An intelligent interaction dynamic resource recommendation method for shared electric bicycles and integrated charging and battery swapping stations

Abstract

The present invention is applicable to the field of resource recommendation technology, and provides a method for intelligent interactive dynamic resource recommendation between shared electric motorcycles and integrated charging and swapping base stations, the method comprising: real-time collection of shared electric vehicle battery status data, user behavior data, and charging demand identification; collection of resource status data of integrated charging and swapping base stations; evaluation of the availability characteristics of available batteries at each base station, and generation of recommendation priorities for each base station; sending a resource lock request containing a dynamic reservation duration to the target base station with the highest priority, wherein the dynamic reservation duration is dynamically adjusted according to the user's real-time location and navigation estimated time, and the recommendation priority is updated when a blocked path or a change in user demand is detected; after the battery swapping operation is completed, the recommendation strategy is optimized based on the actual battery performance data and user operation feedback. The present invention can provide users with highly intelligent and personalized base station recommendations, while optimizing the overall system resource configuration, significantly improving user experience and operational efficiency.

Description

Technical Field

[0001] The present invention relates to the field of resource recommendation technology, and in particular to a method for intelligent interactive dynamic resource recommendation between a shared electric motorcycle and an integrated charging and swapping base station. Background Art

[0002] At present, the shared electric motorcycle system has formed a technical architecture with electric motorcycle terminals, user mobile applications, background management platforms and charging and swapping infrastructure as the core. By integrating GPS positioning, battery management systems and wireless communication modules, electric motorcycles can report their location and battery status in real time; charging and swapping base stations are equipped with charging piles, battery swap cabinets and environmental monitoring sensors to support battery replacement and charging services. Users can use the APP to find available vehicles or base stations nearby, and the background system allocates resources based on set rules. However, because the rules are set based on static factors, there will be delays or large errors in battery status information updates, making it difficult for operations and maintenance to dispatch low-battery vehicles in a timely manner, and users may rent low-quality batteries.

[0003] Furthermore, while some systems have basic data analysis capabilities, they still rely on manual intervention for dynamic scheduling and intelligent recommendations. This results in uneven load distribution across charging and swapping base stations, where users may encounter full charging spots or insufficient batteries upon arrival, resulting in low resource utilization. Furthermore, there is a lack of deep interaction between current e-bikes, base stations, user apps, and backend systems, making it impossible to optimize charging strategies based on real-time grid load, environmental parameters, and user behavior.

[0004] In view of this, a dynamic resource recommendation method for shared electric motorcycles and integrated charging and swapping base stations is proposed. Summary of the Invention

[0005] The present invention provides a method for intelligent interactive dynamic resource recommendation between shared electric motorcycles and integrated charging and swapping base stations, which is used to solve the problem of low resource utilization.

[0006] The present invention provides a method for intelligently interacting with a shared electric motorcycle and an integrated charging and swapping base station to recommend dynamic resources, comprising:

[0007] Real-time collection of shared electric vehicle battery status data, user behavior data, and charging demand identification; at the same time, collection of resource status data of integrated charging and swapping base stations;

[0008] Evaluate availability characteristics of available batteries at each base station based on the resource status data, the availability characteristics including battery health, number of available charging stations, and environmental adaptability;

[0009] Generate a recommended priority for each base station based on the shared electric vehicle's battery status data, user behavior data, charging demand identification, and the availability characteristics;

[0010] Send a resource lock request including a dynamic reservation duration to the target base station with the highest priority. The dynamic reservation duration is dynamically adjusted based on the user's real-time location and estimated navigation time. If the user does not arrive within the dynamic reservation time, the resource lock is automatically released.

[0011] When a path obstruction or a change in user demand is detected, the base station with the lowest availability characteristics and the highest arrival efficiency is re-matched and the recommended priority is updated;

[0012] After the battery replacement operation is completed, the recommendation strategy is optimized based on the actual performance data of the replaced new battery and user operation feedback. The actual performance data of the new battery includes battery health, charge and discharge efficiency and temperature stability corresponding to the availability characteristics.

[0013] Furthermore, the availability characteristics of the available batteries of each base station are evaluated based on the resource status data, and the availability characteristics include battery health, number of available charging positions, and environmental adaptability, including:

[0014]

[0015] in: is the battery health index, is the actual capacity of the battery. is the nominal capacity, is the cumulative number of cycles, To design the maximum cycle life, is the total number of charge and discharge cycles, For the Average temperature and optimal temperature during the charge and discharge cycle The deviation, is the temperature reference period, 、 and are weight coefficients, respectively, based on the priority allocation of capacity attenuation dominating, cycle number secondly, and long-term temperature impact in the shared electric vehicle scenario.

[0016] Furthermore, the availability characteristics of the available batteries of each base station are evaluated based on the resource status data, wherein the availability characteristics include battery health, number of available charging positions, and environmental adaptability, and further include:

[0017]

[0018] in: is the number of available charging stations, is the number of currently available charging slots, is the total number of charging positions, is the total number of charging positions, For the The historical usage frequency of each charging station, Current time With the The most recent usage time of the charging station The interval is the attenuation factor Give more weight to recent use, is the confidence level of demand forecast for a certain period in the future, 、 、 To balance the coefficient, ensure a reasonable proportion of real-time idle rate, historical decay and forecast uncertainty.

