An intelligent scheduling and processing method and system for charging piles

Abstract

The present application provides a method and system for intelligent scheduling and processing of charging piles. Among them, by integrating the remaining power of the vehicle, the occupancy status of the charging pile, and the vehicle type priority label, a dynamic charging demand network topology is constructed. The vehicle position and charging pile status data are updated in real time to predict the vehicle movement trend and generate a preemption conflict warning signal. Based on the multi-agent coordination and allocation mechanism, the matching degree score between the vehicle and the charging pile is calculated in combination with the remaining power decay rate and the departure time constraint, and the allocation strategy is dynamically optimized. Finally, the power distribution gradient is adjusted according to the score ranking, so that the vehicle with high power decay can obtain a higher power boost, realizing efficient resource matching and conflict mitigation. The technical solution provided by the present application intelligently optimizes the charging pile allocation and power adjustment through dynamic prediction and priority matching, significantly improving the charging efficiency of the parking lot and reducing vehicle waiting conflicts.

Description

Technical Field

[0001] This application relates to the technical field of intelligent charging scheduling and energy management, and particularly to a method and system for intelligent scheduling and processing of charging piles. Background Art

[0002] With the popularization of electric vehicles, high-density charging scenarios (such as large parking lots and charging stations) have put forward higher requirements for the collaborative scheduling and energy efficiency optimization of charging pile clusters. Traditional decentralized charging management is difficult to meet the dynamically changing charging demands. There is an urgent need for an intelligent scheduling method that can analyze vehicle charging demands in real time, optimize the allocation of charging piles, and improve the overall energy utilization efficiency to reduce queuing time and avoid waste of charging pile resources.

[0003] Currently, some charging pile management systems adopt a fixed allocation strategy based on static priorities, that is, the priorities are preset according to the remaining battery levels of vehicles or the urgency of charging, and are matched with the real-time occupancy status of charging piles. For example, some systems will preferentially allocate charging piles to vehicles with lower battery levels, but only rely on a single indicator for decision-making, without considering factors such as the dynamic movement trends of vehicles and the adaptive adjustment of charging power.

[0004] However, although this solution is more efficient than manual scheduling, it still has obvious deficiencies: relying only on static priorities, it cannot respond in real time to vehicle movement and changes in charging demands, which easily leads to local congestion or idleness of charging piles. Without adjusting the power allocation in combination with dynamic factors such as vehicle departure time and battery degradation rate, it may result in low charging efficiency or energy waste. Without establishing a charging demand prediction mechanism, it is difficult to detect potential charging pile preemption conflicts in a timely manner, affecting the user experience. Summary of the Invention

[0005] This application provides a method and system for intelligent scheduling and processing of charging piles to solve the problem of low charging efficiency or energy waste in the prior art.

[0006] In a first aspect, this application provides a method for intelligent scheduling and processing of charging piles, including:

[0007] By fusing the remaining battery level data of multiple vehicles read by devices when entering the target parking lot and the occupancy status data of charging piles, and associating the charging priority tags of different vehicle models, a charging demand network topology is established within the target parking lot;

[0008] By real-time updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology to determine the vehicle movement trajectories of multiple vehicles, and determining the motion trend prediction results of multiple vehicles according to the vehicle movement trajectories to generate a warning signal for charging pile preemption conflicts;

[0009] Input the warning signal into the multi-agent coordination and allocation mechanism, and execute a dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle to the target charging pile. The matching degree score is corrected by the product factor of the remaining power decay rate and the vehicle departure time constraint condition;

[0010] Adjust the power allocation gradient of the charging pile according to the sorting result of the matching degree score, where the increase amplitude of the power allocation gradient corresponding to the vehicle and the charging pile is positively correlated with the square root of the remaining power decay rate.

[0011] Optionally, according to the position data of the vehicle nodes in the charging demand network topology, extract the moving direction change amount, moving speed change amount, and continuous increment of the moving speed change of each vehicle in the target parking lot, and combine the position relationship between the vehicle nodes and the charging pile nodes to generate the vehicle moving trajectory corresponding to each vehicle;

[0012] Based on the continuous position change sequence in the vehicle moving trajectory, for each vehicle node, according to the corresponding current moving speed change amount and continuous increment, calculate the first estimated time interval for the vehicle node to reach each charging pile node in the charging area along the current moving direction. The starting point of the first estimated time interval is the minimum arrival time based on the current moving speed, and the end point is the maximum arrival time based on the maximum speed change trend corresponding to the continuous increment;

[0013] Based on the first estimated time interval of each vehicle node, construct a motion trend prediction model for multiple vehicles, and analyze the trajectory crossing probability and time overlap probability of each vehicle node in the charging area through the motion trend prediction model to predict the interaction influence intensity between multiple vehicle nodes as the motion trend prediction result of multiple vehicles;

[0014] Detect conflicts in the case where multiple vehicle nodes select the same target charging pile according to the motion trend prediction result, and generate a warning signal including the charging demand urgency parameter.

[0015] Optionally, for each vehicle node, screen the charging pile node with the smallest starting point of its first estimated time interval as the target charging pile, and record the first estimated time interval as the pre-occupation time window of the vehicle node for the target charging pile;

[0016] For the case where multiple vehicle nodes select the same target charging pile, extract the pre-occupation time windows of each vehicle node, and combine the motion trend prediction results of multiple vehicles. If there is an overlapping part in the pre-occupation time windows of at least two vehicle nodes and the interaction influence coefficient exceeds the set threshold, it is determined that there is a preemption conflict for the target charging pile;

[0017] For a target charging pile with preemption conflicts, calculate the charging demand urgency parameter for each conflicting vehicle node according to the real-time moving speed change amount, continuous increment of each conflicting vehicle node, the interaction influence intensity of the movement trends of each conflicting vehicle node, and the corresponding vehicle type charging priority label. The charging demand urgency parameter is the weighted value of the remaining power decay rate of the vehicle node, the movement trend interaction influence coefficient, and the vehicle type charging priority label;

[0018] Generate a warning signal including the target charging pile identifier, the overlapping range of the preoccupation time window, the interaction influence intensity of the movement trend, and the charging demand urgency parameter of the corresponding conflicting vehicle node.

[0019] Optionally, input the warning signal into a multi-agent coordination and allocation mechanism. Based on the multi-agent coordination and allocation mechanism and according to the preemption conflict relationship between the vehicle and the target charging pile in the warning signal, establish a conflict handling group for each target charging pile. The conflict handling group includes the remaining power decay rate of the vehicle in conflict with the current charging pile, the preset departure time constraint of the vehicle, and the current available power of the charging pile;

[0020] Execute dynamic parameter adjustment on the vehicles in the conflict handling group through the multi-agent coordination and allocation mechanism. The dynamic parameter adjustment includes adjusting the remaining power decay rate of the vehicle in a reverse correlation manner according to the current available power of the charging pile, and at the same time converting the remaining duration of the vehicle departure time constraint into an urgency level according to a preset interval;

[0021] Generate a matching score of the vehicle for the target charging pile according to the adjusted remaining power decay rate and the adjusted urgency level.

[0022] Optionally, fuse the remaining power data of all vehicles in the target parking lot and the current occupancy status data of all charging piles to generate a fusion data set including the real-time status of vehicles and charging piles;

[0023] Based on the vehicle type identifier corresponding to each vehicle in the fusion data set, match a preset charging priority label for each vehicle. The charging priority label is divided into at least three levels according to vehicle type attributes;

[0024] Associate the matched charging priority label with the remaining power of the vehicle and the occupancy status of the charging pile in the fusion data set to generate a target data set including vehicle charging demand characteristics and charging pile adaptation conditions;

[0025] Construct a charging demand network topology according to the vehicle charging demand characteristics and charging pile adaptation conditions in the target data set.

[0026] Optionally, a demand node is generated according to the remaining power of the vehicle and the charging priority label, and an adaptation node is generated according to the occupancy status of the charging pile and the charging priority label;

[0027] Perform priority matching between the demand node and the adaptation node. The priority matching process includes generating a connection channel between the two if the priority range supported by the adaptation node includes the priority label of the demand node, and assigning an initial strength value to the connection channel according to the remaining power value of the demand node;

[0028] Sort the connection channels of multiple demand nodes connected to the same adaptation node according to the initial strength value, and only retain the connection channel with the highest initial strength value;

[0029] Construct a charging demand network topology based on the retained connection channels, the demand nodes, and the adaptation nodes.

[0030] Optionally, according to the sorting result of the matching degree score, generate a vehicle priority sequence from high to low according to the matching degree score, and extract the total allocable power of the current charging pile, so as to allocate a reference power value to each vehicle in the vehicle priority sequence according to the total allocable power;

[0031] Extract the remaining power decay rate of each vehicle through the multi-agent coordinated allocation mechanism, and calculate the square root value of the remaining power decay rate as the rate adjustment base;

[0032] According to the position of the vehicle in the priority sequence, assign a position weight coefficient to each vehicle, and multiply the rate adjustment base by the position weight coefficient to generate a dynamic adjustment coefficient for the vehicle individual;

[0033] Multiply the reference power value by the dynamic adjustment coefficient through the multi-agent coordinated allocation mechanism to generate the actual allocated power value of the vehicle;

[0034] According to the constraint condition that the sum of the actual allocated power values of all vehicles does not exceed the total allocable power of the charging pile, adjust the power allocation gradient of the charging pile proportionally.

[0035] In a second aspect, the present application provides a charging pile intelligent scheduling processing system, including:

[0036] An association module, which establishes a charging demand network topology in a target parking lot by fusing the remaining power data of multiple vehicles and the occupancy status data of the charging piles, and associating the charging priority labels of different vehicle models;

[0037] An early warning module, by updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology in real time, to determine the vehicle movement trajectories of multiple vehicles, and determine the motion trend prediction results of multiple vehicles according to the vehicle movement trajectories, so as to generate an early warning signal for charging pile preemption conflicts;

[0038] A calculation module, input the early warning signal into a multi-agent coordination and allocation mechanism, and execute a dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle to the target charging pile, and the matching degree score is corrected by the product factor of the remaining power decay rate and the vehicle departure time constraint condition;

[0039] An adjustment module, adjusts the power allocation gradient of the charging pile according to the sorting result of the matching degree score, wherein the increase amplitude of the power allocation gradient corresponding to the vehicle and the charging pile is positively correlated with the square root of the remaining power decay rate.

[0040] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a charging pile intelligent scheduling processing method as described in the first aspect above.

[0041] In a fourth aspect, an embodiment of the present application provides a computer storage medium, storing a computer program, and when the computer program is executed by a computer, it implements a charging pile intelligent scheduling processing method as described in the first aspect.

[0042] In the embodiment of the present application, by fusing the remaining power data of multiple vehicles and the occupancy status data of charging piles, and associating the charging priority tags of different vehicle models, a charging demand network topology is established in the target parking lot; by updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology in real time, to determine the vehicle movement trajectories of multiple vehicles, and determine the motion trend prediction results of multiple vehicles according to the vehicle movement trajectories, so as to generate an early warning signal for charging pile preemption conflicts; input the early warning signal into a multi-agent coordination and allocation mechanism, and execute a dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle to the target charging pile, and the matching degree score is corrected by the product factor of the remaining power decay rate and the vehicle departure time constraint condition; adjust the power allocation gradient of the charging pile according to the sorting result of the matching degree score, wherein the increase amplitude of the power allocation gradient corresponding to the vehicle and the charging pile is positively correlated with the square root of the remaining power decay rate.

