Multi-dimensional feature spatio-temporal quantification method for user charging and swapping behavior considering cross-domain factors

Abstract

The application discloses a user charging and changing behavior multi-dimensional feature space-time quantification method considering cross-domain factors, relies on user travel law, charging and changing service level, power grid price level, traffic network data, and meteorological factors of the place, and objectively quantifies the charging behavior of the user. In the method, not only the influence of the user charging behavior on the surrounding road traffic is proposed, the blank in the field is made up, but also the influence of the charging behavior on the area is fully considered, the traditional single evaluation dimension is transcended, the basic ability construction of vehicle-road-network coordination optimization can be better supported, the scientificity and precision of charging and changing infrastructure planning are improved, the user behavior modeling theory is enriched, and cross-domain technology integration is promoted.

Description

Technical Field

[0001] This invention relates to the integration of transportation and energy sectors and the field of user behavior analysis, and in particular to a spatiotemporal quantification method for multidimensional features of user charging and swapping behavior that considers cross-domain factors. Background Technology

[0002] With the continuous growth of electric vehicle ownership, their charging and battery swapping behavior has become a key link in the coupled regulation and control of transportation and power systems. Under the backdrop of "dual-carbon" goals and smart city construction, electric vehicles, as mobile energy storage units, not only affect the spatiotemporal distribution of grid load but are also closely related to traffic efficiency, infrastructure layout, and environmental factors. Among related technologies, a preliminary correlation model between user behavior and system operation has been constructed through the coordinated operation of travel trajectory analysis, charging facility status monitoring, and grid dispatching strategies. Specifically, this technical system covers the entire process from user decision-making mechanisms to charging and battery swapping behavior prediction, including key aspects such as vehicle status identification, charging service evaluation, traffic congestion response, electricity price sensitivity analysis, and meteorological impact modeling, initially achieving multi-dimensional perception of user behavior and partial system coordination.

[0003] However, existing user charging and swapping behavior modeling methods directly use single-dimensional data (such as user preferences, charging convenience, or economics) for analysis, without systematically integrating key cross-domain elements such as "vehicle-charging station-road-network-weather." This may result in insufficient dimensions for behavioral characterization, making it difficult to accurately reflect users' true decision-making mechanisms in complex environments. Specifically, existing technologies typically build models based on static or localized data, lacking collaborative modeling of dynamic factors such as traffic congestion indices, time-of-use electricity pricing, charging station queuing times, and the probability of extreme weather, thus affecting the spatiotemporal accuracy of behavior prediction and system responsiveness. While vehicle status and charging facility service levels have been considered in some studies, their coupling relationship with traffic networks, grid scheduling, and weather conditions has not been effectively quantified, limiting the depth and breadth of collaborative optimization between transportation and energy systems. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] This invention proposes a spatiotemporal quantification method for multidimensional features of user charging and swapping behavior that considers cross-domain factors.

[0006] Another objective of this invention is to propose a spatiotemporal quantification device for multidimensional features of user charging and swapping behavior that takes into account cross-domain factors.

[0007] To achieve the above objectives, this invention proposes a spatiotemporal quantification method for multidimensional features of user charging and swapping behavior that considers cross-domain factors, comprising:

[0008] S1. Construct a multi-dimensional behavioral characteristic indicator system that includes vehicle status, charging facility service level, traffic operation status, grid electricity price level and meteorological factors to quantitatively characterize the impact of each dimension on users' charging and swapping behavior.

[0009] S2, based on the multi-dimensional behavioral feature index system, integrate user travel trajectory, charging pile operation data, traffic congestion index, time-of-use electricity price information and meteorological observation data to perform standardized processing and spatiotemporal alignment of multi-source heterogeneous data;

[0010] S3 utilizes a multinomial logistic regression model and nonlinear least squares method to calibrate and model the probability of users choosing to charge or swap batteries and the probability of choosing charging stations at different times, thereby enabling dynamic prediction of user behavior.

[0011] S4 combines historical user behavior data with real-time environmental data to build a cross-domain collaborative modeling framework to simulate the evolution trend of user charging and swapping behavior in time and space.

[0012] The spatiotemporal quantization method for multidimensional features of user charging and swapping behavior that considers cross-domain factors in this invention may also have the following additional technical features:

[0013] In one embodiment of the present invention, S1 includes:

[0014] S11 defines vehicle status parameters including battery capacity, current remaining charge, acceptable minimum charge threshold, and whether battery swapping is supported, in order to quantify the urgency of recharging for users in different charge states.

[0015] S12, construct service level parameters for charging facilities, including the rated charging power of charging piles, queue length, and queue time, to reflect users' sensitivity to charging efficiency and service accessibility when selecting charging facilities.

[0016] In one embodiment of the present invention, S2 includes:

[0017] S21, a timestamp alignment algorithm is used to match user travel trajectory data with grid time-of-use electricity price data in the time dimension to ensure the synchronization of data in different time domains in behavioral modeling;

[0018] S22 uses spatial interpolation to standardize the spatial dimensions of traffic congestion index and meteorological observation data, so as to unify the expression of data from different geographical locations.

[0019] In one embodiment of the present invention, S3 includes:

[0020] S31 uses nonlinear least squares to fit the regression coefficients and intercept terms in the multinomial logistic regression model to improve the model's accuracy in characterizing user behavior response features.

[0021] S32 dynamically updates model parameters based on users' historical charging and swapping behavior data, real-time electricity prices, traffic congestion, and queuing times to adapt to the evolving trends of user behavior.

[0022] In one embodiment of the present invention, S4 includes:

[0023] S41 introduces a travel chain path accessibility analysis module to evaluate the feasibility and efficiency of the path from the user's current location to each candidate charging station.

[0024] S42. Construct a charging station selection model based on a utility function, where the utility function includes distance, electricity price, queuing time, traffic congestion index, and weighted coefficients of each influencing factor.

[0025] In one embodiment of the present invention, it further includes:

[0026] S5, based on a cross-domain collaborative modeling framework, generates a spatiotemporal heat map of user charging and swapping behavior, which is used to predict the charging and swapping demand density and grid load change trends in various regions within a specific future time period.

[0027] To achieve the above objectives, another aspect of the present invention proposes a spatiotemporal quantification device for multidimensional features of user charging and swapping behavior that considers cross-domain factors, comprising:

[0028] The multi-dimensional feature construction module is used to build a multi-dimensional behavioral feature index system that includes vehicle status, charging facility service level, traffic operation status, grid electricity price level and meteorological factors, and quantitatively characterize the impact of each dimension on users' charging and swapping behavior.

[0029] The data fusion and spatiotemporal alignment module is used to perform standardized processing and spatiotemporal alignment of multi-source heterogeneous data based on the multi-dimensional behavioral feature index system, which integrates user travel trajectory, charging pile operation data, traffic congestion index, time-of-use electricity price information and meteorological observation data.

[0030] The behavioral probability modeling module is used to calibrate and model the probability of users choosing to charge or swap batteries and the probability of choosing charging stations at different times using a multinomial logistic regression model and a nonlinear least squares method, so as to achieve dynamic prediction of user behavior.

