Battery car battery replacement scheduling control system and method based on optimal path planning

Abstract

This invention discloses a battery swapping scheduling and control system and method for electric vehicles based on optimal path planning, belonging to the field of path planning technology. The system includes a network graph management module, a data acquisition module, a database, a model management module, a battery quantity analysis module, and an optimal path planning module. The network graph management module is used to construct a directed weighted network graph; the data acquisition module is used to collect battery status data and battery swapping data; the database is used to store historical data; the model management module is used to predict the battery swapping frequency of each swapping station in the future; the battery quantity analysis module is used to determine the real-time changes in the number of available batteries in each swapping station; and the optimal path planning module is used to automatically plan the optimal path and push it to the user platform. This invention effectively improves the battery swapping success rate for users by combining the real-time available battery quantity of the swapping station, the predicted battery swapping frequency, and the electric vehicle path planning.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, specifically to a battery swapping scheduling control system and method for electric vehicles based on optimal path planning. Background Technology

[0002] With the explosive growth of urban short-distance delivery and instant logistics industries, electric vehicles have become the core transportation capacity for urban last-mile logistics. Compared with the traditional long-term charging mode, the battery swapping mode has become a key infrastructure to ensure delivery timeliness due to its high efficiency and safety. However, in the large-scale scheduling and control of electric vehicle battery swapping, how to achieve efficient coordination between vehicles, batteries and battery swapping stations remains a key technical problem that needs to be solved.

[0003] Chinese invention patent (CN117094538B) discloses "a battery swapping path management method based on battery analysis". This method pre-constructs a map of the target area and generates a predicted operating path based on the real-time battery power information of the electric vehicle and the target location information. It then combines the real-time battery power information and the predicted operating path to generate a predicted battery swapping area on the map. Based on all the predicted battery swapping areas on the map, it generates expected battery swapping information for each battery swapping cabinet. Furthermore, it combines the swappable battery data and battery swapping orders of each battery swapping cabinet to generate a battery swapping priority weight. By analyzing the battery swapping priority weights of each battery swapping cabinet within the electric vehicle's expected battery swapping area, the target battery swapping cabinet for the electric vehicle can be obtained and an order placed. This solves the problem of low efficiency in existing technologies that rely on manual selection of battery swapping cabinets.

[0004] The aforementioned solutions also deeply link the expected battery swapping information of the battery swapping station with the urgency of the battery swapping of the electric bicycle, preventing other electric bicycles with higher urgency from being unable to swap their batteries in time. However, in actual battery swapping scenarios, these methods lack consideration for execution deviations and fail to fully account for riders' subjective intentions and psychological biases during actual operations. Their battery swapping decisions are often influenced by the convenience of traffic around the station and the direction of immediate delivery orders, causing riders to abandon or change the optimal route issued by the system, making it difficult to achieve the global optimal scheduling goal in actual execution. Moreover, most existing technologies use static battery inventory as the basis for route planning, ignoring the dynamic characteristics of the battery electrochemical recharging process. Especially during peak delivery periods, scheduling instructions and actual available resources are prone to time and space misalignment. The system only dispatches orders based on the current inventory status and fails to predict the charging recovery progress of depleted batteries during the vehicle's transit time, resulting in a decrease in resource turnover rate. Summary of the Invention

[0005] The purpose of this invention is to provide a technical solution to solve the problems raised in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a battery swapping scheduling and control method for electric vehicles based on optimal path planning, comprising:

[0007] Construct a directed weighted network graph that integrates battery swapping cabinet nodes and electric vehicle movement paths, and determine the location information of each battery swapping cabinet node and the edge weights of the road segments traversed by each electric vehicle movement path.

[0008] Collect battery status data and battery swapping data from each battery swapping station in the directed weighted network graph; the battery status data includes the charging status and current charge of each battery in the battery swapping station; the battery swapping data includes the swapping timestamp of the battery in the battery swapping station and various swapping impact parameters at different timestamps.