[0043] The technical solution of the present application has the following beneficial effects:

[0044] This application constructs a charging demand network topology by integrating the remaining power of multiple vehicles and the occupancy status data of charging piles and associating with vehicle type priority tags, realizing the global visualization and dynamic association of charging resources; predicts the vehicle movement trend based on the real-time updated vehicle position and charging pile status data and generates a preemption conflict warning signal to improve the system response ability; further inputs the warning signal into a multi-agent coordinated allocation mechanism, and realizes the intelligent matching of charging piles through the matching degree score that combines the remaining power attenuation rate and the departure time constraint; finally, dynamically adjusts the power distribution gradient according to the score ranking, enabling vehicles with high power attenuation to obtain a power increase positively correlated with the square root of the attenuation rate, thereby optimizing the energy utilization rate while ensuring the charging efficiency, and realizing the collaborative scheduling and energy efficiency improvement of the charging pile cluster.

[0045] Furthermore, based on the vehicle position data in the charging demand network topology, extract the moving direction change amount, speed change amount and continuous increment, and generate the vehicle movement trajectory in combination with the position relationship between the vehicle and the charging pile; by analyzing the continuous position change sequence, calculate the time interval (minimum to maximum arrival time) for each vehicle to reach each charging pile at the current speed and the maximum speed change trend, and construct a multi-vehicle movement trend prediction model to evaluate the trajectory intersection probability and time overlap probability to quantify the intensity of the interaction of movement trends; finally, detect the conflict of multiple vehicles selecting the same charging pile based on the prediction result and generate a warning signal containing the charging demand urgency parameter. Realize the early identification and hierarchical warning of charging conflicts, provide decision-making basis for dynamic scheduling in combination with the urgency parameter, effectively reduce the risk of charging pile preemption, and improve the allocation efficiency of charging resources in the parking lot and user satisfaction.

[0046] These aspects or other aspects of this application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0048] Figure 1 Shows a flowchart of a method for intelligent scheduling and processing of charging piles provided by this application;

[0049] Figure 2 Shows a schematic structural diagram of a system for intelligent scheduling and processing of charging piles provided by this application;

[0050] Figure 3 Shows a schematic structural diagram of a computing device provided by this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] To enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this application.

[0052] In some processes described in the specification and claims of this application and the above-mentioned accompanying drawings, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear in this article or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0053] Researchers have found that existing charging pile scheduling systems have problems such as unreasonable resource allocation, lagging conflict warnings, and lack of pertinence in power regulation when dealing with the dynamic charging demands of high-density parking lots, resulting in low charging efficiency and poor user experience. Based on this, a method for intelligent scheduling and processing of charging piles is provided. This method can construct a dynamic charging demand network topology, predict the movement trend of vehicles and generate conflict warning signals, and then combine a multi-agent coordinated allocation mechanism to achieve precise matching and power optimization, significantly improving the utilization rate of charging resources and scheduling efficiency. The technical solution of this application is applicable to high-density electric vehicle charging scenarios such as large parking lots and charging stations, and is particularly suitable for intelligent management scenarios that need to balance the charging demands of multiple vehicles and limited charging resources.

[0054] The entire R & D process reflects a series of progressive technical steps of "data fusion - dynamic prediction - intelligent matching - power optimization", aiming to overcome problems such as static resource allocation, lagging conflict response, and extensive energy efficiency regulation in existing methods. To meet the high standards of intelligent, efficient, and refined management of charging facilities in modern intelligent transportation systems.

[0055] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of this application.

[0056] Figure 1 A flowchart of a method for intelligent scheduling and processing of charging piles is provided for the embodiments of this application, as Figure 1As shown in the figure, the method includes:

[0057] 101. Establish a charging demand network topology in the target parking lot by fusing the remaining power data of multiple vehicles read by the device when entering the target parking lot with the occupancy status data of charging piles, and associating the charging priority tags of different vehicle models.

[0058] In this step, the remaining power data refers to the current battery capacity percentage data of electric vehicles collected in real time by on-vehicle sensors.

[0059] The occupancy status data is the current usage status information of each charging pile obtained through the charging pile management system, including statuses such as idle, charging, and faulty.

[0060] The charging priority tag is an identification of the charging urgency level preset according to the vehicle type. Special vehicles such as taxis and ambulances have a higher priority.

[0061] The target parking lot refers to a specific parking lot area that needs to optimize the charging scheduling.

[0062] The charging demand network topology is a network structure formed by modeling vehicle nodes, charging pile nodes, and their association relationships. This topology structure reflects the supply and demand relationship of charging resources in the current parking lot.

[0063] In the embodiment of this application, first, the RFID device at the parking lot entrance reads the vehicle identity information (such as license plate number), and synchronously calls the on-vehicle battery management system interface to obtain the real-time remaining power data of the vehicle (such as the battery remaining 20%) and the power attenuation rate data (such as decreasing by 0.8% per minute). Secondly, the occupancy status data of the charging pile is collected through the Internet of Things sensor of the charging pile (such as charging pile A is currently idle, and the expected release time is 15 minutes), and the charging pile location distribution map is generated by associating with the parking lot map coordinates. Then, based on the vehicle type database (such as ambulances, taxis, private cars), charging priority tags are bound to the vehicles (priorities are 5, 3, 1 respectively), and a network topology with vehicles and charging piles as nodes and charging demand relationships as edges is constructed through the graph database (Neo4j). The edge weight is calculated by the formula weight = charging priority tag × reciprocal of remaining power × power attenuation rate (such as ambulance weight = 5 × 1 / 20 × 0.8 = 0.2). Finally, the dynamically updated charging demand network topology map is output, providing the basis of vehicle remaining power, power attenuation rate, charging priority tag, and charging pile location data for the vehicle movement trajectory prediction in step 102.

[0064] In the underground parking lot of a large commercial complex, the intelligent charging scheduling system collects the remaining power data of 50 new energy vehicles in real time and combines it with the occupancy status of 12 fast charging piles (6 of which are idle). The system associates the vehicle type database through the vehicle identification code and automatically marks the charging priorities of different vehicles (for example, operating vehicles are given priority when their power is below 30%, and the priority of private cars is increased when their power is below 20%). Based on these data, the system constructs a dynamic charging demand network, where vehicles and charging piles are used as nodes respectively, and calculates the matching weights according to factors such as vehicle power, priority, and distance to form a real-time adjustable charging resource allocation model.

[0065] 102. By updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology in real time, determine the vehicle movement trajectories of multiple vehicles, and based on the vehicle movement trajectories, determine the prediction results of the movement trends of multiple vehicles to generate a warning signal for charging pile preemption conflicts;

[0066] In this step, the vehicle node refers to the entity node representing each electric vehicle in the charging demand network topology.

[0067] The charging pile node refers to the entity node representing each charging pile facility in the charging demand network topology.

[0068] The position data is the real-time coordinate information of the vehicle obtained through in-vehicle GPS or the in-site positioning system.

[0069] The status data includes real-time parameters such as the power output of the charging pile and the estimated full charge time.

[0070] The vehicle movement trajectory is the driving path of the vehicle obtained by analyzing the position data of consecutive time series.

[0071] The prediction result of the movement trend is the future movement direction and speed of the vehicle output by the prediction model established based on historical trajectory data.

[0072] The warning signal for charging pile preemption conflicts is a warning message generated when it is predicted that multiple vehicles will compete for the same charging pile simultaneously, including the pre-occupation time window of the target charging pile and the charging demand urgency parameter of the corresponding vehicle type.

[0073] In the embodiment of the present application, first, based on the charging pile location data in step 101, the vehicle coordinate data is collected in real time through the parking lot ultra-wideband positioning base station (for example, the coordinate of vehicle B is (30, 50)), and the Kalman filter is used to eliminate the positioning jitter to generate smooth vehicle movement trajectory data (for example, vehicle B moves from (30, 50) to (32, 53) within 10 seconds). Secondly, the vehicle movement trajectory data is input into the pre-trained long short-term memory network model to predict the target charging pile and arrival time of the vehicle in the next 5 minutes (for example, vehicle B is expected to arrive at charging pile A in 3 minutes), and the movement trend prediction results of multiple vehicles are determined. Then, combined with the power attenuation rate data in step 101, the estimated remaining power of the vehicle when it arrives at the target charging pile is calculated (for example, the current remaining power of vehicle B is 20%, the power attenuation rate is 0.8% per minute, and the remaining power drops to 17.6% after 3 minutes). If the estimated remaining power is lower than the preset threshold (such as 15%) or the time difference from the arrival time of the conflicting vehicle is less than 30 seconds, a charging pile preemption conflict warning signal is triggered (such as generating a red warning). Finally, the warning signal is bound to the associated vehicle identity information, charging pile identity information, and remaining power risk level, and pushed to the multi-agent coordination and allocation mechanism module in step 103.

[0074] The system uses the precise positioning system of the parking lot to update the vehicle position every 5 seconds and combines the historical trajectory to predict the driving trend in the next few minutes. When it detects that multiple high-priority vehicles are driving towards the same charging pile at the same time and the expected arrival times are close, the system automatically triggers a warning signal. For example, when two operation vehicles with insufficient power will arrive at the same charging pile successively within 30 seconds, the system marks the potential conflict in advance and analyzes the charging piles that will become idle nearby to provide decision-making support for subsequent scheduling.

[0075] 103. Input the warning signal into the multi-agent coordination and allocation mechanism, and execute the dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle to the target charging pile. The matching degree score is corrected by the product factor of the remaining power attenuation rate and the vehicle departure time constraint condition;

[0076] In this step, the multi-agent coordination and allocation mechanism refers to an intelligent scheduling algorithm for handling the competition of multiple vehicles for charging resources.

[0077] The dynamic balance allocation strategy is a decision-making method for adjusting resource allocation according to the real-time changing supply and demand relationship.

[0078] The matching degree score is a numerical index that quantifies the adaptation degree between the vehicle and the charging pile.

[0079] The remaining power attenuation rate is the power consumption value per unit time calculated by monitoring the power change trend.

[0080] The vehicle departure time constraint refers to the time requirement set by the vehicle owner for completing charging and leaving the parking lot.

[0081] In the embodiments of the present application, first, based on the charging priority label, power attenuation rate data in step 101 and the charging pile preemption conflict warning signal in step 102, extract the remaining power data of the conflicting vehicles (such as vehicle B with a remaining power of 17.6%), power attenuation rate data (such as 0.8% per minute), charging priority label (such as priority 3), and the user's preset departure time constraint (such as leaving within 30 minutes). Secondly, based on the multi-agent coordination and allocation mechanism, calculate the matching degree score of the vehicle to the target charging pile through the dynamic balance allocation strategy. The formula is: matching degree score = remaining power attenuation rate × departure time constraint factor × charging priority label + remaining power urgency factor. Among them, the product factor of the vehicle departure time constraint condition = 1 / remaining minutes of departure time, and the remaining power urgency factor = (1 - current remaining power / target power) × 10 (for example, if vehicle B needs to leave after charging to 80% power, then the remaining power urgency factor = (1 - 17.6 / 80) × 10 ≈ 7.8, and the matching degree score = 0.8 × (1 / 30) × 3 + 7.8 ≈ 7.88). Then, use the Hungarian algorithm to perform optimal allocation on the matching degree score matrix of the conflicting vehicles (such as vehicle B is preferentially allocated to charging pile A due to high score). Finally, output the charging pile allocation result and the waiting queue of unmatched vehicles, providing a sorting basis for the power allocation gradient adjustment in step 104.

