changes in the energy-saving adjustment coefficient due to improper setting of a single mapping point in subsequent tasks, thereby causing sudden changes in traction response or sudden drops in accessory power.

[0076] To ensure the traceability of rule updates, after each update of the energy characteristic table, risk calculation parameters, and energy-saving adjustment coefficient mapping relationship, the vehicle controller generates a new rule version identifier and a threshold version identifier. The rule version identifier can be defined as a monotonically increasing number representing the combination of the energy-saving adjustment coefficient mapping relationship and the energy characteristic table rules. The threshold version identifier can be defined as a monotonically increasing number representing the combination of various threshold parameters in the risk calculation rules. The vehicle controller writes the old version identifier, the new version identifier, the update timestamp, the current task session identifier, and the key indicators of the current task (including task completion rate, completion status of various task units, deviation range between the actual remaining energy at the end of the task and the expected remaining energy at the end of the energy envelope, the number of times the high-risk flag is triggered, and the number of times the conservative rollback flag is triggered) into the log area. The log area is preferably deployed in the append-only storage partition of the vehicle controller's non-volatile memory. Each time a log is written in this partition, a new record is added to the end of the existing content, without overwriting or deleting the written records. The vehicle controller does not provide an interface to modify the content of the log area during normal use, so as to ensure that the log area can be used as a chain of evidence in subsequent inspections of the field performance of this method and compliance reviews.

[0077] This method allows the vehicle controller to send the task session identifier, task completion rate, deviation range between the remaining state of charge at the end of the task and the expected remaining energy at the end of the energy envelope, and the latest rule version identifier to the dispatching platform via the communication network after the task is completed. After receiving several consecutive task summaries, the dispatching platform can statistically analyze indicators such as the proportion of timely completion of support units and whether the deviation of the remaining state of charge at the end of the task is maintained within a certain range, in order to evaluate the effectiveness of the energy management method from the perspective of fleet dispatching and decide whether to adjust the task structure.

[0078] In a preferred embodiment, in the industrial park patrol scenario, the most recent patrol tasks can be selected as the historical observation task window. After the vehicle runs multiple task cycles using this method, the vehicle controller can statistically analyze in the log area that the proportion of timely completion of the guarantee unit is significantly improved compared to when no energy-saving adjustment strategy is used. At the end of the task, the deviation between the actual state of charge and the expected value at the end of the energy envelope is basically controlled within a preset small percentage range. After continuously analyzing the task summary for several days, the scheduling platform confirms that the energy use of the patrol route is more stable between different shifts.

[0079] In an alternative approach, instead of explicitly dividing the support unit, compressible unit, and sacrificial unit in the task information, the vehicle controller automatically divides the entire task into several consecutive energy segments based on the travel length or duration before the task begins. Each energy segment corresponds to a continuous distance or continuous time interval. A segment priority weight is assigned to each segment, and the segment priority weights maintain the same high, medium, and low priority levels as the aforementioned task unit categories. High-priority segments are adjusted according to the same rules as the aforementioned support units in terms of traction power limit, regenerative braking torque coefficient, and accessory power limit. Medium-priority segments are adjusted according to the same rules as the aforementioned support units in terms of the aforementioned parameter adjustment rules. The aforementioned compressible unit and low-priority weight segments are equivalent to the aforementioned sacrificial unit in terms of the above parameter adjustment rules. All calculations in energy characteristic statistics, energy envelope construction, task completion risk estimation, and energy-saving adjustment coefficient application are performed using segments as indexes. The upper limit of traction power, regenerative braking torque coefficient, and accessory power limit are adjusted in a hierarchical manner among different segments according to segment priority weights. As long as the segment priority weights are consistent with the priority weights of the aforementioned guarantee unit, compressible unit, and sacrificial unit in terms of calculation order and action method, and the mapping relationship of energy characteristic table, risk calculation parameters, and energy-saving adjustment coefficient is updated, version locking is also performed through rule version identifier and threshold version identifier and written to the log area.