[0009] Establish a database to store the collected battery swapping data as historical data; based on the stored historical battery swapping data, establish a battery swapping prediction model to predict the battery swapping frequency of each battery swapping station in the future.

[0010] The collected battery status data is analyzed, and the changes in the number of available batteries in each battery swapping cabinet in real time are determined based on the charging status and current power of each battery in the future at different time periods.

[0011] When personnel need to go to the battery swapping station to swap batteries, the system obtains the real-time battery level and location information. Based on the real-time battery status data and the established battery swapping prediction model, the system automatically plans the optimal path in the directed weighted network graph and pushes it to the user platform.

[0012] Furthermore, the charging status of the battery is divided into at least two states: charging and fully charged. When the battery is charging, its current charge level is monitored in real time. Batteries in the fully charged state are considered as available batteries in the battery swapping cabinet. Each time a battery swapping operation occurs, the battery swapping cabinet records the timestamp of the swapping operation and the identifiers of the batteries that are put in and taken out, and simultaneously collects the battery swapping impact parameters corresponding to the timestamp.

[0013] Furthermore, the method for predicting the battery swapping frequency of the battery swapping cabinet in the future is as follows:

[0014] S1. Retrieve historical battery swapping data from the database for analysis, determine the timestamp of each battery swapping session in the swapping cabinet and the swapping impact parameters at different timestamps, divide the swapping impact parameters into stationary impact parameters and abrupt impact parameters, and obtain the set of stationary impact parameters and the set of abrupt impact parameters at different timestamps respectively; divide the time period according to the timestamp of each battery swapping session, and count the frequency of battery swapping in the swapping cabinet in each time period;

[0015] S2. Establish a battery swapping prediction model, using the steady-state impact parameters, abrupt change impact parameters, and battery swapping frequency within historical time periods as inputs, and the future battery swapping frequency as output, to predict the battery swapping frequency of the battery swapping station in the future time period, based on the calculation formula:

[0016] ;

[0017] Among them, D T This indicates the battery swapping frequency of the battery swapping station within the predicted time period T; Softplus represents the non-negative smoothing nonlinear activation operator. Indicates linear reference bias; Indicates the degree of influence of the stationary influence parameter i; n represents the number of stationary influence parameters; x i,T tanh represents the mean value of the parameter i that has a steady effect over a time period T; tanh represents the hyperbolic tangent function. Indicates nonlinear reference bias; The expression indicates the degree of influence of the mutation on parameter j; m represents the number of parameters affected by the mutation; x j,T The mean value of the parameter j affected by the sudden change within time period T is represented by ; k represents the periodic correlation coefficient. ε represents the historical average battery swapping frequency during the same period; ε represents the correction term.

[0018] S3. Using the historical battery swapping data analysis results from S1 as training parameters, substitute them into the battery swapping prediction model in S2 for fitting training to obtain the fitting parameters. And the value of k.

[0019] Furthermore, the method for determining the real-time available battery quantity changes of the battery swapping cabinet in different future time periods is as follows: obtain the charging status and current charge of each battery in the battery swapping cabinet, and mark the batteries that are fully charged as currently available batteries; for batteries that are charging, calculate the remaining charging time required for the battery to reach a fully charged state based on its current charge, rated capacity, and charging rate; starting from the current moment, divide the future time axis into several consecutive time periods according to a preset time interval; for each time period, count the sum of the number of batteries that have completed charging before the start time of the time period and the number of currently available batteries, and use this as the number of available batteries in that time period, thereby obtaining the sequence of changes in the real-time available battery quantity of the battery swapping cabinet over time in different future time periods.

[0020] Furthermore, the steps for automatically planning the optimal path in a directed weighted network graph are as follows:

[0021] S10. Determine the current battery location information. Taking the current battery location as the starting point and each battery swapping cabinet node as the candidate endpoint, based on the real-time battery power of the current battery, remove battery swapping cabinet nodes that cannot be reached before the power is exhausted in the directed weighted network graph to obtain the reachable candidate battery swapping cabinet set P.

