Intelligent battery replacement service system supporting battery reservation

Abstract

The application provides an intelligent battery replacement service system supporting battery reservation, relates to the technical field of new energy, and comprises the following steps: extracting position coordinates and load data of surrounding battery replacement stations from the influence range, grouping available battery replacement stations based on the position coordinates and the load data by using a clustering algorithm to obtain resource cluster distribution; for the resource cluster distribution, the nearest resource cluster is matched in combination with position information in a user reservation request, and it is judged whether the reservation load of the corresponding battery replacement station exceeds a capacity threshold, so as to generate a preliminary battery replacement distribution scheme based on the reservation request; the intelligent battery replacement service system supporting battery reservation helps to improve the intelligent level, resource utilization efficiency and service reliability of the battery replacement reservation service in a complex operation scenario.

Description

Technical Field

[0001] This invention relates to the field of new energy technology, and in particular to an intelligent battery swapping service system that supports battery reservation. Background Technology

[0002] With the continuous expansion of urban electrification, battery swapping services are gradually becoming an important technological means to ensure energy replenishment efficiency in high-frequency travel scenarios. Reserving battery swapping resources in advance through appointments helps improve the utilization rate of swapping stations and reduce user waiting time. Therefore, battery swapping service systems that support battery reservations have been widely deployed in practical applications and are gradually forming a battery swapping service network consisting of multiple swapping stations.

[0003] Existing battery swapping reservation systems can typically receive and process user reservation information and complete reservation confirmation and resource allocation based on established rules. However, in actual operation, due to the significant dynamism and uncertainty of battery swapping site operating status, power supply capacity, and user reservation demand, existing technologies exhibit considerable shortcomings in complex operating environments. Especially in situations involving power supply fluctuations, equipment failures, or concentrated reservations in certain areas, the available service capacity of battery swapping sites may change rapidly, making it difficult to effectively execute the original reservation arrangements.

[0004] From a system perspective, existing technologies typically provide a coarse-grained description of the operational status of battery swapping stations, lacking the ability to effectively classify and assess station status based on multi-dimensional operational data. This makes it difficult to promptly identify potential stations with reservation anomalies. Consequently, when a battery swapping station is unable to provide service as scheduled due to insufficient power or other abnormal factors, the system often can only passively cancel reservations or rely on manual intervention. This lack of systematic analysis of the scope of the anomaly's impact makes it difficult to coordinate resources at surrounding stations within the battery swapping network.

[0005] Furthermore, existing battery swapping reservation systems often employ static or localized optimization strategies for resource scheduling, failing to fully integrate the spatial distribution of swapping stations and real-time load conditions for a holistic analysis of available resources. When reservation anomalies occur, the system typically recommends stations based solely on simple distance or fixed priority rules, easily leading to some stations rapidly exceeding their reservation limits while other stations underutilize their resources, thus triggering new service congestion issues.

[0006] Furthermore, in scenarios with significant fluctuations in user reservation volume, existing technologies have limited predictive capabilities for changes in reservation traffic, lacking a mechanism to effectively predict reservation peaks based on historical and real-time operational data. This makes it difficult for the system to adjust reservation allocation strategies in a timely manner when faced with concentrated reservations or sudden anomalies, thus affecting the overall stability and continuity of the service. Simultaneously, at the reservation guidance level, existing systems handle different user reservation needs in a relatively simplistic way, lacking a reasonable differentiation of reservation priorities and an orderly guidance mechanism. When reservations need to be adjusted or diverted, the system struggles to generate a clear and executable guidance sequence, leading to inconsistent user experiences and even raising user doubts about the reliability of the battery swapping service. Summary of the Invention

[0007] This invention provides an intelligent battery swapping service system that supports battery reservation, the method comprising: The state modeling module receives battery swapping reservation requests submitted by users, extracts reservation time, target swapping station and battery demand information from the reservation requests, and combines real-time operation data of the swapping stations to model the operation status of the stations corresponding to the reservation requests, and obtains the current station status classification. The judgment and impact analysis module determines whether the power supply data is lower than a preset threshold based on the site status classification. If it is lower than the preset threshold, the corresponding site is judged as an abnormal reservation site, and the impact range of the abnormal reservation site in the battery swapping network is further determined. The resource cluster analysis module extracts the location coordinates and load data of surrounding battery swapping stations from the influence area, and uses a clustering algorithm to group the available battery swapping stations based on the location coordinates and load data to obtain the resource cluster distribution. The initial battery swapping allocation module, based on the resource cluster distribution, matches the nearest resource cluster with the location information in the user's reservation request, and determines whether the reservation load of the corresponding battery swapping station exceeds the capacity threshold, thereby generating an initial battery swapping allocation scheme based on the reservation request.

[0008] Preferably, the state modeling module receives a battery swapping reservation request submitted by a user, extracts the reservation time, target battery swapping station, and battery demand information from the reservation request, and models the station's operating status corresponding to the reservation request by combining the station's real-time operating data, thus obtaining a current station status classification, including: Obtain the battery swap reservation request message submitted by the user and the real-time operation data of the target swap station. The real-time operation data includes the real-time battery inventory quantity, the set of remaining charging time of batteries currently being charged, and the length of the queue of vehicles. Calculate the estimated net available battery value at the reservation time based on the battery swap reservation request message and the real-time operation data, and generate an operation load feature vector; The operating load feature vector is input into a pre-trained classification model to obtain the current site status classification of the target battery swapping site in response to the battery swapping reservation request message.

[0009] Preferably, the determination and impact analysis module determines whether the power supply data is below a preset threshold based on the site status classification. If it is below the preset threshold, the corresponding site is determined as an abnormal reservation site, and the impact range of the abnormal reservation site in the battery swapping network is further determined, including: Obtain real-time operation logs of the battery swapping station and extract site status and power supply data; use a decision tree algorithm to obtain site status category information. Based on the site status category information, a preset power supply threshold is retrieved. If the power supply data is lower than the preset power supply threshold, the site is marked as an abnormal reservation site. Based on the abnormal reservation sites, a battery swapping network association map is constructed to identify the set of associated sites; Calculate the reservation overflow amount and its diffusion path of the associated site set to determine the impact range of the abnormal reservation sites in the battery swapping network.

[0010] Preferably, the resource cluster analysis module extracts the location coordinates and load data of surrounding battery swapping stations from the influence area, and uses a clustering algorithm based on the location coordinates and load data to group the available battery swapping stations to obtain the resource cluster distribution, including: The influence range is defined based on a preset center point, and the location coordinates and load data of surrounding battery swapping stations are extracted from the influence range. Generate a multidimensional feature vector describing the site status based on the location coordinates and the load data; Calculate the Euclidean distance between the multidimensional feature vectors to establish neighborhood connection relationships, forming an adjacency graph structure that reflects spatial proximity and load similarity; Based on the adjacency graph structure, a clustering algorithm is used to group the available battery swapping sites to obtain a resource cluster distribution.

[0011] Preferably, the preliminary battery swapping allocation module, based on the resource cluster distribution, matches the nearest resource cluster by combining the location information in the user reservation request, and determines whether the reservation load of the corresponding battery swapping station exceeds the capacity threshold, thereby generating a preliminary battery swapping allocation scheme based on the reservation request, including: Obtain the location coordinates from the user's reservation request, and locate the nearest resource cluster based on the location coordinates and the resource cluster distribution topology. Analyze the cumulative reserved load value and capacity threshold setting of the physical battery swapping station in the nearest resource cluster; If the cumulative reserved load value is less than the capacity threshold setting, an availability verification pass signal is generated; In response to the availability verification, a preliminary battery swapping allocation plan is generated via a signal. This preliminary battery swapping allocation plan is based on battery swapping sites with remaining service capacity.