[0080] In the operating scenario shown in this embodiment: Within an industrial park, the park operator has assigned fixed patrol duties to several low-speed four-wheeled electric vehicles. A typical patrol route starts from the east gate of the park, passing through the office area entrance, the warehouse loading / unloading area, the production workshop entrance, the dormitory area entrance, and the west gate, before returning to the east gate, forming a closed patrol loop with six stops. The nominal battery capacity of the vehicles is set at several tens of kilowatt-hours, and the lower limit of battery charge is set at a corresponding state of charge value based on the nominal capacity. In the first few weeks of operation, the vehicles completed several dozen patrol missions along this route, continuously recording vehicle speed, acceleration / deceleration, motor terminal voltage, motor phase current, estimated accessory power, ambient temperature, and mission completion status according to a uniform sampling rhythm. Based on the aforementioned unified time alignment, missing data completion, physical limit elimination, and task unit division rules, the vehicle controller divides each complete journey from the East Gate to the West Gate and back to the East Gate into six task units in these historical tasks. According to the task importance marking of the stop points and the completion ratio of historical tasks, the task units corresponding to the office area entrance, the production workshop entrance, and the dormitory area entrance are divided into support units, the task units corresponding to the warehouse loading and unloading port and the West Gate are divided into compressible units, and an advertising display point set up in the middle is divided into a sacrificial unit. Within each task unit, the vehicle controller accumulates the traction energy and accessory energy according to a predetermined formula. For example, the average traction energy of each task unit within the guarantee unit is in the range of several kilowatt-hours, and the average accessory energy is in the range of several fractions of a kilowatt-hour. The fluctuation range of traction energy is about a fraction of the average value, and the fluctuation range of accessory energy is slightly smaller. The typical peak value of traction power and the typical peak value of accessory power maintain a reasonable proportional relationship with the announced power of the motor and the nominal total power of the accessories, respectively. The vehicle controller writes these average values, fluctuation ranges, and peak values ​​into the energy characteristic table using the task unit number and task unit category as indexes, forming the statistical basis for subsequent energy planning.

[0081] Before the start of an early morning patrol mission, the vehicle is stationary next to a charging station. The battery management module reports that the current state of charge (SOC) is close to full charge. The vehicle controller reads the battery's nominal capacity and lower charge limit from the configuration area. By subtracting the lower charge limit from the current SOC, the releasable SOC difference is obtained. This difference is multiplied by the battery's nominal capacity to obtain the theoretical usable energy. Combined with the efficiency reduction factor statistically obtained within the most recent observation window, the available battery energy for this mission is calculated. For example, the available battery energy is approximately ten kilowatt-hours out of several tens of kilowatt-hours. Following the order of the six stops in the patrol mission, the vehicle controller sequentially reads the average traction energy and average accessory energy of each mission unit from the energy characteristic table. The initial traction energy budget for these six mission units is set to the corresponding average traction energy, and the initial accessory energy budget is set to the corresponding average accessory energy. The initial total energy requirement is then accumulated in memory. At this point, comparing the initial total energy demand with the available battery energy, it is found that the initial total energy demand is slightly higher than the available battery energy. The vehicle controller initiates the energy compression process, allocating accessory energy compression coefficients to compressible and sacrificial units. In one or two compression rounds, the accessory energy budgets of these task units are gradually reduced. If necessary, the traction energy budget of sacrificial units is subject to limited compression, while keeping the traction energy budgets and accessory energy budgets of the three support units as close as possible to their respective energy averages. After several iterations, the sum of the traction energy budgets and accessory energy budgets of all task units no longer exceeds the available battery energy, while each budget value remains no lower than the corresponding average minus the lower limit of the energy fluctuation range. Based on this, the vehicle controller finalizes the traction energy budget and accessory energy budget for this task. Subsequently, the vehicle controller accumulates the expected energy consumption at the end of each task unit according to the task unit sequence. It subtracts the accumulated consumption from the available battery energy, resulting in an energy envelope curve along the task sequence dimension, showing a gradual decrease in battery power from the starting point to a remaining amount of kilowatt-hours at the end of the task. Based on the energy fluctuation range of each task unit in the energy characteristic table, it calculates the upper and lower allowable ranges of remaining energy at the end of each task unit, forming the energy envelope value range. This energy envelope curve is recorded in the task parameter area along with the task session identifier, which is generated by adding the current date and time to the task number, ensuring that this patrol task has a unique identity in subsequent records.

