A method for optimizing energy supplement path based on vehicle-pile space-time load prediction

Abstract

The present application relates to the technical field of new energy vehicle energy supplement path planning, and particularly relates to a kind of energy supplement path optimization method based on vehicle-pile space-time load prediction, which maps vehicle, road, energy supplement resource and surrounding vehicle state to vehicle-pile road space-time state unit, predicts energy supplement demand energy, pile group service load and resource competition load, and determines energy supplement continuous domain and its boundary change accordingly; when energy supplement continuous domain shrinks, breaks or is insufficient in stability, it adjusts arrival time window, road segment access order, target energy supplement amount or temporary occupation load through active manufacturing strategy, and then outputs dynamic driving path, target energy supplement area, target arrival time window and target energy supplement amount. This method helps to improve the adaptability of energy supplement path planning to vehicle-pile load change and the continuity of energy supplement decision.

Description

Technical Field

[0001] This invention relates to the field of new energy vehicle refueling path planning technology, and more specifically, to a refueling path optimization method based on vehicle-charging station spatiotemporal load prediction. Background Technology

[0002] During operation, new energy vehicles typically need to plan their charging routes by considering remaining battery power, road conditions, and available charging resources. This technology involves processes such as vehicle energy consumption prediction, charging resource load assessment, route decision-making, and dynamic adjustment of charging plans.

[0003] Existing charging route planning methods typically select charging stations and routes based on the vehicle's current location, remaining battery power, destination, distance to charging stations, current queuing status, or road travel time, which can meet the needs of vehicles seeking charging resources to a certain extent. However, in actual driving, the status of charging resources is not only affected by the target vehicle itself, but also by already connected vehicles, queued vehicles, reserved vehicles, and potential charging vehicles in the vicinity. If judgment is based solely on the current number of available charging stations or static queuing status, it is difficult to map the energy demand for charging, the service load of the charging station group, and the resource competition load to a unified state formed by the charging area and time slice, and it is also difficult to determine whether adjacent charging states meet the continuity of timing, power, and load. When surrounding vehicles enter the same charging area in close proximity, a previously feasible charging state may lose its carrying capacity due to the overload of the charging station group or the resource competition load, leading to the contraction or breakage of the charging continuity domain boundary. In addition, when the continuous domain of energy replenishment shrinks, the existing processing methods often focus on reselecting stations or routes, but do not adequately coordinate the adjustment of factors such as arrival time window, road segment access sequence, target replenishment power and temporary load occupation. The path output is difficult to continuously correct according to changes in actual power, road conditions and energy replenishment resource status.

[0004] Therefore, this invention proposes a method for optimizing energy replenishment paths based on vehicle-pile spatiotemporal load prediction, in order to make collaborative decisions on energy replenishment feasibility states, energy replenishment continuous domain boundaries, and dynamic driving paths. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the present invention aims to provide a method for optimizing energy replenishment paths based on vehicle-pile spatiotemporal load prediction.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for optimizing energy replenishment paths based on vehicle-pile spatiotemporal load prediction specifically includes the following steps:

[0008] Step S1: Obtain the status of the target vehicle, road, energy replenishment resources and surrounding vehicles, and construct a vehicle-pile-road spatiotemporal status unit according to the energy replenishment area and time slice;

[0009] Step S2: Based on the vehicle-pile-road spatiotemporal state unit, predict the energy demand for energy replenishment, the service load of the pile group, and the resource competition load.

[0010] Step S3: Determine the energy replenishment feasibility state based on the target vehicle's expected energy level, detour cost, charging pile service load, and resource competition load, and connect the energy replenishment feasibility states that meet the requirements of timing, energy level, and load continuity into a continuous energy replenishment domain.

[0011] Step S4: Generate virtual energy replenishment behavior of surrounding vehicles, superimpose the virtual energy replenishment behavior onto the vehicle-pile-road spatiotemporal state unit, and correct the boundary of the energy replenishment continuous domain based on the superimposed load state.

[0012] Step S5: When the corrected energy replenishment continuous domain meets the active generation triggering condition, a candidate active manufacturing strategy is generated, and the expected arrival time of the target vehicle, the target energy replenishment capacity, the expected occupation period and the temporary occupation load are recalculated according to the candidate active manufacturing strategy to obtain the post-manufacturing energy replenishment continuous domain.

[0013] Step S6: Calculate the load replenishment potential energy gradient of the candidate driving direction based on the post-manufacturing replenishment continuous domain;

[0014] Step S7: Output the dynamic driving path, target energy replenishment area, target arrival time window, and target energy replenishment amount based on the load replenishment potential energy gradient.

[0015] Furthermore, in step S4, the virtual charging behavior is generated based on the remaining battery power of surrounding vehicles, current route direction, destination direction, reachable charging area, and charging intention intensity, and each virtual charging behavior includes at least a virtual arrival time slice, candidate charging area, expected charging power, expected service start time, and expected time period.

[0016] Furthermore, in step S4, the virtual energy replenishment behavior is substituted into the calculation of energy replenishment demand, pile group service load and resource competition load; after superimposing the virtual load, the state that loses its bearing capacity due to the pile group service load or resource competition load exceeding the limit will no longer participate in the continuous domain connection, and the boundary of the energy replenishment continuous domain will be re-identified based on the energy replenishment stability and feasible state acceptance relationship.

[0017] Furthermore, in step S4, when the state is located at the necessary connection point leading to the subsequent energy replenishment area, and its energy replenishment stability drops below the stability requirement, the state is marked as a weak state; when the connection between two feasible areas disappears due to the excessive load of the pile group service load or the resource competition load, the fracture location, fracture time slice, and main fracture cause are recorded.

[0018] Furthermore, in step S5, the actively generated triggering conditions include: the coverage of the continuous energy replenishment domain is insufficient to support the current driving task; the energy replenishment stability of the key state unit in the continuous domain is lower than the stability requirement; the connection between two feasible areas is broken due to the overload of the pile group service load or the resource competition load exceeding the limit; the actual energy consumption of the target vehicle is higher than the predicted energy consumption and causes the safe power boundary to move forward; or the virtual energy replenishment behavior of surrounding vehicles is concentrated in the same energy replenishment area and the same arrival time window.

