Charging station planning method and device fusing vehicle behaviors and power grid operation data

Abstract

The invention relates to the technical field of charging planning, and particularly provides a charging station planning method and device fusing vehicle behavior and power grid operation data, and the method comprises the steps: obtaining vehicle behavior and power grid operation fusion data of a fragmented region; taking the vehicle behavior and power grid operation fusion data of the fragmented region as the input of a pre-trained XGBoost regression model, and obtaining the charging amount of the fragmented region output by the pre-trained XGBoost regression model; and configuring charging stations with corresponding capacities in the fragmented areas according to the charging amounts of the fragmented areas. According to the technical scheme provided by the invention, multi-source data are integrated, the XGBoost regression model is adopted to fuse peripheral spatial features, and a high-precision charging demand prediction system is constructed, so that the prediction accuracy, the system-level optimization capability and the commercial feasibility of the planning model are improved, and finally scientific decision making and efficient landing of charging station planning are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of charging planning, in particular to a charging station planning method and device fusing vehicle behavior and power grid operation data. BACKGROUND Charging pile intelligent planning is a complex optimization problem involving multiple factors such as geographic information, user demand, traffic flow and power grid load, and multiple objectives. The existing technology mainly includes four similar schemes: 1) a method based on user demand and behavior, which quantifies charging demand by analyzing user travel data (such as end-of-trip time) and travel trajectory; 2) a strategy based on geographic and traffic information, which prioritizes densely populated areas (such as business centers and office areas) and is close to major roads to improve accessibility; 3) a multi-objective optimization algorithm (such as genetic algorithm, particle swarm optimization), which solves the optimal planning through cost model and demand coverage model; 4) a method based on multi-factor comprehensive evaluation, which integrates POI data, clustering algorithm and fuzzy scoring to divide demand levels.

[0002] However, the data sources (such as vehicle supervision platform, Gaode map POI) relied on by the existing technology have key missing (such as power grid capacity, real-time status of competing stations, dynamic information of road network and vehicle activity trajectory), so the data coverage is not complete and the quality is unreliable. This leads to the prediction model being unable to accurately quantify charging demand (such as potential passenger and casual passenger demand), the measurement being weak, and thus the deviation of the planning prediction being increased. At the same time, because the existing method relies too much on data and algorithm, and does not sufficiently quantify geographic conditions (such as land property conflicts, functional area distribution) and human factors (such as user charging behavior preferences). This leads to the planning decision being unable to adapt to the actual scene (such as high demand in business areas and low demand in industrial areas), resulting in the intelligent site selection theory possibly relying too much on data and algorithm, ignoring the influence of geographic and human factors, and thus reducing the scene adaptability of the model and user satisfaction. SUMMARY

[0003] In order to overcome the above defects, the present application provides a charging station planning method and device fusing vehicle behavior and power grid operation data.

[0004] In a first aspect, a charging station planning method fusing vehicle behavior and power grid operation data is provided, which comprises: obtaining vehicle behavior and power grid operation fusion data of a fragmented area; taking the vehicle behavior and power grid operation fusion data of the fragmented area as input of a pre-trained XGBoost regression model to obtain charging capacity of the fragmented area output by the pre-trained XGBoost regression model; configuring a charging station with corresponding capacity in the fragmented area according to the charging capacity of the fragmented area.

[0005] Preferably, the division of the target area is obtained by the Uber h3 algorithm.

[0006] Preferably, the vehicle behavior and power grid operation fusion data includes at least one of the following: slice area id, number of potential passengers in the slice area, number of potential passengers in the surrounding slice area, number of scattered passengers in the slice area, number of scattered passengers in the surrounding slice area, demand power of potential passengers in the slice area, demand power of potential passengers in the surrounding slice area, demand power of scattered passengers in the slice area, demand power of scattered passengers in the surrounding slice area, number of existing direct current piles in the slice area, number of existing direct current piles in the surrounding slice area, number of existing alternating current piles in the slice area, number of existing alternating current piles in the surrounding slice area, number of large shopping malls in the slice area, number of large shopping malls in the surrounding slice area, number of residential buildings in the slice area, number of residential buildings in the surrounding slice area, number of parking lots in the slice area, number of parking lots in the surrounding slice area.