[0082] After the energy envelope is constructed, the vehicle controller generates a set of control parameters for this task. For the first support unit from the east gate to the office area entrance, the vehicle controller calculates the expected travel time based on the estimated mileage in the task information and the unit's historical average speed in the energy characteristic table. The traction energy budget of this unit is divided by the expected travel time to obtain the average traction power boundary. This boundary is then compared with the vehicle's announced power, the motor's long-term allowable power, and the temperature rise constraint power derived from the relationship between winding temperature and output power under the current ambient temperature. The minimum value is taken as the upper limit of the traction power for this task unit. For example, the upper limit of the traction power may be limited to a certain percentage of the motor's announced power. For compressible units traveling from the office area to the warehouse area and sacrificial units traveling from the dormitory area to the west gate, the vehicle controller follows the same principle when calculating the upper limit of the traction power. However, considering that these task units can reduce their dwell time or skip entirely when necessary, a larger reduction will be given in the subsequent energy-saving adjustment phase. In the regenerative braking section, the vehicle controller calculates the energy redundancy of each task unit and subsequent task units based on the energy envelope range. For example, the energy redundancy is significantly smaller for sacrificial units near the end of the patrol, while it is more sufficient for support units in the first half of the patrol. Combining this with the gradient difficulty level obtained from map data or historical acceleration / deceleration statistics, the energy redundancy index and gradient difficulty level are used as independent variables to find the basic regenerative braking torque coefficient in a two-dimensional mapping table. Then, based on the current charging current limit given by the battery management module, the basic coefficient is trimmed according to the aforementioned rules to form a regenerative braking torque coefficient that improves energy recovery without exceeding the charging safety boundary. Regarding accessory power limits, the vehicle controller classifies headlights, brake lights, and windshield defoggers as safety-related accessories, and air conditioning compressors, heaters, and advertising displays as comfort-related accessories. In each task unit, the power limit of safety-related accessories is first ensured to be no lower than the average value statistically analyzed in the energy characteristic table. Then, based on the accessory energy budget and energy redundancy of that unit, an adjustable power limit is assigned to comfort-related accessories. Higher comfort accessory permissions are reserved in support units, while a larger downward adjustment space is allowed in sacrificial units. The control parameters such as the upper limit of traction power, regenerative braking torque coefficient and accessory power limit of each task unit are written into the parameter area along with the task unit number, task session identifier and parameter version identifier, and are sent to the traction control module and accessory control module through the in-vehicle communication network. Each module compares the parameters according to the parameter version identifier and only accepts the current or higher version of the control parameters to take effect, thereby ensuring the consistency of control strategies between different modules.

[0083] After the driver confirms the start of the task on the onboard human-machine interface, the vehicle departs from the east gate and enters the first task unit. During task execution, the vehicle controller collects the traction energy increment and accessory energy increment within the current task unit from the traction control module and accessory control module at a rhythm of once per second. The traction energy increment and accessory energy increment are accumulated cycle by cycle within this task unit to form the cumulative traction energy value and cumulative accessory energy value. The difference between these values ​​and the corresponding traction energy budget and accessory energy budget for this task unit is used to calculate the traction energy deviation ratio and accessory energy deviation ratio. Simultaneously, at the end of each sampling cycle, the vehicle controller reads the updated state of charge (SOC) from the battery management module, converts it to the current actual remaining energy in the same way as calculating the battery's available energy, maps the current actual remaining energy and task unit number to the energy envelope curve to find the expected remaining energy, and then subtracts the expected remaining energy from the actual remaining energy and divides it by the difference between the upper and lower bounds of the energy envelope value range at that position to obtain the remaining energy deviation, which is used to characterize the deviation of the current task execution from the energy plan. When the vehicle encounters temporary construction detours while traveling to the storage area in the second task unit, resulting in a slight increase in travel distance and frequent rapid acceleration by the driver to ensure on-time arrival, the vehicle controller calculates that the traction energy deviation ratio for this task unit is negative and its absolute value is close to several percent. The accessory energy deviation ratio is essentially zero, while the remaining energy deviation is negative and close to half of the lower limit of the energy envelope range. At this point, in its risk calculation logic, the vehicle controller simulates the remaining energy changes from the start of the current task unit to the end of each subsequent task unit. Considering that the actual energy consumption caused by the detour is higher than the budget, and combining the energy fluctuation range recorded in the energy characteristic table, the expected remaining energy is obtained for each subsequent task unit under pessimistic assumptions, and the task completion risk index for each subsequent task unit is calculated through a mapping table. The calculation results show that, without adjusting the energy usage strategy, the task completion risk index of the last sacrificial unit rapidly rises to close to one, while the risk index of the penultimate guaranteed unit also rises from close to zero to a moderate level. The vehicle controller writes these task completion risk indices into the risk index distribution recording area according to the task order, providing a basis for subsequent energy-saving adjustments.