[0031] The cross-domain collaborative modeling module is used to combine historical user behavior data with real-time environmental data to build a cross-domain collaborative modeling framework to simulate the evolution trend of user charging and swapping behavior in time and space.

[0032] The spatiotemporal quantification method and apparatus for multidimensional features of user charging and swapping behavior that considers cross-domain factors in this invention can systematically integrate multi-source heterogeneous data from "vehicle-charging pile-road-network-meteorology" to achieve accurate quantification and dynamic modeling of user charging and swapping behavior in the spatiotemporal dimension, thereby improving the accuracy of behavior prediction and the collaborative optimization capability of the transportation energy system.

[0033] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0034] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0035] Figure 1 This is a flowchart of a spatiotemporal quantification method for multidimensional features of user charging and swapping behavior that considers cross-domain factors, according to an embodiment of the present invention.

[0036] Figure 2 This is another flowchart of the spatiotemporal quantification method for multidimensional features of user charging and swapping behavior that considers cross-domain factors according to an embodiment of the present invention;

[0037] Figure 3 This is a structural diagram of a spatiotemporal quantification device for multidimensional features of user charging and swapping behavior that considers cross-domain factors, according to an embodiment of the present invention. Detailed Implementation

[0038] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0039] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0040] The following describes, with reference to the accompanying drawings, a method and apparatus for spatiotemporal quantification of multidimensional features of user charging and swapping behavior considering cross-domain factors, according to embodiments of the present invention.

[0041] Example 1

[0042] Figure 1 This is a flowchart of a spatiotemporal quantization method for multidimensional features of user charging and swapping behavior that considers cross-domain factors, according to an embodiment of the present invention. Figure 1As shown, it includes:

[0043] S1 constructs a multi-dimensional behavioral characteristic indicator system that includes vehicle status, charging facility service level, traffic operation status, grid electricity price level and meteorological factors, and quantitatively represents the impact of each dimension on users' charging and swapping behavior.

[0044] Specifically, this step aims to construct a multi-dimensional behavioral characteristic indicator system encompassing vehicle status, charging facility service levels, traffic conditions, grid electricity prices, and meteorological factors, in order to quantify the impact of each dimension on users' charging and swapping behavior. This system systematically integrates five key elements—vehicle, charging station, road, grid, and weather—to achieve multi-dimensional, spatiotemporal dynamic modeling of user behavior, thereby improving the accuracy of behavior prediction and decision support capabilities.

[0045] This step first defines and extracts key metrics for each dimension. For example, the vehicle state dimension includes battery capacity (Ah), current remaining charge (SOC, percentage), acceptable minimum charge threshold (SOC_min, typically set at 15%-20%), and whether battery swapping is supported (a binary variable, 1 indicating no support, 2 indicating support). The service level of charging facilities is quantified by metrics such as the rated power of charging piles (kW), queue length (vehicles), and queue time (minutes). The calculation of queue time can be based on queuing theory models, such as the M / M / n model, deriving the system idle probability, average queue length, and average queue time through arrival rate (λ, vehicles / hour) and service rate (μ, vehicles / hour).

[0046] Regarding traffic conditions, the Traffic Congestion Index (TTI) is used as the core indicator, defined as the ratio of peak-hour travel time to off-peak travel time, representing the accessibility and travel cost for users to charging stations. Electricity prices are modeled using a Time-of-Use (TOU) mechanism, incorporating a price sensitivity index to calculate changes in user charging behavior during different electricity price periods, revealing its economic drivers. For meteorological factors, an extreme weather probability model is constructed using logistic regression. Input variables include temperature (°C) and humidity (%RH), with the output being the probability (0-1 range) of weather types such as heavy rain and typhoons.

[0047] This step, by constructing a multi-dimensional indicator system, provides a structured and standardized data foundation for subsequent behavior modeling and strategy optimization, enables collaborative analysis of cross-domain factors, significantly improves the quantitative accuracy and model generalization ability of user charging and swapping behavior, and has important engineering application value and theoretical research significance.

[0048] Furthermore, S1 includes:

[0049] S11 defines vehicle status parameters including battery capacity, current remaining charge, minimum acceptable charge threshold, and whether battery swapping is supported, in order to quantify the urgency of recharging for users in different charge states.

[0050] Specifically, this step aims to quantitatively assess the urgency of recharging for users under different battery charge states by defining vehicle state parameters (including battery capacity, current remaining charge, acceptable minimum state of charge threshold SOC_min, and whether battery swapping is supported). This step is a key input module in the entire multidimensional feature spatiotemporal quantification method, providing fundamental data support for subsequent charging and swapping behavior modeling and decision support.

[0051] The collection and modeling of vehicle status parameters requires combining onboard OBD data, user historical charging records, and a vehicle type database. Battery capacity is typically measured in ampere-hours (Ah) or kilowatt-hours (kWh), reflecting the vehicle's maximum energy storage capacity. State of Charge (SOC) is expressed as a percentage, such as 80% SOC indicating 80% battery capacity remaining. The acceptable minimum charge threshold (SOC_min) is the lowest tolerable charge level for the user, typically set between 15% and 20% SOC to avoid operational interruptions due to insufficient charge during driving. Whether battery swapping is supported is encoded using a binary variable, such as "1" indicating no swapping and "2" indicating support. This parameter directly affects the user's choice of charging methods and response speed.

[0052] The SOC_min setting needs to be dynamically adjusted based on the vehicle's range and the user's travel habits. For example, for a user with an average daily mileage of 50km and a vehicle range of 300km, SOC_min can be set to 15%; while for users who travel frequently and over long distances, SOC_min may need to be increased to 25% to ensure travel continuity. Furthermore, the difference between SOC and SOC_min (ΔSOC) can serve as a key indicator of the urgency of refueling; the smaller the ΔSOC, the stronger the user's need for refueling.

[0053] This step is widely applicable to scenarios such as electric vehicle user behavior prediction, charging load scheduling, and battery swapping station site optimization. For example, in urban traffic management systems, by monitoring the difference between a user's State of Charge (SOC) and SOC_min in real time, their charging demand at different road sections and time periods can be dynamically predicted, thereby optimizing charging resource allocation and traffic guidance strategies.

[0054] S12, construct service level parameters for charging facilities, including the rated charging power of charging piles, queue length, and queue time, to reflect users' sensitivity to charging efficiency and service accessibility when selecting charging facilities.

[0055] Specifically, this step aims to construct service level parameters for charging facilities, including the rated charging power of charging piles, queue length, and queue time, to quantify users' sensitivity to charging efficiency and service accessibility when choosing charging facilities. In some implementations, this step is based on queuing theory and service system modeling principles, combined with charging station operation data and user behavior characteristics, to construct a multi-dimensional evaluation index system, thereby achieving dynamic evaluation of the service performance of charging facilities.