[0022] S20. Determine the time required for personnel to travel to each battery swapping station in set P. Based on real-time battery status data, determine the change in the number of available batteries in each battery swapping station over time during the time period required for personnel to travel to each battery swapping station.

[0023] S30. Obtain the current battery swapping impact parameters. Using the time required for personnel to travel to each battery swapping station as the time window length, extract the battery swapping frequency within a historical time period of the same length from historical battery swapping data. Input the current battery swapping impact parameters and the battery swapping frequency within the historical time period into the battery swapping prediction model to predict the battery swapping frequency of each battery swapping station within the time period required for personnel to travel to each battery swapping station, and automatically plan the optimal path to ensure that the optimal path meets the following conditions:

[0024] ;

[0025] Where F represents the battery swapping success rate index for the optimal path; s∈P; The time T required for personnel to travel to the battery swapping cabinet s is indicated. s The number of batteries available in real time; The time T required for personnel to travel to the battery swapping cabinet s is indicated. s Battery swapping frequency within a given time period.

[0026] Furthermore, road nodes and battery swapping cabinet nodes are treated as nodes in a directed weighted network graph, and the directed road segments between adjacent nodes along the road direction are treated as directed edges. Each directed road segment is assigned an edge weight based on its distance.

[0027] The battery swapping scheduling and control system for electric vehicles based on optimal path planning includes a network graph management module, a data acquisition module, a database, a model management module, a battery quantity analysis module, and an optimal path planning module.

[0028] The network graph management module is used to construct a directed weighted network graph that integrates the battery swapping cabinet nodes and the electric vehicle movement paths, and to determine the location information of each battery swapping cabinet node and the edge weights of the road segments traversed by each electric vehicle movement path.

[0029] The data acquisition module is used to collect battery status data and battery swapping data of each battery swapping cabinet in the directed weighted network graph. The battery status data includes the charging status and current power of each battery in the battery swapping cabinet. The battery swapping data includes the swapping timestamp of the battery in the battery swapping cabinet and various swapping impact parameters at different timestamps. The acquired battery status data is sent to the battery quantity analysis module, and the acquired battery swapping data is sent to the database and model management module.

[0030] The database is used to store the collected battery swapping data as historical data.

[0031] The model management module is used to establish a battery swapping prediction model based on stored historical battery swapping data, and to predict the battery swapping frequency of each battery swapping cabinet in the future time period.

[0032] The battery quantity analysis module is used to analyze the collected battery status data, determine the real-time available battery quantity changes of each battery in the battery swapping cabinet in different future time periods based on the charging status and current power of each battery in the battery swapping cabinet, and send the real-time available battery quantity changes of each battery in the battery swapping cabinet in different future time periods to the optimal path planning module.

[0033] The optimal path planning module is used to obtain the real-time battery level and location information of the current battery when personnel need to go to the battery swapping station. Based on the changes in the number of available batteries in each battery swapping station in real time over different future time periods and the established battery swapping prediction model, it automatically plans the optimal path in the directed weighted network graph and pushes it to the user platform.

[0034] Furthermore, a user platform is provided, through which users can view the optimal push path and the real-time changes in the number of available batteries corresponding to each battery swapping cabinet node in the directed weighted network graph.

[0035] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: 1. By establishing a battery swapping prediction model, the frequency of battery swapping at battery swapping stations in the future is predicted, accurately predicting the battery replacement demand of each battery swapping station in the future. Furthermore, the periodic and nonlinear fluctuation characteristics of battery swapping demand are fully explored, eliminating the subjective intentions and psychological biases of riders in actual operations. By anticipating differences in battery supply and demand under different scenarios, the problem of battery supply and demand mismatch is effectively avoided, improving battery turnover efficiency and the system's battery swapping service response capability. 2. By determining the real-time available battery quantity of each battery swapping station over time, and considering the dynamic energy replenishment resources generated by the battery swapping station at the time of vehicle arrival, the real-time available battery quantity of the battery swapping station, the predicted battery swapping frequency, and the electric vehicle route planning are combined, effectively improving the user's battery swapping success rate. By accurately guiding users, the ineffective occupation of battery swapping stations and repeated battery swapping behavior are reduced, thus lowering the number of ineffective battery turnovers. Attached Figure Description