[0084] At the end of each energy-saving adjustment cycle consisting of several sampling periods, the vehicle controller selects the maximum value from the traction energy deviation ratio, accessory energy deviation ratio, and task completion risk index corresponding to all support units in the risk index distribution of the current task unit. It then compares whether this maximum task completion risk index shows an upward or downward trend in the recent energy-saving adjustment cycles. These three quantities are used as independent variables in the energy-saving adjustment coefficient mapping table. The basic energy-saving adjustment coefficient is retrieved from the mapping table, and then fine-tuned based on the trend of risk index changes. When, under the aforementioned detour scenario, the traction energy deviation ratio is significantly negative, and the maximum task completion risk index of the support unit rapidly approaches a medium-high level from a low level and still shows an upward trend, the vehicle controller increases the energy-saving adjustment coefficient from a weak intervention level close to zero to a level close to the median or even above in that energy-saving adjustment cycle. Subsequently, the vehicle controller adjusts the control parameter sets of the current and several subsequent task units in stages according to the current task unit category and energy-saving adjustment coefficient: for the current compressible unit, the upper limit of traction power is reduced by a certain proportion, and the speed limit and acceleration limit are lowered accordingly, making it difficult for the vehicle to continue to accelerate rapidly for a long time in subsequent road sections; the regenerative braking torque coefficient is appropriately amplified according to the energy-saving adjustment coefficient and regenerative adjustment weight, and the proportion of motor braking in downhill and deceleration stages is increased on the premise of ensuring that the battery charging current does not exceed the limit; the power limit of comfort-related accessories is significantly reduced according to the accessory compression ratio. In units with tight energy redundancy indicators, the vehicle controller periodically reduces the duty cycle of the air conditioning compressor through the body control module, and even temporarily shuts down the advertising display screen in the sacrificial unit corresponding to the advertising display point, and prompts the driver on the vehicle human-machine interface that the current operation is in the energy-saving stage. When the patrol enters the last two task units and the risk index distribution shows that the task completion risk index of a certain support unit is approaching the preset high-risk threshold, and the energy-saving adjustment coefficient remains close to one for several consecutive energy-saving adjustment cycles, the vehicle controller records a high-risk flag in the operating status area. At the same time, after the traction control module and accessory control module read this flag in their respective cycle tasks, they will reduce the local traction power limit and accessory power limit to a conservative level to avoid excessive consumption of remaining energy, and prompt the driver to end the patrol or return to the charging area as soon as possible, thereby maintaining the safe operation of the vehicle under conditions of extreme energy shortage and large changes in the driving environment.