[0019] Furthermore, the availability characteristics of the available batteries of each base station are evaluated based on the resource status data, wherein the availability characteristics include battery health, number of available charging positions, and environmental adaptability, and further include:

[0020]

[0021] in: is the environmental adaptability index, is the real-time temperature, is the safety temperature threshold, The real-time load of the power grid, is the reference load, For weather warning levels, 、 、 is a dynamic weight, indicating the priority allocation of high temperature, grid load, and weather warning. 、 is the standard deviation, based on historical data statistics.

[0022] Furthermore, generating the recommendation priority of each base station based on the battery status data, user behavior data, charging demand identification and availability characteristics of the shared electric vehicle includes:

[0023] Obtain the current battery charge and battery health index, and generate a battery status score based on a battery health quantification model that incorporates voltage, cycle count, and temperature history data;

[0024] Generate a user behavior adaptation score based on historical riding route preferences, time sensitivity, and battery selection tendencies in user behavior data;

[0025] Parse the charging demand identifier and generate a charging demand urgency score based on the urgency of the user's order;

[0026] Combined calculation and calculated , performing weighted summation on the battery status score, user behavior adaptation score, and charging demand urgency score to generate a total recommendation score for each base station.

[0027] Furthermore, the user behavior adaptation score is generated based on the historical riding route preference, time sensitivity, and battery selection tendency in the user behavior data, including:

[0028] Define the user behavior state space, including historical riding path sets, time sensitivity intervals, and battery health selection preferences;

[0029] The action space is defined as base station selection behavior, and the reward function is dynamically calculated based on the matching degree between the user's final base station and the system's recommended base stations, order completion efficiency, and battery life satisfaction.

[0030] The preset reinforcement learning model is trained through user behavior data to output the probability distribution of user preference for base stations;

[0031] According to the preference probability distribution, user behavior patterns are divided into high-frequency optimization type, regular type and random type, and mapped to corresponding behavior adaptation scores; among them, the high-frequency optimization type has the highest score and the random type has the lowest score.

[0032] Furthermore, generating the recommended priority of each base station includes:

[0033] Sort all base stations according to the total recommendation score and divide them into priority levels;

[0034] Optimizing priority results based on a dynamic weighting strategy, including increasing the weight of charging station availability during peak hours, increasing the weight of environmental adaptability during extreme weather conditions, and increasing the weight of user behavior adaptation for VIP users;

[0035] Multi-dimensional verification of the priority results, including whether the battery health meets the minimum threshold, the probability of charging slot reservation conflict, and path accessibility;

[0036] Output the final priority list and push the real-time status and navigation path of the recommended base station to the user terminal.

[0037] Furthermore, the sending of a resource lock request including a dynamic reservation duration to the target base station with the highest priority, wherein the dynamic reservation duration is dynamically adjusted according to the user's real-time location and the estimated navigation time, and the automatic release of the resource lock if the user does not arrive within the reserved time, includes:

[0038] Collect historical usage data of the target base station, including charging station occupancy rate, user arrival interval, and battery replacement time at different time periods;

[0039] Build a time series model, extract periodic and trend features through a sliding window algorithm, and predict the probability of base station congestion within a preset time period in the future;

[0040] Dynamically calculate the reserved time based on the congestion probability and the estimated navigation time from the user's real-time location to the base station;

[0041] If the predicted congestion probability of the target base station during the user arrival period is higher than a preset threshold, the reservation duration is shortened to avoid idle resources; otherwise, the reservation duration is extended.

[0042] Furthermore, the dynamic reservation duration is dynamically adjusted based on the user's real-time location and estimated navigation time. If the user does not arrive within the reserved time, the resource lock is automatically released, including:

[0043] Real-time monitoring of user travel speed, traffic conditions and base station resource status updates;

[0044] If the user's travel speed is lower than the preset ratio of the navigation estimate, the reserved time is extended according to the delay ratio, and the new estimated arrival time is pushed through the user terminal;

[0045] If the base station resource status update shows that the reserved charging position is unexpectedly occupied, re-match the available base stations and update the reservation duration;

[0046] If an external event triggers a sudden change in the base station load, the reserved time will be compressed or extended according to the real-time load rate.

[0047] Furthermore, when a path obstruction or a change in user demand is detected, re-matching a base station that meets the minimum availability characteristics and has the highest arrival efficiency and updating the recommendation priority includes:

[0048] Based on a preset minimum availability threshold, select candidate base stations that meet the following requirements: battery health ≥ preset health threshold, charging position idle rate ≥ preset idle rate threshold, and environmental adaptability ≥ preset adaptability threshold;

[0049] Calculate the shortest estimated arrival time for each base station based on the location of the candidate base stations, real-time traffic conditions, and user travel speed;

[0050] Sort candidate base stations in ascending order by the shortest estimated arrival time, generate an updated recommendation priority list, and push the adjusted recommendation results and navigation path to the user terminal;

[0051] If the original target base station no longer meets the conditions, its reserved resources are released and a dynamic resource locking request is sent to the new base station with the highest priority.