[0036] Figure 1 This is a schematic diagram of the steps of the battery swapping scheduling and control method for electric vehicles based on optimal path planning according to the present invention;

[0037] Figure 2 This is a schematic diagram of the structure of the battery swapping scheduling and control system for electric vehicles based on optimal path planning, as proposed in this invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Please see Figure 1-2 The present invention provides the following technical solution:

[0040] Please see Figure 1 In this first embodiment: the battery swapping scheduling and control method for electric vehicles based on optimal path planning includes:

[0041] Construct a directed weighted network graph that integrates battery swapping cabinet nodes and electric vehicle movement paths, and determine the location information of each battery swapping cabinet node and the edge weights of the road segments traversed by each electric vehicle movement path.

[0042] Furthermore, road nodes and battery swapping cabinet nodes are treated as nodes in a directed weighted network graph, and the directed road segments between adjacent nodes along the road direction are treated as directed edges. Each directed road segment is assigned an edge weight based on its distance.

[0043] In this embodiment, electronic map road network data within the range of the directed weighted network graph is obtained. Road intersections are taken as road nodes, the location information of each battery swapping cabinet node is taken as battery swapping cabinet node, and each battery swapping cabinet node is topologically associated with the nearest road node to be integrated into the directed weighted network graph. The directed road segments between adjacent nodes along the road direction are taken as directed edges, and the direction of each directed edge is consistent with the allowed driving direction of the road segment.

[0044] The battery status data and battery swapping data of each battery swapping cabinet in the directed weighted network graph are collected. The battery status data includes the charging status and current power of each battery in the battery swapping cabinet. The battery swapping data includes the battery swapping timestamp and various battery swapping impact parameters at different timestamps.

[0045] Furthermore, the charging status of the battery is divided into at least two states: charging and fully charged. When the battery is charging, its current charge level is monitored in real time. Batteries in the fully charged state are considered as available batteries in the battery swapping cabinet. Each time a battery swapping operation occurs, the battery swapping cabinet records the timestamp of the swapping operation and the identifiers of the batteries that are put in and taken out, and simultaneously collects the battery swapping impact parameters corresponding to the timestamp.

[0046] It should be noted that the battery swapping impact parameters refer to external environmental parameters that affect users' battery swapping demand; these parameters are divided into stable impact parameters and abrupt impact parameters; the stable impact parameters refer to parameters that have a regularity over a time scale and have a continuous impact on users' battery swapping demand, including but not limited to date type, time period type, and season; the abrupt impact parameters refer to parameters that do not have a regularity and cause instantaneous fluctuations in users' battery swapping demand, including but not limited to sudden weather changes and surges in regional population flow.

[0047] A database is established to store the collected battery swapping data as historical data; based on the stored historical battery swapping data, a battery swapping prediction model is established to predict the battery swapping frequency of each battery swapping station in the future.

[0048] Specifically, the method and steps for predicting the battery swapping frequency of battery swapping cabinets in the future are as follows:

[0049] S1. Retrieve historical battery swapping data from the database for analysis, determine the timestamp of each battery swapping session in the swapping cabinet and the swapping impact parameters at different timestamps, divide the swapping impact parameters into stationary impact parameters and abrupt impact parameters, and obtain the set of stationary impact parameters and the set of abrupt impact parameters at different timestamps respectively; divide the time period according to the timestamp of each battery swapping session, and count the frequency of battery swapping in the swapping cabinet in each time period;

[0050] S2. Establish a battery swapping prediction model, using the steady-state impact parameters, abrupt change impact parameters, and battery swapping frequency within historical time periods as inputs, and the future battery swapping frequency as output, to predict the battery swapping frequency of the battery swapping station in the future time period, based on the calculation formula:

[0051] ;

[0052] Among them, D T This indicates the battery swapping frequency of the battery swapping station within the predicted time period T; Softplus represents the non-negative smoothing nonlinear activation operator. Indicates linear reference bias; Indicates the degree of influence of the stationary influence parameter i; n represents the number of stationary influence parameters; x i,Ttanh represents the mean value of the parameter i that has a steady effect over a time period T; tanh represents the hyperbolic tangent function. Indicates nonlinear reference bias; The expression indicates the degree of influence of the mutation on parameter j; m represents the number of parameters affected by the mutation; x j,T The mean value of the parameter j affected by the sudden change within time period T is represented by ; k represents the periodic correlation coefficient. ε represents the historical average battery swapping frequency during the same period; ε represents the correction term.