[0082] When the warning signal is triggered, the system starts the intelligent allocation mechanism, comprehensively considering factors such as the remaining power of the vehicle, power attenuation speed, and departure time, and calculates the matching priority of each vehicle to the target charging pile. For example, a net car with almost exhausted power and needing to leave as soon as possible will have a higher matching priority than a private car with higher power and no time constraint. The system automatically allocates charging piles accordingly and recommends the optimal alternative solutions for vehicles that fail to match successfully, such as guiding them to nearby charging piles that will be available soon.

[0083] 104. Adjust the power allocation gradient of the charging pile according to the sorting result of the matching degree score, where the increase amplitude of the power allocation gradient corresponding to the vehicle and the charging pile is positively correlated with the square root of the remaining power attenuation rate.

[0084] In this step, the power allocation gradient refers to the differential configuration scheme of the charging pile output power.

[0085] The increase amplitude refers to the increase amount compared to the basic power configuration.

[0086] The positive correlation relationship of the square root means the mathematical relationship between the power increase amplitude and the power attenuation rate, and the square root function is used to ensure the rationality of the adjustment amplitude.

[0087] The sorting result refers to the priority sequence formed by arranging the matching degree scores of all vehicles in descending order of numerical values.

[0088] In the embodiments of the present application, first, according to the sorting result of the matching degree scores in step 103 and the power decay rate data in step 101, a vehicle priority ranking list is generated (for example, vehicle B has a matching degree score of 7.88 and ranks first). Secondly, the power distribution gradient of the charging pile is dynamically adjusted based on the square root of the power decay rate. The calculation formula is: allocated power = base power × √(power decay rate) (for example, if the power decay rate of vehicle B is 0.8% per minute and the base power is 22 kW, then the allocated power = 22 × √0.8 ≈ 22 × 0.89 ≈ 19.6 kW). Here, √ represents taking the square root. The increase in the power distribution gradient of the charging pile corresponding to the vehicle is positively correlated with the square root of the remaining power decay rate. Then, the linear programming model is used to verify whether the total power exceeds the upper limit of the parking lot grid load (such as 200 kW). If it exceeds the limit, the vehicle power is reduced from low to high according to the matching degree score (for example, reducing the power of a private car from 22 kW to 11 kW). Finally, a power distribution instruction is sent to the charging pile controller through the Modbus protocol, and the adjustment result is transmitted back to the central database in real time to complete the closed loop of dynamic power distribution.

[0089] The system dynamically adjusts the output power of the charging pile according to the matching priority. For high-priority vehicles, the charging power is appropriately increased to shorten the charging time; for low-priority vehicles, the charging power is correspondingly reduced to ensure the overall power load balance. For example, an operating vehicle in urgent need of charging may obtain a higher charging rate than the standard power, while a private car with a relatively sufficient battery level uses a slightly lower charging power. Through this intelligent adjustment, while ensuring rapid energy replenishment for key vehicles, the system optimizes the overall charging efficiency and reduces resource contention problems.

[0090] In summary, through steps 101 to 104, the dynamic optimal allocation of charging resources in the parking lot is achieved. The system constructs a charging demand network topology by integrating multi-source heterogeneous data, real-time tracks the changes in vehicle positions and charging pile statuses, and intelligently predicts potential charging pile preemption conflicts. The power gradient allocation algorithm is innovatively introduced to dynamically adjust the charging power according to the urgency of the vehicle, achieving the optimal allocation of resources while ensuring the charging efficiency. This technical solution breaks through the limitations of the traditional static charging scheduling mode and constructs a full-process optimization mechanism of "demand prediction - conflict warning - intelligent matching - power adjustment", significantly improving the overall operation efficiency of the charging facilities in the parking lot.

[0091] To solve the problem of early warning of competition conflicts among multiple electric vehicle charging piles, in some embodiments, in step 102, by updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology in real time to determine the vehicle movement trajectories of multiple vehicles, and determining the predicted results of the movement trends of multiple vehicles based on the vehicle movement trajectories to generate an early warning signal for charging pile preemption conflicts, including:

[0092] 201. According to the position data of vehicle nodes in the charging demand network topology, extract the change amount of the moving direction, the change amount of the moving speed, and the continuous increment of the change in the moving speed of each vehicle within the target parking lot, and combine the positional relationship between the vehicle node and the charging pile node to generate the vehicle movement trajectory corresponding to each vehicle;

[0093] In step 201, the charging demand network topology refers to the network structure of the supply and demand relationship of charging resources within the parking lot constructed by vehicle nodes and charging pile nodes. The position data of vehicle nodes is the vehicle coordinate information obtained in real time through in-vehicle GPS or in-parking lot positioning systems. The change amount of the moving direction is the angular change value of the driving direction of the vehicle within a continuous time interval, reflecting the turning situation of the vehicle. The change amount of the moving speed is the change value of the speed magnitude of the vehicle per unit time, indicating the acceleration or deceleration state. The continuous increment of the change in the moving speed is the acceleration change rate obtained by analyzing the speed change trend within a continuous time period. The vehicle movement trajectory is the predicted result of the vehicle driving path generated by integrating the above parameters and the vehicle-charging pile relative positional relationship, including a position sequence, a speed curve, and direction change characteristics.

[0094] In the embodiments of the present application, first, the position data of vehicle nodes in the charging demand network topology (such as the coordinates of vehicle A being (10, 20)) is collected in real time through ultra-wideband positioning base stations within the parking lot, and the change amount of the moving direction (such as the change amount of the moving direction of vehicle A from time t1 to t2 being 5°) and the change amount of the moving speed (such as the speed of vehicle A increasing from 2 m / s to 3 m / s) are calculated based on the difference in position data at adjacent times. Secondly, the continuous increment of the change in the moving speed is statistically analyzed through a sliding window (such as the speed of vehicle A increasing by 0.5 m / s, 0.3 m / s, and 0.2 m / s respectively within 3 consecutive time windows, and the continuous increment being 1.0 m / s). Then, in combination with the positional relationship between the vehicle node and the charging pile node (such as vehicle A being 50 meters away from charging pile B), a vehicle movement trajectory including the change amount of the moving direction, the change amount of the moving speed, the continuous increment, and the target charging pile is generated. Finally, the movement trajectory data of each vehicle is output, providing a basis for the continuous position change sequence for step 202.

[0095] 202. Based on the continuous position change sequence in the vehicle movement trajectory, for each vehicle node, according to the corresponding current moving speed change amount and continuous increment, calculate the first estimated time interval for the vehicle node to reach each charging pile node within the charging area along the current moving direction. The starting point of the first estimated time interval is the minimum arrival time based on the current moving speed, and the ending point is the maximum arrival time based on the maximum speed change trend corresponding to the continuous increment;

[0096] In step 202, the continuous position change sequence refers to the set of coordinate point data arranged in chronological order in the vehicle movement trajectory. The current moving speed change amount is the instantaneous speed change value of the vehicle within the recent time window. The continuous increment is a parameter reflecting the current acceleration change trend of the vehicle. The charging area refers to the specific area range in the parking lot where charging piles are centrally arranged. The first estimated time interval is the time range predicted to reach the charging pile through the current motion state. Its starting point is the shortest possible arrival time calculated using the current speed, and the ending point considers the longest possible arrival time under the maximum acceleration change trend. The minimum arrival time is the arrival time assuming the vehicle travels at a constant speed maintaining the current speed, and the maximum arrival time is the arrival time considering the possible acceleration of the vehicle.

[0097] In the embodiment of the present application, first, based on the continuous position change sequence in the vehicle movement trajectory generated in step 201, extract the current moving speed change amount (for example, the current speed of vehicle A is 3 m / s) and the continuous increment of the moving speed change (for example, 1.0 m / s). Secondly, predict the maximum speed change trend corresponding to the continuous increment through a linear regression model (for example, the function of the speed of vehicle A changing with time is v(t)=3 + 0.2t), and correct the speed upper limit in combination with the parking lot speed limit rule (for example, the maximum speed is 5 m / s). Then, according to the real-time distance between the vehicle node and the charging pile node (for example, vehicle A is 50 meters away from charging pile B), calculate the following parameters respectively: Minimum arrival time: calculated based on the current moving speed (for example, 50 / 3≈16.7 seconds); Maximum arrival time: calculated based on the maximum speed change trend corresponding to the continuous increment (for example, when the speed increases to 5 m / s, the time = 50 / 5 = 10 seconds). Finally, generate the first estimated time interval (for example, the interval for vehicle A to reach charging pile B is [10 seconds, 16.7 seconds]), providing input data for the motion trend prediction model in step 203.

[0098] 203. Based on the first estimated time intervals of each vehicle node, construct a motion trend prediction model for multiple vehicles. Analyze the trajectory crossing probability and time overlap probability of each vehicle node within the charging area through the motion trend prediction model to predict the intensity of the interactive influence of the motion trends between multiple vehicle nodes as the motion trend prediction result of multiple vehicles;

[0099] In step 203, the motion trend prediction model for multiple vehicles is a mathematical model that analyzes the interactive influence of vehicle motions. The trajectory crossing probability is to predict the possibility of the driving paths of different vehicles intersecting within the charging area. The time overlap probability is to predict the coincidence degree of the time windows for different vehicles to reach the same location. The intensity of the interactive influence of motion trends is a comprehensive index that quantifies the degree of motion interference between vehicles, combining the prediction results in two dimensions of spatial crossing and time overlap. This model evaluates the dynamic behavior characteristics of the multi-vehicle system through methods such as Monte Carlo simulation or Markov chain.

[0100] In the embodiment of the present application, first, based on the first estimated time intervals of each vehicle node in step 202, the overlapping situation of the time windows for multiple vehicles to reach the same target charging pile is statistically analyzed (for example, the overlapping duration of the interval [10s, 16.7s] of vehicle A and the interval [12s, 18s] of vehicle C is 4.7 seconds). Secondly, the intensity of the interactive influence of motion trends is analyzed by constructing a motion trend prediction model for multiple vehicles: Calculation of the trajectory crossing probability: Based on the change amount of the moving direction in step 201 and the position of the target charging pile, it is judged whether the vehicle paths cross (for example, the heading angle difference between vehicle A and C is <10°, and the path coincidence degree is 80%), and the trajectory crossing probability is obtained through historical data statistics (for example, 80%); Calculation of the time overlap probability: It is calculated according to the proportion of the overlapping duration of the time windows (for example, 4.7 / (16.7 - 10) ≈ 70%); Generation of the interactive influence intensity: The trajectory crossing probability and the time overlap probability are multiplied (for example, 0.8×0.7 = 0.56) as the intensity of the interactive influence of the motion trends of multiple vehicle nodes. Finally, an intensity list is output (for example, the interactive influence intensity between vehicle A and C is 0.56) to predict the intensity of the interactive influence of the motion trends between multiple vehicle nodes as the motion trend prediction result of multiple vehicles.

[0101] 204. Perform conflict detection on the situation where multiple vehicle nodes select the same target charging pile according to the motion trend prediction result, and generate a warning signal including a charging demand urgency parameter.