[0052] It can be seen from the above technical solutions that the present invention has the following advantages:

[0053] The present invention collects battery status, user behavior and base station resource data in real time to build a multi-source dynamic database, providing a highly timely and accurate data foundation for subsequent analysis; converts heterogeneous data into standardized indicators based on quantitative models; generates personalized recommendation priorities by dynamically weighting and fusing user needs, battery health and base station availability characteristics, effectively improving resource matching accuracy; dynamically adjusts resource reservation duration through time series prediction and real-time monitoring mechanisms to balance resource utilization efficiency and user experience; when paths are blocked or demand changes, quickly rematches base stations through threshold screening and shortest arrival time calculation to ensure system flexibility and service continuity; finally, based on battery performance data after battery replacement and user feedback, utilizes reinforcement learning to iteratively optimize recommendation strategies. The present invention can provide users with highly intelligent and personalized base station recommendations, while optimizing overall system resource allocation and significantly improving user experience and operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart of an embodiment of a method for intelligent interactive dynamic resource recommendation between a shared electric motorcycle and an integrated charging and swapping base station in the present invention. DETAILED DESCRIPTION

[0055] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "corresponding to" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0056] Example 1

[0057] The implementation method in this embodiment can be implemented in the system, in the server, or in the terminal, and there is no specific limitation. The following will introduce the intelligent interactive dynamic resource recommendation method for shared electric motorcycles and charging and swapping integrated base stations in this application from the perspective of system implementation. Figure 1 The method provided in the embodiment of the present application includes the following steps:

[0058] S11. Real-time collection of shared electric vehicle battery status data, user behavior data, and charging demand identification; and simultaneous collection of resource status data from integrated charging and swapping base stations;

[0059] In this embodiment, the motorcycle's built-in battery management system collects battery status data, user behavior data, and parameters corresponding to the charging requirement indicator in real time, and transmits them to the backend management platform via the cellular network or Bluetooth. Under normal conditions, the data is reported once a minute, and abnormal conditions trigger real-time reporting. Battery status data includes the current percentage of remaining battery charge, reflecting available battery life; the degree of battery capacity decay and life status; the real-time voltage and current during the charge and discharge process, used to monitor the battery's operating status; the internal and external ambient temperature of the battery to avoid overheating risks; the cumulative number of charge and discharge cycles of the battery to assess the degree of aging; and abnormal conditions detected by the battery management system, such as overvoltage, undervoltage, and short circuit.

[0060] User behavior data is collected through the user app, including riding trajectories, battery swap operation logs, and order data. This data is then combined with a backend database to analyze user behavior patterns and generate structured behavior tags, such as "high-frequency optimization user" and "time-sensitive user." This user behavior data includes commonly used riding routes, starting and ending point distribution; average battery swap intervals and preferred swap times; order punctuality and riding speed fluctuations; the proportion of users selecting high-health batteries; and the frequency and response time of battery swap requests initiated through the app.

[0061] The charging demand identifier categorizes the user's current charging needs and is used to dynamically adjust resource allocation strategies. The system automatically generates an identifier based on the user's real-time battery level and app actions (such as selecting "Emergency Battery Replacement"). The identifier is transmitted to the backend management platform via an API. Emergency needs, triggered automatically by the system or manually marked by the user, have the highest priority; high needs, generated automatically by the system when the battery level is between 15% and 30% and there is no emergency identifier; and normal needs, generated when the battery level is above 30% and there is no special identifier, have the lowest priority.

[0062] The resource status data of the integrated charging and swapping base station is collected in real time through a network of sensors at the base station terminals, such as photoelectric sensors, temperature and humidity sensors, and current transformers, which monitor the physical state. Real-time load and electricity price data is obtained through the grid interface and uploaded to the backend management platform via the base station controller's communication module. This resource status data includes the current number of available charging spots and their occupancy time; the health, charge level, and cycle count of the base station's replaceable batteries; the base station's internal temperature, humidity, and smoke detection status; real-time grid power, time-of-use electricity pricing information, and power supply stability; and alarms indicating charging station fault codes and abnormal battery compartment lock status.

[0063] S12. Evaluate the availability characteristics of the available batteries at each base station based on the resource status data, including battery health, number of available charging stations, and environmental adaptability;

[0064] This step is based on the battery status data, user behavior data, and base station resource status data collected in real time in step S11 above. It quantitatively evaluates the availability characteristics of the integrated charging and swapping base station, specifically including three core indicators: battery health, number of available charging stations, and environmental adaptability. Through mathematical modeling and dynamic weight allocation, complex multi-source data is converted into comparable quantitative indicators. The calculation formulas are as follows:

[0065] Battery health:

[0066]

[0067] in: is the battery health index, is the actual capacity of the battery. is the nominal capacity, is the cumulative number of cycles, To design the maximum cycle life, is the total number of charge and discharge cycles, For the Average temperature and optimal temperature during the charge and discharge cycle The deviation, is the temperature reference period, 、 and The weighting factor is based on the priority of capacity decay in shared electric vehicle scenarios, followed by cycle count and the long-term impact of temperature. For example, a 60% weighting is set for capacity decay to reflect the battery's current available capacity, which directly determines battery life. A 30% weighting is set for cycle life to indicate the degree of battery aging; the closer the cycle count is to the design life, the lower the score. A 10% weighting is set for temperature deviation to reflect the cumulative time the temperature deviates from the optimal value (e.g., 25°C). High temperatures accelerate battery aging, requiring long-term correction.