[0085] When the patrol mission ended and the vehicle returned to the east gate to park, the state of charge reported by the battery management module was still several percent higher than the lower limit. Based on the mission information and parking events, the vehicle controller determined that all the stops corresponding to the support units in this mission were completed as planned, while the advertising display points corresponding to the sacrificial units were prompted to skip once during the energy-saving adjustment phase. The onboard human-machine interface displayed a message indicating that there was detour and energy-saving intervention during the execution of this mission. After the mission ended, the vehicle controller, according to the aforementioned closed-loop update logic, wrote the actual traction energy and actual accessory energy of all mission units, mission completion markers, high-risk markers, and conservative rollback markers into the historical observation task window corresponding to this mission type. It then weighted and updated the average traction energy and accessory energy of each mission unit in the energy characteristic table according to a smoothing factor, and made minor adjustments to the upper and lower limits of the fluctuation range of traction energy and accessory energy within the limited range. In response to the situation where multiple detours in this task caused the energy consumption of a certain compressible unit to frequently approach the upper limit of historical fluctuations, the vehicle controller statistically analyzed the proportion of incomplete events and the deviation of expected remaining energy in the historical task observation window. It then made minor adjustments to the safety factor in the risk calculation rules, ensuring that the risk index for task completion of this unit would increase earlier under similar energy margin conditions in the future, prompting the system to strengthen energy-saving intervention in advance. Regarding the energy-saving adjustment coefficient mapping relationship, the vehicle controller extracted the sequence of energy-saving adjustment coefficients, energy deviation ratios, and risk indices within the current task from the operating status region. When a local jump in the energy-saving adjustment coefficient was detected near a certain input combination, linear interpolation plus moving average was used to smooth the output in that region, ensuring the continuity of the trend of energy-saving adjustment coefficients with changes in energy deviation and risk index in subsequent tasks. After each update of the energy characteristic table, risk calculation parameters, and energy-saving adjustment coefficient mapping relationship, the vehicle controller generated a new rule version identifier and threshold version identifier, and wrote the old version identifier, new version identifier, task session identifier, task completion rate, the deviation range between the actual remaining energy and the expected value at the end of the energy envelope at the end of the task, the number of times the high-risk flag was triggered, and the number of times the conservative backoff flag was triggered into the log area. Several weeks later, the operators could see from the statistics on the dispatch platform that during the period when the energy management and range optimization method was adopted, the on-time completion rate of the key patrol points corresponding to the support unit was significantly improved. At the end of the mission, the actual state of charge was generally within the energy envelope value or slightly higher than the median of the envelope. Only in a few missions with extreme temperatures and abnormal driving habits were the actual remaining energy close to the lower limit of the envelope but still did not trigger the depletion of power. This verified that the method has clear feasibility, stability and range optimization effect in the real scenario of park patrol.

[0086] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0087] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0088] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0089] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0091] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0093] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0094] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0095] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for overall energy management and range optimization of a low-speed four-wheeled electric vehicle, characterized in that, include: S1. Obtain mission information and historical data, divide the mission into support units, compressible units and sacrificial units, and statistically analyze the traction energy and accessory energy characteristics of each mission unit. S2. Under the constraints of available battery energy and lower charge limit, allocate traction energy budget and accessory energy budget to each task unit according to energy characteristics and priority to form an accumulated energy envelope. S3. Based on the energy envelope, generate a set of control parameters for each task unit, consisting of the upper limit of traction power, the regenerative braking torque coefficient, and the power limit of accessories, and store them into the vehicle controller in sequence. S4. During the execution process, periodically collect the traction energy and accessory energy of each task unit, calculate the actual and budgeted deviations, and combine the remaining energy and the energy budget of subsequent task units to obtain the risk index distribution of subsequent task completion. S5. Based on the energy deviation and the risk index distribution of subsequent task completion, obtain the energy-saving adjustment coefficient, scale the upper limit of traction power, the speed limit and the acceleration limit according to the energy-saving adjustment coefficient, increase the regenerative braking torque coefficient and reduce the accessory power limit; S6. At the end of the task, update the energy characteristic table, risk calculation parameters and energy saving adjustment coefficient mapping table according to the actual energy of each task unit and the task completion status, so that they can be called when constructing the energy envelope and generating the control parameter set.

2. The method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to claim 1, characterized in that, S1 includes: The vehicle controller retrieves task information and historical task records, aligns the time sequence of the measurement signals according to the sampling rhythm, and marks records that exceed the physical limit range as invalid measurement points; Based on the order of stops and the status of the doors, historical mission records are divided into mission units. Based on the mission importance and the amount of mission completion, the mission units are further divided into support units, compressible units, and sacrificial units. The traction energy, accessory energy, and corresponding power characteristics of various task units are statistically analyzed and written into the energy characteristic table according to task unit category and task unit order.