[0102] In step 204, conflict detection refers to identifying the competitive situation where multiple vehicles may select the same charging pile simultaneously. The target charging pile is a specific charging facility preferentially selected by the vehicle according to its own needs. The charging demand urgency parameter is an emergency degree index calculated by comprehensively considering factors such as the remaining power of the vehicle, the arrival time, and the charging priority. The warning signal is a warning message generated when a potential charging conflict is detected, including key data such as a list of conflicting vehicles, the expected conflict time, and the urgency ranking. This signal triggers the subsequent charging resource coordination and allocation mechanism to ensure the orderly and efficient charging process.

[0103] In the embodiments of the present application, first, according to the motion trend prediction result of step 203, vehicle node pairs with an intensity exceeding a preset threshold (such as 0.5) are screened (for example, the intensity of vehicle A and C is 0.56). Secondly, the remaining power data (such as vehicle A remaining 15%), the power attenuation rate data (such as 1.2% per minute), and the charging priority label (such as priority 5) of the vehicle nodes in the charging demand network topology of step 101 are called, and the charging demand urgency parameter is calculated according to the formula: Urgency = charging priority label × (1 / remaining power) × power attenuation rate (such as Urgency = 5 × (1 / 15) × 1.2 = 0.4). Then, if the urgency parameter exceeds the threshold (such as 0.3), a warning signal including the target charging pile, the conflicting vehicle, and the urgency level is generated (such as "Vehicle A and C are competing for charging pile B, orange warning, urgency 0.4"). Finally, the warning signal is pushed to the dynamic scheduling system and the real-time database is updated to complete the conflict closed-loop response.

[0104] The following is a specific example:

[0105] During the dynamic scheduling process of the intelligent charging parking lot management system, the system first tracks the moving trajectories of each electric vehicle in real time based on high-precision positioning data: by analyzing the driving data of vehicle A during the time period from T0 to T1 (the direction change amount is 15°, the speed accelerates from 8 km / h to 12 km / h and the continuous increment is 0.5 m / s²), and combining its 20-meter straight-line distance from charging pile C3, a moving trajectory model including steering characteristics is generated; subsequently, the system calculates the time interval for vehicle A to reach each charging pile. For example, based on the current speed of 12 km / h, it takes 6 seconds (the minimum value) to reach C3, and considering the acceleration, it is calculated to be 4.8 seconds (the maximum value) according to the maximum possible speed of 15 km / h, forming the first estimated time interval of [4.8s, 6s]; through the analysis of the multi-vehicle motion trend prediction model, it is found that the estimated interval of vehicle B [5.2s, 6.5s] has a 68% time overlap probability with vehicle A at charging pile C3, and the 22° intersection angle of the two vehicle trajectories results in a 42% space conflict risk; when a selection conflict between the two vehicles for charging pile C3 is detected, the system comprehensively considers the data of vehicle A with a remaining power of 8% (high urgency) and vehicle B with a remaining 15%, generates a yellow warning signal for vehicle B and recommends that it change its route to charging pile C7, while reserving charging pile C3 for vehicle A and triggering a path guidance instruction to achieve the optimal allocation of charging resources based on multi-objective dynamic game.

[0106] In summary, the accurate prediction of multi-vehicle charging behavior and conflict warning are achieved through steps 201 to 204. The system constructs a vehicle motion trend prediction model based on spatio-temporal probability by analyzing the direction changes, speed changes, and acceleration characteristics in the real-time vehicle movement trajectories. This model can accurately calculate the time intervals for each vehicle to reach each charging pile and intelligently evaluate the spatio-temporal overlap probability of multi-vehicle trajectories. This technical solution innovatively combines vehicle dynamics characteristics with charging demand prediction, realizing the intelligent conversion from raw trajectory data to charging conflict warning, and providing accurate decision-making basis for subsequent optimal allocation of charging resources.

[0107] To solve the problem of accurate warning of multi-vehicle charging pile preemption conflicts, in some embodiments, step 204 of detecting conflicts for the situation where multiple vehicle nodes select the same target charging pile according to the motion trend prediction result and generating a warning signal including a charging demand urgency parameter includes:

[0108] 301. For each vehicle node, screen the charging pile node with the smallest starting point of its first estimated time interval as the target charging pile, and record the first estimated time interval as the pre-occupation time window of the vehicle node for the target charging pile;

[0109] In step 301, the vehicle node refers to the entity node representing each electric vehicle in the charging demand network topology. The first estimated time interval is the time range predicted to reach the charging pile by analyzing the current motion state of the vehicle, including two boundary values: the minimum arrival time and the maximum arrival time. The charging pile node with the smallest starting point refers to the charging facility with the earliest estimated arrival time of the vehicle among all optional charging piles. The target charging pile is the charging device initially selected by the vehicle according to the principle of optimal arrival time. The pre-occupation time window is the time period information associated with the charging pile by recording the first estimated time interval of the vehicle, indicating the potential time range for the vehicle to occupy the charging pile.

[0110] In the embodiments of the present application, first, by traversing the first estimated time intervals of each vehicle node generated in step 202 (such as the interval [10s, 16.7s] of vehicle A for charging pile B and the interval [15s, 20s] of charging pile C), use the linear search algorithm to screen the charging pile node with the smallest starting time (such as the starting point of 10 seconds of charging pile B). Secondly, mark the first estimated time interval (such as [10s, 16.7s]) as the pre-occupation time window of the vehicle node for the target charging pile, and store the binding relationship between the vehicle node and the target charging pile through a hash table. Then, if there are multiple charging piles with the same starting time, select the nearest charging pile based on the Euclidean distance priority strategy. Finally, generate a list of pre-occupation time windows of target charging piles for all vehicle nodes, providing input data for conflict detection in step 302.

[0111] 302. For the case where multiple vehicle nodes select the same target charging pile, extract the pre-occupation time windows of each vehicle node. Combining the motion trend prediction results of multiple vehicles, if there is an overlapping part in the pre-occupation time windows of at least two vehicle nodes and the interaction influence coefficient exceeds the set threshold, it is determined that there is a preemption conflict for the target charging pile;

[0112] In step 302, the overlapping of pre-occupation time windows means that there is an intersection in the estimated usage time periods of different vehicles for the same charging pile. The interaction influence coefficient is a quantitative value of the degree of motion interference between vehicles calculated by the motion trend prediction model, reflecting the comprehensive influence of vehicle trajectory intersection and time overlap. The set threshold is a conflict determination standard value preset according to the actual operation requirements of the parking lot. The preemption conflict refers to the charging resource competition state determined by the system when the pre-occupation time windows of multiple vehicles overlap and the interaction is significant. This step accurately identifies potential charging pile usage conflicts through spatio-temporal overlap analysis and influence coefficient evaluation.

[0113] In the embodiment of the present application, first, based on the pre-occupation time window list in step 301, aggregate the associated time windows by the charging pile node ID through the hash table grouping algorithm (such as the charging pile B is associated with vehicle A [10s, 16.7s] and vehicle C [12s, 18s]). Secondly, use the interval intersection detection algorithm to calculate the overlapping part of the time windows pair by pair (such as the overlapping interval [12s, 16.7s] between vehicle A and C). If the overlapping duration is greater than zero, it is determined that there is a time conflict. Then, call the motion trend interaction influence intensity output in step 203 (such as the intensity between vehicle A and C is 0.56). If the intensity exceeds the set threshold (such as 0.5), it is marked as a preemption conflict. Finally, output the list of conflict charging pile nodes and the associated vehicle node information, providing conflict vehicle parameters for step 303.

[0114] 303. For the target charging pile with a preemption conflict, calculate the charging demand urgency parameter of each conflict vehicle node according to the real-time moving speed change amount, continuous increment of each conflict vehicle node, the motion trend interaction influence intensity of each conflict vehicle node, and the corresponding vehicle type charging priority label. The charging demand urgency parameter is the weighted value of the remaining power attenuation rate of the vehicle node, the motion trend interaction influence coefficient, and the vehicle type charging priority label;

[0115] In step 303, the conflicting vehicle nodes refer to multiple electric vehicles participating in the competition for the same charging pile. The real-time moving speed change reflects the current acceleration state of the vehicle. The continuous increment represents the change trend of the vehicle's acceleration. The motion trend interaction influence intensity is a parameter quantifying the degree of dynamic influence between vehicles. The vehicle type charging priority label is an identification of the charging priority level set according to the vehicle's use. The charging demand urgency parameter is the weighted calculation result of comprehensively considering the remaining power decay rate (reflecting the power consumption speed), the interaction influence coefficient (reflecting the moving urgency), and the vehicle type priority (reflecting the social importance), and is used to objectively evaluate the urgency of the charging demands of each conflicting vehicle. The weighted value is a calculation method of standardizing and summing up each factor through a preset weight coefficient.

[0116] In the embodiment of the present application, first, extract the real-time moving speed change (such as the speed change of vehicle A is 1 m / s) and the continuous increment (1.0 m / s) of the conflicting vehicle nodes from step 201, obtain the vehicle type charging priority label (such as priority 5) from step 101, and combine the motion trend interaction influence intensity (0.56) in step 203. Secondly, calculate the parameter value through the formula urgency parameter = remaining power decay rate × motion trend interaction influence intensity × vehicle type charging priority label (such as the decay rate of vehicle A is 1.2% per minute, urgency = 1.2 × 0.56 × 5 = 3.36). Then, normalize the urgency parameter (such as scaling it to the range of 0 - 10), and set the warning level threshold (such as red ≥ 3, orange ≥ 2). Finally, output the urgency parameter and the level list of the conflicting vehicles, providing a basis for generating the warning signal in step 304.

[0117] 304. Generate a warning signal including the target charging pile identifier, the overlapping range of the pre-occupied time window, the motion trend interaction influence intensity, and the charging demand urgency parameter of the corresponding conflicting vehicle nodes.

[0118] In step 304, the target charging pile identifier is the number or location code uniquely identifying the conflicting charging pile. The overlapping range of the pre-occupied time window is the start and end times of the intersection of the expected usage time periods of the conflicting vehicles. The motion trend interaction influence intensity is the quantified value of the motion interference between the conflicting vehicles. The charging demand urgency parameter is the scoring of the charging urgency of each conflicting vehicle calculated by the system. The warning signal is a standardized conflict alarm generated by integrating the above key parameters, including structured data such as conflict location information, time characteristics, influence degree, and urgency evaluation, providing a complete basis for subsequent charging scheduling decisions. This signal is pushed to the parking lot management system in real time through a visual interface or an API interface.

[0119] In the embodiments of the present application, first, the conflicting target charging pile identifier (such as charging pile B) in step 302, the overlapping range of the pre-occupation time window ([12s, 16.7s]), the interactive influence intensity of the motion trend in step 203 (0.56), and the charging demand urgency parameter in step 303 (such as vehicle A: 3.36) are integrated. Secondly, the JSON data encapsulation technology is used to generate a structured warning signal (such as {"conflicting charging pile": "B", "urgency parameter": 3.36, "warning level": "red"}), and it is pushed to the dynamic scheduling system through the MQTT protocol. Then, a visual warning is triggered according to the urgency level (such as a red warning is displayed as a flashing icon). Finally, the warning signal is written into the MongoDB real-time database, and the timestamp and processing status are recorded to complete the closed-loop data update.