[0019] Furthermore, in step S5, after triggering active generation processing, the main cause of the breakage is identified; when the service load of the pile group exceeds the limit, the arrival time window or target replenishment power is adjusted first; when the resource competition load exceeds the limit, the access order of the road segment, the candidate replenishment area or the occupied time period is adjusted first; when the expected arrival power is insufficient, short-term and appropriate replenishment is arranged first.

[0020] Furthermore, in step S5, the candidate active manufacturing strategy includes at least one control quantity among the target driving speed range, road segment access sequence, candidate energy replenishment area arrival window, target energy replenishment capacity, and temporary load occupation.

[0021] Furthermore, in step S5, when adjusting the arrival time window of the candidate energy replenishment area, an executable speed range is selected within the constraints of road speed limit, traffic safety and driving comfort, and the time slice for the target vehicle to arrive at the energy replenishment area is recalculated; when adjusting the access order of road segments, a sequence of candidate road segments that meets the constraints of destination direction and detour cost is selected in the road topology, so that the time slice for the vehicle to enter the energy replenishment area falls into the time window with low service load and low competition load.

[0022] Furthermore, in step S5, when adjusting the target replenishment power, the minimum available power required after leaving the station is calculated in reverse from the subsequent available state, and the target replenishment power is determined accordingly; when adjusting the temporary occupancy load, the expected replenishment power and expected occupancy period of the target vehicle in the candidate arrival time window are projected into the replenishment demand energy, pile group service load and resource competition load, and the corresponding temporary projection is removed or transferred when the candidate strategy is not adopted or the path is deviated.

[0023] Furthermore, in step S7, the vehicle continuously receives the actual location, actual battery level, actual speed, road conditions, and energy replenishment resource status during its journey. When the actual battery level is lower than the predicted battery level, the estimated arrival time deviates from the target arrival window, the virtual energy replenishment behavior of surrounding vehicles is concentrated in the same energy replenishment area, or the target vehicle is about to leave the energy replenishment continuum, the energy replenishment continuum boundary correction and active manufacturing strategy are re-executed.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] 1. To address the difficulty in uniformly describing the states of vehicles, roads, energy replenishment resources, and surrounding vehicles in energy replenishment path planning, this invention constructs a vehicle-pile-road spatiotemporal state unit based on energy replenishment area and time slice. Based on this state unit, it predicts the energy demand for energy replenishment, the service load of the pile group, and the resource competition load. Furthermore, it determines the feasible energy replenishment state based on the target vehicle's expected arrival energy, detour cost, pile group service load, and resource competition load, and connects feasible energy replenishment states that satisfy the requirements of timing, energy supply, and continuous load acceptance into a continuous energy replenishment domain. This approach enables vehicle range safety, energy replenishment resource service capacity, and vehicle quantity competition to participate in path determination within the same spatiotemporal framework, helping to improve the adaptability of energy replenishment path planning to load changes.

[0026] 2. To address the issue that potential refueling behavior from surrounding vehicles may lead to changes in the boundary of the refueling continuum, this invention generates virtual refueling behavior based on the remaining battery power of surrounding vehicles, current route direction, destination direction, reachable refueling area, and refueling intention intensity. The virtual arrival time slice, candidate refueling area, expected refueling power, expected service start time, and expected occupancy period are then superimposed onto the vehicle-pile-road spatiotemporal state unit. After superimposing the virtual load, states that lose their carrying capacity due to exceeding the service load of the pile group or resource competition load limits are no longer included in the continuum connection. The boundary of the refueling continuum is re-identified based on refueling stability and the feasible state continuity. This method helps to reflect the impact of concentrated refueling by surrounding vehicles on the feasible refueling state before path output.

[0027] 3. To address the issues of insufficient coverage of the energy replenishment continuum, decreased stability of key state units, or breaks in feasible regions, this invention generates candidate active manufacturing strategies when the active generation trigger conditions are met. It then recalculates the expected arrival time of the target vehicle, the target energy replenishment, the expected occupation period, and the temporary occupation load using at least one control variable among the target driving speed range, road segment access sequence, candidate energy replenishment area arrival window, target energy replenishment, and temporary occupation load, thus obtaining the post-manufacturing energy replenishment continuum. Based on this post-manufacturing energy replenishment continuum, it calculates the load energy replenishment potential gradient of the candidate driving direction, outputting the dynamic driving path, target energy replenishment area, target arrival window, and target energy replenishment. This processing method helps to allow the path, window, and energy replenishment to be adjusted dynamically according to the load status and the actual vehicle status. Attached Figure Description

[0028] Figure 1 This is a flowchart of an energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction;

[0029] Figure 2 This is a flowchart of the continuous domain boundary correction steps driven by the vehicle energy replenishment competition relationship of the present invention.

[0030] Figure 3 This is a flowchart of the active generation step of the energy replenishment feasible domain under load time-series shaping according to the present invention. Detailed Implementation

[0031] Example, refer to Figure 1 The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction in this embodiment specifically includes the following steps:

[0032] Step S1: Discretization representation of the spatiotemporal state unit of vehicle-pile-road.

[0033] The feasibility of vehicle recharging is related to road conditions, vehicle remaining battery power, charging resource occupancy, and the arrival trend of surrounding vehicles. Vehicle status, road conditions, and recharging resource status are described in a unified time slice.

[0034] S11. Vehicle and Road Status Acquisition:

[0035] The system acquires the target vehicle's current location, destination, remaining available battery power, available battery capacity, current speed, vehicle load, air conditioning load, historical energy consumption characteristics, and future driving needs. Historical energy consumption characteristics are obtained by recording the vehicle's energy consumption per unit distance under similar road conditions, speed ranges, external temperatures, and load conditions. If the target vehicle lacks sufficient driving records, the energy consumption of the same vehicle model can be used, and adjustments can be made based on the current level of congestion, road gradient, and onboard electrical load.

[0036] Roadside data includes road topology, segment length, speed limit, real-time travel time, predicted travel time, congestion level, traffic control status, and segment gradient. For each segment in the candidate path, a base energy consumption is calculated based on segment length and energy consumption per unit distance. This base energy consumption is then adjusted by incorporating factors such as speed deviation, gradient, start-stop frequency, and vehicle load. After accumulating data segment by segment along the candidate path, the estimated energy consumption for the vehicle to travel from its current location to the candidate refueling area is obtained.