[0068] Number of charging stations available:

[0069]

[0070] in: is the number of available charging stations, is the number of currently available charging slots, is the total number of charging positions, is the total number of charging positions, For the The historical usage frequency of each charging station, Current time With the The most recent usage time of the charging station The interval is the attenuation factor Give more weight to recent use, is the confidence level of demand forecast for a certain period in the future, 、 、 To balance the coefficients, we ensure a reasonable proportion of real-time idle rate, historical decay, and forecast uncertainty. For example, real-time idle rate (primary) directly reflects current availability, with an idle rate ≥50% receiving the highest score. Historical usage decay (secondary) assigns higher weight to recently frequently used charging stations to avoid interference from older data. Forecast confidence correction (secondary) reduces the score if uncertainty in future demand forecasts is high to prevent resource overload.

[0071] Environmental adaptability:

[0072]

[0073] in: is the environmental adaptability index, is the real-time temperature, is the safety temperature threshold, The real-time load of the power grid, is the reference load, For weather warning levels, 、 、 is a dynamic weight, indicating the priority allocation of high temperature, grid load, and weather warning. 、 The standard deviation is based on historical data. For example, if the temperature safety setting is set to 50%, the score will be significantly reduced when the real-time temperature exceeds the safety threshold (such as 40°C), triggering the cooling system or charging reduction. If the grid load setting is set to 30%, the score will be reduced when the load exceeds the baseline value (such as the historical average), and charging costs will be optimized based on time-of-use electricity prices. If the weather warning setting is set to 20%, the score will be reduced in the event of extreme weather such as heavy rain or strong winds, and protected base stations will be recommended first.

[0074] The quantitative indicators in this step can comprehensively reflect the overall performance of the base station and avoid single-dimensional deviation. Through quantitative scoring, the system can prioritize the recommendation of high-health batteries, low-load base stations and environmentally stable resources to improve user experience and operational efficiency.

[0075] S13. Generate a recommended priority for each base station based on the shared electric vehicle's battery status data, user behavior data, charging demand identification, and availability characteristics;

[0076] This step generates the recommended priority of each base station through a multi-dimensional scoring model based on the battery status, user behavior, charging demand identification collected in step S11 and the base station availability characteristics quantified in step S12, namely battery health, number of available charging positions, and environmental adaptability.

[0077] S131. Obtain the current battery charge and battery health index, generate a battery status score based on the battery health quantification model, and integrate the voltage, cycle count, and temperature history data into the quantification model;

[0078] The battery status score is based on the battery power, health, voltage, cycle times and temperature history data collected in real time in step S11. The quantification model adopts the above battery health quantification model to calculate the battery health index. , give priority to recommending batteries with long battery life and low aging degree to reduce users' riding interruptions caused by battery problems.

[0079] S132. Generate a user behavior adaptation score based on historical riding route preferences, time sensitivity, and battery selection tendencies in the user behavior data;

[0080] 1. Define the user behavior state space, including historical riding path sets, time sensitivity intervals, and battery health selection preferences;

[0081] 2. Define the action space as base station selection behavior. The reward function is dynamically calculated based on the matching degree between the user's final base station selection and the system's recommended base stations, order completion efficiency, and battery life satisfaction.

[0082] 3. Train a preset reinforcement learning model using historical user behavior data to output a probability distribution of user preferences for base stations;

[0083] 4. Based on the preference probability distribution, user behavior patterns are divided into high-frequency optimization, regular, and random types, and mapped to corresponding behavior adaptation scores; among them, high-frequency optimization has the highest score and random has the lowest score.

[0084] Specifically, user behavior data includes historical riding routes, time sensitivity, and battery selection preferences collected in step S11. Historical riding routes include frequently used routes and frequently visited base stations. Time sensitivity ranges include an on-time rate ≥ 95% for high sensitivity. Battery health preference represents the percentage of users selecting high-SoH batteries. The action space represents the user's base station selection behavior. The reward function dynamically calculates the reward value based on the degree of compatibility between the user's final base station selection and the system's recommendations, order fulfillment efficiency (e.g., on-time delivery rate), and actual battery life satisfaction (the percentage of remaining battery charge after the user's ride).

[0085] The total reward value is determined by the matching reward , Order efficiency rewards And battery life satisfaction rewards Perform weighted calculation to obtain the result, where if the base station finally selected by the user is consistent with the system recommendation, , otherwise ; ; .

[0086] A deep Q reinforcement learning model is trained using user behavior data to output a probability distribution of user preferences for base stations. The state space consists of the user's historical riding routes, time sensitivity intervals, and battery health preferences; the action space represents the user's base station selection behavior, i.e., the identifiers of all available base stations in the recommendation list. The input layer of the Q network structure is the user state vector (dimension = number of historical routes + number of time sensitivity levels + number of battery preference parameters); the hidden layer is a 3-Chan-connected neural network with a Reluctant Linear Unit (ReLU) activation function; and the output layer is the Q value for each base station. The model training process includes: 1. Experience Recycling: storing user state, action, reward, and next state tuples in a recycling pool; 2. Action Selection: using the ε-greedy strategy; 3. Network Update: sampling batches of data from the replay pool every 1000 steps, calculating the time-delay error, and updating the network parameters using the Adam optimizer; 4. Target Network Synchronization: copying the master network parameters to the target network every 10,000 steps.