[0120] The following is a specific example:

[0121] In the conflict detection module of the intelligent charging scheduling system, the system first selects the optimal charging pile for each electric vehicle: for example, the earliest arrival time of vehicle A (remaining battery level 10%) at charging pile C3, which is 4.8 seconds, is recorded as the starting point of the pre-occupation time window. At the same time, the pre-occupation time window of vehicle B (remaining battery level 15%) for charging pile C3 is 5.2 - 6.5 seconds. When the system detects that there is a 1.3-second overlap in the time windows of the two vehicles (4.8 - 6.0 seconds and 5.2 - 6.5 seconds) and the interactive influence coefficient reaches 0.68 (exceeding the threshold of 0.5), a preemption conflict determination is triggered. The system then calculates the charging demand urgency of the conflicting vehicles: based on the remaining battery level decay rate of vehicle A at 0.8% / min, the interactive influence coefficient of 0.68, and the weighted value of the ambulance priority (weight 0.6), the urgency parameter is obtained as 82; while for vehicle B, due to the priority weight of 0.3 for ordinary private cars, the final urgency parameter is 45. The system generates a structured warning signal, including the identifier of charging pile C3, the overlapping range of the time window 5.2 - 6.0 seconds, the interactive intensity of 0.68, and the comparison data of the urgency of the two vehicles. This signal triggers the scheduling center to preferentially allocate charging pile C3 to vehicle A and dynamically plan an alternative path for vehicle B to charging pile C5. At the same time, real-time scheduling suggestions are pushed through the in-vehicle terminal to achieve the optimal allocation of charging resources based on multi-dimensional conflict assessment.

[0122] In summary, the accurate identification and hierarchical early warning of charging pile preemption conflicts are achieved through steps 301 to 304. The system intelligently identifies potential charging pile preemption conflicts by analyzing the optimal charging pile selection and preoccupation time window of each vehicle, combined with the intensity of the interactive influence of the movement trends. For conflict scenarios, the system comprehensively considers multi-dimensional parameters such as the dynamic movement characteristics of the vehicle, the battery power attenuation rate, and the vehicle type priority to accurately calculate the urgency of the charging demand of each vehicle. This technical solution innovatively constructs a hierarchical early warning mechanism based on spatio-temporal overlap detection and urgency assessment, which can accurately predict the severity of charging pile usage conflicts and generate early warning information including conflict details and priority rankings, providing a scientific basis for subsequent intelligent scheduling decisions.

[0123] To solve the problem of dynamic allocation of charging pile resources, in some embodiments, inputting the early warning signal into the multi-agent coordination and allocation mechanism, and executing a dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle to the target charging pile includes:

[0124] 401. Input the early warning signal into the multi-agent coordination and allocation mechanism. Based on the multi-agent coordination and allocation mechanism and according to the preemption conflict relationship between the vehicle and the target charging pile in the early warning signal, establish a conflict handling group for each target charging pile. The conflict handling group includes the remaining power attenuation rate of the vehicle in conflict with the current charging pile, the preset departure time constraint of the vehicle, and the current available power of the charging pile.

[0125] In step 401, the multi-agent coordination and allocation mechanism refers to an intelligent scheduling system for coordinating multiple vehicles competing for charging resources. The early warning signal is alarm data containing key information on charging pile preemption conflicts. The preemption conflict relationship is the associated information recording that multiple vehicles simultaneously compete for the same charging pile. The conflict handling group is a special data structure established for the target charging pile with competition. The remaining power attenuation rate is a parameter reflecting the urgency of the vehicle's power consumption. The preset departure time constraint of the vehicle is the time requirement set by the vehicle owner to complete charging. The current available power of the charging pile is the real-time power output capacity parameter of the charging pile.

[0126] In the embodiments of the present application, first, by analyzing the warning signals generated in step 304, based on the multi-agent coordinated allocation mechanism, conflict charging pile identifiers (such as charging pile B) and their associated conflict vehicle node lists (such as vehicle A and vehicle C) are extracted; second, the remaining power attenuation rate of the vehicle (such as 1.2% per minute for vehicle A) is obtained from the real-time database in step 101, the preset departure time constraint of the vehicle in step 102 (such as vehicle A needs to depart within 30 minutes), and the current available power of the charging pile in step 104 is read (such as the available power of charging pile B is 22 kW); then, based on the hash table nested structure, the charging pile identifier, the remaining power attenuation rate, the preset departure time constraint of the vehicle, and the current available power of the charging pile are bound to generate a conflict handling group data structure (example: {"charging pile B": {"attenuation rate": [1.2, 0.8], "departure time": [30, 45], "available power": 22}}); finally, the conflict handling group is pushed to the multi-agent coordinated allocation mechanism to provide dynamic parameter adjustment input for step 402.

[0127] 402. Perform dynamic parameter adjustment on the vehicles in the conflict handling group through the multi-agent coordinated allocation mechanism. The dynamic parameter adjustment includes adjusting the remaining power attenuation rate of the vehicle in a reverse association manner according to the current available power of the charging pile, and at the same time converting the remaining duration of the vehicle departure time constraint into an urgency level according to a preset interval.

[0128] In step 402, the dynamic parameter adjustment is an optimization process for the parameters of conflict vehicles by the multi-agent coordinated allocation mechanism. The reverse association method is a calculation method for adjusting vehicle requirements according to the charging pile power in reverse. The remaining duration is the remaining time amount of the vehicle departure time constraint. The preset interval is a standardized time range divided by the remaining duration. The urgency level is an emergency degree identifier converted according to the remaining duration interval.

[0129] In the embodiments of the present application, first, based on the available power of the charging pile (such as 22 kW) in the conflict handling group and the total required power of the conflict vehicles (such as vehicle A requires 24 kW, vehicle C requires 11 kW, and the total is 35 kW) through the multi-agent coordinated allocation mechanism, the remaining power attenuation rate is adjusted proportionally by the reverse association weight allocation algorithm (formula: adjusted rate = original rate × (available power / total required power), such as the rate of vehicle A = 1.2 × 22 / 35 ≈ 0.75% per minute); second, the remaining duration of the vehicle departure time constraint (such as vehicle A has 30 minutes remaining) is converted into an urgency level according to a preset interval (rule: ≤ 15 minutes, level 5; 15 - 30 minutes, level 4; 30 - 60 minutes, level 3; other levels, level 1); finally, the adjusted parameter table is output (such as the adjusted rate of vehicle A is 0.75% per minute and the urgency level is 4) to provide a matching degree calculation basis for step 403.

[0130] 403. Generate a matching degree score of the vehicle to the target charging pile according to the adjusted remaining power attenuation rate and the adjusted urgency level.

[0131] In step 403, the matching degree score is a comprehensive evaluation index quantifying the adaptation degree between the vehicle and the charging pile. The adjusted remaining power attenuation rate is a power consumption rate parameter dynamically optimized. The adjusted urgency level is a charging urgency identification after standardization processing. The target charging pile is the specific charging device that the vehicle is currently competing for.

[0132] In the embodiment of the present application, first, based on the adjusted remaining power attenuation rate (such as 0.75% / minute) and the urgency level (such as 4) in step 402, calculate the score according to the formula matching degree score = adjusted rate × urgency level (such as vehicle A score = 0.75 × 4 = 3.0); second, scale the original score to the range of 0-10 through the Min-Max normalization algorithm (such as vehicle A score 3.0 corresponds to the normalized value 7.5); finally, generate a vehicle priority ranking list (such as vehicle A score 7.5 ranks first), and push it to the dynamic scheduling system to execute the charging pile resource allocation, completing the multi-agent coordination closed loop.

[0133] The following is a specific example:

[0134] In the multi-vehicle collaborative decision-making module of the intelligent charging scheduling system, the system first inputs the warning signal into the multi-agent coordination and allocation engine: for the preemption conflict of charging pile C3 (involving vehicles A and B), the system establishes a conflict handling group including parameters such as the remaining power of vehicle A, the departure time constraint, and the current available power of pile C3. Through the dynamic parameter adjustment algorithm, the system distributes power according to the power upper limit of pile C3, allocates higher power to vehicle A to increase its charging speed, and allocates lower power to vehicle B at the same time; the system also marks the remaining departure time of vehicle A as the "urgent" level and that of vehicle B as the "normal" level. Based on the adjusted parameters, the system calculates the matching degree score: vehicle A obtains a higher score and vehicle B obtains a lower score. This scoring result triggers the system to automatically lock pile C3 for vehicle A and start the fast charging mode, and at the same time guide vehicle B to other available charging piles. Through this dual optimization mechanism of dynamic power distribution and time constraint conversion, the system significantly improves the utilization rate of charging facilities and ensures that emergency vehicles complete charging first.

[0135] In summary, the dynamic optimization decision-making in the multi-vehicle conflict scenario is achieved through steps 401 to 403. By constructing a conflict handling group mechanism, the system converts the warning signal into quantifiable scheduling parameters. For each charging pile with preemption conflicts, the system intelligently aggregates the real-time status data of relevant vehicles, including key indicators such as the power decay situation and the departure time requirement. Based on the multi-agent collaborative decision-making algorithm, the system dynamically adjusts the parameter evaluation weights of each vehicle, and precisely matches the available power of the charging pile with the vehicle demand. By performing a standardized conversion on the remaining power decay rate and the urgency of the departure time, the system can generate a scientific and objective matching score, providing data support for subsequent charging resource allocation and ensuring the optimal allocation of charging resources even in conflict situations.

[0136] To solve the problem of networked modeling of charging demand, in some embodiments, in step 101, by fusing the remaining power data of multiple vehicles and the occupancy status data of charging piles, and associating the charging priority labels of different vehicle models, a charging demand network topology is established in the target parking lot, including:

[0137] 501. Fuse the remaining power data of all vehicles in the target parking lot with the current occupancy status data of all charging piles to generate a fusion data set containing the real-time status of vehicles and charging piles;

[0138] In step 501, the target parking lot refers to a specific parking lot area that needs to optimize charging scheduling. The remaining power data is the current battery capacity percentage data of electric vehicles collected in real time by on-vehicle sensors. The charging pile occupancy status data is the current usage status information of each charging pile obtained through the charging pile management system, including states such as idle, charging, and faulty. The fusion data set is a structured data set formed by integrating vehicle power information and charging pile status information, reflecting the real-time supply and demand situation of charging resources in the parking lot.

[0139] In the embodiments of the present application, first, the current occupancy status data of all charging piles in the parking lot is collected in real time through the Internet of Things sensors of the charging piles (such as charging pile A is idle, charging pile B is in use), and at the same time, the remaining power data of all vehicles is obtained through the on-vehicle battery management system interface (such as vehicle X has 20% remaining, vehicle Y has 35% remaining); second, using the timestamp synchronization mechanism of the central processing platform, align the vehicle remaining power and the charging pile occupancy status data according to a unified time window; then, merge the two types of data through the key-value pair association technology (Key = vehicle / charging pile ID, Value = real-time data) to generate a fusion data set containing vehicle ID, remaining power, charging pile ID, and occupancy status; finally, store it in a distributed database (such as HBase) to provide basic data for step 502.

[0140] 502. Based on the vehicle type identifiers corresponding to each vehicle in the fusion data set, match a preset charging priority label for each vehicle. The charging priority label is divided into at least three levels according to vehicle type attributes.

[0141] In step 502, the vehicle type identifier is a unique coding identifier used to distinguish different vehicle types. The charging priority label is a classification identifier for the charging urgency preset according to vehicle types, and at least includes three levels: high, medium, and low. Vehicle type attributes refer to the usage nature and functional characteristics of vehicles, such as classification attributes like taxis, private cars, and special vehicles.