[0037] In some embodiments, energy consumption per unit distance can be selected according to road grade and speed range. When the actual vehicle speed is lower than the road segment's operating speed, the energy consumption correction caused by the increased start-stop frequency is included in the energy consumption of that road segment; when the vehicle passes through an uphill section, the additional energy consumption corresponding to the slope is included in the energy consumption of that road segment; when the vehicle is in low temperature or high air conditioning load conditions, the on-board electrical load is converted into energy consumption per unit distance.

[0038] S12, Energy Resource Status Acquisition:

[0039] The energy replenishment resource side records the location of charging facilities, the number of available charging piles, the rated power of the charging piles, the effective service power, fault status, occupancy status, number of vehicles in queue, the estimated departure time of connected vehicles, and the station's power limitation status. The effective service power is determined by the executable power after being jointly limited by the vehicle's current acceptable battery power, the rated power of the charging pile, and the station's allocable power; when a vehicle has not yet connected to a specific charging pile, the executable power of the matching charging piles in that energy replenishment area within the corresponding time slice is used.

[0040] For vehicles already connected, the estimated departure time is calculated based on the current battery level, the target departure battery level, and the available charging power. For vehicles in the queue, the estimated service start time is calculated based on the queue order, reservation window, available charging station type, and the estimated departure time of the preceding vehicle. For vehicles with confirmed reservations, the estimated service start time is determined if the reservation window is earlier than a service availability window, and later than a service availability window. For nearby vehicles that have not yet confirmed their charging activity, their current location, remaining battery level, current direction of travel, destination, and route status are recorded for subsequent virtual charging activity prediction.

[0041] S13. Spatiotemporal state unit division:

[0042] The future prediction time domain is divided into multiple time slices, and the length of each time slice is denoted as . , The time scale is determined jointly based on three types of time scales: station-end state changes, road state changes, and vehicle-side rolling decision-making. The station-end state change time scale is denoted as... The time interval is determined by the charging pile status reporting timestamp within the charging area, the station's power limitation status update time, the estimated departure time of connected vehicles, and the reservation window boundary. For the same charging area, if there are no estimated departure, reservation start, or reservation end events between two adjacent station status reports, the time interval between these reports can be used as the station status maintenance duration. If such events exist, the reporting interval is further divided based on the event time, and the shorter segment is used as a candidate value for the station status change time scale.

[0043] The time scale for changes in road condition is denoted as The estimated travel time is determined by the update interval of the predicted travel time of the main road segments in the candidate path, the estimated travel time of vehicles entering adjacent road segments from the current road segment, and the estimated travel time of vehicles from the entrance road segment of the energy replenishment area to the entrance of the station. For road segments with rapidly changing congestion, the adjacent timestamp interval of the road's predicted travel time update is used; for road segments with relatively stable traffic conditions, the estimated travel time required for vehicles to pass through that road segment is used as the candidate value.

[0044] The rolling decision time scale is denoted as The trigger interval for recalculating the recharging path is determined by the vehicle side. This trigger interval can be triggered by vehicle location updates, changes in battery charge exceeding re-estimation conditions allowed by battery management, deviation from the target arrival window, updates to road traffic conditions, or updates to the recharging area status. The vehicle side also has a minimum executable control duration. Its resolution is determined by the shortest execution time required for the vehicle to complete speed range adjustment, route switching prompts, and driving safety constraints; the data timestamp resolution is denoted as... The accuracy is determined by the timestamps used to collect data from vehicles, roads, and stations.

[0045] In implementations using equal-length slices, Select as not less than and and not greater than , and The executable time length is set at a relatively small time scale. This ensures that major changes in station-side available power, road arrival times, or vehicle rollover decisions are not missed within a single time slice, and that the vehicle side has sufficient time to perform speed adjustments and route switching. If the above conditions cannot be simultaneously met in specific scenarios, priority is given to ensuring... The duration is not less than the minimum executable control duration on the vehicle side, and is segmented and converted within the same time slice according to the original timestamps of the station status events and road status events, so that the charging power, the number of available charging piles and the passage time participate in the load calculation according to their actual effective time period.

[0046] Recharge area With Time Slice The combination forms a vehicle-pile-road spatiotemporal state unit, denoted as A charging station replenishment area can correspond to a single charging station or several charging stations with strong road accessibility, overlapping service radii, and unified scheduling capabilities. Each state unit records the energy demand for replenishment, road access concentration, charging station service load, and resource competition load. Road access concentration can be obtained by accumulating the virtual arrival probabilities of vehicles entering the charging station replenishment area's entrance road segment and normalizing this probability in conjunction with the entrance road segment's traffic capacity.

[0047] Step S2: Prediction of vehicle-pile spatiotemporal load evolution based on virtual arrival behavior.

[0048] After completing the state unit division, the tendency of the vehicle group to generate energy replenishment behavior in the future time slice is projected into the energy replenishment area. The prediction results cover the energy demand for energy replenishment, the occupancy of charging pile services, and the competition for the number of vehicles.

[0049] S21. Determination of the set of candidate energy replenishment spatiotemporal states:

[0050] For vehicles and refueling area Calculate the distance from the vehicle's current location to the refueling area. Expected arrival time and expected power The expected energy level and the set of candidate recharge spatiotemporal states are determined by the following formula:

[0051] ;

[0052] ;

[0053] In the formula, For vehicles Current remaining available battery power; For vehicles From current location to refueling area The estimated energy consumption is calculated by accumulating the length, gradient, congestion status, average speed, and vehicle energy consumption characteristics of each segment in the candidate path. Specifically, the candidate path is divided into multiple consecutive, ordered segments based on road topology nodes, with consistent road grade, speed limit, gradient, and congestion level within each segment. For a single segment, the segment length is first multiplied by the baseline energy consumption per unit distance at the corresponding road grade and baseline speed to obtain the basic energy consumption for that segment. Then, gradient correction, speed and congestion correction, and vehicle energy consumption characteristics are added separately. The load correction includes three types of adjustments: for uphill sections, the additional energy consumption is calculated based on the gradient angle and the total vehicle mass; for downhill sections, a negative deduction is made according to the energy feedback ratio; the energy consumption deviation caused by start-stop idling or high-speed wind resistance is calculated based on the deviation between the actual average speed of the road section and the reference speed; the continuous power of the air conditioner and vehicle electrical appliances is converted into additional energy consumption per unit distance based on the average speed of the road section; after correction, the total power consumption of a single road section is obtained, which is accumulated segment by segment along the driving direction starting from the current location of the vehicle to finally obtain the total estimated power consumption for driving.