[0012] Preferably, it also includes a dynamic reservation allocation path generation module. When the reserved load in the initial battery swapping allocation scheme exceeds the capacity threshold, it is used to build a prediction model based on historical reservation data and real-time operation data to predict the peak user reservation traffic and determine the adjusted dynamic reservation allocation path accordingly. Specifically, it includes: When the reserved load value in the initial battery swapping allocation plan exceeds the preset capacity threshold, historical reservation data and real-time operation data are obtained. Time-series feature extraction is performed on the historical reservation data and the real-time operation data to construct a feature vector sequence; The feature vector sequence is input into a preset time series prediction model to obtain the peak user reservation traffic; The congestion coefficient is calculated based on the peak user reservation traffic, and the path weight values ​​in the battery swapping network topology are updated based on the congestion coefficient. Based on the path weight values, search for the minimum cost path and determine the adjusted dynamic reservation allocation path.

[0013] Preferably, it also includes a reservation guidance sequence generation module, which extracts reservation request priority information from user reservation records based on the dynamic reservation allocation path, sorts the reservation requests, and generates a reservation guidance sequence, specifically including: Obtain user reservation records, calculate a basic credit score based on the historical default rate and service type weight in the user reservation records, and generate reservation request priorities by combining the basic credit score with time sensitivity features; Obtain the dynamic reservation allocation path and path congestion index, map and compare the reservation request priority with the path congestion index, and calculate the matching correlation between the reservation request and the currently available resources. The ranking weight is determined based on the matching correlation, and the reservation requests are reorganized and sorted according to the ranking weight to generate a reservation guidance sequence.

[0014] Preferably, it also includes a reservation guidance and service record module, which pushes battery swapping guidance or backup battery swapping notifications to the user terminal according to the reservation guidance sequence, and updates the system operation log after the user responds and confirms, so as to obtain the corresponding reservation service continuity record, specifically including: Based on the resource matching status results generated from the reservation guidance sequence and the battery swapping station load data, a standardized guidance data package containing battery swapping guidance information or backup battery swapping notification is constructed. The standardized boot data packet is sent to the user terminal, and a response confirmation signal is received from the user terminal based on the standardized boot data packet.

[0015] Preferably, the appointment guidance and service record module pushes battery swapping guidance or backup battery swapping notifications to the user terminal according to the appointment guidance sequence, and updates the system operation log after the user responds and confirms, so as to obtain the corresponding appointment service continuity record, and further includes: The system operation log is updated based on the response confirmation signal, and the logical closed-loop features in the updated system operation log are extracted to generate a reservation service continuity record.

[0016] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0017] This invention effectively improves the stability and continuity of reservation services under dynamic changes and abnormal conditions in the operation of battery swapping stations. By modeling real-time operational data of battery swapping stations and classifying and judging station status, abnormal reservation situations such as insufficient power supply can be identified in a timely manner, preventing reservation resources from being ineffectively occupied at unavailable stations. Simultaneously, by uniformly analyzing surrounding battery swapping stations within the impact range of abnormal stations and dividing resources into clusters based on spatial location and load conditions, it helps to achieve more reasonable resource coordination and allocation at the battery swapping network level, thereby reducing the risk of local congestion. Furthermore, by introducing a reservation traffic prediction mechanism based on historical reservation data and real-time operational data, potential reservation peaks can be detected in advance, and reservation allocation paths can be dynamically adjusted, improving the system's responsiveness to sudden changes in demand. In addition, by prioritizing user reservation requests and generating a clear reservation guidance sequence, orderly diversion and clear guidance for users can be achieved, reducing uncertainty in the reservation adjustment process and improving user experience and trust in the battery swapping service. Overall, this invention helps improve the intelligence level, resource utilization efficiency, and service reliability of battery swapping reservation services in complex operating scenarios. Attached Figure Description

[0018] Figure 1 This is a connection diagram of the system modules of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] like Figure 1 As shown in this embodiment, a smart battery swapping service system supporting battery reservation is described, the method of which includes: The state modeling module receives battery swapping reservation requests submitted by users, extracts reservation time, target swapping station and battery demand information from the reservation requests, and combines real-time operation data of the swapping stations to model the operation status of the stations corresponding to the reservation requests, and obtains the current station status classification. The judgment and impact analysis module determines whether the power supply data is lower than a preset threshold based on the site status classification. If it is lower than the preset threshold, the corresponding site is judged as an abnormal reservation site, and the impact range of the abnormal reservation site in the battery swapping network is further determined. The resource cluster analysis module extracts the location coordinates and load data of surrounding battery swapping stations from the influence area, and uses a clustering algorithm to group the available battery swapping stations based on the location coordinates and load data to obtain the resource cluster distribution. The preliminary battery swapping allocation module, based on the resource cluster distribution, matches the nearest resource cluster with the location information in the user reservation request, and determines whether the reservation load of the corresponding battery swapping station exceeds the capacity threshold, thereby generating a preliminary battery swapping allocation scheme based on the reservation request. The dynamic reservation allocation path generation module is used to build a prediction model based on historical reservation data and real-time operation data when there is a situation in the preliminary battery swapping allocation scheme where the reserved load exceeds the capacity threshold, to predict the peak user reservation traffic, and to determine the adjusted dynamic reservation allocation path accordingly. The appointment guidance sequence generation module extracts appointment request priority information from user appointment records based on the dynamic appointment allocation path, sorts appointment requests, and generates an appointment guidance sequence. The appointment guidance and service record module pushes battery swapping guidance or backup battery swapping notification to the user terminal according to the appointment guidance sequence, and updates the system operation log after the user responds and confirms, so as to obtain the corresponding appointment service continuity record.

[0021] In this embodiment, the intelligent battery swapping service system supporting battery reservations achieves predictive assessment and dynamic scheduling of battery swapping service capabilities through joint modeling of user reservation requests and the operational status of battery swapping stations. When a user submits a battery swapping reservation request via a terminal, the system first parses the reservation time, target battery swapping station, battery demand information, and user location information in the request, and simultaneously acquires real-time operational data of the target battery swapping station. This real-time operational data includes information such as the current available battery inventory, the remaining charging time set of batteries currently being charged, and the number of vehicles currently queuing at the station. Based on the time difference between the reservation time and the current time, the system calculates the availability of batteries currently being charged at the reservation time, selects the number of batteries that can complete charging before the reservation time, and combines this number with the existing available battery inventory to obtain the estimated net available battery value at the reservation time. Furthermore, the system will construct operational load characteristics by combining the expected net available battery value with operational status parameters such as queue length of vehicles and historical service rate, and input them into a pre-trained site status classification model to output the site status classification result of the target battery swapping site at the reservation time, thereby realizing a forward-looking judgment on the site's service capacity.

[0022] After obtaining the site status classification results, the system further determines the feasibility of reservations based on the power supply capacity of the battery swapping stations. The system retrieves the power supply data of the target battery swapping station at the reservation time and compares the power supply data with a preset power supply safety threshold. When the available power supply capacity of a station at the reservation time is detected to be lower than the preset threshold, the system marks the station as a reservation anomaly. Subsequently, the system is not limited to processing a single station, but constructs a battery swapping network association map based on the overall structure of the battery swapping network. Using the reservation anomaly station as the central node, it identifies a set of surrounding battery swapping stations that are related to it in terms of geographical location, service coverage, or user traffic. By combining historical reservation overflow data with the relationships between stations, the system further analyzes the diffusion trend of reservation demand in the battery swapping network, thereby determining the potential impact range of the reservation anomaly station on surrounding stations and providing a basis for subsequent resource coordination.