[0007] Further, the potential passenger in the slice area is a user who has a non-charging stay record in the slice area for not less than 15 days, with an average stay of more than 30 minutes, and without a private charging pile in the slice area. The scattered passenger in the slice area is a user who has a non-charging stay record in the slice area for more than 1 day and less than 15 days, with an average stay of more than 30 minutes, and without a private charging pile in the slice area.

[0008] Further, the surrounding slice area is six adjacent hexagonal slice areas of the division area.

[0009] Preferably, the training process of the pre-trained XGBoost regression model includes: constructing training data using historical vehicle behavior and power grid operation fusion data of the division area and corresponding charging data thereof; training an initial XGBoost regression model using the training data to obtain the pre-trained XGBoost regression model.

[0010] Further, the evaluation index in the training process of the pre-trained XGBoost regression model includes at least one of the following: root mean square error index, determination coefficient index.

[0011] In a second aspect, a charging station planning device that fuses vehicle behavior and power grid operation data is provided, and the charging station planning device that fuses vehicle behavior and power grid operation data includes: an acquisition module configured to acquire vehicle behavior and power grid operation fusion data of a division area; an analysis module configured to take the vehicle behavior and power grid operation fusion data of the division area as an input of a pre-trained XGBoost regression model, and obtain charging capacity of the division area output by the pre-trained XGBoost regression model; A planning module is configured to configure charging stations with corresponding capacities in the sub-regions according to the charging amounts of the sub-regions.

[0012] In a third aspect, a computer device is provided, comprising: one or more processors; The processor is configured to execute one or more programs. When the one or more programs are executed by the one or more processors, the charging station planning method that integrates vehicle behavior and power grid operation data is implemented.

[0013] In a fourth aspect, a computer readable storage medium is provided, which has a computer program stored thereon, and the computer program, when executed, implements the charging station planning method that integrates vehicle behavior and power grid operation data.

[0014] The one or more technical solutions of the present application have at least one or more of the following beneficial effects: The present application relates to the technical field of charging planning, and specifically provides a charging station planning method and device that integrates vehicle behavior and power grid operation data, comprising: obtaining vehicle behavior and power grid operation integrated data of a sub-region; taking the vehicle behavior and power grid operation integrated data of the sub-region as input of a pre-trained XGBoost regression model to obtain a charging amount of the sub-region output by the pre-trained XGBoost regression model; and configuring charging stations with corresponding capacities in the sub-region according to the charging amount of the sub-region. The technical solution provided by the present application integrates multiple sources of data, uses an XGBoost regression model to integrate surrounding spatial features, and constructs a high-precision charging demand prediction system, thereby improving the prediction accuracy of the planning model, the system-level optimization capability, and the commercial feasibility, and ultimately achieving scientific decision-making and efficient landing of the charging station planning. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 FIG. 1 is a main step flow diagram of the charging station planning method that integrates vehicle behavior and power grid operation data according to an embodiment of the present application. DETAILED DESCRIPTION

[0016] The specific embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0018] Embodiment 1 Referring to the drawings Figure 1 , Figure 1 is the main step flow diagram of the charging station planning method of fusing vehicle behavior and power grid operation data according to an embodiment of the present application. As shown in Figure 1 , the charging station planning method of fusing vehicle behavior and power grid operation data according to an embodiment of the present application mainly includes the following steps: Step S101: Obtain vehicle behavior and power grid operation fusion data of a fragmented area; Step S102: Take the vehicle behavior and power grid operation fusion data of the fragmented area as the input of a pre-trained XGBoost regression model to obtain the charging capacity of the fragmented area output by the pre-trained XGBoost regression model; Step S103: Configure a charging station with a corresponding capacity in the fragmented area according to the charging capacity of the fragmented area.