[0053] S3. Using the historical battery swapping data analysis results from S1 as training parameters, substitute them into the battery swapping prediction model in S2 for fitting training to obtain the fitting parameters. And the value of k.

[0054] It should be noted that the non-negative smoothing nonlinear activation operator Softplus is used to strictly constrain the prediction results within the non-negative interval, ensuring the output battery swapping frequency D. T It conforms to the actual physical meaning; among them, ; u represents the input value of the battery swapping prediction model.

[0055] In this embodiment, a battery swapping prediction model is established, and the analysis results of historical battery swapping data are substituted into the calculation formula for fitting and training. Based on the set of stationary impact parameters and the set of abrupt impact parameters at different time stamps, the mean values ​​of each stationary impact parameter and each abrupt impact parameter within each time period are obtained. The mean value of each stationary impact parameter within each time period is used as x. i,T The training parameters are used, with the mean of each mutation impact parameter within each time period as x. j,T The training parameters; based on the timestamp, the frequency of battery swapping at the battery swapping station within each time period is used as... and D T The training parameters are used to fit the parameters using the least squares method. The values ​​of k and k; where the time period is divided into 15-minute intervals; and the historical average battery swapping frequency during the same period. With D T The historical cycle is 7 days, such as the historical average battery swapping frequency within a certain time period on Sunday. By predicting the battery swapping frequency of the swapping cabinet in the future, the battery replacement demand of each swapping cabinet in the future can be accurately predicted. The periodic and non-linear fluctuation characteristics of the swapping demand are fully explored, eliminating the subjective intentions and psychological biases of riders in actual operation. By perceiving the differences in battery supply and demand in different scenarios in advance, the battery supply and demand mismatch problem is effectively avoided, and the battery turnover efficiency and the system's battery swapping service response capability are improved.

[0056] The collected battery status data is analyzed, and the real-time changes in the number of available batteries in each battery swapping cabinet are determined based on the charging status and current power of each battery in the future at different time periods.

[0057] Specifically, the method for determining the real-time available battery quantity changes of the battery swapping cabinet in different future time periods is as follows: The charging status and current charge level of each battery in the battery swapping cabinet are obtained, and batteries that are fully charged are marked as currently available batteries. For batteries that are charging, the remaining charging time required to reach a fully charged state is calculated based on their current charge level, rated capacity, and charging rate. Starting from the current moment, the future time axis is divided into several consecutive time periods according to a preset time interval. For each time period, the sum of the number of batteries that have completed charging before the start time of that time period and the number of currently available batteries is counted as the number of available batteries in that time period, thus obtaining the sequence of changes in the real-time available battery quantity of the battery swapping cabinet over time in different future time periods.

[0058] When personnel need to go to the battery swapping station to swap batteries, the system obtains the real-time battery level and location information. Based on the real-time battery status data and the established battery swapping prediction model, the system automatically plans the optimal path in the directed weighted network graph and pushes it to the user platform.

[0059] Specifically, the steps for automatically planning the optimal path in a directed weighted network graph are as follows:

[0060] S10. Determine the current battery location information. Taking the current battery location as the starting point and each battery swapping cabinet node as the candidate endpoint, based on the real-time battery power of the current battery, remove battery swapping cabinet nodes that cannot be reached before the power is exhausted in the directed weighted network graph to obtain the reachable candidate battery swapping cabinet set P.