[0087] Finally, users are divided into the following categories based on the probability distribution of preferences: high-frequency optimization type corresponds to a preference probability ≥ 70%, with a score of +30%; conventional type corresponds to a probability of 30%≤<70%, with a score of +15%; and random type corresponds to a probability of <30%, with a score of +0%.

[0088] S133. Analyze the charging demand identifier and generate an emergency charging demand score based on the urgency of the user's order;

[0089] The charging demand indicator is generated in step S11 based on the user's real-time battery level and the urgency of the order. Urgent demand is assigned a score of +25%, with a SoC < 20% or a remaining time of less than 10 minutes; high demand is assigned a score of +10%, with a SoC ≤ 20% < 50% and no urgency indicator; and normal demand is assigned a score of +0%, with a SoC ≥ 50%. By responding to the user's urgency, resources are prioritized for low-battery or expedited orders.

[0090] S134. Combined calculation and calculated , the battery status score, user behavior adaptation score and charging demand urgency score are weighted and summed to generate the recommended total score of each base station.

[0091] The battery status score (weighted 30%), user behavior adaptation score (weighted 20%), charging demand urgency score (weighted 25%), charging station availability score (weighted 15%), and environmental adaptability score (weighted 10%) are weighted and summed. Dynamic weighting is used to adapt to real-time scenario needs and optimize resource allocation efficiency. Specifically, the recommended priority of each charging station is optimized, including the following:

[0092] 1. Sort all base stations according to the total recommendation score and divide them into priority levels;

[0093] 2. Optimize priority results based on a dynamic weighting adjustment strategy. This strategy includes increasing the weight of charging station availability during peak hours, environmental adaptability during extreme weather conditions, and user behavior adaptation for VIP users.

[0094] 3. Verify the priority results in multiple dimensions, including whether the battery health meets the minimum threshold, the probability of charging slot reservation conflict, and path accessibility;

[0095] 4. Output the final priority list and push the real-time status and navigation path of the recommended base station to the user terminal.

[0096] Specifically, dynamic filtering includes user identity priority: VIP users are automatically promoted to the top 10% of the list; real-time traffic conditions: if the route is congested or under construction, the corresponding base station priority is lowered; weather warning level: in the event of a heavy rain warning, base stations with canopies or indoor base stations are preferred. Battery health thresholds exclude base stations with a SoH < 75%; charging slot reservation conflicts: if the target base station's reserved slot is preempted, switching to a suboptimal base station; and path accessibility: using the navigation API to verify whether the route is blocked or restricted. Finally, a final priority list is pushed to the user terminal, including base station scores, estimated arrival times, and navigation links, with resources allocated to the highest-priority base stations.

[0097] The above steps combine user behavior preferences, battery health, and real-time needs to provide personalized base station recommendations. Through weight adjustment and multi-dimensional verification, it adapts to complex scenarios such as traffic congestion and extreme weather. It can avoid the allocation of low-health batteries and base station overload, improving operational efficiency and user satisfaction.

[0098] S14 sends a resource lock request containing a dynamic reservation duration to the target base station with the highest priority. The dynamic reservation duration is dynamically adjusted according to the user's real-time location and navigation estimated time. If the user does not arrive within the reserved time, the resource lock is automatically released;

[0099] This step sends a resource lock request to the target base station with the highest priority based on the recommended priority list generated in step S13, and dynamically adjusts the reserved duration based on the user's real-time location and estimated navigation time to ensure efficient resource utilization. If the user does not arrive within the reserved time, the lock is automatically released to avoid idle resources. The details are as follows:

[0100] S141. Collect historical usage data of the target base station, including charging position occupancy rate, user arrival interval and battery replacement time at different time periods;

[0101] Collect historical usage data of the target base station, including the occupancy rate of charging stations during different time periods (such as morning peak hours and evening peak hours), the average time interval for users to arrive at the base station, and the average time required for a single battery replacement operation.

[0102] S142. Build a time series model, extract periodicity and trend characteristics through a sliding window algorithm, and predict the probability of base station congestion within a preset time period in the future;

[0103] A sliding window algorithm is used to construct a time series model. The specific operations are as follows: 1. Extract periodic features (such as peak occupancy during daily peak hours); 2. Extract trend features (such as a 20% increase in charging demand on weekends compared to weekdays); 3. Predict the probability of base station congestion within a preset time period (e.g., within the next hour). For example, if the predicted congestion probability for the next hour is 80%, the base station is considered to be overloaded.

[0104] S143. Dynamically calculate the reservation duration based on the congestion probability and the estimated navigation time from the user's real-time location to the base station; if the predicted congestion probability of the target base station during the user's arrival period is higher than a preset threshold, the reservation duration is shortened to avoid idle resources; otherwise, the reservation duration is extended.

[0105] The reserved time is calculated by the map API in real time based on the time required for the user to reach the base station, the basic reserved buffer time, and the critical value of the congestion probability: Reserved time = Navigation estimated time + Basic reserved buffer time (1 - congestion probability / preset threshold). The preset threshold is the critical congestion probability value (e.g., 60%), which is used to determine whether to shorten or extend the reservation duration. The threshold is optimized through historical data analysis to balance resource utilization and user experience. If the threshold is exceeded, the reservation duration is shortened. For high congestion probability ≥ 60%, the reservation duration is shortened. For example, if the user is expected to arrive in 10 minutes, the reservation duration is set to 8 minutes. For low congestion probability < 60%, the reservation duration is extended. For example, if the user is expected to arrive in 10 minutes, the reservation duration is set to 15 minutes.