[0142] In the embodiment of the present application, first, based on the vehicle type identifiers in the fusion data set of step 501 (such as ambulances, taxis, private cars), query the preset vehicle type - priority mapping table (such as ambulance priority 5, taxi 3, private car 1); second, use the hash table lookup algorithm to match the corresponding charging priority label for each vehicle node (such as vehicle X is an ambulance → label 5); then, if the vehicle type identifier is not defined in the mapping table, assign the lowest priority according to the default rule (such as label 1); finally, bind the priority label to the vehicle ID in the fusion data set to generate an extended data set (fields: vehicle ID, remaining power, charging pile status, priority label), and push it to step 503.

[0143] 503. Associate the matched charging priority label with the remaining power of the vehicle and the occupancy status of the charging pile in the fusion data set to generate a target data set including vehicle charging demand characteristics and charging pile adaptation conditions.

[0144] In step 503, the charging demand characteristic is a comprehensive parameter reflecting the charging urgency of the vehicle, including elements such as remaining power and priority label. The charging pile adaptation condition is a parameter set describing the matching requirements between the charging pile and the vehicle, including conditions such as power matching and interface compatibility. The target data set is a complete scheduling decision data set integrating vehicle demand characteristics and charging pile adaptation conditions.

[0145] In the embodiment of the present application, first, based on the extended data set of step 502, extract the remaining power of the vehicle node (such as vehicle X has 20% remaining), the occupancy status of the charging pile node (such as charging pile A is idle), and the charging priority label (such as label 5); second, use the multi - field association algorithm to bind the vehicle charging demand characteristics (remaining power + priority) and the charging pile adaptation conditions (occupancy status + power level) according to the spatial position relationship (such as vehicle X is 30 meters away from charging pile A); then, filter out invalid data (such as vehicles with remaining power > 80% are not included in the demand network); finally, generate a target data set including vehicle charging demand characteristics (low power + high priority) and charging pile adaptation conditions (idle + adapted power) to provide topological construction input for step 504.

[0146] 504. Construct a charging demand network topology according to the vehicle charging demand characteristics and charging pile adaptation conditions in the target data set.

[0147] In step 504, the charging demand network topology is a network structure formed by modeling vehicle nodes, charging pile nodes, and their association relationships. This topology structure includes node attributes such as vehicle location, power demand, and charging pile status, as well as edge attributes such as supply-demand matching relationships, constituting a complete charging resource scheduling network model.

[0148] In the embodiment of the present application, first, based on the target data set in step 503, vehicle nodes and charging pile nodes are modeled as vertices in a graph structure, and vertex attributes include remaining vehicle power, priority labels, and charging pile occupancy status; second, a weighted edge modeling algorithm is used to define the demand relationship from vehicles to charging piles, and the edge weight calculation formula is: weight = priority label × (1 / remaining power) (for example, vehicle X with priority 5 and remaining power 20% → weight = 5 × 1 / 20 = 0.25); then, a graph database (Neo4j) is used to store the topology structure, and the vehicle location and charging pile status are updated in real time; finally, a dynamic charging demand network topology is output to provide a global view for conflict prediction and resource scheduling.

[0149] The following is a specific example:

[0150] In the real-time decision-making process of the intelligent charging management system, the system first integrates the remaining power data of all electric vehicles in the parking lot (such as vehicle A with a remaining range of 30 kilometers and vehicle B with a remaining range of 50 kilometers) and the charging pile status information (such as pile C3 is in use and pile C5 is idle) to form a real-time updated integrated data set. Based on vehicle registration information, the system automatically labels special vehicle types such as ambulances and logistics vehicles with "urgent" priority, taxis with "priority" level, and private cars with "regular" level. After these priority labels are intelligently associated with the actual remaining power of the vehicles, the power and occupancy of surrounding charging piles (such as fast charging pile power of 150 kW and slow charging pile power of 60 kW), a structured target data set is generated. Finally, the system converts these data elements into a network topology model: using vehicles as dynamic nodes (the size of the nodes represents the degree of power urgency), charging piles as fixed nodes (the color of the nodes distinguishes the available status), and connection lines representing reachable relationships (the thickness of the lines reflects the adaptation degree), a visual charging demand network is constructed to provide a complete decision-making basis for subsequent intelligent scheduling.

[0151] In summary, through steps 501 to 504, the dynamic modeling and optimal matching of charging demand and resource status are achieved. This technical solution innovatively conducts networked modeling on discrete vehicle charging demands and charging pile resource statuses, constructs a three-layer topology structure including real-time status monitoring, demand hierarchical evaluation, and resource adaptation analysis, provides a comprehensive and accurate data basis for subsequent intelligent scheduling decisions, and realizes the intelligent conversion from raw data to executable strategies.

[0152] To solve the problem of the accuracy of charging supply and demand network modeling, in some embodiments, constructing a charging demand network topology according to the vehicle charging demand characteristics and charging pile adaptation conditions in step 504 includes:

[0153] 601. Generate a demand node based on the remaining battery power of the vehicle and the charging priority label, and generate an adaptation node based on the occupancy status of the charging pile and the charging priority label;

[0154] In step 601, the remaining battery power value refers to the specific percentage value of the remaining battery power of the vehicle at present. The charging priority label is a classification identifier for the charging urgency preset according to the vehicle type. The demand node is a network node representing the vehicle charging demand, including two core attributes: the remaining battery power value and the charging priority label. The adaptation node is a network node representing the charging pile adaptation condition, including two core attributes: the occupancy status of the charging pile and the supported priority range.

[0155] In the embodiment of the present application, first, based on the remaining battery power value in step 501 (such as vehicle A remaining 20%) and the charging priority label in step 502 (such as priority 5), a demand node is generated through structured data encapsulation technology, and the node attributes include the remaining battery power, priority label, and vehicle position coordinates; secondly, according to the occupancy status of the charging pile in step 501 (such as charging pile B is idle) and the charging priority label in step 502, an adaptation node is generated, and the node attributes include the occupancy status, supported priority range (such as charging pile B supports priority ≥ 3), and power level; finally, a list of demand nodes (such as {"demand node A": {"remaining battery power": 20, "priority": 5}}) and a list of adaptation nodes (such as {"adaptation node B": {"occupancy status": "idle", "supported priority": "≥ 3"}}) are output, providing matching inputs for step 602.

[0156] 602. Perform priority matching on the demand node and the adaptation node. The priority matching process includes that if the supported priority range of the adaptation node contains the priority label of the demand node, a connection channel is generated between the two, and an initial intensity value is assigned to the connection channel according to the remaining battery power value of the demand node;

[0157] In step 602, priority matching refers to the process of associating a demand node with an adaptation node according to priority rules. The priority range is the value range of the charging priority labels supported by the adaptation node. The connection channel is a virtual association path established between the demand node and the adaptation node. The initial strength value is a quantitative parameter reflecting the matching degree between the demand node and the adaptation node, and its numerical value is determined by the remaining power value of the demand node.

[0158] In the embodiment of the present application, first, traverse the list of demand nodes and adaptation nodes, and use the range check algorithm to determine whether the supported priority range of the adaptation node contains the priority label of the demand node (for example, the adaptation node B supports a priority ≥ 3, and the priority of the demand node A is 5 → matching is successful); second, generate a connection channel for the successfully matched node pair, and assign an initial strength value to the channel through the weight calculation formula (formula: initial strength value = remaining power × priority label, such as the remaining power of the demand node A is 20 × priority 5 = 100); finally, store the connection channel data (such as {"adaptation node B → demand node A": "initial strength value 100"}), providing sorting input for step 603.

[0159] 603. Sort the connection channels of multiple demand nodes connected to the same adaptation node according to the initial strength value, and only retain the connection channel with the highest initial strength value;

[0160] In step 603, the initial strength value sorting refers to arranging the matching degree parameters of all demand nodes connected to the same adaptation node from high to low. The retention operation refers to the screening process of selecting the optimal match and eliminating the sub-optimal match among multiple connection channels. The highest initial strength value refers to the matching degree parameter with the largest numerical value among all connection channels of the same adaptation node.

[0161] In the embodiment of the present application, first, for the connection channels of multiple demand nodes connected to the same adaptation node (such as the adaptation node B is associated with the demand nodes A and C, and the initial strength values are 100 and 80 respectively), use the quicksort algorithm to sort them in descending order according to the initial strength value; second, only retain the connection channel ranked first through the threshold truncation strategy (such as the adaptation node B retains the channel of the demand node A); finally, output the list of screened connection channels (such as {"adaptation node B": "demand node A → strength 100"}), providing topology construction data for step 604.

[0162] 604. Construct a charging demand network topology based on the retained connection channels, the demand nodes, and the adaptation nodes.

[0163] In step 604, the charging demand network topology is a network structure formed by correlating demand nodes and adaptation nodes with each other through reserved connection channels. This topological structure completely describes the optimal matching relationship between the vehicle charging demands in the parking lot and the charging pile adaptation conditions, providing networked data support for subsequent charging scheduling decisions.

[0164] In the embodiment of the present application, first, the connection channels reserved in step 603, the demand nodes and adaptation nodes in step 601 are modeled as graph structure vertices and edges. The vertex attributes include the remaining power and priority of the demand nodes, and the occupancy status and supported priority range of the adaptation nodes. Secondly, the network topology is stored through a graph database (Neo4j), and the edge attributes include the initial strength value and the channel status (such as activated / inactivated). Finally, a dynamically updated charging demand network topology is output, providing a global visual decision basis for resource scheduling.

[0165] The following is a specific example:

[0166] In the network modeling stage of the intelligent charging scheduling system, the system first converts the real-time status of electric vehicles in the parking lot into network nodes: a circular demand node is generated with vehicle A (remaining power 20%, "urgent" priority), and a square adaptation node is generated with charging pile C3 (150kW fast charge, supporting the "urgent" level). Through the priority matching algorithm, when it is detected that the priority range of adaptation node C3 includes the label of vehicle A, the system automatically establishes a connection channel between the two and assigns a strength value of 85 (in the range of 0-100) to this channel according to the power criticality of vehicle A. At this time, if it is found that charging pile C3 has also established a connection channel (strength value 60) with vehicle B (remaining power 35%, "priority" level), the system will automatically retain the connection channel of vehicle A with a higher strength value. Finally, the system integrates all the verified connection channels (such as vehicle A-C3 strength 85, vehicle C-C5 strength 72, etc.) to construct a complete charging demand network topology diagram. In this topological structure, the high-strength red connection lines represent urgent charging demands, the medium-strength yellow connection lines represent priority demands, and the size and color depth of the network nodes intuitively reflect the priorities and matching degrees of each element, providing visual decision support for dynamic scheduling.

[0167] In summary, the optimal allocation of charging resources based on dynamic priority matching is achieved through steps 601 to 604. By establishing an intelligent matching mechanism between demand nodes and adaptation nodes, the system automatically generates candidate connection channels and assigns an initial weight value based on the remaining battery power. In the case of multiple candidate connections, the system adopts a competitive selection mechanism to retain only the optimal matching connection. The finally constructed charging demand network topology can intuitively reflect the optimal matching relationship between the charging demand and resource supply in the current parking lot, providing visual support for subsequent intelligent scheduling decisions. This technical solution innovatively transforms discrete charging demands into a networked model, achieving global optimization of charging resource allocation.