[0054] To determine the minimum safe charge level upon arrival, the larger of the reserve charge required by the battery management strategy and the charge required to travel from the candidate charging area to the nearest backup charging area can be used, and adjustments can be made based on the risks of low temperature, congestion, and detours. For time slices The corresponding time interval; For vehicles Arrive at the refueling area Compared to the detour cost incurred by the original driving task; For vehicles The maximum acceptable detour cost can be determined by user settings, task time limits, or operational vehicle dispatch constraints. A status unit that meets the above conditions enters the vehicle. Candidate energy replenishment spatiotemporal state set . When the value is empty, the vehicle will not participate in the candidate energy replenishment state allocation during the current rolling prediction period.

[0055] S22. Determination of the intensity of the energy replenishment intention:

[0056] Considering that not all surrounding vehicles require refueling, the intention to introduce refueling before load accumulation is emphasized. For vehicles whose refueling activity has not been confirmed. ,That Calculate using the following formula:

[0057] ;

[0058] In the formula, For vehicles The intensity participating in the superposition of future energy replenishment loads is limited to a value within [specific range]. ; The predicted energy required for a vehicle to travel from its current location to its destination or the next task point can be accumulated segment by segment along the current feasible path; This represents the vehicle's current remaining battery power. This is a minimum positive energy value, used to avoid invalidation of the denominator; This is the driving direction correction factor, and its value is limited to... . To restrict the results within the parentheses A cutoff function within a specified range.

[0059] The calculation is based on the degree of deviation between the vehicle's current direction of travel and the direction of the charging area entrance. Specifically, the directional quantity is calculated based on the vehicle's current direction of travel. The directional quantity is formed by the feasible road direction from the vehicle's current position to the entrance of the refueling area. .when and When the included angle is small, and reaching the entrance does not require a significant U-turn or a large detour, Take the higher value; when the vehicle's direction of travel is away from the entrance to the reachable charging area, or when the current remaining battery power is sufficient to complete the subsequent tasks. Take the lower value. In practical scenarios, the directional angles can be divided into several levels in ascending order. Then, the level values ​​are subtracted based on the costs of turning around and detouring. The subtracted result is truncated. Within the scope, vehicles that are already connected, queued, or have confirmed reservations will be treated as definitive energy replenishment activities. Take 1.

[0060] S23. Virtual arrival probability calculation:

[0061] For vehicles with candidate refueling status, Calculate the conditional probability of it entering different energy replenishment regions and time slices:

[0062]

[0063] ;

[0064] In the formula, For vehicles Under conditions with the intention of replenishing energy, in time slice Entering the energy replenishment area And form the conditional probability of energy replenishment demand; denominator traversal Candidate states within . For the vehicle to enter the status unit The virtual arrival cost. The normalized estimated travel time is denoted as... It is obtained by the ratio of the estimated arrival time to the predicted time domain length; the normalized arrival range risk is denoted as... ,by The corresponding safety margin is the input; the smaller the safety margin, the higher the risk value. When the safety margin reaches the preset redundancy limit, the risk value approaches zero. The normalized basic load state is denoted as... The current number of occupied charging piles, the number of queued vehicles, the number of vehicles with confirmed reservations, and the number of vehicles continuing to occupy the charging piles from the previous rolling cycle can be converted into station-end occupancy contributions, and then normalized according to the number of available charging piles. The conversion method is as follows: the currently connected and occupied charging piles are directly counted as 1 unit of occupancy contribution per pile; queued vehicles are converted in sequence according to the queuing order, and the equivalent occupancy contribution of a single queued vehicle is the proportion of the overlap between the vehicle's expected occupancy time and the current time slice, with the value limited to the range of 0 to 1; the equivalent occupancy contribution of vehicles with confirmed reservations is converted according to the proportion of the overlap between their reserved service time and the current time slice; the equivalent occupancy contribution of vehicles that continued to occupy the charging piles from the previous rolling cycle is converted according to the ratio of their remaining expected occupancy time to the length of the current time slice. The weights are dimensionless and adjusted according to vehicle task constraints. When task time is tight, the weight of the time item is increased; when the risk of low battery is high, the weight of the range risk item is increased; and when there is a queue in the charging area, the weight of the base load item is increased.

[0065] S24. Future Time Slice Load Calculation:

[0066] Once the vehicle arrives at the charging area, its charging process will span one or more time slices. If in time slice Entering the energy replenishment area Its charging process is projected onto a time slice. The expected energy replenishment is:

[0067] ;

[0068] In the formula, For vehicles In time slice The expected replenishment of energy formed internally; For vehicles In the energy replenishment area The expected charging power is an executable value that is constrained by the vehicle battery's acceptable power, the allocatable charging pile power, and the station's power limit. For vehicles In the energy replenishment area The estimated time period to be occupied; For time slices Time interval; For occupied time periods and time slices The duration of overlap.

[0069] The starting point is the estimated service start time. For vehicles already connected, this starting point is the actual connection time; for queued vehicles, this starting point is calculated based on the estimated departure time of the preceding vehicle and the release time of the available charging pile; for virtual charging vehicles, this starting point is the later of the vehicle's estimated arrival time and the available service gap time. The end point of the occupied period is the time when the vehicle has accumulated the target charging power; the occupied period ends when the vehicle's available power limit, the user-set departure power, or the minimum departure power required by the operational task is reached.

[0070] Recharge area In time slice The predicted energy demand for replenishment that the internal system needs to support is:

[0071] ;

[0072] In the formula, To predict the energy demand for replenishment; To predict the potential impact on the energy replenishment area within the time domain The load includes the target vehicle, potential nearby vehicles for refueling, vehicles already connected, vehicles in the queue, and vehicles with confirmed reservations; Intent to replenish vehicle power; The conditional probability of a vehicle entering the corresponding spatiotemporal state for energy replenishment when it has the intention to do so. The process of recharging the vehicle in time slices The energy projection in the image. For vehicles that have been confirmed to be occupied, the actual occupancy status can be projected. Set it to 1, and set the probability of other candidate states to 0.