[0106] Specifically, the dynamic reservation duration is dynamically adjusted based on the user's real-time location and estimated navigation time. If the user does not arrive within the reserved time, the resource lock is automatically released, including the following:

[0107] 1. Real-time monitoring of user travel speed, traffic conditions and base station resource status updates;

[0108] 2. If the user's speed falls below a preset percentage of the navigation estimate, the reserved time is extended by the delay ratio, and the new estimated arrival time is pushed to the user terminal;

[0109] 3. If the base station resource status update shows that the reserved charging position is unexpectedly occupied, re-match the available base station and update the reserved duration;

[0110] 4. If an external event triggers a sudden change in the base station load, the reserved time will be compressed or extended based on the real-time load rate.

[0111] Real-time monitoring parameters include the user's travel speed, which is calculated by comparing GPS positioning data with historical speed and calculating the deviation between actual speed and estimated speed; access to the traffic API to obtain real-time congestion, construction or accident information and other traffic condition changes; and monitoring of charging station occupancy, battery health and environmental parameters.

[0112] If the actual speed falls below a preset percentage (80%) of the estimated speed, the reservation duration is extended by the delay ratio, using the formula: New Reservation Duration = Original Reservation Duration ⋅ (1 + Estimated Time - Actual Estimated Time). For example, if a user's original arrival time is 10 minutes, but due to congestion, it actually takes 15 minutes, the reservation duration is extended to 15 minutes, and an update notification is sent. If other users or system failures occupy the reserved space, the resource is immediately released and a new available base station is matched. If a sudden downpour causes a sudden increase in base station load, the reservation duration is compressed to the minimum threshold to accelerate resource turnover. If a red alert for heavy rain or high temperatures is triggered, protected base stations are prioritized and their reservation durations are extended. If the grid load exceeds a safety threshold, charging operations are suspended and the reservation strategy is adjusted.

[0113] S15. When a path obstruction or a change in user demand is detected, the base station with the lowest availability and highest arrival efficiency is re-matched and the recommended priority is updated;

[0114] This step is used to re-screen eligible base stations and update the recommendation priority when a route obstruction such as traffic congestion or road construction is detected, or when user demand changes such as urgent order status adjustment or sudden power drop, to ensure that users always have access to the best battery swap resources.

[0115] 1. Based on a preset minimum availability threshold, select candidate base stations that meet the following requirements: battery health ≥ preset health threshold, charging station idle rate ≥ preset idle rate threshold, and environmental adaptability ≥ preset adaptability threshold;

[0116] Path obstruction detection uses an integrated map service API to obtain real-time traffic event data and combines it with the user's GPS location data to determine whether the current navigation route is feasible. If the estimated arrival time exceeds a preset deviation percentage from the original value, the route is considered obstructed. User demand change detection updates the charging demand indicator to a higher urgency level by monitoring manual user actions or automatic system triggers.

[0117] Candidate base stations are screened based on minimum availability thresholds to ensure that recommended base stations meet basic availability requirements. The battery health threshold, based on historical data analysis, states that batteries with a SoH ≥ 75% guarantee at least 80% of their rated range, preventing user interruptions and excluding base stations with a SoH < 75%. The charging spot availability threshold ensures at least one available charging spot, preventing users from arriving without a spot to change to. This is used to screen base stations with an availability rate ≥ 10%. The environmental adaptability threshold is set as follows: temperature ≤ 40°C and humidity ≤ 80% for safe battery operation. Exceeding the threshold may cause performance degradation or failure, and is used to exclude base stations with temperatures > 40°C or humidity > 80%.

[0118] 2. Calculate the shortest estimated arrival time for each base station based on the location of the candidate base stations, real-time traffic conditions, and user speed;

[0119] Based on the geographic location of the candidate base stations, real-time traffic conditions, and the user's real-time travel speed, the map API (such as AutoNavi route planning) is called to obtain the real-time navigation route and estimated time. If the user's travel speed is lower than a preset percentage of the historical average (such as 80%), the estimated time is adjusted proportionally. Calculation formula: Shortest estimated time = Navigation API time ⋅ (1 + ), for example, if the navigation API estimates an arrival time of 8 minutes and the user's real-time speed is only 70% of the historical average, the corrected time is 10.4 minutes.

[0120] 3. Sort candidate base stations in ascending order by shortest estimated arrival time, generate an updated recommendation priority list, and push the adjusted recommendation results and navigation path to the user terminal;

[0121] The candidate base stations are sorted in ascending order by the shortest estimated arrival time to generate an updated priority list; the adjusted recommendation results are pushed to the user terminal, including: the battery health and charging status of the new base station; the revised navigation path and estimated arrival time; and a prompt indicating the reason why the original base station is unavailable.

[0122] 4. If the original target base station no longer meets the conditions, its reserved resources are released and a dynamic resource locking request is sent to the new base station with the highest priority.

[0123] If the original target base station no longer meets the conditions due to path obstruction or resource occupation, its reservation lock is immediately released; a dynamic resource lock request is sent to the base station with the highest new priority, and the reservation duration is recalculated according to the logic of step S14.