[0023] After clarifying the scope of impact, the system performs a unified resource analysis on the battery swapping sites within that scope. The system extracts the spatial coordinates and real-time load data of each battery swapping site within the scope of impact, and standardizes this data to construct a multi-dimensional feature vector describing the spatial attributes and service load characteristics of the sites. Based on this multi-dimensional feature vector, the system calculates the distance metric between sites and constructs an adjacency graph structure reflecting the spatial proximity and load similarity of the sites. Based on this adjacency graph structure, the system runs a clustering algorithm to group multiple battery swapping sites that are spatially close and have similar service load levels into the same resource cluster. This integrates the originally scattered battery swapping sites into resource units with collaborative service capabilities, thereby improving the flexibility and efficiency of overall resource scheduling.

[0024] After completing the resource cluster construction, the system further generates a battery swapping allocation scheme for user reservation requests. Based on the location information in the user reservation request, the system matches the resource cluster closest to the user's location in the resource cluster distribution topology and traverses the physical battery swapping sites within that resource cluster to obtain the cumulative reservation load value and corresponding capacity threshold of each site at the reservation time. When a battery swapping site is detected whose cumulative reservation load does not exceed the capacity threshold, the system uses this site as a candidate service node and generates a preliminary battery swapping allocation scheme. When the reservation load of all candidate sites in the preliminary battery swapping allocation scheme exceeds the capacity threshold, the system builds a prediction model based on historical reservation data and real-time operational data to predict the peak user reservation traffic in the future period and calculates the congestion coefficient of different paths in the battery swapping network accordingly. The system dynamically updates the path weights in the battery swapping network topology based on the congestion coefficient and searches for the minimum cost path in the updated topology to determine the adjusted dynamic reservation allocation path, thereby achieving dynamic guidance and distribution of reservation traffic.

[0025] After determining the dynamic reservation allocation path, the system further guides reservation requests in an orderly manner by combining user reservation records. The system extracts reservation request priority information reflecting the user's historical fulfillment status and service type characteristics from the user reservation records, and comprehensively matches the reservation request priority with the path status corresponding to the dynamic reservation allocation path. Multiple reservation requests are then sorted to generate a reservation guidance sequence. Subsequently, the system pushes battery swapping guidance information or backup battery swapping notifications to the user terminal according to the reservation guidance sequence, and updates the system operation log after receiving the user's confirmation response, thereby forming a complete reservation service continuity record and achieving closed-loop management of the entire battery swapping reservation service process.

[0026] The state modeling module receives battery swapping reservation requests submitted by users and extracts reservation time, target swapping station, and battery demand information from the requests. Combined with real-time operational data from the swapping stations, it models the operational status of the corresponding stations to obtain current station status classifications, including: Obtain the battery swap reservation request message submitted by the user and the real-time operation data of the target swap station. The real-time operation data includes the real-time battery inventory quantity, the set of remaining charging time of batteries currently being charged, and the length of the queue of vehicles. Calculate the estimated net available battery value at the reservation time based on the battery swap reservation request message and the real-time operation data, and generate an operation load feature vector; The operating load feature vector is input into a pre-trained classification model to obtain the current site status classification of the target battery swapping site in response to the battery swapping reservation request message.

[0027] In this embodiment, the state modeling module is used to quantitatively determine the actual service capacity of the target battery swapping station at the user's reservation time. Its operation is triggered by the user's submitted battery swapping reservation request message. The reservation request message is sent from the user terminal to the system backend via the communication network. The message includes at least reservation time information, target battery swapping station identification information, and battery demand information, where the battery demand information represents the number or type of batteries required by the user in this reservation. Upon receiving the reservation request message, the system first parses the message content and uses the parsed reservation time as the time reference for subsequent operational status prediction.

[0028] While parsing the reservation request message, the system acquires the real-time operational data of the target battery swapping station through the data acquisition interface corresponding to that station. This real-time operational data includes three core components: first, the real-time battery inventory quantity, indicating the number of batteries that have completed charging and are immediately available for swapping; second, the set of remaining charging time for batteries currently charging, recording the remaining time required for all batteries in the station to complete charging at the current moment; and third, the queue length of vehicles, reflecting the number of vehicles currently waiting for battery swapping services at the station, thus characterizing the station's immediate service pressure level. All of the above real-time operational data originates from the station control system or back-end management system of the battery swapping station and undergoes timestamp verification upon acquisition to ensure data consistency with the time of reservation request processing.

[0029] After receiving the reservation request message and real-time operational data, the state modeling module uses the reservation time as the prediction target to estimate the available resources at the site at that reservation time. Specifically, the system first evaluates each battery's remaining charging time based on the time difference between the reservation time and the current time, including batteries with remaining charging time less than or equal to the time difference in the estimated number of batteries that can be charged. This process is completed by iterating through the set of remaining charging times and does not involve any formula calculations; instead, it uses logical filtering based on the time sequence. Subsequently, the system merges the estimated number of batteries that can be charged with the real-time battery inventory to obtain the estimated total number of available batteries at the reservation time.

[0030] After obtaining the estimated total available battery capacity, the system further deducts this amount from the battery demand information in the reservation request message to obtain the estimated net available battery value at the reservation time. This estimated net available battery value reflects the remaining battery capacity of the target battery swapping station after fulfilling the current reservation request, which can still be used for subsequent services. This net value is an integer, and its magnitude directly affects whether the station has sufficient service capacity.

[0031] After calculating the estimated net available battery value, the state modeling module uniformly organizes the estimated net available battery value, queue length of vehicles, and battery demand information from reservation requests, generating an operational load feature vector according to preset data structure rules. These data structure rules were determined during the system deployment phase to ensure consistent data format across different reservation requests when inputting into the classification model. Each data item in the operational load feature vector originates from the aforementioned calculation or data collection results, without introducing additional inference information, thus ensuring that the feature vector accurately reflects the station's operational status.

[0032] Subsequently, the state modeling module inputs the operational load feature vector into a pre-trained classification model. This classification model is trained before the system goes live based on historical battery swapping station operation data, historical reservation records, and actual service results. Its training samples clearly label the corresponding station state categories under different operational load conditions. After model training is complete, its parameters are fixed and deployed in the system operating environment. During actual use, only the inference process is executed, and no further online training is performed. The classification model analyzes and processes the operational load feature vector, outputting the current station state classification result corresponding to the reservation request.

[0033] The site status classification results are used to characterize the carrying capacity level of the target battery swapping site to meet user battery swapping demand at the scheduled time. For example, it is used to distinguish different states such as being able to normally handle reservations, having limited carrying capacity, or having service risks. This classification result serves as the input basis for subsequent judgment and impact analysis modules, used to further determine whether abnormal handling or resource adjustment of reservation requests is necessary. Through the above continuous and explicit calculation and judgment process, predictable and quantifiable modeling of the operating status of battery swapping sites is achieved, providing reliable data support for the stable operation of the entire intelligent battery swapping reservation service system.