[0073] The service load of the pile group is calculated using the following formula:

[0074] ;

[0075] In the formula, For energy replenishment area In time slice The service load of the pile group; For energy replenishment area A collection of internal charging stations; For charging piles In time slice The serviceability status is set to 1 when the service is available and 0 when the service is unavailable. For charging piles In time slice Effective service power; The time slice length; It represents an extremely small positive energy value.

[0076] The resource competition load in terms of vehicle quantity is:

[0077] ;

[0078] ;

[0079] In the formula, For vehicles In time slice The contribution of internal resources to the occupation of charging resources is limited to a value within... ; For time slices Internal energy replenishment area The number of available charging stations can be obtained by counting the number of charging stations that are available for service and not currently occupied. It is a very small positive number. Corresponding vehicle In time slice The expected contribution still occupies replenishment resources. (Through) It can identify the risk of competition for pile positions caused by concentrated vehicle arrivals when the energy demand for replenishment has not yet exceeded the service capacity of the pile group.

[0080] Step S3: Determination of the continuous energy replenishment domain under the constraints of range safety and load bearing.

[0081] After obtaining the future load, it is determined whether the target vehicle still possesses the conditions for energy security, service capacity, and vehicle competition when entering a certain refueling state. Next, it is determined whether multiple feasible states can be continuously connected along the time, road, and power supply relationships.

[0082] S31. Determining the Feasibility of Energy Supplementation:

[0083] For the target vehicle Calculate its arrival at candidate state unit The estimated remaining power at that time Detour costs Service load of pile groups and resource competition load Here and The prediction results include the projected load of the target vehicle. In other words, when determining whether the target vehicle can enter the state, the expected replenishment power and the occupied time period of the target vehicle are included in the corresponding time slice.

[0084] The feasibility of energy replenishment is determined by the following formula:

[0085] ;

[0086] In the formula, For the target vehicle The set of feasible states for energy replenishment; Minimum safe battery level upon arrival at the destination; The service load threshold for the charging pile group can be calculated based on the allowable queuing scale at the station, the number of charging piles that can be served, and the service saturation control requirements. The full-load output energy of all available charging piles in the replenishment area within a single time slice is used as the benchmark service capacity, with a corresponding benchmark load ratio of 1. Based on the allowable queuing vehicle scale at the station, the equivalent load ratio corresponding to a single queuing vehicle is calculated by the ratio of the typical target replenishment power of a single vehicle to the service power of a single pile in a single time slice. This ratio is then multiplied by the allowable queuing vehicle number to obtain the queuing equivalent load increment. Finally, a preset proportion of service redundancy is reserved according to the service saturation control requirements. The service load threshold for the charging pile group is obtained by subtracting the redundancy ratio from the sum of the benchmark load and the queuing equivalent load increment. The resource competition load threshold can be calculated based on the number of available charging piles and the allowable degree of simultaneous competition. The number of available charging piles in the corresponding time slot is used as the baseline load factor, combined with the allowable simultaneous competition coefficient set at the station. The allowable simultaneous competition coefficient represents the equivalent number of competing vehicles that a single available charging pile can withstand in a single time slot. Its value is determined based on the station's queuing tolerance and the fluctuation range of the average charging time occupied by vehicles, and the value is not less than 1. Multiply the number of available charging piles by the allowable simultaneous competition coefficient to obtain the total number of equivalent competing vehicles allowed in that time slot. Then divide this total number by the number of available charging piles to obtain the resource competition load threshold. This represents the maximum acceptable detour cost for the vehicle. The four criteria correspond to range safety, energy service capacity, vehicle competition capacity, and detour acceptability, respectively.

[0087] S32, Energy-supplemented continuous domain connection:

[0088] Whether two feasible states can be continuously connected depends on the simultaneous fulfillment of timing, power, and load conditions. For feasible states... and The succession relationship is determined by the following formula:

[0089] ;

[0090] In the formula, For state Able to continuously receive until the state ; The target vehicle has arrived. The expected time; For the target vehicle in the state The estimated departure time after the energy replenishment is completed is obtained from the estimated service start time and the time required to complete the target energy replenishment. To replenish energy from the region To the refueling area The estimated travel time; For the target vehicle in the state The estimated departure power after recharging is obtained by accumulating the arrival power and the planned recharge power, and then subjecting it to the upper limit of the vehicle's available power. For the target vehicle from the area Driving towards the area The projected energy consumption; and Entering the target vehicle state The subsequent service load and resource competition load of the pile group. The energy replenishment continuum consists of feasible states that satisfy the continuous bearing relationship. This continuum corresponds to the spatiotemporal range in which the target vehicle can continuously maintain energy security and obtain energy replenishment opportunities in the future predicted time domain.

[0091] Step S4: Continuous domain boundary correction driven by vehicle refueling competition relationship.

[0092] like Figure 2 As shown, the target vehicle's refueling continuum is affected by the virtual refueling behavior of surrounding vehicles. Some states are not yet in the queue at this stage, but will lose their carrying capacity in subsequent time slices due to the concentrated arrival of other vehicles. Therefore, it is necessary to correct the continuum boundary before the path output.

[0093] S41, Virtual energy replenishment behavior superposition:

[0094] For surrounding vehicles, the data includes remaining battery power, current route direction, destination, available charging areas, and the intensity of the charging intention. This generates virtual charging actions. Each virtual charging action includes at least a virtual arrival time slice, candidate charging areas, estimated charging capacity, estimated service start time, and estimated occupancy period. When a vehicle has explicitly selected a charging station, the corresponding status can be processed as a deterministic occupancy record; when the vehicle has not explicitly indicated its charging intention, it is processed through... Control its intensity of participation in future load superposition.

[0095] The aforementioned virtual behavior was substituted , and The calculation is as follows: Even if a certain state unit meets the endurance safety conditions, it will still lose its carrying capacity due to the overload of the charging pile group service load or the resource competition load exceeding the limit after the virtual load is superimposed. At this time, the state will no longer participate in the continuous domain connection. For states that have not yet actually queued but whose future service load continues to approach the threshold, their stability is reduced to prevent vehicles from entering the soon-to-be-shrinking energy replenishment space.