[0061] S20. Determine the time required for personnel to travel to each battery swapping station in set P. Based on real-time battery status data, determine the change in the number of available batteries in each battery swapping station over time during the time period required for personnel to travel to each battery swapping station.

[0062] S30. Obtain the current battery swapping impact parameters. Using the time required for personnel to travel to each battery swapping station as the time window length, extract the battery swapping frequency within a historical time period of the same length from historical battery swapping data. Input the current battery swapping impact parameters and the battery swapping frequency within the historical time period into the battery swapping prediction model to predict the battery swapping frequency of each battery swapping station within the time period required for personnel to travel to each battery swapping station, and automatically plan the optimal path to ensure that the optimal path meets the following conditions:

[0063] ;

[0064] Where F represents the battery swapping success rate index for the optimal path; s∈P; The time T required for personnel to travel to the battery swapping cabinet s is indicated. s The number of batteries available in real time; The time T required for personnel to travel to the battery swapping cabinet s is indicated. s Battery swapping frequency within a given time period.

[0065] In this embodiment, based on the reachable candidate battery swapping cabinet set P, the battery swapping success index value is calculated when personnel go to each battery swapping cabinet to swap batteries. The system sets a threshold for successful battery swapping, classifies the success rate of battery swapping based on the threshold, and displays it on the user platform. These levels include high success rate, medium success rate, and low success rate. Users can choose to go to the corresponding battery swapping station based on the planned optimal route or success rate, thereby reducing the risk of no available batteries upon arrival and improving the battery swapping experience.

[0066] It should be noted that by determining the change in the number of available batteries in each battery swapping station over time, and taking into account the dynamic energy replenishment resources generated by the battery swapping station at the time of vehicle arrival, the system combines the number of available batteries in the battery swapping station, the predicted battery swapping frequency, and the electric vehicle route planning. This effectively improves the success rate of battery swapping for users. By accurately guiding users, it reduces the ineffective occupation of battery swapping stations and repeated battery swapping behavior, and reduces the number of ineffective battery turnovers.

[0067] Please see Figure 2 In this second implementation: an electric vehicle battery swapping scheduling and control system based on optimal path planning is provided. The system includes a network graph management module, a data acquisition module, a database, a model management module, a battery quantity analysis module, and an optimal path planning module.

[0068] The network graph management module is used to construct a directed weighted network graph that integrates the battery swapping cabinet nodes and the electric vehicle movement paths, and to determine the location information of each battery swapping cabinet node and the edge weights of the road segments traversed by each electric vehicle movement path.

[0069] The data acquisition module is used to collect battery status data and battery swapping data of each battery swapping cabinet in the directed weighted network graph. The battery status data includes the charging status and current power of each battery in the battery swapping cabinet. The battery swapping data includes the swapping timestamp of the battery in the battery swapping cabinet and various swapping impact parameters at different timestamps. The acquired battery status data is sent to the battery quantity analysis module, and the acquired battery swapping data is sent to the database and model management module.

[0070] The database is used to store the collected battery swapping data as historical data.

[0071] The model management module is used to establish a battery swapping prediction model based on stored historical battery swapping data, and to predict the battery swapping frequency of each battery swapping cabinet in the future time period.

[0072] The battery quantity analysis module is used to analyze the collected battery status data, determine the real-time available battery quantity changes of each battery in the battery swapping cabinet in different future time periods based on the charging status and current power of each battery in the battery swapping cabinet, and send the real-time available battery quantity changes of each battery in the battery swapping cabinet in different future time periods to the optimal path planning module.

[0073] The optimal path planning module is used to obtain the real-time battery level and location information of the current battery when personnel need to go to the battery swapping station. Based on the changes in the number of available batteries in each battery swapping station in real time over different future time periods and the established battery swapping prediction model, it automatically plans the optimal path in the directed weighted network graph and pushes it to the user platform.

[0074] Furthermore, a user platform is provided, through which users can view the optimal push path and the real-time changes in the number of available batteries corresponding to each battery swapping cabinet node in the directed weighted network graph.