[0034] The judgment and impact analysis module classifies the power supply data based on the site status to determine whether it is below a preset threshold. If it is below the preset threshold, the corresponding site is judged as an abnormal reservation site, and the impact range of the abnormal reservation site in the battery swapping network is further determined, including: Obtain real-time operation logs of the battery swapping station and extract site status and power supply data; use a decision tree algorithm to obtain site status category information. Based on the site status category information, a preset power supply threshold is retrieved. If the power supply data is lower than the preset power supply threshold, the site is marked as an abnormal reservation site. Based on the abnormal reservation sites, a battery swapping network association map is constructed to identify the set of associated sites; Calculate the reservation overflow amount and its diffusion path of the associated site set to determine the impact range of the abnormal reservation sites in the battery swapping network.

[0035] In this embodiment, the judgment and impact analysis module is used to further judge the power supply reliability of the battery swapping station under the reservation scenario after the site status modeling is completed, and to identify the abnormal reservation sites and their impact range in the battery swapping network accordingly. The processing flow of this module is based on the real-time operation logs of the target battery swapping station, and completes the entire process from anomaly identification to impact range determination through step-by-step analysis and calculation.

[0036] In the specific implementation process, the system first obtains the real-time operation logs of the target battery swapping station from the station's backend management system. These real-time operation logs are continuously generated sets of data records, each containing the station's operational status identifier and power supply data at the corresponding time point. The power supply data reflects the station's ability to stably output power for battery charging and battery swapping equipment operation at that time. After obtaining the operation logs, the system performs time sorting and integrity verification on the log data to ensure that the selected log data covers the time of the reservation request processing and the necessary time intervals before and after it.

[0037] After acquiring the log data, the judgment and impact analysis module parses the operation logs, extracting status data describing the site's operational status and power supply data describing its power supply capacity. Subsequently, the system organizes the status data according to preset data mapping rules, forming a set of status feature data that can be processed by the algorithm. All data in this set originates directly from the operation logs, without manual intervention or subjective correction, ensuring the objectivity of the status determination.

[0038] After obtaining the state feature data set, the system inputs this data set into a pre-trained decision tree algorithm model. This decision tree algorithm model is trained on a large amount of historical operation log data during the system deployment phase. Each branch node corresponds to a certain operation state judgment condition, and each leaf node corresponds to a specific site state category. During actual operation, the system judges the state feature data layer by layer according to the decision tree structure. By matching the order of each state condition, it finally outputs the site state category information corresponding to the current operation log. This process does not involve dynamic learning; it only executes a predetermined judgment path, thereby ensuring the stability and consistency of the judgment results.

[0039] After obtaining the site status category information, the system retrieves the corresponding preset power supply threshold based on this information. This preset power supply threshold is determined during the system configuration phase based on multiple factors, including the battery swapping station's rated power supply capacity, historical peak load levels, equipment safety operation requirements, and minimum service guarantee needs. Specifically, the system statistically analyzes historical operating data to determine the minimum stable power supply level of the site under normal service conditions, and reserves a safety redundancy range based on this, thereby determining the power supply threshold used to assess power supply risk. Different site status categories correspond to different power supply threshold settings to reflect the minimum power supply capacity requirements under different operating conditions.

[0040] Subsequently, the system compares the power supply data extracted from the real-time operation log with the preset power supply threshold item by item. When the current power supply data is detected to be lower than the corresponding preset power supply threshold, the system determines that the battery swapping site has a risk of insufficient power supply in the current reservation scenario and marks the site as a reservation abnormal site. This marking result is used to trigger the subsequent impact range analysis process without directly interrupting the reservation service, thereby providing the system with room for further scheduling and adjustment.

[0041] After marking the abnormal reservation sites, the judgment and impact analysis module further analyzes the propagation impact of this abnormal state in the battery swapping network. The system uses the abnormal reservation sites as the starting point for analysis and constructs a battery swapping network association graph based on the service relationships, geographical proximity relationships, and reservation transfer rules between sites in the network. In this association graph, each battery swapping site exists as an independent node, and the connections between nodes represent the potential paths through which user reservation requests may be transferred in the event of an abnormal reservation.

[0042] After constructing the battery swapping network association map, the system identifies sites directly connected to or connected via limited redirection relationships with sites experiencing reservation anomalies, forming a set of associated sites. Subsequently, the system combines historical reservation records and the current number of reservation requests to statistically analyze the number of reservation requests that the sites experiencing reservation anomalies cannot handle within the current time period, thus obtaining the reservation overflow. The reservation overflow represents the scale of reservation requests that need to be reallocated due to site anomalies.

[0043] After obtaining the reservation overflow, the system analyzes the distribution of the overflow across the set of associated sites based on the service capacity limitations and historical transfer data of each connection path in the battery swapping network's association map. Specifically, the system allocates reservation requests to associated sites step by step according to the priority order of the associated paths, and records the transfer direction of the reservation requests in each allocation step. By continuously tracking the allocation process of the reservation overflow, the system can identify the main diffusion paths of reservation requests in the battery swapping network.

[0044] Ultimately, the system comprehensively considers the range of the associated site set, the scale of the reservation overflow, and the hierarchical depth of the diffusion path to determine the impact range of abnormal sites in the battery swapping network. This impact range characterizes the actual degree of impact of the abnormal site on surrounding battery swapping services and serves as an important input for subsequent resource cluster analysis and dynamic reservation allocation.

[0045] The resource cluster analysis module extracts the location coordinates and load data of surrounding battery swapping stations within the influence area, and uses a clustering algorithm based on the location coordinates and load data to group available battery swapping stations, obtaining the resource cluster distribution, including: The influence range is defined based on a preset center point, and the location coordinates and load data of surrounding battery swapping stations are extracted from the influence range. Generate a multidimensional feature vector describing the site status based on the location coordinates and the load data; Calculate the Euclidean distance between the multidimensional feature vectors to establish neighborhood connection relationships, forming an adjacency graph structure that reflects spatial proximity and load similarity; Based on the adjacency graph structure, a clustering algorithm is used to group the available battery swapping sites to obtain a resource cluster distribution.

[0046] In this embodiment, the resource cluster analysis module is used to perform structured analysis on battery swapping sites within the affected area, given that the impact range of the abnormal reservation site has been determined. This analysis identifies sets of sites that are spatially close and have similar operating loads, thereby generating resource cluster distribution results that can be used for subsequent reservation allocation and route adjustment. The overall processing flow of this module starts with the impact range information as the input and completes the generation of resource clusters through continuous data filtering, feature construction, relationship calculation, and grouping processing.

[0047] In its implementation, the system first determines a preset center point for resource analysis based on the impact range information output by the judgment and impact analysis module. The method for determining this preset center point is set during the system configuration phase; typically, the geographical coordinates of the abnormal booking site are selected as the center point. In the case of multiple abnormal booking sites, the system can calculate their spatial center position based on the location coordinates of these sites, using this as a unified analysis center point. This center point ensures a clear spatial reference benchmark for subsequent impact range delineation and site selection processes.

[0048] After determining the preset center point, the system spatially filters the sites in the battery swapping network according to pre-configured influence range delineation rules. These influence range delineation rules can be set based on the straight-line distance between sites, road service radius, or historical reservation transfer range; the specific values ​​are determined during the system deployment phase based on the battery swapping network's coverage density and service capacity. Starting from the preset center point, the system compares the location coordinates of all battery swapping sites one by one, retaining only those surrounding sites within the influence range to form a candidate site set for subsequent analysis.