[0019] In this embodiment, the fragmented area is obtained by dividing the target area through the Uber h3 algorithm. Algorithm advantage: convert the GPS latitude and longitude into a geographic index (h3 index) of a regular hexagon with a side length of about 460 meters. Through the h3 index, the proximity relationship of the "people-vehicle-pole-grid" coordinated interaction data system of the location and the location can be well reflected. At the same time, the regular hexagon is the closest seamless regular polygon to a circle, which can well reflect the service range after the station is built. Peripheral area: 6 hexagons adjacent to each hexagon. Considering that the charging station service not only covers the current one area, but also radiates to the surrounding area, the peripheral area characteristics are aggregated into the station building selection consideration factor. This function module will be developed for various application scenarios.

[0020] In this embodiment, the vehicle behavior and power grid operation fusion data includes at least one of the following: area id, number of potential passengers in the area, number of potential passengers in the surrounding area, number of scattered passengers in the area, number of scattered passengers in the surrounding area, demand power of potential passengers in the area, demand power of potential passengers in the surrounding area, demand power of scattered passengers in the area, demand power of scattered passengers in the surrounding area, number of existing direct current poles in the area, number of existing direct current poles in the surrounding area, number of existing alternating current poles in the area, number of existing alternating current poles in the surrounding area, number of large shopping malls in the area, number of large shopping malls in the surrounding area, number of residential buildings in the area, number of residential buildings in the surrounding area, number of parking lots in the area, number of parking lots in the surrounding area.

[0021] In one embodiment, the potential passengers in the area are users whose non-charging stay records in the area are not less than 15 days, the average stay is more than 30 minutes, and the vehicle has no private charging pile in the area. The scattered passengers in the area are users whose non-charging stay records in the area are more than 1 day and less than 15 days, the average stay is more than 30 minutes, and the vehicle has no private charging pile in the area.

[0022] In one embodiment, the peripheral area is 6 hexagonal sub-regions adjacent to the sub-region.

[0023] In this embodiment, the training process of the pre-trained XGBoost regression model includes: The training data is constructed by using the historical vehicle behavior and power grid operation fusion data of the sub-region and the corresponding charging amount data; The initial XGBoost regression model is trained using the training data to obtain the pre-trained XGBoost regression model.

[0024] In one embodiment, the evaluation index in the training process of the pre-trained XGBoost regression model includes at least one of the following: root mean square error index, determination coefficient index.

[0025] The root mean square error index is the square root of the square of the deviation between the predicted value and the true value and the number of observations n. When RMSE=0, it means that the predicted value is completely consistent with the actual value without error. The theoretical value of the RMSE of the model is 0-1, and the actual RMSE value is 0.07, i.e. the error between the station site selection scoring result and the true value is about 7%, and the fitting effect is good.

[0026] The determination coefficient index reflects the proportion of the total variation of the dependent variable that can be explained by the regression relationship of the independent variable. When R2_score=1, the predicted value and the true value in the sample are completely equal, without any error, indicating that the better the independent variable explains the dependent variable in the regression analysis. The R2_score of the model is 0.87, the correlation between the model features and the prediction results is very strong, and the interpretability is good.

[0027] The commonly used development languages for XGBoost regression model building, training and debugging are Python, R language, C++ and the like, and the most commonly used platform development language is Java, so the online deployment of the algorithm model needs a "bridge" to realize cross-language calling.

[0028] The predictive model markup language PMML (Predictive Model Markup Language) is a model representation language independent of platform and environment, and is the current standard for representing machine learning models. The trained machine learning can be applied and deployed across platforms by exporting PMML. PMML is a standardized language that contains information needed for model application, including feature name, feature type, model name, trained parameters, etc.