[0056] The rated charging power (unit: kW) of a charging pile is a core parameter for measuring its service capacity. It is typically set based on the type of charging pile (e.g., fast charging, slow charging) and equipment specifications. For example, the power range for fast charging piles is 60-360kW, while for slow charging piles it is 7-21kW. Queue length (unit: vehicles) indicates the number of vehicles currently waiting to charge at the charging station. This can be collected through a real-time monitoring system or predicted and modeled based on historical data. Queue time (unit: minutes) reflects the average waiting time for users at the charging station. Its calculation depends on the arrival rate (λ, unit: vehicles / hour) and service rate (μ, unit: vehicles / hour), and is combined with the number of service counters (n) for system stability analysis.

[0057] This step can be widely applied to urban charging infrastructure planning, charging station operation optimization, and user behavior prediction systems. For example, during peak traffic hours or holidays, the queue length and time at charging stations increase significantly, and users become more sensitive to charging efficiency. At this time, the system can dynamically adjust charging strategies or guide users to alternative stations to improve the overall service experience.

[0058] By introducing quantitative indicators, the service capabilities and user perceptions of charging facilities can be accurately reflected, providing key support for subsequent user behavior modeling and cross-system collaborative optimization, and improving the intelligence level and operational efficiency of the charging service system.

[0059] This invention is the first to systematically introduce five key elements: vehicle, charging pile, road, network, and weather. It integrates user travel trajectories, charging pile status, electricity price curves, traffic flow, and weather factors to achieve unified modeling of multi-source heterogeneous data. Breaking through the limitations of traditional single-dimensional modeling, it enables comprehensive quantification, dynamic analysis, and spatiotemporal prediction of user charging / swapping behavior, thereby improving modeling accuracy.

[0060] S2, based on the multi-dimensional behavioral feature index system, integrates user travel trajectory, charging pile operation data, traffic congestion index, time-of-use electricity price information and meteorological observation data to perform standardized processing and spatiotemporal alignment of multi-source heterogeneous data.

[0061] Traffic congestion indexes typically use real-time traffic indices provided by Baidu or Gaode Maps, with values ​​ranging from 0 to 10. Higher values ​​indicate more congested roads. Time-of-use electricity pricing information is based on time-of-use pricing standards published by the State Grid or local power companies (e.g., peak-valley price ratio of 1.6:1:0.4), and is mapped to time-series data. Meteorological data, including temperature, rainfall, and wind speed, usually comes from the China Meteorological Administration or the OpenWeatherMap API and requires timestamp alignment and spatial interpolation to match the spatiotemporal resolution of user behavior data.

[0062] Furthermore, this step uses spatiotemporal alignment algorithms (such as sliding matching based on time windows or spatial buffer matching based on GIS) to precisely correlate user travel trajectories with charging behavior events in both time and space. For example, a time window of 15 minutes and a spatial buffer of 500 meters are set to identify a user's charging intentions at specific times and locations. Through this processing, a unified spatiotemporal feature vector can be constructed, providing a high-quality, structured data foundation for subsequent user behavior modeling and decision analysis, significantly improving the model's generalization ability and prediction accuracy.

[0063] Furthermore, S2 includes:

[0064] S21 uses a timestamp alignment algorithm to match user travel trajectory data with grid time-of-use electricity price data in the time dimension, ensuring the synchronization of data from different time domains in behavioral modeling.

[0065] Specifically, this step employs a timestamp alignment algorithm to precisely match user travel trajectory data with grid time-of-use electricity price data in the time dimension, ensuring the synchronization and consistency of traffic behavior and energy consumption behavior over time during the behavior modeling process. This step is a crucial link in achieving multi-dimensional feature fusion modeling of "vehicle-charging station-road-grid-weather," and its technical implementation is based on time-series data interpolation, resampling, and time window matching strategies.

[0066] In some implementations, user travel trajectory data is typically stored as a spatiotemporal point sequence recorded by GPS or collected by an OBU (On-Board Unit), with timestamp accuracy down to the millisecond level and sampling frequency generally between 1 and 10 seconds. Grid time-of-use electricity price data, on the other hand, is released at fixed time intervals (e.g., 15 minutes, 30 minutes, or 1 hour), exhibiting lower temporal resolution. To achieve alignment between the two in the time dimension, this invention employs an interpolation alignment algorithm based on a sliding time window. This algorithm maps each time point in the user's trajectory to the nearest electricity price period and performs linear interpolation or nearest-neighbor interpolation based on the time difference to obtain the electricity value at that moment. For example, if a user performs a charging operation at 14:07, and the electricity price data is divided into 15-minute units, this time point can be mapped to the 14:00-14:15 period, and interpolation is performed based on the time offset (7 minutes).

[0067] Furthermore, to improve alignment accuracy, a timestamp calibration mechanism can be introduced to remove or correct abnormal data through timestamp consistency checks (such as timestamp continuity and sampling interval stability). Simultaneously, a time tolerance threshold (such as ±30 seconds) can be set to address data acquisition errors or system latency issues.

[0068] This step is widely applicable in practical scenarios such as user behavior-based charging and swapping prediction, load scheduling optimization, and traffic-energy coordinated control. By aligning timestamps, it is possible to effectively identify user charging behavior responses under different electricity price periods, providing high-quality time-synchronized data support for the subsequent construction of user behavior models based on electricity price sensitivity, thereby improving the model's prediction accuracy and decision support capabilities.

[0069] S22 uses spatial interpolation to standardize the spatial dimensions of traffic congestion index and meteorological observation data, so as to unify the expression of data from different geographical locations.

[0070] Specifically, this step standardizes the spatial dimensions of traffic congestion index and meteorological observation data using spatial interpolation methods. This aims to unify the representation of data from different geographical locations, providing a consistent spatial data foundation for subsequent multi-dimensional feature modeling of user charging and swapping behavior. In some implementations, this step employs spatial statistical methods such as Kriging Interpolation or Inverse Distance Weighting (IDW) to transform discrete traffic congestion index and meteorological observation data (such as temperature, rainfall, and wind speed) into a continuous spatial distribution field, thereby eliminating data representation differences caused by uneven geographical distribution.

[0071] This step is widely used in the coordinated optimization of urban transportation and energy systems, especially in scenarios such as electric vehicle user behavior prediction, charging station site selection evaluation, and emergency dispatch under extreme weather conditions. By spatially standardizing traffic and meteorological data, it is possible to achieve unified modeling of the influencing factors of user charging and swapping behavior in different regions and at different times, thereby improving the spatial adaptability and generalization ability of the prediction model.

[0072] This step effectively solves the problem of spatial inconsistency in multi-source heterogeneous data, enhances the model's sensitivity to geospatial features, and provides key support for building a collaborative modeling framework of "vehicle-pile-road-network-meteorology", thereby improving the accuracy and reliability of spatiotemporal quantitative analysis of user charging and swapping behavior.

[0073] This invention constructs multi-dimensional user behavior characteristic indicators, including vehicle status, charging facility service level, price sensitivity, traffic operation status, and meteorological influencing factors. Quantitatively characterizing the behavioral patterns of different user groups in different situations helps in the classification of charging behavior profiles and the formulation of personalized scheduling strategies.

[0074] S3 utilizes a multinomial logistic regression model and nonlinear least squares method to calibrate and model the probability of users choosing to charge or swap batteries and the probability of choosing charging stations at different times, thereby enabling dynamic prediction of user behavior.