[0049] After site selection, the resource cluster analysis module extracts the location coordinates and load data of each battery swapping site in the candidate site set from the system operation data. Location coordinates characterize the spatial location of the site within the battery swapping network and typically include the site's fixed coordinates on a map. Load data reflects the site's operational pressure level during the current or predicted period; this data can be determined by a combination of factors, including the reserved load ratio, the number of battery swapping requests being processed, and battery turnover intensity. The selection criteria for load data are determined during the system initialization phase to ensure the comparability of load indicators between different sites.

[0050] After acquiring location coordinate data and load data, the system processes and organizes this data. Specifically, following a preset data structure, the system combines the location coordinate data of each battery swapping station with its corresponding load data to generate a multi-dimensional feature vector describing the overall operating status of the station. Each dimension of this multi-dimensional feature vector is numerical data, and each dimension corresponds to a specific station attribute, reflecting both the station's spatial location and operational load characteristics. The construction of this feature vector does not involve any subjective weight adjustments; all data originates directly from system acquisition or calculation results.

[0051] After constructing the multidimensional feature vectors, the system enters the stage of calculating relationships between sites. The resource cluster analysis module compares the multidimensional feature vectors in the candidate site set pairwise. By calculating the Euclidean distance between the multidimensional feature vectors of different sites, it quantifies the comprehensive proximity of sites in terms of spatial location and load characteristics. This distance calculation process is achieved by uniformly measuring the numerical differences of each dimension, and the result is a numerical distance result. The smaller the distance value, the closer the corresponding sites are in spatial location and the more similar their load levels; the larger the distance value, the greater the differences between sites in terms of geographical location or operational load.

[0052] After obtaining the distance calculation results between stations, the system establishes connection relationships between stations according to preset neighborhood determination rules. These neighborhood determination rules determine which stations should be considered related, typically based on whether the distance results are within a reasonable proximity range set by the system. This proximity range threshold is determined during the system configuration phase based on the average distribution density of battery swapping stations and acceptable service transfer costs, to avoid including stations that are too far apart or have significantly different loads in the same analysis scope. Stations that meet the neighborhood determination criteria are connected, thus forming a set of adjacency relationships reflecting the spatial proximity and load similarity of the stations.

[0053] After establishing the neighborhood relationships, the system constructs a complete adjacency graph structure, using battery swapping stations as nodes and the neighborhood connections as the connecting edges between nodes. This adjacency graph structure reflects the degree of association between battery swapping stations within the influence area from an overall perspective, providing structured input for subsequent grouping processing.

[0054] Subsequently, the resource cluster analysis module uses a clustering algorithm based on the adjacency graph structure to group the candidate site set. During clustering, the system primarily relies on the connectivity and distance characteristics between nodes in the adjacency graph, gradually merging sites that are spatially close and have similar load levels into the same group. Simultaneously, for sites with weak connectivity or significant distance differences, the system assigns them to different groups to ensure high consistency within each group. This grouping process terminates when a clustering termination condition is met, which is typically set during the system configuration phase based on the number of sites and cluster stability requirements.

[0055] Through the aforementioned continuous, explicit, and reproducible processing steps, the system ultimately obtains a resource cluster distribution result that reflects the spatial distribution characteristics and load distribution characteristics of battery swapping sites within the affected area. This resource cluster distribution result serves as the input basis for the subsequent preliminary battery swapping allocation module and the dynamic reservation allocation path generation module, thereby realizing intelligent battery swapping reservation scheduling based on regional resource collaboration.

[0056] The initial battery swapping allocation module, based on the resource cluster distribution, matches the nearest resource cluster with the location information in the user's reservation request, and determines whether the reservation load of the corresponding battery swapping station exceeds the capacity threshold, thereby generating an initial battery swapping allocation plan based on the reservation request, including: Obtain the location coordinates from the user's reservation request, and locate the nearest resource cluster based on the location coordinates and the resource cluster distribution topology. Analyze the cumulative reserved load value and capacity threshold setting of the physical battery swapping station in the nearest resource cluster; If the cumulative reserved load value is less than the capacity threshold setting, an availability verification pass signal is generated; In response to the availability verification, a preliminary battery swapping allocation plan is generated via a signal. This preliminary battery swapping allocation plan is based on battery swapping sites with remaining service capacity.

[0057] In this embodiment, the preliminary battery swapping allocation module is used to perform preliminary battery swapping site matching and allocation for reservation requests submitted by individual users, provided that the resource cluster distribution is already clear. Its core purpose is to determine a practically feasible battery swapping site plan for the user while meeting the service capacity constraints of the battery swapping stations. The processing flow of this module is triggered by the user's reservation request and is completed step by step based on the existing resource cluster analysis results of the system.

[0058] In its implementation, the system first parses and retrieves location information from the user's submitted reservation request. This location information is stored as coordinates, representing the user's geographical location at the time the reservation request was initiated. These coordinates are typically obtained by the user's terminal via location services and uploaded to the system along with the reservation request. When parsing this location information, the system performs integrity checks and validity assessments on the coordinate data to ensure it can be used for subsequent spatial matching calculations.

[0059] After obtaining the user's location coordinates, the initial battery swapping allocation module compares and analyzes these coordinates with the resource cluster distribution topology. This topology contains multiple resource clusters, each corresponding to a specific spatial coverage area and associated with several physical battery swapping sites. The system compares the spatial distance between the user's location coordinates and the representative locations of each resource cluster according to preset spatial matching rules. These representative locations can be determined during the system configuration phase, typically using the concentrated location of sites within the resource cluster. By comparing the spatial proximity between each resource cluster and the user's location, the system identifies the resource cluster closest to the user's location as the priority matching resource cluster for this reservation request.

[0060] After identifying the nearest resource cluster, the initial battery swapping allocation module further analyzes each physical battery swapping site within that cluster. For each physical battery swapping site, the system extracts its cumulative reservation load value within the reservation time window from the reservation management data. This cumulative reservation load value represents the total number of reservation requests allocated to the battery swapping site during the current reservation period. This value is updated in real time as reservation requests are confirmed or canceled, accurately reflecting the site's current reservation capacity status.

[0061] Meanwhile, the system retrieves the corresponding capacity threshold setting for each physical battery swapping station. This capacity threshold setting is determined during system deployment or station configuration, taking into account multiple factors, including the number of battery swapping devices at the station, battery storage capacity, the time required for a single battery swapping service, the station's concurrent service capacity, and stable operating data during historical peak periods. The system typically uses the maximum reservation load that the station can consistently and stably serve in its historical operation as a basis, and reserves a safety redundancy space on top of this, thus forming the capacity threshold setting used to determine whether the station can continue to handle reservation requests. Different battery swapping stations can correspond to different capacity thresholds to adapt to differences in station hardware conditions and operational capabilities.

[0062] After obtaining the cumulative reserved load value and the capacity threshold setting, the system compares and judges them site by site. Specifically, the system determines whether the cumulative reserved load value of a certain battery swapping station is less than its corresponding capacity threshold setting. When the judgment result is that the cumulative reserved load value is less than the capacity threshold setting, the system determines that the battery swapping station still has available remaining service capacity in the current reservation scenario and can accept new reservation requests. Based on this judgment result, the system generates a corresponding availability verification pass signal to identify the battery swapping station as a selectable target site.

[0063] After generating an availability verification pass signal, the initial battery swapping allocation module responds to the signal by adding the verified battery swapping sites to the allocable site set. Within this set, the system prioritizes battery swapping sites located within the nearest resource cluster that have passed availability verification as the target site for this reservation request. If multiple battery swapping sites simultaneously meet the criteria, the system can select one according to a preset priority rule, such as prioritizing sites with lower cumulative reservation loads or sites geographically closer to the user.