3. The method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to claim 1, characterized in that, S2 include: The vehicle controller determines the available energy of the battery based on the battery state of charge, battery nominal capacity, lower charge limit, and efficiency reduction factor. Read the average traction energy and average accessory energy from the energy characteristic table in the order of task units as the initial values ​​of traction energy budget and accessory energy budget, and calculate the initial total energy demand. When the initial total energy demand is greater than the available energy of the battery, the accessory energy compression coefficient and the traction energy compression coefficient are set for the compressible unit and the sacrificial unit. The compression coefficient is reduced according to the compression round. The final energy budget is obtained under the constraint that the traction energy budget and accessory energy budget of each task unit are not lower than their respective safe energy consumption lower bound. The vehicle controller constructs an energy envelope curve based on the final energy budget and stores the energy envelope curve and the task session identifier in the task parameter area.

4. The method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to claim 1, characterized in that, S3 includes: The vehicle controller determines the expected travel time of each task unit based on the expected mileage in the task information and the statistical results of the vehicle speed of each task unit in the energy characteristic table. The average traction power boundary is obtained by dividing the traction energy budget of each task unit by the expected driving time. It is then compared with the vehicle's announced power, the motor's long-term allowable power, and the temperature rise constraint power one by one, and the minimum value is taken as the upper limit of the traction power of the task unit. Write the upper limit of traction power, as well as the speed limit and acceleration limit that match the upper limit of traction power, into the control parameter set in the task parameter area.

5. The method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to claim 4, characterized in that: The vehicle controller determines the energy redundancy of each task unit based on the energy envelope value range, inputs the energy redundancy and slope difficulty level into the mapping table to select the regenerative braking torque coefficient, and corrects the regenerative braking torque coefficient under the constraint of the upper limit of battery charging current. The vehicle controller classifies on-board accessories into safety-related accessories and comfort-related accessories based on mission information. It determines the average total power limit of the accessory system based on accessory energy budget and expected duration, and determines the power limits of safety-related accessories and comfort-related accessories according to the priority principle of safety-related accessories. The upper limit of traction power, regenerative braking torque coefficient, and accessory power limit are associated with the task session identifier and parameter version identifier and then sent to the traction control module and accessory control module.

6. The method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to claim 1, characterized in that, S4 include: The vehicle controller acquires the cumulative traction energy and accessory energy values ​​according to the sampling period; The energy deviation ratio is calculated based on the traction energy budget and the accessory energy budget, and the remaining energy is obtained based on the battery state of charge and the battery available energy and compared with the energy envelope curve. The vehicle controller generates a risk index for subsequent task completion based on the energy deviation ratio, the degree of deviation between the remaining energy and the envelope curve, and the subsequent energy budget according to the risk mapping relationship. When it detects that the traction control module information and accessory control module information have not been updated, a conservative strategy is adopted to reduce the upper limit of traction power and the power limit of accessories.

7. The method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to claim 1, characterized in that, S5 include: At the end of the energy-saving adjustment cycle, the vehicle controller reads the traction energy deviation ratio, the accessory energy deviation ratio, and the maximum task completion risk index of the support unit. The basic energy-saving adjustment coefficient is obtained by looking up the energy-saving adjustment coefficient mapping table, and then the energy-saving adjustment coefficient is corrected according to the changing trend of the maximum task completion risk index.

8. The method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to claim 7, characterized in that: The vehicle controller reduces the upper limit of traction power, vehicle speed limit and acceleration limit according to the task unit category and energy-saving adjustment coefficient by traction adjustment weight, amplifies the regenerative braking torque coefficient by regenerative adjustment weight and is constrained by the upper limit of battery charging current, and reduces the power limit of comfort-related accessories by accessory compression ratio according to accessory compression ratio. When recording excessively high risk markers, the upper limit of traction power and the power limit of comfort-related accessories will be reverted to the conservative upper limit of traction power and the conservative power limit of accessories.

9. The method for energy management and range optimization of a low-speed four-wheeled electric vehicle according to claim 1, characterized in that, S6 include: After each task is completed, the vehicle controller updates the energy characteristic table and energy fluctuation range in the historical observation task window based on the task operation record; Adjust the safety factor in the risk calculation rules according to the incomplete task status, and interpolate and smooth the energy-saving adjustment factor mapping table according to the trajectory of the energy-saving adjustment factor changing with the input amount; generate rule version identifier and threshold version identifier after each update, and write the new and old version identifiers and key task indicators into the log area with only append write mode to realize version locking and parameter traceability.

Images