[0075] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A battery swapping scheduling and control method for electric vehicles based on optimal path planning, characterized in that: include: Construct a directed weighted network graph that integrates battery swapping cabinet nodes and electric vehicle movement paths, and determine the location information of each battery swapping cabinet node and the edge weights of the road segments traversed by each electric vehicle movement path. Collect battery status data and battery swapping data from each battery swapping station in the directed weighted network graph; the battery status data includes the charging status and current charge of each battery in the battery swapping station; the battery swapping data includes the swapping timestamp of the battery in the battery swapping station and various swapping impact parameters at different timestamps. Establish a database to store the collected battery swapping data as historical data; based on the stored historical battery swapping data, establish a battery swapping prediction model to predict the battery swapping frequency of each battery swapping station in the future. The collected battery status data is analyzed, and the changes in the number of available batteries in each battery swapping cabinet in real time are determined based on the charging status and current power of each battery in the future at different time periods. When personnel need to go to the battery swapping station to swap batteries, the system obtains the real-time battery level and location information. Based on the real-time battery status data and the established battery swapping prediction model, the system automatically plans the optimal path in the directed weighted network graph and pushes it to the user platform.

2. The battery swapping scheduling and control method for electric vehicles based on optimal path planning according to claim 1, characterized in that: The battery charging status is divided into at least two states: charging and fully charged. When the battery is charging, its current charge level is monitored in real time. Batteries in the fully charged state are designated as available batteries in the battery swapping cabinet. Each time a battery swapping operation occurs, the battery swapping cabinet records the timestamp of the swapping operation and the identifiers of the batteries that are placed in and removed, and simultaneously collects the battery swapping impact parameters corresponding to the timestamp.

3. The battery swapping scheduling and control method for electric vehicles based on optimal path planning according to claim 2, characterized in that: The method for predicting the battery swapping frequency of battery swapping cabinets in the future is as follows: S1. Retrieve historical battery swapping data from the database for analysis, determine the timestamp of each battery swapping session in the swapping cabinet and the swapping impact parameters at different timestamps, divide the swapping impact parameters into stationary impact parameters and abrupt impact parameters, and obtain the set of stationary impact parameters and the set of abrupt impact parameters at different timestamps respectively; divide the time period according to the timestamp of each battery swapping session, and count the frequency of battery swapping in the swapping cabinet in each time period; S2. Establish a battery swapping prediction model, using the steady-state impact parameters, abrupt change impact parameters, and battery swapping frequency within historical time periods as inputs, and the future battery swapping frequency as output, to predict the battery swapping frequency of the battery swapping station in the future time period, based on the calculation formula: ； Among them, D T This indicates the battery swapping frequency of the battery swapping station within the predicted time period T; Softplus represents the non-negative smoothing nonlinear activation operator. Indicates linear reference bias; Indicates the degree of influence of the stationary influence parameter i; n represents the number of stationary influence parameters; x i,T tanh represents the mean value of the parameter i that has a steady effect over a time period T; tanh represents the hyperbolic tangent function. Indicates nonlinear reference bias; The expression indicates the degree of influence of the mutation on parameter j; m represents the number of parameters affected by the mutation; x j,T The mean value of the parameter j affected by the sudden change within time period T is represented by ; k represents the periodic correlation coefficient. ε represents the historical average battery swapping frequency during the same period; ε represents the correction term. S3. Using the historical battery swapping data analysis results from S1 as training parameters, substitute them into the battery swapping prediction model in S2 for fitting training to obtain the fitting parameters. And the value of k.

4. The battery swapping scheduling and control method for electric vehicles based on optimal path planning according to claim 3, characterized in that: The method to determine the changes in the number of available batteries in the battery swapping cabinet in real time during different time periods in the future is as follows: obtain the charging status and current power of each battery in the battery swapping cabinet, and mark the batteries with the charging status of fully charged as currently available batteries. For a battery that is charging, calculate the remaining charging time required for the battery to reach a fully charged state based on its current charge level, rated capacity, and charging rate. Starting from the current moment, the future time axis is divided into several consecutive time periods according to a preset time interval. For each time period, the number of batteries that have been charged before the start of the time period is counted and the number of currently available batteries is counted as the number of available batteries in that time period. This gives the sequence of changes in the number of real-time available batteries in the battery swapping cabinet over time in different future time periods.