[0064] Finally, the system generates a preliminary battery swapping allocation plan based on the selected battery swapping sites. This preliminary allocation plan clarifies the battery swapping execution target of the user's reservation request at the current stage, and its content includes at least the target battery swapping site identification information and the corresponding reservation association information. This preliminary allocation plan serves as the input basis for the subsequent dynamic reservation allocation path generation module.

[0065] The dynamic reservation allocation path generation module, when the reserved load in the initial battery swapping allocation scheme exceeds the capacity threshold, is used to build a prediction model based on historical reservation data and real-time operation data to predict the peak user reservation traffic and determine the adjusted dynamic reservation allocation path accordingly. Specifically, it includes: When the reserved load value in the initial battery swapping allocation plan exceeds the preset capacity threshold, historical reservation data and real-time operation data are obtained. Time-series feature extraction is performed on the historical reservation data and the real-time operation data to construct a feature vector sequence; The feature vector sequence is input into a preset time series prediction model to obtain the peak user reservation traffic; The congestion coefficient is calculated based on the peak user reservation traffic, and the path weight values ​​in the battery swapping network topology are updated based on the congestion coefficient. Based on the path weight values, search for the minimum cost path and determine the adjusted dynamic reservation allocation path.

[0066] In this embodiment, the dynamic reservation allocation path generation module is used to predictively adjust the allocation path of reservation requests when the preliminary battery swapping allocation plan has been generated but cannot meet the service capacity constraints of the battery swapping stations, in order to avoid local congestion at battery swapping stations caused by concentrated reservations. This module uses the verification result of the preliminary battery swapping allocation plan as the trigger condition, and through comprehensive analysis of historical and real-time data, it achieves advance judgment of reservation traffic trends and path optimization.

[0067] In its implementation, the system first performs a load check on the preliminary battery swapping allocation scheme. This check is performed by comparing the reserved load value of each battery swapping station in the preliminary scheme with the pre-set capacity threshold for that station. The reserved load value represents the number of reservation requests allocated to a particular battery swapping station within a reservation time window, and the capacity threshold represents the maximum number of reservations that the station can stably handle per unit time. When the system detects that the reserved load value of any battery swapping station exceeds its corresponding capacity threshold, it determines that the preliminary battery swapping allocation scheme has a risk of insufficient service capacity, thereby triggering the dynamic reservation allocation path generation module.

[0068] Once the triggering conditions are met, the system first acquires historical reservation data and real-time operational data related to the current reservation scenario. The historical reservation data reflects changes in reservation traffic over multiple time periods, typically including the number of reservation requests in different time periods, the historical load levels of each battery swapping station, and the transfer of reservations between stations. The time span of the historical reservation data is set during the system configuration phase to ensure that the data covers typical operating scenarios such as peak and off-peak periods. The real-time operational data reflects the current instantaneous operating status of the battery swapping network, including the current reservation load at each battery swapping station, equipment availability, and the reachability of each path in the network topology.

[0069] After acquiring the aforementioned data, the dynamic reservation allocation path generation module performs unified time-series processing on historical reservation data and real-time operational data. The system first aligns the data according to chronological order, combining reservation traffic data, site load data, and operational status data within the same time point or time window. Subsequently, the system extracts key time-series features reflecting reservation traffic trends from the processed data, such as changes in the number of reservation requests over time, the continuity of load increases or decreases, and the timing of peak occurrences. Through this processing, the system constructs a feature vector sequence that continuously reflects the changing patterns of reservation behavior.

[0070] After constructing the feature vector sequence, the system inputs this sequence into a pre-configured time series prediction model. This model is trained on a large amount of historical reservation data during system deployment, and its parameters are fixed after training. It is used to predict and analyze the input feature vector sequence during operation. Through this prediction model, the system obtains the peak user reservation traffic that may occur within the prediction time window. This peak traffic represents the maximum number of reservation requests the battery swapping network may face in the near future.

[0071] After obtaining the peak user reservation traffic, the system quantitatively assesses the congestion risk of the battery swapping network based on this peak traffic. Specifically, the system compares the predicted peak reservation traffic with the reservation capacity that the battery swapping network can handle under the current structure and service capabilities, and calculates a congestion coefficient to characterize the degree of network congestion. The congestion coefficient is a numerical parameter, and its value reflects the concentration of reservation requests and the level of service pressure during the predicted period. The calculation rules for the congestion coefficient are determined during the system configuration phase, typically using historical high-load scenario operating data as a reference benchmark, thereby ensuring that the coefficient accurately reflects the network congestion risk.

[0072] After calculating the congestion coefficient, the dynamic reservation and allocation path generation module updates the path weights in the battery swapping network topology based on this coefficient. The battery swapping network topology describes the service relationships between battery swapping sites, where each path corresponds to a weight value representing the service cost. The system adjusts the path weights according to the congestion levels of the sites associated with different paths, increasing the weights of paths with high congestion risk and keeping the weights of paths with low congestion risk at a lower level. In this way, the path weights dynamically reflect the differences in service risk among paths within the prediction period.

[0073] After updating the path weights, the system performs a path search based on the updated battery swapping network topology. Starting with a user's reservation request and ending with a battery swapping station with service capabilities, the system searches for the service path with the lowest overall cost within the topology. This search process uses path weights as the primary evaluation criterion, prioritizing paths with lower service load and higher stability. Through this search process, the system ultimately determines the adjusted dynamic reservation allocation path. This dynamic allocation path replaces the original preliminary battery swapping allocation scheme, guiding reservation requests to battery swapping stations with lower loads and service capabilities, thus proactively adjusting resource allocation before the anticipated reservation peak.

[0074] The appointment guidance sequence generation module extracts appointment request priority information from user appointment records based on the dynamic appointment allocation path, sorts the appointment requests, and generates an appointment guidance sequence, specifically including: Obtain user reservation records, calculate a basic credit score based on the historical default rate and service type weight in the user reservation records, and generate reservation request priorities by combining the basic credit score with time sensitivity features; Obtain the dynamic reservation allocation path and path congestion index, map and compare the reservation request priority with the path congestion index, and calculate the matching correlation between the reservation request and the currently available resources. The ranking weight is determined based on the matching correlation, and the reservation requests are reorganized and sorted according to the ranking weight to generate a reservation guidance sequence.

[0075] In this implementation, the system first retrieves the user reservation record corresponding to the currently pending reservation request from the reservation management database. This user reservation record is a long-term accumulated dataset that includes at least records of the user's fulfillment of reservations in history, historical default records, and information on the service types previously selected. The system parses the user reservation record to extract historical default rate data. The historical default rate represents the proportion of times a user has failed to complete a service as scheduled in their reservations. It is calculated based on the relationship between the number of historical defaults and the total number of historical reservations. This percentage is maintained numerically within the system and updated after each reservation is completed or a default occurs.

[0076] While extracting historical default rates, the system retrieves pre-set service type weights based on the service type information corresponding to the reservation request. These service type weights are set during the system configuration phase according to the importance of different battery swapping services in the overall operation; for example, regular battery swapping services and emergency battery swapping services may correspond to different weight values. The criteria for setting service type weights typically include the degree of impact of the service on user travel continuity, the operational risks that service failure may bring, and the priority level of the service in the business system. These weights remain stable during system operation and are only updated when operational strategies are adjusted.