[0096] S42, State Margin Normalization:

[0097] To compare different sources of risk, the state unit The driving range margin, the charging pile service load margin, and the resource contention load margin are normalized:

[0098] ;

[0099] In the formula, For normalized range margin; To normalize the service load margin of the pile group; To normalize the resource competition load margin; State unit Energy replenishment stability; The target vehicle's available battery capacity is the maximum, and it meets the following requirements: ; and All are positive thresholds. The minimum value is used for calculation. This reflects the characteristic that the failure of the energy replenishment state is usually triggered by the weakest constraint.

[0100] S43. Continuous domain boundary correction:

[0101] in accordance with The boundary of the continuous domain is re-identified based on the relationship between feasible states. A certain state is located at a necessary transition point leading to the subsequent energy replenishment region, and its... When the elevation falls below the stability requirement, the location is marked as weak. The connection between two feasible regions is determined by… or When a condition disappears due to exceeding the limit, record the fracture location, fracture time slice, and main fracture cause. This record is only used in subsequent strategy calculations within the current rolling prediction cycle and is not retained as a fixed rule after the rolling cycle ends.

[0102] Step S5: Active generation of the energy replenishment feasible domain under load time-series shaping.

[0103] like Figure 3 As shown, when a continuous domain shrinks or breaks, path adjustment is not limited to replacing charging stations, but also includes rearranging vehicle arrival times, changing the access sequence of road segments, adjusting the target replenishment power, and transferring temporarily occupied loads.

[0104] S51, Actively Generated Trigger:

[0105] Active generation processing is initiated when the following conditions occur: the coverage area of ​​the recharge continuum is insufficient to support the current driving task; or the critical state units in the continuum... Below stability requirements; due to the distance between two feasible regions or Exceeding limits and breaking; the actual energy consumption of the target vehicle is higher than the predicted energy consumption, causing the safe power boundary to shift forward; the virtual energy replenishment behavior of surrounding vehicles is concentrated in the same energy replenishment area and similar arrival time windows; the similar arrival time windows refer to the time when multiple vehicles participating in load superposition are expected to arrive at the energy replenishment area, which falls within a preset number of consecutive time slices. In this embodiment, the preset number is 1 to 2, that is, the maximum difference between the expected arrival times of each vehicle is no more than 1 to 2 times the length of the time slice.

[0106] Identify the primary cause of the breakage after triggering. If the limit is exceeded, it indicates that the energy service capacity is insufficient. Priority should be given to adjusting the arrival time window or the target power replenishment. If the limit is exceeded, it indicates that the vehicles are arriving too densely. Priority should be given to adjusting the access order of road segments, candidate charging areas, or time periods occupied. When the power supply is insufficient, priority should be given to short-term and appropriate power replenishment to enable the vehicle to obtain the power supply capacity to enter the next feasible area.

[0107] S52, Generation of candidate active manufacturing strategies:

[0108] Candidate Active Manufacturing Strategy It includes at least one of the following control variables: target driving speed range, road segment access sequence, arrival window of candidate energy replenishment area, target energy replenishment capacity, and temporary load occupation.

[0109] When adjusting the arrival time window, for each segment of the candidate path, an executable speed range is selected within constraints of road speed limits, traffic safety, and driving comfort. The arrival time slice for the target vehicle at the refueling area is then recalculated. The adjusted arrival time slice avoids high-speed areas. or high When the interval is reached, the state re-enters the feasibility assessment.

[0110] When adjusting the access order of road segments, candidate road segment sequences that still meet the constraints of destination direction and detour cost are selected from the road topology. This sequence does not solely aim for the shortest distance, but rather ensures that the time slice for vehicles to enter the charging area falls within a window with low service load and low competing load. Candidate road segment sequences can be obtained by expanding adjacent road segments in the road topology; during the expansion process, road segment sequences that can reach the destination direction, do not exceed the maximum detour cost, and are expected to reach an energy level no lower than the safe energy level are retained, and the remaining sequences are not included in the strategy evaluation.

[0111] When adjusting the target replenishment capacity, the minimum available power required after departure is calculated in reverse from the subsequent available charging state. This minimum available power is determined by the energy consumption required for the target vehicle to travel from the current candidate charging area to the next available charging state, the minimum safe charging capacity upon arrival in the next state, and road detour redundancy. If the expected arrival power is lower than this minimum available power, the difference is used as the target replenishment capacity, and is constrained by the vehicle's battery charging range and the user's charging preferences. If the expected arrival power already meets the subsequent charging requirements, the target replenishment capacity can be processed according to the user's departure preferences or the station's minimum service requirements. This reduces the time occupied while ensuring the safety of the next segment's range.

[0112] When adjusting temporary load occupancy, the expected replenishment power and expected occupancy period of the target vehicle within the candidate arrival time window are projected onto... , and In the middle. When a candidate strategy is not adopted, the temporary projection within the corresponding time window is removed; when the path shifts or a more stable time window appears, the temporary occupancy load is transferred to a new candidate state.

[0113] S53. Evaluation of Active Manufacturing Strategy:

[0114] Candidate strategy The evaluation value is calculated using the following formula:

[0115] ;

[0116] In the formula, For proactive manufacturing strategy Evaluation value; To determine the improvement in stability of the continuous domain after implementing the strategy, the value within the continuous domain after manufacturing can be taken. The difference between the low quantile value and the corresponding low quantile value before manufacturing; The improvement in connectivity of the feasible fracture region can be obtained by normalizing the difference in duration of the longest acceptable state sequence before and after manufacturing according to the predicted time domain length. The normalized value for the additional detour distance can be obtained by dividing the detour distance introduced by the strategy by the maximum allowed detour distance; The arrival time window offset normalized value can be obtained by dividing the arrival time offset by the acceptable time window width; The normalized value for adjusting the power supply or temporarily occupying the load can be obtained by dividing the adjustment amount by the target power supply demand of the vehicle. and It is a dimensionless weight. When operational tasks are urgent, it can be increased. Corresponding weight; when the pile group is under heavy load, it can be increased. Corresponding weight; when the risk of low battery is high, it can be increased. Corresponding weights. All candidate strategies use the same set of weights and the same normalization method within the same rolling period to avoid inconsistencies in evaluation criteria between different strategies.