5. The battery swapping scheduling and control method for electric vehicles based on optimal path planning according to claim 3, characterized in that: The steps for automatically planning the optimal path in a directed weighted network graph are as follows: S10. Determine the current battery location information. Taking the current battery location as the starting point and each battery swapping cabinet node as the candidate endpoint, based on the real-time battery power of the current battery, remove battery swapping cabinet nodes that cannot be reached before the power is exhausted in the directed weighted network graph to obtain the reachable candidate battery swapping cabinet set P. S20. Determine the time required for personnel to travel to each battery swapping station in set P. Based on real-time battery status data, determine the change in the number of available batteries in each battery swapping station over time during the time period required for personnel to travel to each battery swapping station. S30. Obtain the current battery swapping impact parameters. Using the time required for personnel to travel to each battery swapping station as the time window length, extract the battery swapping frequency within a historical time period of the same length from historical battery swapping data. Input the current battery swapping impact parameters and the battery swapping frequency within the historical time period into the battery swapping prediction model to predict the battery swapping frequency of each battery swapping station within the time period required for personnel to travel to each battery swapping station, and automatically plan the optimal path to ensure that the optimal path meets the following conditions: ； Where F represents the battery swapping success rate index for the optimal path; s∈P; The time T required for personnel to travel to the battery swapping cabinet s is indicated. s The number of batteries available in real time; The time T required for personnel to travel to the battery swapping cabinet s is indicated. s Battery swapping frequency within a given time period.

6. The battery swapping scheduling and control method for electric vehicles based on optimal path planning according to any one of claims 1-5, characterized in that: The road nodes and battery swapping cabinet nodes are treated as nodes in a directed weighted network graph. The directed road segments between adjacent nodes along the road direction are treated as directed edges, and each directed road segment is assigned an edge weight based on its distance.

7. A battery swapping dispatch and control system for electric vehicles based on optimal path planning, characterized in that: The system includes a network graph management module, a data acquisition module, a database, a model management module, a battery quantity analysis module, and an optimal path planning module; The network graph management module is used to construct a directed weighted network graph that integrates the battery swapping cabinet nodes and the electric vehicle movement paths, and to determine the location information of each battery swapping cabinet node and the edge weights of the road segments traversed by each electric vehicle movement path. The data acquisition module is used to collect battery status data and battery swapping data of each battery swapping cabinet in the directed weighted network graph. The battery status data includes the charging status and current power of each battery in the battery swapping cabinet. The battery swapping data includes the swapping timestamp of the battery in the battery swapping cabinet and various swapping impact parameters at different timestamps. The acquired battery status data is sent to the battery quantity analysis module, and the acquired battery swapping data is sent to the database and model management module. The database is used to store the collected battery swapping data as historical data. The model management module is used to establish a battery swapping prediction model based on stored historical battery swapping data, and to predict the battery swapping frequency of each battery swapping cabinet in the future time period. The battery quantity analysis module is used to analyze the collected battery status data, determine the real-time available battery quantity changes of each battery in the battery swapping cabinet in different future time periods based on the charging status and current power of each battery in the battery swapping cabinet, and send the real-time available battery quantity changes of each battery in the battery swapping cabinet in different future time periods to the optimal path planning module. The optimal path planning module is used to obtain the real-time battery level and location information of the current battery when personnel need to go to the battery swapping station. Based on the changes in the number of available batteries in each battery swapping station in real time over different future time periods and the established battery swapping prediction model, it automatically plans the optimal path in the directed weighted network graph and pushes it to the user platform.

8. The electric vehicle battery swapping dispatch control system based on optimal path planning according to claim 7, characterized in that: A user platform is provided, through which users can view the optimal push path and the real-time changes in the number of available batteries corresponding to each battery swapping cabinet node in the directed weighted network graph.

Images