[0077] After obtaining the historical default rate and service type weights, the system comprehensively processes these two types of data to calculate the user's basic credit score. This basic credit score reflects the user's overall reliability level in the booked service, and its value is positively correlated with the user's historical performance; that is, users with lower historical default rates and higher service type weights have relatively higher basic credit scores. This calculation process is completed internally by the system through predetermined data mapping and weighting rules, without human intervention, thus ensuring the consistency of the credit assessment results.

[0078] After generating a basic credit score, the system further acquires time sensitivity features related to the appointment request. These time sensitivity features describe the degree to which the appointment request depends on service timeliness, and are determined based on factors including the proximity of the appointment time to the current time and whether there are explicit time constraints. The system grades the time sensitivity based on the time interval between the appointment time and the current processing time; the shorter the time interval, the higher the time sensitivity of the appointment request. The system then integrates these time sensitivity features with the basic credit score to generate an appointment request priority that characterizes the combined urgency and reliability of a single appointment request.

[0079] After generating the priority of reservation requests, the reservation guidance sequence generation module further obtains the dynamic reservation allocation path information output by the dynamic reservation allocation path generation module. This dynamic reservation allocation path information includes the candidate battery swapping site path corresponding to the current reservation request and the operational status evaluation result of that path within the predicted time window. The system extracts a path congestion index related to the path. This path congestion index reflects the service pressure level that the current path may bear within the predicted time period, and its value is derived from the comprehensive evaluation result of the reservation traffic peak and path weight during the aforementioned dynamic path generation process.

[0080] After obtaining the reservation request priority and the path congestion index, the system performs a mapping comparison process between the two. This mapping comparison process is used to evaluate the rationality of prioritizing the execution of a single reservation request under the current path conditions. Specifically, the system determines the importance of the reservation request based on its priority, and simultaneously considers the path congestion index to determine whether the current path has sufficient capacity. When the reservation request has a high priority and the path congestion index is within an acceptable range, the system considers the reservation request to have a high degree of matching with the currently available battery swapping resources; otherwise, it considers the matching degree to be low. Through the above mapping comparison process, the system calculates the matching correlation degree used to quantify this relationship.

[0081] After obtaining the matching correlation score, the system determines a corresponding ranking weight for each reservation request based on the matching correlation score. This ranking weight establishes a unified ranking standard among multiple reservation requests, and its value directly reflects the priority order in which the reservation request is processed under the current network conditions. The system regroups and sorts all pending reservation requests according to their ranking weights, placing reservation requests with higher ranking weights first and those with lower ranking weights last, thus forming an ordered queue of reservation requests.

[0082] Finally, the system generates a reservation guidance sequence based on the above sorting results. This reservation guidance sequence serves as the basis for the execution of subsequent reservation guidance and notification push modules, ensuring that reservation requests are guided and responded to in a reasonable, fair, and controllable order when battery swapping resources are limited or there is a risk of path congestion.

[0083] The appointment guidance and service record module pushes battery swapping guidance or backup battery swapping notifications to the user terminal according to the appointment guidance sequence, and updates the system operation log after the user confirms the response, so as to obtain the corresponding appointment service continuity record, including: Based on the resource matching status results generated from the reservation guidance sequence and the battery swapping station load data, a standardized guidance data package containing battery swapping guidance information or backup battery swapping notification is constructed. The standardized boot data packet is sent to the user terminal, and a response confirmation signal is received from the user terminal based on the standardized boot data packet. The system operation log is updated based on the response confirmation signal, and the logical closed-loop features in the updated system operation log are extracted to generate a reservation service continuity record.

[0084] In this embodiment, the reservation guidance and service record module is used to guide users to battery swapping or push backup battery swapping notifications, provided that the reservation guidance sequence has been generated. After the user confirms the response, the module records and marks the entire reservation execution process in a closed loop, thereby forming a reservation service continuity record that can be used for subsequent analysis and traceability. This module is a key component of the entire intelligent battery swapping reservation service system for realizing service implementation and result feedback.

[0085] In its implementation, the system first uses the reservation guidance sequence as the processing basis, reading each reservation request within the current processing cycle one by one. The reservation guidance sequence has already sorted the reservation requests according to their ranking weights, so the system processes each reservation request sequentially according to this sequence. For the currently processed reservation request, the system simultaneously obtains the associated target battery swapping station information and further obtains the real-time load data of that battery swapping station at the current time or within a predicted time period. This real-time load data reflects the current service capacity level occupied by the station; the data originates from the battery swapping station operation monitoring system and is maintained numerically within the system.

[0086] After acquiring real-time load data of the target battery swapping station, the system compares this load data with the corresponding capacity threshold. This capacity threshold is determined during the system configuration phase based on the station's equipment scale, the number of battery swapping channels that can be served simultaneously, battery turnover capacity, and historical stable operation data. Its value is determined by the principle of maximizing the number of scheduled services the station can handle per unit time while ensuring service stability. The system then determines whether the current real-time load data is within the allowable range of the capacity threshold and generates a resource matching status result that characterizes the matching relationship between the current scheduling request and battery swapping resources.

[0087] When the resource matching status indicates that the target battery swapping site has available battery swapping resources, the system marks the reservation request as ready for direct guidance; when the resource matching status indicates that the target battery swapping site has limited resources or is nearing its load limit, the system marks the reservation request as requiring a backup plan for guidance. This resource matching status result serves as the direct basis for generating subsequent guidance content.

[0088] After obtaining the resource matching status result, the reservation guidance and service record module constructs a standardized guidance data packet based on the result. This standardized guidance data packet is a data structure defined internally by the system, and its field content is uniformly determined during the system design phase to ensure consistent parsing and display of guidance information across different user terminals. When the reservation request is in a directly accessible guidance state, the standardized guidance data packet contains explicit battery swapping guidance information, which includes at least the identification information of the target battery swapping station and battery swapping execution prompts. When the reservation request is in a backup plan guidance state, the standardized guidance data packet contains backup battery swapping notification information, which explains the current site resource status to the user and provides optional backup battery swapping plans or a prompt for delayed execution.

[0089] After constructing the standardized guidance data packet, the system sends the packet to the corresponding user terminal via a preset communication interface. Upon receiving the packet, the user terminal prompts the user based on the guidance or notification content and generates a response confirmation signal after the user performs a clear action. This response confirmation signal indicates the user's acceptance of the current battery swapping guidance or backup plan, and its content includes at least the reservation request identifier and the user's confirmation result.

[0090] Upon receiving a response confirmation signal from the user terminal, the appointment guidance and service record module updates the system operation log. The system operation log is a collection of log data used by the system to record the entire appointment service process, and its log entries are predefined during system initialization. When updating the operation log, the system associates and stores information such as the confirmation result and confirmation time from the response confirmation signal with the corresponding appointment request, thereby forming a complete service execution record. This update process is completed through appending or status changes to ensure the integrity and traceability of historical records.

[0091] After updating the system operation log, the appointment guidance and service record module further analyzes and processes the updated system operation log to extract logical closed-loop features that characterize the complete execution of the appointment service. These logical closed-loop features reflect whether the appointment request has undergone a complete processing flow. The criteria for determination include whether the appointment request successfully generated a guidance data packet, whether the guidance data packet was successfully sent to the user terminal, and whether the user returned a valid response confirmation signal. Only when all of the above steps are completed does the system determine that the appointment request has formed a complete logical closed loop.