[0168] To solve the problem of optimizing the fairness and efficiency of charging power distribution, in some embodiments, adjusting the power distribution gradient of the charging pile according to the sorting result of the matching degree score in step 104 includes:

[0169] 701. According to the sorting result of the matching degree score, generate a vehicle priority sequence from high to low according to the matching degree score, and extract the total allocable power of the current charging pile, so as to allocate a reference power value to each vehicle in the vehicle priority sequence according to the total allocable power;

[0170] In step 701, the matching degree score is a quantitative score value calculated by comprehensively considering factors such as the urgency of vehicle charging demand, charging priority label, and charging pile adaptation conditions, and is used to evaluate the priority level of a vehicle to obtain charging resources. The vehicle priority sequence is an ordered list generated by sorting all vehicles to be charged from high to low according to the matching degree score, and this sequence determines the order in which vehicles obtain charging resources. The total allocable power refers to the total maximum output power that the target charging pile can provide in the current state, and this value is jointly affected by the rated power of the charging pile and the already allocated power. The reference power value is the basic charging power value initially allocated to each vehicle according to the total allocable power and the vehicle priority sequence, and its allocation principle needs to ensure that high-priority vehicles obtain sufficient power supply.

[0171] In the embodiment of the present application, first, based on the sorting result of the matching degree score generated in step 403 (such as vehicle A scoring 7.5 and vehicle B scoring 5.0), generate a vehicle priority sequence from high to low according to the score (such as sequence order: vehicle A → vehicle B); secondly, extract the total allocable power of the current charging pile (such as the total power of charging pile B is 22kW), and calculate the reference power value according to the length of the priority sequence (formula: reference power = total allocable power / number of vehicles, such as the number of vehicles is 2, then reference power = 22 / 2 = 11kW); finally, allocate the reference power value to each vehicle (such as 11kW for both vehicle A and B), providing initial parameters for subsequent dynamic adjustment.

[0172] 702. Extract the remaining power decay rate of each vehicle through the multi-agent coordinated allocation mechanism, and calculate the square root value of the remaining power decay rate as the rate adjustment base;

[0173] In step 702, the remaining power decay rate is the power consumption value per unit time calculated by monitoring the change trend of the vehicle battery power. This parameter directly reflects the emergency consumption status of the vehicle power. The square root value is an adjustment reference value obtained by mathematically processing the remaining power decay rate. Using the square root operation can effectively balance the magnitude differences between different decay rates. The rate adjustment base is a standardized parameter calculated based on the square root value and is used for the dynamic adjustment process of subsequent power distribution to ensure that the adjustment amplitude has a reasonable corresponding relationship with the power decay degree.

[0174] In the embodiment of the present application, first, extract the remaining power decay rate of each vehicle from step 402 (for example, the decay rate of vehicle A is 1.2% per minute), and calculate the rate adjustment base through the square root operation (formula: adjustment base = √decay rate, for example, the adjustment base of vehicle A = √1.2 ≈ 1.095); second, bind the adjustment base to the reference power value (for example, vehicle A binds the base 1.095 and the reference power 11kW) to provide calculation input for step 703.

[0175] 703. According to the position of the vehicle in the priority sequence, assign a position weight coefficient to each vehicle, and multiply the rate adjustment base by the position weight coefficient to generate a dynamic adjustment coefficient for the individual vehicle;

[0176] In step 703, the position weight coefficient is a weight value dynamically assigned according to the specific sorting position of the vehicle in the priority sequence. This coefficient is designed in a non-linear decreasing manner to ensure that the vehicles in the front section of the priority sequence obtain a significantly higher weight advantage. The dynamic adjustment coefficient is a composite parameter generated by combining the rate adjustment base and the position weight coefficient through multiplication. This coefficient takes into account both the objective urgency of power decay and the subjective decision-making factor of priority sorting, realizing personalized customization of power distribution.

[0177] In the embodiment of the present application, first, according to the position of the vehicle in the priority sequence in step 701 (for example, vehicle A ranks first and vehicle B ranks second), assign the position weight coefficient according to the inverse weighting rule (formula: weight coefficient = 1 / rank, for example, the weight of vehicle A = 1 / 1 = 1.0, the weight of vehicle B = 1 / 2 = 0.5); second, multiply the rate adjustment base in step 702 by the position weight coefficient to generate the dynamic adjustment coefficient (for example, the coefficient of vehicle A = 1.095 × 1.0 ≈ 1.095, the coefficient of vehicle B = √0.8 × 0.5 ≈ 0.447); finally, output the adjustment coefficient list to provide a power correction parameter for step 704.

[0178] 704. Multiply the reference power value by the dynamic adjustment coefficient through the multi-agent coordination and allocation mechanism to generate the actual allocated power value of the vehicle;

[0179] In step 704, the actual allocated power value is the specific power value finally determined for each vehicle to perform the charging operation, which is calculated by multiplying the reference power value by the dynamic adjustment coefficient. This calculation process ensures that vehicles with high priority and fast battery power decay obtain a larger power allocation, while ensuring that the power allocation result matches the overall power supply capacity of the charging pile. The power allocation scheme needs to be updated in real time to adapt to changes in vehicle status and charging pile conditions.

[0180] In the embodiment of the present application, first, multiply the reference power value in step 701 (such as 11 kW) by the dynamic adjustment coefficient in step 703 (such as the coefficient of vehicle A being 1.095) to generate the actual allocated power value (formula: actual power = reference power × adjustment coefficient, such as the power of vehicle A = 11 × 1.095 ≈ 12.0 kW, the power of vehicle B = 11 × 0.447 ≈ 4.9 kW); second, check whether the actual power exceeds the upper limit of the single-pile power of the charging pile (such as the upper limit of charging pile B being 24 kW, and the power of vehicle A being 12.0 kW not exceeding the limit); finally, output the actual power allocation table of the vehicle to provide global adjustment input for step 705.

[0181] 705. According to the constraint condition that the sum of the actual allocated power values of all vehicles does not exceed the total allocable power of the charging pile, adjust the power allocation gradient of the charging pile proportionally.

[0182] In step 705, the power allocation gradient refers to the differential allocation scheme of the charging pile output power among different vehicles, which is achieved by dynamically adjusting the actual allocated power values of each vehicle. The constraint condition requires that the cumulative sum of the actual allocated power values of all vehicles must be strictly less than or equal to the total allocable power of the charging pile, and this limit ensures the safe and stable operation of the charging system. Proportional adjustment means that when the sum of the initially calculated actual allocated powers exceeds the total allocable power, the system automatically reduces it proportionally according to the proportion of the original allocated power of each vehicle. This adjustment method can not only ensure the fairness of power allocation but also maintain the original priority order. The finally formed power allocation gradient scheme will be synchronously updated to the execution unit of the charging management system.

[0183] In the embodiments of the present application, first, the sum of the actual allocated power values of all vehicles is calculated (for example, Vehicle A 12.0 kW + Vehicle B 4.9 kW = 16.9 kW). If the sum exceeds the total allocable power of the charging pile (such as 22 kW), the power is adjusted according to the difference ratio (for example, the difference = 22 - 16.9 = 5.1 kW, and the allocation is based on the weight: Vehicle A increases by 5.1×(12.0 / 16.9)=3.6 kW, and the final power = 12.0 + 3.6 = 15.6 kW; Vehicle B increases by 1.5 kW, and the final power = 4.9 + 1.5 = 6.4 kW); if it does not exceed the limit, it takes effect directly. Finally, the adjusted power allocation gradient table is output (such as Vehicle A 15.6 kW, Vehicle B 6.4 kW), completing the closed-loop of dynamic power allocation.

[0184] The following is a specific example:

[0185] In the power allocation stage of the intelligent charging scheduling system, the system first generates a vehicle priority sequence according to the matching degree score: for example, Ambulance A (score 92 points) ranks first, Taxi B (score 85 points) second, and Private Car C (score 68 points) third. Based on the total power of the charging pile of 150 kW, the system allocates benchmark power values to the top three vehicles (Ambulance A 60 kW, Taxi B 50 kW, Private Car C 40 kW). Subsequently, the system calculates the dynamic adjustment parameters of each vehicle: it is detected that the remaining power attenuation rate of Ambulance A is 0.5% / min, and the square root value of 0.707 is taken as the rate adjustment base number. Combining its first position weight coefficient of 1.2, a dynamic adjustment coefficient of 0.85 is generated. Multiply the benchmark power of 60 kW by this coefficient to obtain the actual allocated power of 51 kW. Similarly, it is calculated that Taxi B obtains 43 kW and Private Car C obtains 32 kW, and the total of 126 kW does not exceed the total power of 150 kW. Finally, the system proportionally increases the power allocation gradient so that Ambulance A actually obtains 60 kW, Taxi B 55 kW, and Private Car C 35 kW, realizing the optimal configuration of priority full-power charging for emergency vehicles while ensuring the total power constraint.

[0186] In summary, through steps 701 to 705, the power optimization allocation based on dynamic priority and power attenuation characteristics is achieved. The vehicle priority sequence is established according to the matching degree score, and the benchmark allocation scheme is determined based on the total power of the charging pile. By introducing the square root of the remaining power attenuation rate as an adjustment factor and combining the priority position weight, the system can intelligently calculate the dynamic power demand of each vehicle. This technical solution innovatively constructs a two-layer power allocation mechanism of "benchmark allocation + dynamic adjustment", realizing the power tilt allocation for high-priority and high-attenuation rate vehicles while ensuring the total power constraint, and significantly improving the utilization efficiency of charging resources and user satisfaction.

[0187] Figure 2The following is a schematic structural diagram of an intelligent scheduling processing system for charging piles provided by an embodiment of the present application. As Figure 2 shown, the system includes:

[0188] An association module 21, which establishes a charging demand network topology in a target parking lot by fusing the remaining power data of multiple vehicles and the occupancy status data of charging piles, and associating the charging priority tags of different vehicle models;

[0189] An early warning module 22, which determines the vehicle movement trajectories of multiple vehicles by real-time updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology, and determines the prediction results of the movement trends of multiple vehicles based on the vehicle movement trajectories to generate an early warning signal for charging pile preemption conflicts;

[0190] A calculation module 23, which inputs the early warning signal into a multi-agent coordination and allocation mechanism, and executes a dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle to a target charging pile, and the matching degree score is corrected by a product factor of the remaining power attenuation rate and the vehicle departure time constraint condition;

[0191] An adjustment module 24, which adjusts the power allocation gradient of the charging pile according to the sorting result of the matching degree score, where the increase amplitude of the power allocation gradient corresponding to the vehicle and the charging pile is positively correlated with the square root of the remaining power attenuation rate.

[0192] Figure 2 The described intelligent scheduling processing system for charging piles can execute Figure 1 the intelligent scheduling processing method for charging piles described in the embodiment shown. The implementation principle and technical effects will not be elaborated further. For the intelligent scheduling processing system for charging piles in the above embodiment, the specific manners in which each module and unit perform operations have been described in detail in the embodiment related to the method, and will not be elaborated in detail here.

[0193] In a possible design, Figure 2 the intelligent scheduling processing system for charging piles in the embodiment shown can be implemented as a computing device. As Figure 3 shown, the computing device can include a storage component 31 and a processing component 32;

[0194] The storage component 31 stores one or more computer instructions, where the one or more computer instructions are called and executed by the processing component 32.

[0195] The processing component 32 is used for the intelligent scheduling processing method for charging piles in the above Figure 1 embodiment.

[0196] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above methods. Of course, the processing component may also be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above methods.

[0197] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0198] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0199] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0200] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0201] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above processing component, storage component, etc. may be basic server resources leased or purchased from a cloud computing platform.