[0117] After selecting a strategy that meets the evaluation requirements and can restore the continuous connection relationship, recalculate the target vehicle's estimated arrival time, target replenishment power, estimated occupation period, and temporary load occupation. (Original state due to...) When exceeding the limit is not feasible, shifting the arrival time window later can reduce the service load of that time slice to within the threshold; the original state is due to When exceeding the limit is not feasible, shortening the target replenishment power can reduce the occupied period and bring the vehicle number competition load back below the threshold. After this processing, the post-manufacturing replenishment continuous domain is obtained.

[0118] Step S6: Solve for the load replenishment potential function and directional gradient.

[0119] In the post-engineering energy replenishment continuum, the selection of the next travel direction is affected not only by distance and travel time, but also by whether the energy replenishment state remains stable after the vehicle enters that direction. Therefore, a conservative evaluation of the potential energy in candidate directions is performed to avoid individual high-potential-energy states masking weak points in the path.

[0120] S61. Calculation of the state potential function:

[0121] For state units Calculate the risk of load contraction in the neighborhood and the potential energy for load replenishment:

[0122] ;

[0123] ;

[0124] In the formula, State unit The risk of load shrinkage; State unit The set of states of adjacent time slices that are later in time and adjacent energy replenishment areas that can be received on the road; Adjacent states The service load of the pile group; Current state The service load of the pile group; State unit The load replenishment potential energy; , and These are respectively: range margin, charging pile service load margin, and resource contention load margin; and The weights are dimensionless. When the service load of the pile group in an adjacent future state is higher than that in the current state, rise, Lowering the vehicle makes it less likely to be guided to a refueling space where instability is imminent. If If there is no acceptable state, then that state will not participate in the calculation of the positive potential energy increase of the candidate direction, or it will be treated as a high-load contraction risk.

[0125] S62. Calculation of gradient in candidate directions:

[0126] Starting from the target vehicle's current position, several candidate driving directions are generated along the road topology. Each candidate direction... Corresponding short-time trajectory segment This trajectory segment can be formed by adjacent road segments or several consecutive reachable road segments. When mapping to state units in the post-manufacturing energy replenishment continuum, the corresponding time slice is matched according to the vehicle's expected arrival time along the trajectory segment, and the corresponding energy replenishment region is matched according to the reachability relationship between the trajectory segment and the entrance road of the energy replenishment region. Trajectory segments that cannot be mapped to the post-manufacturing energy replenishment continuum are not included in the high potential energy direction comparison, or are processed as low potential energy states.

[0127] The potential energy gradient of the load replenishment is calculated using the following formula:

[0128] ;

[0129] In the formula, Candidate driving directions The load replenishment potential energy gradient; This is the state unit corresponding to the current position of the target vehicle. For low quantile operators, low quantile proportion Set according to the degree of path conservatism; This is a conservative representative value of the load replenishment potential energy in the candidate trajectory segment; The normalized road segment cost can be obtained by normalizing the travel distance, estimated travel time, or a combination of both for the candidate directions. It is a very small positive number. The higher the value, the easier it is for the vehicle to enter a stable energy replenishment continuum after traveling along that candidate direction; When the value is negative, this direction will cause the vehicle to approach areas with low range margin, high service load, or high competition load.

[0130] Step S7: Rolling closed-loop decision-making regarding the energy replenishment path, arrival window, and replenishment capacity.

[0131] During vehicle operation, the load in the energy replenishment area, road conditions, and the vehicle's actual battery level will change. The output path, arrival window, and replenishment power are updated in a rolling manner to avoid the failure of a fixed path due to subsequent load changes.

[0132] S71, Route and Refueling Plan Output:

[0133] Based on candidate directions The system generates a continuous charging domain and defines the current driving task, outputting a dynamic driving path, target charging area, target arrival time window, suggested speed range, and target charging capacity. The target charging area can be a single charging station or multiple charging areas with an interchangeable relationship. The target arrival time window constrains the time when the vehicle enters the charging area, ensuring it avoids time slots with high service load and resource contention load. The target charging capacity ensures the vehicle meets the range safety boundary for the next segment and reduces the possibility of prolonged single-time occupation of charging resources.

[0134] Multiple candidate directions Upon approach, continue comparing detour costs, arrival window offset, and disturbances to the load in the energy replenishment area. If none of the candidate directions can maintain the stability of the energy replenishment continuum, return to S5 to re-execute the active generation of the energy replenishment feasible domain.

[0135] S72, Rolling Updates and Anomaly Corrections:

[0136] During vehicle operation, it continuously receives information on actual location, actual battery level, actual speed, road conditions, and charging resource status. If the actual battery level is lower than the predicted level, it recalculates. It also checks whether the refueling continuum has shrunk due to the forward shift of the range safety boundary. If traffic changes cause the estimated arrival time to deviate from the target arrival window, it recalculates. , , and When the virtual charging behavior of surrounding vehicles is concentrated in the same charging area, the boundary of the continuous domain is revised, and it is determined whether the temporary load occupied by the target vehicle needs to be released or transferred.

[0137] When the target vehicle is still within the post-manufacturing refueling continuum, only minor adjustments are made to the speed range, arrival window, or target refueling capacity. When the target vehicle is about to leave the refueling continuum, or when a critical state unit within the continuum... When the battery level falls below stability requirements, the active manufacturing strategy is re-executed to enable the vehicle to obtain a new rechargeable state before the low battery warning. If the original target recharge area fails due to load impact, a backup recharge area that is connected to the current recharge continuum is selected first to avoid static screening from the global site again.

[0138] Through the detailed description of the above embodiments, the present invention provides a vehicle-pile spatiotemporal load prediction-based energy replenishment path optimization method. This method incorporates the target vehicle state, road state, energy replenishment resource state, and surrounding vehicle state into a vehicle-pile-road spatiotemporal state unit composed of the energy replenishment area and time slice, enabling spatiotemporal prediction of energy replenishment demand, pile group service load, and resource competition load. Based on this, the method determines the feasibility of energy replenishment by combining the expected arrival power, detour cost, and load carrying capacity, and forms a continuous energy replenishment domain through continuous connection relationships. Furthermore, the present invention superimposes the virtual energy replenishment behavior of surrounding vehicles into the load calculation process, corrects the boundary of the continuous energy replenishment domain, and generates candidate active manufacturing strategies by controlling variables such as arrival time window, road segment access sequence, target energy replenishment power, and temporary load occupation when the continuous domain shrinks, breaks, or lacks stability. Subsequently, the load energy replenishment potential gradient is calculated based on the post-manufacturing continuous energy replenishment domain, and is continuously updated during vehicle travel by combining actual location, actual power, road state, and energy replenishment resource state, thereby forming a collaborative decision-making process for energy replenishment path, target energy replenishment area, arrival time window, and energy replenishment power.