[0092] After identifying the logical closed-loop characteristics, the system generates a corresponding appointment service continuity record based on these characteristics. This appointment service continuity record identifies the execution continuity status of a single appointment request within the system, and its content can be used for subsequent service quality assessment, user behavior analysis, and system scheduling strategy optimization.

[0093] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A smart battery swapping service system supporting battery reservation, characterized in that, The method includes: The state modeling module receives battery swapping reservation requests submitted by users, extracts reservation time, target swapping station and battery demand information from the reservation requests, and models the station operation status corresponding to the reservation request by combining the real-time operation data of the swapping station to obtain the current station status classification. The judgment and impact analysis module determines whether the power supply data is lower than a preset threshold based on the site status classification. If it is lower than the preset threshold, the corresponding site is judged as an abnormal reservation site, and the impact range of the abnormal reservation site in the battery swapping network is further determined. The resource cluster analysis module extracts the location coordinates and load data of surrounding battery swapping stations from the influence area, and uses a clustering algorithm to group the available battery swapping stations based on the location coordinates and load data to obtain the resource cluster distribution. The initial battery swapping allocation module, based on the resource cluster distribution, matches the nearest resource cluster with the location information in the user's reservation request, and determines whether the reservation load of the corresponding battery swapping station exceeds the capacity threshold, thereby generating an initial battery swapping allocation scheme based on the reservation request.

2. The intelligent battery swapping service system supporting battery reservation according to claim 1, characterized in that, The state modeling module receives battery swapping reservation requests submitted by users, extracts reservation time, target swapping station, and battery demand information from the reservation requests, and models the station's operational status corresponding to the reservation request by combining it with the station's real-time operational data, thus obtaining a current station status classification, including: Obtain the battery swap reservation request message submitted by the user and the real-time operation data of the target swap station. The real-time operation data includes the real-time battery inventory quantity, the set of remaining charging time of batteries currently being charged, and the length of the queue of vehicles. Calculate the estimated net available battery value at the scheduled time based on the battery swapping reservation request message and the real-time operating data, and generate an operating load feature vector. The operating load feature vector is input into a pre-trained classification model to obtain the current site status classification of the target battery swapping station in response to the battery swapping reservation request message.

3. The intelligent battery swapping service system supporting battery reservation according to claim 1, characterized in that, The judgment and impact analysis module determines whether the power supply data is below a preset threshold based on the site status classification. If it is below the preset threshold, the corresponding site is judged as an abnormal reservation site, and the impact range of the abnormal reservation site in the battery swapping network is further determined, including: Obtain real-time operation logs of the battery swapping station and extract site status and power supply data; use a decision tree algorithm to obtain site status category information. Based on the site status category information, a preset power supply threshold is retrieved. If the power supply data is lower than the preset power supply threshold, the site is marked as an abnormal reservation site. Based on the abnormal reservation sites, a battery swapping network association map is constructed to identify the set of associated sites; Calculate the reservation overflow amount and its diffusion path of the associated site set to determine the impact range of the abnormal reservation sites in the battery swapping network.

4. The intelligent battery swapping service system supporting battery reservation according to claim 1, characterized in that, The resource cluster analysis module extracts the location coordinates and load data of surrounding battery swapping stations within the influence area, and uses a clustering algorithm based on the location coordinates and load data to group available battery swapping stations to obtain the resource cluster distribution, including: The influence range is defined based on a preset center point, and the location coordinates and load data of surrounding battery swapping stations are extracted from the influence range. Generate a multidimensional feature vector describing the site status based on the location coordinates and the load data; Calculate the Euclidean distance between the multidimensional feature vectors to establish neighborhood connection relationships, forming an adjacency graph structure that reflects spatial proximity and load similarity; Based on the adjacency graph structure, a clustering algorithm is used to group the available battery swapping sites to obtain a resource cluster distribution.

5. The intelligent battery swapping service system supporting battery reservation according to claim 1, characterized in that, The preliminary battery swapping allocation module, based on the resource cluster distribution, matches the nearest resource cluster by combining the location information in the user's reservation request, and determines whether the reservation load of the corresponding battery swapping station exceeds the capacity threshold, thereby generating a preliminary battery swapping allocation scheme based on the reservation request, including: Obtain the location coordinates from the user's reservation request, and locate the nearest resource cluster based on the location coordinates and the resource cluster distribution topology. Analyze the cumulative reserved load value and capacity threshold setting of the physical battery swapping station in the nearest resource cluster; If the cumulative reserved load value is less than the capacity threshold setting, an availability verification pass signal is generated; In response to the availability verification, a preliminary battery swapping allocation plan is generated via a signal. This preliminary battery swapping allocation plan is based on battery swapping sites with remaining service capacity.

6. The intelligent battery swapping service system supporting battery reservation according to claim 1, characterized in that, It also includes a dynamic reservation allocation path generation module. When the reserved load in the initial battery swapping allocation scheme exceeds the capacity threshold, it is used to build a prediction model based on historical reservation data and real-time operation data to predict the peak user reservation traffic and determine the adjusted dynamic reservation allocation path accordingly. Specifically, it includes: When the reserved load value in the initial battery swapping allocation plan exceeds the preset capacity threshold, historical reservation data and real-time operation data are obtained. Time-series feature extraction is performed on the historical reservation data and the real-time operation data to construct a feature vector sequence; The feature vector sequence is input into a preset time series prediction model to obtain the peak user reservation traffic; The congestion coefficient is calculated based on the peak user reservation traffic, and the path weight values ​​in the battery swapping network topology are updated based on the congestion coefficient. Based on the path weight values, search for the minimum cost path and determine the adjusted dynamic reservation allocation path.

7. The intelligent battery swapping service system supporting battery reservation according to claim 6, characterized in that, It also includes a reservation guidance sequence generation module, which extracts reservation request priority information from user reservation records based on the dynamic reservation allocation path, sorts the reservation requests, and generates a reservation guidance sequence, specifically including: Obtain user reservation records, calculate a basic credit score based on the historical default rate and service type weight in the user reservation records, and generate reservation request priorities by combining the basic credit score with time sensitivity features; Obtain the dynamic reservation allocation path and path congestion index, map and compare the reservation request priority with the path congestion index, and calculate the matching correlation between the reservation request and the currently available resources. The ranking weight is determined based on the matching correlation, and the reservation requests are reorganized and sorted according to the ranking weight to generate a reservation guidance sequence.

8. The intelligent battery swapping service system supporting battery reservation according to claim 7, characterized in that, It also includes a reservation guidance and service record module, which pushes battery swapping guidance or backup battery swapping notifications to user terminals according to the reservation guidance sequence, and updates the system operation log after the user responds and confirms, so as to obtain the corresponding reservation service continuity record, specifically including: Based on the resource matching status results generated from the reservation guidance sequence and the battery swapping station load data, a standardized guidance data package containing battery swapping guidance information or backup battery swapping notification is constructed. The standardized boot data packet is sent to the user terminal, and a response confirmation signal is received from the user terminal based on the standardized boot data packet.

9. A smart battery swapping service system supporting battery reservation according to claim 8, characterized in that, The reservation guidance and service record module pushes battery swapping guidance or backup battery swapping notifications to the user terminal according to the reservation guidance sequence, and updates the system operation log after the user confirms the response to obtain the corresponding reservation service continuity record, and also includes: The system operation log is updated based on the response confirmation signal, and the logical closed-loop features in the updated system operation log are extracted to generate a reservation service continuity record.

Images