[0075] Specifically, the core of this step lies in using a multinomial logistic regression model and nonlinear least squares method to calibrate and model the probability of users choosing charging / swapping and the probability of choosing charging stations at different times, thereby achieving dynamic prediction of user behavior. In some implementations, this modeling process first constructs a quantitative index system of behavioral influencing factors based on multidimensional characteristics such as user vehicle status (e.g., SOC, battery capacity, whether battery swapping is supported), charging facility service level (e.g., charging pile power, queuing time, number of service counters), traffic conditions (e.g., congestion index), grid electricity price level (e.g., time-of-use pricing, user sensitivity), and meteorological factors (e.g., temperature, rainfall, probability of extreme weather).

[0076] Furthermore, a multinomial logistic regression (MNL) model is used to model user choice behavior across multiple charging / battery swapping periods. The model input consists of the aforementioned multidimensional feature variables, and the output is the probability that a user chooses charging or battery swapping during period T.

[0077] To improve the model's fitting accuracy and generalization ability, this invention further employs nonlinear least squares (NLS) to calibrate the model parameters. This method optimizes the estimation of regression coefficients by minimizing the sum of squared errors between the predicted probabilities and actual user behavior data.

[0078] In practical applications, this step can be deployed in city-level charging and battery swapping scheduling systems. By combining real-time traffic flow, grid load, and weather forecast data, it dynamically predicts users' charging and battery swapping behavior at different times and locations, providing decision support for charging station load management, traffic resource optimization scheduling, and grid demand response. This modeling method significantly improves the spatiotemporal resolution and accuracy of user behavior prediction, providing key technical support for achieving collaborative optimization of multiple systems including vehicles, charging stations, roads, networks, and weather.

[0079] Furthermore, S3 includes:

[0080] S31 uses nonlinear least squares to fit the regression coefficients and intercept terms in the multinomial logistic regression model to improve the model's accuracy in characterizing user behavior response features.

[0081] Specifically, in this invention, the use of nonlinear least squares to fit the regression coefficients and intercept terms in a multinomial logistic regression model is one of the key steps in improving the accuracy of multidimensional characterization of user charging and swapping behavior. This step is based on the nonlinear relationship between historical user charging and swapping behavior data and multi-source heterogeneous influencing factors (such as vehicle status, charging facility service level, grid electricity price, traffic congestion index, and weather conditions). By optimizing model parameters, the goodness of fit and predictive ability of the behavior response model are improved.

[0082] At the technical implementation level, this step first constructs a multi-logistic regression model, with the dependent variable being the probability that a user chooses to charge or swap batteries at different time periods T, and the independent variables including key features such as remaining battery SOC, charging price, queuing time, traffic congestion index, and weather coefficient.

[0083] This step is widely used in scenarios such as optimizing the site selection of city-level charging and swapping facilities, constructing user behavior profiles, and forecasting power grid load. The regression coefficients obtained through fitting can reflect the sensitivity of users to variables such as electricity prices, queuing times, and traffic congestion, providing a quantitative basis for formulating differentiated service strategies and dynamic pricing mechanisms. This step significantly improves the model's accuracy in characterizing user behavior response features. Compared to traditional linear regression or simple logistic regression methods, nonlinear least squares can more accurately capture the nonlinear interaction relationships between variables, improving the model's goodness of fit. 2 This enhances the system's ability to predict user behavior and its response level by measuring the Area Under the ROC Curve (AUC).

[0084] S32 dynamically updates model parameters based on users' historical charging and swapping behavior data and real-time data such as electricity prices, traffic congestion, and queuing times to adapt to the evolving trends of user behavior.

[0085] Specifically, the core of this step lies in dynamically updating model parameters based on historical user charging and battery swapping behavior data and multi-source dynamic data such as real-time electricity prices, traffic congestion, and queuing times to adapt to the evolving trends of user behavior. In some implementations, this process employs an online learning mechanism and a multi-source data fusion strategy. By collecting and processing data in real time from vehicle OBD systems, charging station operation platforms, power grid dispatching systems, traffic flow monitoring systems, and meteorological databases, dynamic feature vectors are constructed and input into the behavior prediction model. Specifically, historical user data includes, but is not limited to, charging time, location, SOC change curve, vehicle type, and battery swapping support status, while real-time data covers current electricity prices (such as time-of-use pricing and peak pricing), traffic congestion indices (such as the road segment travel time ratio calculated based on floating car data), charging station queue length and waiting time (calculated through queuing theory models, such as the average queuing time and system idle probability in the M / M / n model), etc.

[0086] This invention employs nonlinear least squares or online gradient descent algorithms to dynamically fit regression coefficients in the model (such as charging price sensitivity coefficient, traffic congestion impact coefficient, and queuing time weight). Model input features include: whether the remaining SOC is below a set threshold (0 or 1), the current electricity price (yuan / kWh), the traffic congestion index (e.g., 0.8 indicates slight congestion, 1.5 indicates severe congestion), and queuing time (minutes). The output is the probability that a user will choose to charge or swap batteries during a specific time period. By continuously updating model parameters, the system can capture the dynamic changes in user behavior over time, space, and the external environment, improving prediction accuracy and responsiveness.

[0087] In practical applications, this step can be deployed in city-level charging and swapping scheduling platforms or smart grid management systems. By combining real-time data stream processing technologies (such as Apache Kafka and Flink) with distributed computing frameworks (such as Spark), it can achieve model parameter updates at the minute or even second level. Its technical effect lies in significantly improving the adaptability and robustness of user behavior prediction models, providing data support for dynamic scheduling of charging facilities, load forecasting, and optimization of user response mechanisms, thereby enhancing the coordinated control capabilities of the transportation-energy system.

[0088] This invention establishes a unified collaborative modeling framework to collaboratively couple modeling modules for subsystems such as transportation, energy, electricity, and meteorology. It quantifies user charging / swapping behavior from the perspective of collaborative control and coordinated scheduling within complex urban systems, providing support for charging infrastructure site selection and user-level response optimization.

[0089] S4 combines historical user behavior data with real-time environmental data to build a cross-domain collaborative modeling framework to simulate the evolution trend of user charging and swapping behavior in time and space.

[0090] Specifically, the core of this step lies in combining historical user behavior data with real-time environmental data to construct a cross-domain collaborative modeling framework, simulating the evolutionary trend of user charging and swapping behavior in time and space. This step, through the fusion modeling of multi-source heterogeneous data, achieves dynamic prediction and spatiotemporal quantification of user behavior, and is a key link in this invention to realize multi-dimensional feature-driven behavior modeling.

[0091] At the technical implementation level, this step first extracts key features from historical user behavior data, including the vehicle's battery capacity, current remaining charge (SOC), acceptable minimum charge threshold (SOC_min), and whether battery swapping is supported (Binary Flag). Simultaneously, it integrates real-time environmental data, such as the Traffic Congestion Index (TTI), charging station service rate (λ), queuing time (Wq), grid time-of-use pricing (TOU), and meteorological parameters (temperature T, humidity H, and extreme weather probability P_weather). By constructing a quantitative representation system of five influencing factors—vehicle-charging station-road-grid-meteorology—the above data is mapped into a unified feature vector and input into a cross-domain collaborative modeling framework.