[0139] The preset parameters in the above formulas shall be set by those skilled in the art according to the actual situation.

[0140] 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 wired or wireless (e.g., infrared, wireless, microwave, etc.) means. 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 that includes 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.

[0141] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0142] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0143] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0144] 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 units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units 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 units may be electrical, mechanical, or other forms.

[0145] If the aforementioned functions are implemented as software functional units 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.

[0146] 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.

Claims

1. A method for optimizing energy replenishment paths based on vehicle-pile spatiotemporal load prediction, characterized in that, The method flow is as follows: Step S1: Obtain the status of the target vehicle, road, energy replenishment resources and surrounding vehicles, and construct a vehicle-pile-road spatiotemporal status unit according to the energy replenishment area and time slice; Step S2: Based on the vehicle-pile-road spatiotemporal state unit, predict the energy demand for energy replenishment, the service load of the pile group, and the resource competition load. Step S3: Determine the energy replenishment feasibility state based on the target vehicle's expected energy level, detour cost, charging pile service load, and resource competition load, and connect the energy replenishment feasibility states that meet the requirements of timing, energy level, and load continuity into a continuous energy replenishment domain. Step S4: Generate virtual energy replenishment behavior of surrounding vehicles, superimpose the virtual energy replenishment behavior onto the vehicle-pile-road spatiotemporal state unit, and correct the boundary of the energy replenishment continuous domain based on the superimposed load state. Step S5: When the corrected energy replenishment continuous domain meets the active generation triggering condition, a candidate active manufacturing strategy is generated, and the expected arrival time of the target vehicle, the target energy replenishment capacity, the expected occupation period and the temporary occupation load are recalculated according to the candidate active manufacturing strategy to obtain the post-manufacturing energy replenishment continuous domain. Step S6: Calculate the load replenishment potential energy gradient of the candidate driving direction based on the post-manufacturing replenishment continuous domain; Step S7: Output the dynamic driving path, target energy replenishment area, target arrival time window, and target energy replenishment amount based on the load replenishment potential energy gradient.

2. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 1, characterized in that, In step S4, the virtual charging behavior is generated based on the remaining battery power of surrounding vehicles, current route direction, destination direction, reachable charging area, and charging intention intensity. Each virtual charging behavior includes at least a virtual arrival time slice, candidate charging area, expected charging power, expected service start time, and expected time period.

3. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 2, characterized in that, In step S4, the virtual energy replenishment behavior is substituted into the calculation of energy replenishment demand, pile group service load and resource competition load; after superimposing the virtual load, the state that loses its bearing capacity due to the pile group service load or resource competition load exceeding the limit will no longer participate in the continuous domain connection, and the boundary of the energy replenishment continuous domain will be re-identified based on the energy replenishment stability and feasible state acceptance relationship.

4. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 3, characterized in that, In step S4, when the state is located at the necessary connection point leading to the subsequent energy replenishment area and its energy replenishment stability drops below the stability requirement, the state is marked as a weak state; when the connection between two feasible areas disappears due to the excessive load of the pile group service load or the resource competition load, the fracture location, fracture time slice and main fracture cause are recorded.

5. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 4, characterized in that, In step S5, the active generation triggering conditions include: the coverage of the continuous energy replenishment domain is insufficient to support the current driving task; the energy replenishment stability of the key state unit in the continuous domain is lower than the stability requirement; the two feasible areas are disconnected due to the overload of the pile group service load or the resource competition load exceeding the limit; the actual energy consumption of the target vehicle is higher than the predicted energy consumption and the safe power boundary is moved forward; or the virtual energy replenishment behavior of surrounding vehicles is concentrated in the same energy replenishment area and the same arrival time window.

6. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 5, characterized in that, In step S5, the main cause of the breakage is identified after the active generation process is triggered; when the service load of the pile group exceeds the limit, the arrival time window or target replenishment power is adjusted first; when the resource competition load exceeds the limit, the access order of the road section, the candidate replenishment area or the occupied time period is adjusted first; when the expected arrival power is insufficient, short-term and appropriate replenishment is arranged first.

7. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 6, characterized in that, In step S5, the candidate active manufacturing strategy includes at least one control quantity among the target driving speed range, road segment access sequence, candidate energy replenishment area arrival window, target energy replenishment capacity, and temporary load occupation.

8. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 7, characterized in that, In step S5, when adjusting the arrival time window of the candidate energy replenishment area, an executable speed range is selected within the constraints of road speed limit, traffic safety and driving comfort, and the time slice for the target vehicle to arrive at the energy replenishment area is recalculated; when adjusting the access sequence of road segments, a candidate road segment sequence that meets the constraints of destination direction and detour cost is selected in the road topology, so that the time slice for the vehicle to enter the energy replenishment area falls into the time window with low service load and low competition load.

9. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 7, characterized in that, In step S5, when adjusting the target replenishment power, the minimum available power required after leaving the station is calculated in reverse from the subsequent available state, and the target replenishment power is determined accordingly; when adjusting the temporary load, the expected replenishment power and expected occupancy period of the target vehicle in the candidate arrival time window are projected into the replenishment demand energy, pile group service load and resource competition load, and the corresponding temporary projection is removed or transferred when the candidate strategy is not adopted or the path is deviated.

10. The energy replenishment path optimization method based on vehicle-pile spatiotemporal load prediction according to claim 7, characterized in that, In step S7, the vehicle continuously receives the actual location, actual battery level, actual speed, road conditions, and energy replenishment resource status during its journey. When the actual battery level is lower than the predicted battery level, the estimated arrival time deviates from the target arrival window, the virtual energy replenishment behavior of surrounding vehicles is concentrated in the same energy replenishment area, or the target vehicle is about to leave the energy replenishment continuum, the energy replenishment continuum boundary correction and active manufacturing strategy are re-executed.

Images