[0124] The following scenarios illustrate this: Scenario 1: A user originally planned to travel to base station A (SoH = 80%, idle rate 20%), but the route was blocked due to road construction. The system screened out base stations B (SoH = 78%, idle rate 15%) and C (SoH = 85%, idle rate 25%), calculated that base station C had the shortest arrival time (8 minutes), updated its recommendation, and locked in resources. Scenario 2: The user's battery level suddenly dropped from 25% to 8%, and their needs became urgent. The system immediately screened for base stations with SoH ≥ 75% and idle rates ≥ 10%, prioritizing the recommendation of base station D, which was the closest and had the lowest load.

[0125] S16. After the battery replacement operation is completed, optimize the recommendation strategy based on the actual battery performance data and user operation feedback.

[0126] In this embodiment, the actual battery performance data such as the battery health SoH, charge and discharge efficiency, and temperature stability after battery replacement are collected in real time, and user operation feedback such as battery replacement time, interface operation satisfaction score, and whether to manually adjust the recommendation results are collected; if the battery health of a base station is continuously lower than the preset threshold, it is marked as a "low-quality resource" and its recommendation weight is reduced; if the user frequently refuses to recommend a base station, such as refusing to accept the recommendation due to complex paths or insufficient charging stations, it is recorded as a "low-trust base station" and its exposure priority is reduced; a reinforcement learning model is used to dynamically adjust the score weight according to the degree of match between the user's actual choice and the recommendation result. For example, when the user chooses a non-recommended base station, the distance weight is reduced and the battery health weight is increased; the user retention rate, order completion efficiency and resource utilization under the new and old strategies are compared to select the optimal model version; finally, the model is retrained regularly, the latest battery performance data and user feedback are integrated to generate an updated recommendation strategy; the effect of the new strategy is gradually verified through grayscale release to avoid global risks.

[0127] The above-mentioned embodiments improve the utilization rate of shared electric motorcycle charging and swapping resources and user experience through data-driven dynamic resource allocation. At the same time, through intelligent algorithms and real-time response capabilities, they significantly reduce operation and maintenance costs and enhance the system's adaptability to complex urban scenarios.

[0128] It is understandable that those skilled in the art can, under the guidance of the above embodiments, combine various implementation methods in the above embodiments to obtain technical solutions of multiple implementation methods.

[0129] 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 in the scope of protection of the present invention.

Claims

1. A method for intelligent interactive dynamic resource recommendation between shared electric motorcycles and charging and swapping integrated base stations, characterized in that: include: Real-time collection of shared electric vehicle battery status data, user behavior data, and charging demand indicators; At the same time, the resource status data of the integrated charging and swapping base station is collected; Evaluate availability characteristics of available batteries at each base station based on the resource status data, the availability characteristics including battery health, number of available charging stations, and environmental adaptability; Generate a recommended priority for each base station based on the shared electric vehicle's battery status data, user behavior data, charging demand identification, and the availability characteristics; Send a resource lock request including a dynamic reservation duration to the target base station with the highest priority. The dynamic reservation duration is dynamically adjusted based on the user's real-time location and estimated navigation time. If the user does not arrive within the dynamic reservation time, the resource lock is automatically released. The step of sending a resource locking request including a dynamic reservation duration to a target base station with the highest priority, wherein the dynamic reservation duration is dynamically adjusted according to the user's real-time location and the estimated navigation time, and automatically releasing the resource lock if the user does not arrive within the reserved time, includes: Collect historical usage data of the target base station, including charging station occupancy rate, user arrival interval, and battery replacement time at different time periods; Build a time series model, extract periodic and trend features through a sliding window algorithm, and predict the probability of base station congestion within a preset time period in the future; Dynamically calculate the reserved time based on the congestion probability and the estimated navigation time from the user's real-time location to the base station; If the predicted congestion probability of the target base station during the user arrival period is higher than a preset threshold, the reservation duration is shortened to avoid idle resources; otherwise, the reservation duration is extended. When a path obstruction or a change in user demand is detected, the base station with the lowest availability characteristics and the highest arrival efficiency is re-matched and the recommended priority is updated; After the battery replacement operation is completed, the recommendation strategy is optimized based on the actual performance data of the replaced new battery and user operation feedback. The actual performance data of the new battery includes battery health, charge and discharge efficiency and temperature stability corresponding to the availability characteristics.

2. The method for intelligent interactive dynamic resource recommendation of shared electric motorcycles and charging and swapping integrated base stations according to claim 1 is characterized in that: The evaluation of the availability characteristics of the available batteries of each base station based on the resource status data, wherein the availability characteristics include battery health, the number of available charging positions, and environmental adaptability, includes: in: is the battery health index, is the actual capacity of the battery. is the nominal capacity, is the cumulative number of cycles, To design the maximum cycle life, is the total number of charge and discharge cycles, For the Average temperature and optimal temperature during the charge and discharge cycle The deviation, is the temperature reference period, 、 and is the weight coefficient, based on the priority allocation in the shared tram scenario where capacity decay is dominant, cycle number is second, and temperature has a long-term impact.