[0092] At the parameter level, the SOC threshold is usually set at 20% to 30% to reflect the user's sensitivity to range anxiety; the electricity price sensitivity ε is fitted through historical data and ranges from 0.5 to 1.2; the traffic congestion index TTI uses real-time data provided by Gaode Map or Baidu Map API and ranges from 0 to 10, with higher values ​​indicating more severe congestion; the extreme weather probability P_weather is based on historical meteorological data and LSTM model prediction, with an accuracy of over 90%.

[0093] This step can be widely applied to scenarios such as optimizing the layout of urban charging and battery swapping facilities, forecasting power grid load, and controlling traffic flow. For example, during peak hours or under extreme weather conditions, the system can dynamically adjust the charging pile scheduling strategy based on the evolution of user behavior, thereby improving overall service efficiency and power grid stability.

[0094] Furthermore, S4 includes:

[0095] S41 introduces a travel chain path accessibility analysis module to evaluate the feasibility and efficiency of the path from the user's current location to each candidate charging station.

[0096] Specifically, the step of "introducing a travel chain path accessibility analysis module to evaluate the feasibility and efficiency of the path from the user's current location to each candidate charging station" is one of the key steps in realizing the spatiotemporal quantitative modeling of user charging and swapping behavior in this invention. Its technical implementation principle is based on traffic network topology analysis and path planning algorithms, combined with real-time traffic flow, road restriction rules, and Geographic Information System (GIS) data, to dynamically evaluate the accessibility between the user's current location and candidate charging stations, thereby providing a basic input for subsequent probabilistic modeling of charging station selection.

[0097] In some implementations, this module first obtains the user's real-time location coordinates via GPS or an onboard positioning system, and combines this with geographic coordinate information from a charging station database to construct a spatial topology relationship between the user and each candidate charging station. Subsequently, Dijkstra's algorithm or A* algorithm is used for path planning to calculate the shortest or optimal path from the user's current location to each charging station. During path planning, factors such as road type (e.g., highway, main road, secondary road), restricted hours, traffic congestion index (e.g., real-time travel time based on floating car data or traffic camera data), and potential weather impacts (e.g., road closures due to heavy rain) must be comprehensively considered.

[0098] Furthermore, route accessibility assessment includes two core indicators: route feasibility and traffic efficiency. Route feasibility is represented by a Boolean variable; if there are impassable sections in the route (such as construction, traffic restrictions, or closures), the route is unreachable and is recorded as 0; otherwise, it is recorded as 1. Traffic efficiency is quantified by the ratio of route travel time (in minutes) to the historical average travel time, i.e., the traffic congestion coefficient, which ranges from [1, ∞). A higher value indicates lower traffic efficiency.

[0099] Optionally, the module can also introduce time window constraints, which determine whether a user can arrive at the charging station on time within a specific time period. For example, if a user plans to complete the trip within 30 minutes, only routes with a travel time of less than 30 minutes will be retained as feasible routes. In addition, energy accessibility can be determined by combining the vehicle's range and remaining battery charge (SOC) to ensure that the user can complete the journey without charging.

[0100] In practical applications, this step can be deployed in intelligent charging and swapping scheduling platforms or vehicle-to-everything (V2X) systems, enabling users to obtain optimal charging station recommendations in real time while on the move. The introduction of this module effectively improves the spatiotemporal accuracy of user behavior modeling, enhances the system's predictive ability for users' actual travel and charging behaviors, and provides reliable data support for subsequent multi-dimensional feature fusion and collaborative optimization.

[0101] S42. Construct a charging station selection model based on a utility function, where the utility function includes distance, electricity price, queuing time, traffic congestion index, and weighted coefficients of each influencing factor.

[0102] Specifically, this step constructs a charging station selection model based on a utility function. Its core lies in quantifying the multidimensional influencing factors users consider when choosing a charging station, including distance, electricity price, queuing time, and traffic congestion index, and introducing weighting coefficients for each factor to reflect their relative importance. This model employs a Multinomial Logit Model (MNL) from discrete choice theory, using a utility function to probabilistically model user selection behavior among multiple candidate charging stations, thereby achieving accurate prediction and optimized guidance of user behavior.

[0103] This model can be embedded in intelligent charging scheduling systems, user behavior prediction platforms, or urban transportation-energy collaborative management systems for real-time recommendation of optimal charging stations, prediction of charging demand distribution, and optimization of charging pile scheduling strategies. Especially during peak hours or under extreme weather conditions, the model can dynamically adjust weights, improving the adaptability and response efficiency of the recommendation system. By introducing multi-dimensional influencing factors and a weighting mechanism, the model significantly improves the modeling accuracy and explanatory power of user charging station selection behavior, providing crucial support for subsequent charging load forecasting, facility site optimization, and transportation-energy collaborative regulation.

[0104] This invention introduces spatiotemporal trajectory analysis technology, combining users' historical trajectories with real-time locations to dynamically identify potential charging demand points and their accessibility. This enhances the predictive model's sensitivity to temporal changes in user behavior and its adaptability to geographical distribution, achieving more accurate simulation of charging / battery swapping behavior.

[0105] Also includes:

[0106] S5, based on a cross-domain collaborative modeling framework, generates a spatiotemporal heat map of user charging and swapping behavior, which is used to predict the charging and swapping demand density and grid load change trends in various regions within a specific future time period.

[0107] Specifically, this step, based on a cross-domain collaborative modeling framework, generates a spatiotemporal heatmap of user charging and swapping behavior to predict the charging and swapping demand density and grid load trends in various regions over a specific future time period. Its technical implementation principle integrates multi-source heterogeneous data such as traffic trajectories, grid electricity prices, charging and swapping facility status, traffic congestion index, and meteorological environment. Through spatiotemporal modeling and behavior prediction algorithms, it achieves dynamic characterization of user behavior and load prediction.

[0108] In some implementations, this step first constructs a spatiotemporal behavioral feature vector based on users' historical travel and charging / swapping data, including user location coordinates, timestamps, SOC status, charging power, queuing time, and traffic congestion index. Then, a grid-based spatial clustering method (such as the H3 hexagonal grid system) is used to divide the city into several spatial units, aggregating user behavioral features within each unit to form a regional-level charging / swapping behavior heatmap. In the time dimension, a sliding time window (such as 15 minutes or 1 hour) is used to dynamically update user behavior, and combined with external factors such as time-of-use electricity pricing and weather changes, a time series prediction model (such as LSTM, ARIMA, or Prophet) is constructed.

[0109] At the parameter level, the generation of heatmaps relies on key indicators such as the SOC threshold (usually set at 20%), charging power (generally 7kW to 120kW), queuing time (in minutes), traffic congestion index (0-10 range, with 10 indicating severe congestion), and electricity price fluctuation coefficient (e.g., peak-valley electricity price ratio of 1.6:1:0.4). Furthermore, the probability of extreme weather events (such as heavy rain and typhoons) is predicted using meteorological models (such as random forests or XGBoost) and used as input variables for the behavioral response model.