[0202] The embodiment of the present application also provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the Figure 1 intelligent scheduling processing method of a charging pile shown in the above embodiments.

[0203] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0204] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative work.

[0205] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A method for intelligent scheduling and processing of charging piles, characterized in that, Including: Establish a charging demand network topology in the target parking lot by fusing the remaining power data of multiple vehicles read by the device when entering the target parking lot with the occupancy status data of charging piles and associating charging priority tags for different vehicle models; Determine the vehicle movement trajectories of multiple vehicles by real-time updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology, and determine the motion trend prediction results of multiple vehicles according to the vehicle movement trajectories to generate a warning signal for charging pile preemption conflicts; Input the warning signal into a multi-agent coordination and allocation mechanism, and execute a dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle for the target charging pile, and the matching degree score is corrected by the product factor of the remaining power attenuation rate and the vehicle departure time constraint condition; Adjust the power allocation gradient of the charging pile according to the sorting result of the matching degree score, where the increase amplitude of the power allocation gradient corresponding to the vehicle for the charging pile is positively correlated with the square root of the remaining power attenuation rate; The step of determining the vehicle movement trajectories of multiple vehicles by real-time updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology, and determining the motion trend prediction results of multiple vehicles according to the vehicle movement trajectories to generate a warning signal for charging pile preemption conflicts includes: Extract the moving direction change amount, moving speed change amount, and continuous increment of moving speed change of each vehicle in the target parking lot according to the position data of vehicle nodes in the charging demand network topology, and generate the vehicle movement trajectory corresponding to each vehicle in combination with the position relationship between the vehicle node and the charging pile node; Based on the continuous position change sequence in the vehicle movement trajectory, for each vehicle node, calculate the first estimated time interval for the vehicle node to reach each charging pile node in the charging area along the current moving direction according to the corresponding current moving speed change amount and continuous increment, where the starting point of the first estimated time interval is the minimum arrival time based on the current moving speed, and the end point is the maximum arrival time based on the maximum speed change trend corresponding to the continuous increment; Construct a motion trend prediction model for multiple vehicles based on the first estimated time intervals of each vehicle node, and analyze the trajectory crossing probability and time overlap probability of each vehicle node in the charging area through the motion trend prediction model to predict the intensity of the interaction of motion trends between multiple vehicle nodes as the motion trend prediction results of multiple vehicles; Perform conflict detection on the situation where multiple vehicle nodes select the same target charging pile according to the motion trend prediction results, and generate a warning signal including a charging demand urgency parameter; Input the warning signal into a multi-agent coordination and allocation mechanism, and execute a dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle for the target charging pile, including: Input the warning signal into the multi-agent coordination and allocation mechanism. Based on the multi-agent coordination and allocation mechanism and according to the preemption conflict relationship between the vehicle and the target charging pile in the warning signal, establish a conflict handling group for each target charging pile. The conflict handling group includes the remaining power attenuation rate of the vehicle in conflict with the current charging pile, the preset departure time constraint of the vehicle, and the current available power of the charging pile; Execute dynamic parameter adjustment on the vehicles in the conflict handling group through the multi-agent coordination and allocation mechanism. The dynamic parameter adjustment includes adjusting the remaining power attenuation rate of the vehicle in an inverse correlation manner according to the current available power of the charging pile, and at the same time converting the remaining duration of the vehicle departure time constraint into an urgency level according to a preset interval; Generate a matching degree score of the vehicle to the target charging pile according to the adjusted remaining power attenuation rate and the adjusted urgency level.

2. The method according to claim 1, characterized in that, The conflict detection for the situation where multiple vehicle nodes select the same target charging pile according to the motion trend prediction result, and generate a warning signal including the charging demand urgency parameter, includes: For each vehicle node, screen the charging pile node with the smallest starting point of its first estimated time interval as the target charging pile, and record the first estimated time interval as the pre-occupation time window of the vehicle node for the target charging pile; For the situation where multiple vehicle nodes select the same target charging pile, extract the pre-occupation time windows of each vehicle node, and combine the motion trend prediction results of multiple vehicles. If there is an overlapping part in the pre-occupation time windows of at least two vehicle nodes and the interaction influence coefficient exceeds the set threshold, it is determined that there is a preemption conflict for the target charging pile; For the target charging pile with preemption conflict, calculate the charging demand urgency parameter of each conflict vehicle node according to the real-time moving speed change amount, continuous increment of each conflict vehicle node, the motion trend interaction influence intensity of each conflict vehicle node, and the corresponding vehicle type charging priority label. The charging demand urgency parameter is the weighted value of the remaining power attenuation rate of the vehicle node, the motion trend interaction influence coefficient, and the vehicle type charging priority label; Generate a warning signal including the target charging pile identifier, the overlapping range of the pre-occupation time window, the motion trend interaction influence intensity, and the charging demand urgency parameter of the corresponding conflict vehicle node.

3. The method according to claim 1, characterized in that, Establish a charging demand network topology in the target parking lot by fusing the remaining power data of multiple vehicles and the occupancy status data of the charging piles, and associating the charging priority labels of different vehicle types, including: Fuse the remaining power data of all vehicles in the target parking lot and the current occupancy status data of all charging piles to generate a fusion data set including the real-time status of vehicles and charging piles; Based on the vehicle type identifier corresponding to each vehicle in the fusion data set, match a preset charging priority label for each vehicle. The charging priority label is divided into at least three levels according to vehicle type attributes; Associate the matched charging priority label with the remaining power of the vehicle and the occupancy status of the charging pile in the fusion data set to generate a target data set including the vehicle charging demand characteristics and the charging pile adaptation conditions; Construct a charging demand network topology according to the vehicle charging demand characteristics and charging pile adaptation conditions in the target dataset.

4. The method according to claim 3, wherein Construct a charging demand network topology according to the vehicle charging demand characteristics and charging pile adaptation conditions in the target dataset, including: Generate demand nodes based on the remaining power of the vehicle and the charging priority label, and generate adaptation nodes based on the occupancy status of the charging pile and the charging priority label; Perform priority matching between the demand nodes and the adaptation nodes. The priority matching process includes generating a connection channel between the two if the priority range supported by the adaptation node contains the priority label of the demand node, and assigning an initial strength value to the connection channel according to the remaining power value of the demand node; Sort the connection channels of multiple demand nodes connected to the same adaptation node according to the initial strength value, and only retain the connection channel with the highest initial strength value; Construct a charging demand network topology based on the retained connection channels, the demand nodes, and the adaptation nodes.

5. The method according to claim 1, wherein Adjust the power distribution gradient of the charging pile according to the sorting result of the matching degree score, including: According to the sorting result of the matching degree score, generate a vehicle priority sequence from high to low according to the matching degree score, and extract the total allocable power of the current charging pile, so as to allocate a reference power value for each vehicle in the vehicle priority sequence according to the total allocable power; Extract the remaining power decay rate of each vehicle through the multi-agent coordinated allocation mechanism, and calculate the square root value of the remaining power decay rate as the rate adjustment base; According to the position of the vehicle in the priority sequence, assign a position weight coefficient to each vehicle, and multiply the rate adjustment base by the position weight coefficient to generate a dynamic adjustment coefficient for the individual vehicle; Multiply the reference power value by the dynamic adjustment coefficient through the multi-agent coordinated allocation mechanism to generate the actual allocated power value of the vehicle; According to the constraint condition that the sum of the actual allocated power values of all vehicles does not exceed the total allocable power of the charging pile, adjust the power distribution gradient of the charging pile proportionally.

6. An intelligent scheduling and processing system for a charging pile, characterized in that, Including: An association module that establishes a charging demand network topology in the target parking lot by fusing the remaining power data of multiple vehicles and the occupancy status data of the charging piles, and associating the charging priority labels of different vehicle models; An early warning module that determines the vehicle movement trajectories of multiple vehicles by real-time updating the position data of the vehicle nodes and the status data of the charging pile nodes in the charging demand network topology, and generates an early warning signal for charging pile preemption conflicts according to the vehicle movement trajectories; A calculation module that inputs the early warning signal into the multi-agent coordinated allocation mechanism, and executes a dynamic balance allocation strategy through the multi-agent coordinated allocation mechanism to calculate the matching degree score of each vehicle to the target charging pile, and the matching degree score is corrected by the product factor of the remaining power decay rate and the vehicle departure time constraint condition; An adjustment module that adjusts the power distribution gradient of the charging pile according to the sorting result of the matching degree score, where the increase amplitude of the power distribution gradient corresponding to the vehicle to the charging pile is positively correlated with the square root of the remaining power decay rate; By updating the position data of vehicle nodes and the status data of charging pile nodes in the charging demand network topology in real time to determine the vehicle movement trajectories of multiple vehicles, and determining the motion trend prediction results of multiple vehicles based on the vehicle movement trajectories to generate a warning signal for charging pile preemption conflicts, including: According to the position data of vehicle nodes in the charging demand network topology, extract the change amount of the moving direction, the change amount of the moving speed, and the continuous increment of the change of the moving speed of each vehicle within the target parking lot, and combine the positional relationship between the vehicle nodes and the charging pile nodes to generate the vehicle movement trajectory corresponding to each vehicle; Based on the continuous position change sequence in the vehicle movement trajectory, for each vehicle node, calculate the first estimated time interval for the vehicle node to reach each charging pile node within the charging area along the current moving direction according to the corresponding current change amount of the moving speed and the continuous increment. The starting point of the first estimated time interval is the minimum arrival time based on the current moving speed, and the end point is the maximum arrival time based on the maximum speed change trend corresponding to the continuous increment; Based on the first estimated time intervals of each vehicle node, construct a motion trend prediction model for multiple vehicles, and analyze the trajectory crossing probability and time overlap probability of each vehicle node within the charging area through the motion trend prediction model to predict the intensity of the interactive influence of the motion trends between multiple vehicle nodes as the motion trend prediction results of multiple vehicles; Perform conflict detection on the situation where multiple vehicle nodes select the same target charging pile according to the motion trend prediction results, and generate a warning signal including a charging demand urgency parameter; Input the warning signal into a multi-agent coordination and allocation mechanism, and execute a dynamic balance allocation strategy through the multi-agent coordination and allocation mechanism to calculate the matching degree score of each vehicle for the target charging pile, including: Input the warning signal into a multi-agent coordination and allocation mechanism, and based on the multi-agent coordination and allocation mechanism and according to the preemption conflict relationship between the vehicle and the target charging pile in the warning signal, establish a conflict handling group for each target charging pile. The conflict handling group includes the remaining power attenuation rate of the vehicle in conflict with the current charging pile, the vehicle's preset departure time constraint, and the current available power of the charging pile; Execute dynamic parameter adjustment on the vehicles within the conflict handling group through the multi-agent coordination and allocation mechanism. The dynamic parameter adjustment includes adjusting the remaining power attenuation rate of the vehicle in a reverse association manner according to the current available power of the charging pile, and at the same time converting the remaining duration of the vehicle departure time constraint into an urgency level according to a preset interval; Generate the matching degree score of the vehicle for the target charging pile according to the adjusted remaining power attenuation rate and the adjusted urgency level.

7. A computing device, characterized in that, Includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a charging pile intelligent scheduling processing method as described in any one of claims 1 to 5.

8. A computer storage medium, characterized in that, Stores a computer program, and when the computer program is executed by a computer, it implements a charging pile intelligent scheduling processing method as described in any one of claims 1 to 5.

Images