3. The method for intelligent interactive dynamic resource recommendation of shared electric motorcycles and charging and swapping integrated base stations according to claim 2 is characterized in that: The availability characteristics of the available batteries of each base station are evaluated based on the resource status data, wherein the availability characteristics include battery health, number of available charging stations, and environmental adaptability, and further include: in: is the number of available charging stations, is the number of currently available charging slots, is the total number of charging positions, is the total number of charging positions, For the The historical usage frequency of each charging station, Current time With the The most recent usage time of the charging station The interval is the attenuation factor Give more weight to recent use, is the confidence level of demand forecast for a certain period in the future, 、 、 To balance the coefficient, ensure a reasonable proportion of real-time idle rate, historical decay and forecast uncertainty.

4. The method for intelligent interactive dynamic resource recommendation of shared electric motorcycles and charging and swapping integrated base stations according to claim 3 is characterized in that: The availability characteristics of the available batteries of each base station are evaluated based on the resource status data, wherein the availability characteristics include battery health, number of available charging stations, and environmental adaptability, and further include: in: is the environmental adaptability index, is the real-time temperature, is the safety temperature threshold, The real-time load of the power grid, is the reference load, For weather warning levels, 、 、 is a dynamic weight, indicating the priority allocation of high temperature, grid load, and weather warning. 、 Standard deviation, based on historical data statistics.

5. The method for intelligent interactive dynamic resource recommendation between a shared electric motorcycle and an integrated charging and swapping base station according to any one of claims 1 to 4, characterized in that: Generating the recommended priority of each base station based on the battery status data, user behavior data, charging demand identification, and availability characteristics of the shared electric vehicle includes: Obtain the current battery charge and battery health index, and generate a battery status score based on a battery health quantification model that incorporates voltage, cycle count, and temperature history data; Generate a user behavior adaptation score based on historical riding route preferences, time sensitivity, and battery selection tendencies in user behavior data; Parse the charging demand identifier and generate a charging demand urgency score based on the urgency of the user's order; Combined calculation and calculated , performing weighted summation on the battery status score, user behavior adaptation score, and charging demand urgency score to generate a total recommendation score for each base station.

6. The method for intelligent interactive dynamic resource recommendation of shared electric motorcycles and charging and swapping integrated base stations according to claim 5 is characterized in that: The user behavior adaptation score is generated based on the historical riding route preference, time sensitivity, and battery selection tendency in the user behavior data, including: Define the user behavior state space, including historical riding path sets, time sensitivity intervals, and battery health selection preferences; The action space is defined as base station selection behavior, and the reward function is dynamically calculated based on the matching degree between the user's final selected base station and the system's recommended base station, order completion efficiency, and battery life satisfaction. The preset reinforcement learning model is trained through user behavior data to output the probability distribution of user preference for base stations; According to the preference probability distribution, user behavior patterns are divided into high-frequency optimization type, regular type and random type, and mapped to corresponding behavior adaptation scores; among them, the high-frequency optimization type has the highest score and the random type has the lowest score.

7. The method for intelligent interactive dynamic resource recommendation of shared electric motorcycles and charging and swapping integrated base stations according to claim 5 is characterized in that: Generating the recommended priority of each base station includes: Sort all base stations according to the total recommendation score and divide them into priority levels; Optimizing priority results based on a dynamic weighting strategy, including increasing the weight of charging station availability during peak hours, increasing the weight of environmental adaptability during extreme weather conditions, and increasing the weight of user behavior adaptation for VIP users; Multi-dimensional verification of the priority results, including whether the battery health meets the minimum threshold, the probability of charging slot reservation conflict, and path accessibility; Output the final priority list and push the real-time status and navigation path of the recommended base station to the user terminal.

8. The method for intelligent interactive dynamic resource recommendation of shared electric motorcycles and charging and swapping integrated base stations according to claim 1 is characterized in that: The dynamic reservation duration is dynamically adjusted based on the user's real-time location and estimated navigation time. If the user does not arrive within the reserved time, the resource lock is automatically released, including: Real-time monitoring of user travel speed, traffic conditions and base station resource status updates; If the user's travel speed is lower than the preset ratio of the navigation estimate, the reserved time is extended according to the delay ratio, and the new estimated arrival time is pushed through the user terminal; If the base station resource status update shows that the reserved charging position is unexpectedly occupied, re-match the available base stations and update the reservation duration; If an external event triggers a sudden change in the base station load, the reserved time will be compressed or extended according to the real-time load rate.

9. The method for intelligent interactive dynamic resource recommendation of shared electric motorcycles and charging and swapping integrated base stations according to claim 1 is characterized in that: When a path obstruction or a change in user demand is detected, re-matching a base station that meets the minimum availability characteristics and has the highest arrival efficiency and updating the recommendation priority includes: Based on a preset minimum availability threshold, select candidate base stations that meet the following requirements: battery health ≥ preset health threshold, charging position idle rate ≥ preset idle rate threshold, and environmental adaptability ≥ preset adaptability threshold; Calculate the shortest estimated arrival time for each base station based on the location of the candidate base stations, real-time traffic conditions, and user travel speed; Sort candidate base stations in ascending order by the shortest estimated arrival time, generate an updated recommendation priority list, and push the adjusted recommendation results and navigation path to the user terminal; If the original target base station no longer meets the conditions, its reserved resources are released and a dynamic resource locking request is sent to the new base station with the highest priority.

Images