[0110] In practical applications, this step can be deployed in city-level charging and battery swapping dispatch platforms or power grid load forecasting systems. By combining real-time traffic flow, power grid operation data, and weather forecasts, it enables dynamic forecasting of charging and battery swapping demand in various regions for the next day or the next few hours. In terms of technical effectiveness, this step significantly improves the spatiotemporal resolution and accuracy of user behavior prediction, providing data support for power grid load dispatching, charging and battery swapping station site optimization, and traffic resource coordination, thereby enhancing the coordinated control capabilities of urban energy and transportation systems.

[0111] This invention presents a preprocessing and fusion strategy adapted to different data structures and granularities across "vehicle-charging station-road-network-weather" systems. This enables comparability and interconnected analysis of data from different domains, providing a unified data foundation for the subsequent training and evaluation of user charging behavior-derived models.

[0112] This invention proposes a spatiotemporal quantification method for multidimensional characteristics of user charging and swapping behavior, which comprehensively considers cross-domain factors such as travel behavior, grid price levels, charging and swapping infrastructure service levels, road traffic network conditions, and meteorological environment. By fusing multi-source heterogeneous data, it achieves accurate characterization and dynamic modeling of user charging and swapping behavior in both time and space dimensions, overcoming the limitations of existing methods in depicting single-factor dependencies and coupling relationships. This invention provides data support and decision-making basis for the scientific layout of charging and swapping facilities, charging station load management, coordinated optimization of transportation and energy systems, and the design of vehicle-grid interaction mechanisms. It enhances the coordinated control capabilities of multiple systems including vehicles, charging piles, roads, networks, and meteorology, contributing to the deep integration and development of smart transportation and smart grids.

[0113] Example 2

[0114] This invention proposes a spatiotemporal quantization method for multidimensional features of user charging and swapping behavior that considers cross-domain factors. The calculation process is as follows: Figure 2 As shown in the figure, this method first constructs a multi-dimensional behavioral characteristic index system, systematically integrating heterogeneous data from multiple sources such as user vehicle status, charging facility service level, grid price level, traffic operation status, and meteorological factors. This comprehensively quantifies the temporal and spatial characteristics of user charging and swapping behavior, extracting key indicators such as battery capacity, queuing time, and traffic congestion coefficient. Based on this, a cross-domain collaborative modeling strategy is introduced, combining user travel status and environmental constraints to construct a charging selection behavior model, achieving dynamic simulation and detailed characterization of the spatiotemporal evolution trend of charging and swapping behavior. Through the above process, a systematic and scalable quantitative evaluation method for user behavior is formed, providing high-quality data support and decision-making basis for optimizing charging and swapping facility site selection, charging station load management, traffic resource scheduling, and vehicle-grid collaborative control, promoting the deep integration and intelligent collaborative development of transportation and energy.

[0115] Preferably, the spatiotemporal quantification method for multidimensional characteristics of user charging and swapping behavior that considers cross-domain factors should first establish a quantitative characterization system for multidimensional influencing factors of "vehicle-charging station-road-grid-meteorology" to systematically evaluate the key factors affecting user charging and swapping behavior. This evaluation system mainly covers the following dimensions: user vehicle status, charging facility service level, traffic operation status, grid electricity price level, and meteorological factors, in order to comprehensively characterize the external driving mechanism of user decision-making behavior.

[0116] Preferably, regarding the user's vehicle status, the impact on their charging / swapping selection behavior is mainly reflected in the remaining battery state and its related characteristics. Therefore, this can be achieved by constructing a system that includes battery capacity (E...). ba ), Current remaining battery power (SOC) ini Acceptable minimum charge threshold (SOC) minThe system includes an indicator framework that quantifies the impact of vehicle status on users' charging and battery swapping decisions, including parameters such as whether battery swapping is supported (1 for no support and 2 for support). These parameters not only reflect the urgency of users' charging needs under different remaining battery levels but also provide fundamental data support for subsequent charging behavior modeling.

[0117] Preferably, regarding the service level of charging facilities, to accurately assess its impact on users' charging and battery swapping decisions, quantitative modeling is needed from two dimensions: facility operating efficiency and service experience. Specifically, the rated charging power P of the charging pile can be selected. char Key parameters such as charging efficiency per unit time, queue length (representing the number of vehicles currently waiting to charge), and queue time (measuring the average waiting cost for users at charging stations) are used as core evaluation indicators. These indicators can comprehensively reflect users' sensitivity to charging speed and service accessibility when choosing charging facilities, providing effective support for subsequent behavior modeling and system optimization.

[0118] In system performance analysis based on queuing theory, the system idle probability P0 represents the probability that the system is in an idle state (i.e., there are no tasks in the system); the average queue length L... q The average number of tasks waiting for service in the system; the average queuing time W. q This represents the average time a task waits in the queue. The formulas for calculating the above parameters are as follows:

[0119] The system idle probability P0 satisfies the following relationship.

[0120]

[0121] In the formula: λ: arrival rate, the number of vehicles arriving at the charging station per unit time (vehicles / hour), which follows a Poisson distribution. During holidays, the rate needs to be multiplied by a coefficient k = 2-3 for highway service areas, and is 1 during normal times.

[0122] μ: Service rate, the number of vehicles served by a single charging pile per unit time (vehicles / hour), and the service time follows an exponential distribution.

[0123] c: Number of service counters, which represents the total number of charging stations.

[0124] ρ: Service counter utilization rate The system must satisfy ρ < 1 to ensure stability.

[0125] P0: System idle probability, representing the initial state probability that all charging stations are unoccupied.

[0126] Average queue length L q It satisfies the following relationship.

[0127]

[0128] In the formula: L q The average number of vehicles waiting in the queue at a charging station is strongly correlated with the arrival rate and the number of service counters.

[0129] Average queuing time W q It satisfies the following relationship.

[0130]

[0131] In the formula: W q It reflects the average waiting time in the charging station's queue and is related to arrival efficiency and the number of service counters.

[0132] Preferably, regarding traffic conditions, to comprehensively assess the impact between road traffic conditions and user charging / swapping behavior, key parameters reflecting road traffic efficiency should be introduced. The traffic congestion coefficient, as an important indicator of road operating conditions, can effectively characterize the saturation level of road traffic in different time periods and areas. Dynamic monitoring and analysis of the traffic congestion coefficient can help identify potential obstacles users may face during their travels, thereby more accurately depicting their choice behavior in charging / swapping decisions.

[0133]

[0134] In the formula: t h During peak hours, the time when roads near charging stations are open to traffic;

[0135] t p Off-peak hours: Traffic hours on roads near charging stations;

[0136] Preferably, regarding grid electricity prices, to quantify the impact of the pricing mechanism on users' charging and battery swapping behavior, key parameters reflecting changes in charging behavior and charging costs need to be introduced. Time-of-use pricing, as an important means of reflecting the relationship between electricity supply and demand and dispatch strategies, directly affects users' charging costs and behavioral preferences due to price fluctuations at different times. Quantitative analysis of the sensitivity of time-of-use pricing to users' charging and battery swapping behavior can reveal the economic driving factors behind users' choice of charging times at different times.

[0137] The user sensitivity index satisfies the following relationship.

[0138]

[0139] In the formula: Δp: represents the percentage change in electricity price, Δp=p1×(p2-p1), p1 represents the charging price in the first period, and p2 represents the charging price in the second period.

[0140] ΔE: Represents the percentage change in charging amount, ΔE=E1×(E2-E1), where E1 represents the charging price in the first period and E2 represents the charging price in the second period.

[0141] Preferably, regarding meteorological elements, to comprehensively assess the impact of environmental conditions on users' charging and battery swapping behavior, key indicators such as temperature, rainfall, and the probability of extreme weather events should be introduced. A quantitative model of the probability of extreme weather types is constructed using indicators such as temperature and humidity. The probability of extreme weather events (such as typhoons, high temperatures, and heavy rain) determines, to a certain extent, the avoidance characteristics of users' travel behavior and the urgency of their charging and battery swapping behavior. The introduction of these meteorological parameters helps to build a more realistic and adaptable behavioral response model, improving the accuracy and robustness of user behavior modeling. Therefore, this invention quantifies the probability of weather occurrence.

[0142]

[0143] In the formula: y represents the extreme weather type, 0 = normal weather, 1 = rainstorm, 2 = typhoon;

[0144] K: Number of days in the weather observation data sample size;

[0145] X = [X1, X2, ..., X p ] T These are independent variables (eigenvectors), including temperature, humidity, etc.

[0146] β i =[β i0 ,β i1 ,…,β ip ] T It is the coefficient vector of the i-th type, representing variable X in the i-th type of weather. i The intensity of the impact.

[0147] Preferably, to achieve accurate modeling and dynamic evaluation of user charging and swapping behavior, this invention constructs a multi-dimensional feature-driven user behavior representation method based on five key dimensions: vehicle, charging station, road, network, and weather. This method fully considers factors such as electric vehicle technical parameters, charging infrastructure service levels, road traffic conditions, grid electricity prices, and external meteorological conditions, as shown in Table 1.

[0148] Table 1. Factors affecting electric vehicle user charging

[0149]

[0150]

[0151] Through a systematic design of indicators, a comprehensive characterization of the mechanisms influencing user behavior is achieved. Specifically, based on vehicle status, charging service level, grid electricity price level, and weather conditions, a multinomial logistic regression model is used to quantify the probability of a user choosing a charging / battery swapping time period T, as shown in the following formula.

[0152]

[0153] In the formula: k soc Indicates whether the remaining SOC is lower than the SOC threshold. min The value of P may be insufficient to meet driving requirements; if it does, the value is 1, otherwise it is 0. t K represents the charging price during time period t, including electricity and service fees; w K represents a weather coefficient; the higher the probability of extreme weather, the lower the value. w =1-p(y=w ext ); β kj Represents the regression coefficient, indicating the relationship between the independent variable and β. kj The degree of influence, β k0 This represents the intercept term.

[0154] Expected charging time:

[0155]

[0156] Where: SOC end : Indicates the expected SOC upon completion of charging; SOC end : Indicates the initial state of charge (SOC); E ba : Indicates capacitance; P char : Indicates charging power.

[0157] Expected battery swapping time:

[0158] T swap =t char

[0159] In the formula: T swap This indicates the actual time required for battery swapping.

[0160] Preferably, the probability of choosing a charging station needs to be combined with the user's travel chain, traffic congestion, and queuing for charging and battery swapping facilities to construct a probability model for the user's choice of charging location.

[0161] The probability that a user chooses charging station i:

[0162]

[0163] In the formula: U i : Represents the utility function of charging station i Charging station i's attractiveness model to users: A i =k1 / di +k2 / p t +k3 / t q +k4 / TTI;

[0164] In the formula: d i : Distance from the user to the charging station; p t The charging price during time period t; t q : Indicates queue time; TTI: Traffic Congestion Index; k i Impact coefficient.

[0165] Preferably, by comprehensively considering the user's immediate charging needs and the selection preferences of each candidate charging station, the probability of the user choosing the i-th charging station can be further determined. This probability, as one of the core parameters for measuring user charging behavior, provides a basis for subsequent charging behavior modeling and station power load management. Its calculation method is shown below.

[0166] P = P(T = k) × P(A = i)

[0167] In the formula: P(T=k): the probability that the user chooses to charge or swap batteries within the time period T; P(A=i): the probability that the user charges at the i-th charging station.

[0168] Preferably, to more accurately quantify the key factors influencing users' charging and swapping behavior, a nonlinear least squares method is used to fit the model parameters based on a probability model of users choosing to charge or swap at different time periods T and users' historical charging behavior data. By calibrating the model parameters, user charging characteristics can be quantified through parameter values. This not only characterizes users' preferences in the time dimension but also reflects their response characteristics to factors such as electricity prices and convenience, thus providing a quantitative basis for user behavior modeling and strategy optimization.

[0169]

[0170] In the formula: x i : Represents the independent variable in the model, x i = [x1, x2… x n ];y i b: represents the dependent variable in the model; β: represents the parameter that needs to be calibrated in the model. i =[β k0 ,β k1 …β k4 ];k i = [k1,k2…k4].

[0171] In summary, this invention is the first to systematically introduce multi-source core elements of "vehicle-charging station-road-network-meteorology," integrating heterogeneous data from multiple sources such as user travel trajectories, charging infrastructure, grid electricity prices, traffic congestion, and meteorological conditions. By constructing a multi-dimensional behavioral characteristic index system and a cross-domain collaborative modeling framework, it achieves spatiotemporal dynamic simulation and quantitative analysis of user charging and swapping behavior. Compared with existing research, this method not only improves the quantitative accuracy of user charging behavior characteristics but also provides a unified modeling foundation and decision support for practical applications such as charging facility layout optimization, charging station load control, traffic resource coordination, and extreme weather response. It significantly promotes the evolution of intelligent sensing and collaborative control technologies in the context of deep integration of user charging behavior with multiple systems such as the power grid and transportation.

[0172] Example 3

[0173] To achieve the above embodiments, such as Figure 3 As shown, this embodiment also provides a spatiotemporal quantization device 10 for multidimensional features of user charging and swapping behavior that considers cross-domain factors, including:

[0174] The multi-dimensional feature construction module 100 is used to construct a multi-dimensional behavioral feature index system that includes vehicle status, charging facility service level, traffic operation status, power grid price level and meteorological factors, and quantitatively characterize the impact of each dimension on users' charging and swapping behavior.

[0175] The data fusion and spatiotemporal alignment module 200 is used to perform standardized processing and spatiotemporal alignment of multi-source heterogeneous data based on the multi-dimensional behavioral feature index system, which integrates user travel trajectory, charging pile operation data, traffic congestion index, time-of-use electricity price information and meteorological observation data.

[0176] The behavioral probability modeling module 300 is used to calibrate and model the probability of users choosing to charge or swap batteries and the probability of choosing charging stations at different times using a multinomial logistic regression model and nonlinear least squares method, so as to realize dynamic prediction of user behavior.

[0177] The cross-domain collaborative modeling module 400 is used to combine historical user behavior data with real-time environmental data to build a cross-domain collaborative modeling framework to simulate the evolution trend of user charging and swapping behavior in time and space.

[0178] Furthermore, the multidimensional feature building module is also used for:

[0179] Vehicle status parameters are defined, including battery capacity, current remaining charge, minimum acceptable charge threshold, and whether battery swapping is supported, in order to quantify the urgency of recharging for users in different charge states.

[0180] The service level parameters for charging facilities include the rated charging power of the charging pile, queue length, and queue time, to reflect users' sensitivity to charging efficiency and service accessibility when choosing charging facilities.

[0181] Furthermore, the data fusion and spatiotemporal alignment module is also used for:

[0182] A timestamp alignment algorithm is used to match user travel trajectory data with grid time-of-use electricity price data in the time dimension, ensuring the synchronization of data from different time domains in behavioral modeling;

[0183] The spatial interpolation method is used to standardize the spatial dimension of traffic congestion index and meteorological observation data to unify the expression of data from different geographical locations.

[0184] Furthermore, the behavioral probability modeling module is also used for:

[0185] The nonlinear least squares method is used to fit the regression coefficients and intercept term in the multinomial logistic regression model to improve the model's accuracy in characterizing user behavior response features.

[0186] Based on users' historical charging and swapping behavior data, along with real-time electricity prices, traffic congestion, and queuing times, the model parameters are dynamically updated to adapt to the evolving trends in user behavior.

[0187] The spatiotemporal quantification device for multidimensional features of user charging and swapping behavior, which considers cross-domain factors, in this invention can systematically integrate multi-source heterogeneous data from "vehicle-charging pile-road-network-meteorology" to achieve accurate quantification and dynamic modeling of user charging and swapping behavior in the spatiotemporal dimension, thereby improving the accuracy of behavior prediction and the ability of multi-system collaborative optimization.

[0188] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0189] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A spatiotemporal quantification method for multidimensional features of user charging and swapping behavior considering cross-domain factors, characterized in that, include: S1. Construct a multi-dimensional behavioral characteristic indicator system that includes vehicle status, charging facility service level, traffic operation status, grid electricity price level and meteorological factors to quantitatively characterize the impact of each dimension on users' charging and swapping behavior. S2, based on the multi-dimensional behavioral feature index system, integrate user travel trajectory, charging pile operation data, traffic congestion index, time-of-use electricity price information and meteorological observation data to perform standardized processing and spatiotemporal alignment of multi-source heterogeneous data; S3 utilizes a multinomial logistic regression model and nonlinear least squares method to calibrate and model the probability of users choosing to charge or swap batteries and the probability of choosing charging stations at different times, thereby enabling dynamic prediction of user behavior. S4 combines historical user behavior data with real-time environmental data to build a cross-domain collaborative modeling framework to simulate the evolution trend of user charging and swapping behavior in time and space. in, The S2 includes: S21, using a timestamp alignment algorithm to match user travel trajectory data with grid time-of-use electricity price data in the time dimension to ensure the synchronization of data from different time domains in behavior modeling; S22, using spatial interpolation methods to standardize traffic congestion index and meteorological observation data in the spatial dimension to unify the expression of data from different geographical locations. S3 includes: S31, using nonlinear least squares method to fit the regression coefficients and intercept terms in the multinomial logistic regression model to improve the model's accuracy in characterizing user behavior response features; S32, dynamically updating model parameters based on users' historical charging and swapping behavior data and real-time electricity prices, traffic congestion, and queuing time to adapt to the evolution trend of user behavior.

2. The method as described in claim 1, characterized in that, S1 includes: S11 defines vehicle status parameters including battery capacity, current remaining charge, acceptable minimum charge threshold, and whether battery swapping is supported, in order to quantify the urgency of recharging for users in different charge states. S12, construct service level parameters for charging facilities, including the rated charging power of charging piles, queue length, and queue time, to reflect users' sensitivity to charging efficiency and service accessibility when selecting charging facilities.

3. The method as described in claim 1, characterized in that, The S4 includes: S41 introduces a travel chain path accessibility analysis module to evaluate the feasibility and efficiency of the path from the user's current location to each candidate charging station. S42. Construct a charging station selection model based on a utility function, where the utility function includes distance, electricity price, queuing time, traffic congestion index, and weighted coefficients of each influencing factor.

4. The method as described in claim 1, characterized in that, Also includes: S5, based on a cross-domain collaborative modeling framework, generates a spatiotemporal heat map of user charging and swapping behavior, which is used to predict the charging and swapping demand density and grid load change trends in various regions within a specific future time period.

5. A spatiotemporal quantization device for multidimensional features of user charging and swapping behavior considering cross-domain factors, characterized in that, include: The multi-dimensional feature construction module is used to build a multi-dimensional behavioral feature index system that includes vehicle status, charging facility service level, traffic operation status, grid electricity price level and meteorological factors, and quantitatively characterize the impact of each dimension on users' charging and swapping behavior. The data fusion and spatiotemporal alignment module is used to perform standardized processing and spatiotemporal alignment of multi-source heterogeneous data based on the multi-dimensional behavioral feature index system, which integrates user travel trajectory, charging pile operation data, traffic congestion index, time-of-use electricity price information and meteorological observation data. The behavioral probability modeling module is used to calibrate and model the probability of users choosing to charge or swap batteries and the probability of choosing charging stations at different times using a multinomial logistic regression model and a nonlinear least squares method, so as to achieve dynamic prediction of user behavior. The cross-domain collaborative modeling module is used to combine historical user behavior data with real-time environmental data to build a cross-domain collaborative modeling framework to simulate the evolution trend of user charging and swapping behavior in time and space. in, The data fusion and spatiotemporal alignment module is also used to: match user travel trajectory data with power grid time-of-use electricity price data in the time dimension using a timestamp alignment algorithm to ensure the synchronization of data from different time domains in behavioral modeling; and standardize traffic congestion index and meteorological observation data in the spatial dimension through spatial interpolation methods to unify the expression of data from different geographical locations. The behavior probability modeling module is also used to: fit the regression coefficients and intercept terms in the multinomial logistic regression model using the nonlinear least squares method, so as to improve the model's accuracy in characterizing user behavior response features. Based on users' historical charging and swapping behavior data, along with real-time electricity prices, traffic congestion, and queuing times, the model parameters are dynamically updated to adapt to the evolving trends in user behavior.

6. The apparatus as claimed in claim 5, characterized in that, The multidimensional feature construction module is also used for: Vehicle status parameters are defined, including battery capacity, current remaining charge, minimum acceptable charge threshold, and whether battery swapping is supported, in order to quantify the urgency of recharging for users in different charge states. The service level parameters for charging facilities include the rated charging power of the charging pile, queue length, and queue time, to reflect users' sensitivity to charging efficiency and service accessibility when choosing charging facilities.

Images