[0029] After the model training is completed through the machine learning algorithm, use the tool or library that supports PMML export to convert the model into a PMML file, which contains all the necessary information of the model and can be read by different platforms and languages. Deploy the generated PMML file to a server or cloud environment, ensuring that the Java program can access it. Develop a Java program using the pmml4j Java library that supports PMML to load the PMML file. In the Java program, use the above-mentioned library to load the PMML file to obtain the internal representation of the model. Then implement a prediction interface that accepts input data, uses the loaded model to make predictions, and returns the prediction results. According to the needs, optimize the performance of the Java program to ensure the response time and accuracy of online prediction. Integrate the Java program with existing systems or services and conduct comprehensive testing to ensure its stability and reliability. Deploy the integrated and tested Java program to the production environment and set up monitoring mechanisms to track the performance of the model and the health status of the system.

[0030] Embodiment 2 Based on the same inventive concept, the application also provides a charging station planning device that fuses vehicle behavior and power grid operation data, comprising: An acquisition module is configured to acquire vehicle behavior and power grid operation fusion data of a fragmented area. An analysis module is configured to take the vehicle behavior and power grid operation fusion data of the fragmented area as an input of a pre-trained XGBoost regression model to obtain a charging capacity of the fragmented area output by the pre-trained XGBoost regression model. A planning module is configured to configure a charging station with a corresponding capacity in the fragmented area according to the charging capacity of the fragmented area.

[0031] Preferably, the fragmented area is obtained by dividing the target area through the Uber h3 algorithm.

[0032] Preferably, the vehicle behavior and power grid operation fusion data includes at least one of the following: a slice area id, a number of potential passengers in the slice area, a number of potential passengers in a surrounding slice area, a number of scattered passengers in the slice area, a number of scattered passengers in the surrounding slice area, a demand power of potential passengers in the slice area, a demand power of potential passengers in the surrounding slice area, a demand power of scattered passengers in the slice area, a demand power of scattered passengers in the surrounding slice area, a number of existing direct current piles in the slice area, a number of existing direct current piles in the surrounding slice area, a number of existing alternating current piles in the slice area, a number of existing alternating current piles in the surrounding slice area, a number of large-scale shopping malls in the slice area, a number of large-scale shopping malls in the surrounding slice area, a number of residential buildings in the slice area, a number of residential buildings in the surrounding slice area, a number of parking lots in the slice area, and a number of parking lots in the surrounding slice area.

[0033] Further, the potential customer in the slice is a user whose non-charging stay record in the slice is not less than 15 days, the average stay is more than 30 minutes, and the vehicle has no private charging pile in the slice; and the scattered customer in the slice is a user whose non-charging stay record in the slice is more than 1 day and less than 15 days, the average stay is more than 30 minutes, and the vehicle has no private charging pile in the slice.

[0034] Further, the peripheral slice is six hexagonal sub-regions adjacent to the sub-region.

[0035] Preferably, the training process of the pre-trained XGBoost regression model comprises: The training data is constructed by using the historical vehicle behavior and power grid operation fusion data of the sub-region and the corresponding charging amount data; The initial XGBoost regression model is trained by using the training data, and the pre-trained XGBoost regression model is obtained.

[0036] Further, the evaluation index in the training process of the pre-trained XGBoost regression model comprises at least one of the following: root mean square error index, determination coefficient index.

[0037] Embodiment 3 Based on the same inventive concept, the application further provides a computer device, which comprises a processor and a memory, the memory is used to store a computer program, the computer program comprises program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method process or corresponding function, so as to implement the steps of the charging station planning method based on the fusion of vehicle behavior and power grid operation data in the above-mentioned embodiments.

[0038] Embodiment 4 Based on the same inventive concept, the application further provides a storage medium, specifically a computer readable storage medium (Memory), which is a memory device in a computer device, used for storing programs and data. It can be understood that the computer readable storage medium herein can include an internal storage medium in the computer device, and of course can also include an extended storage medium supported by the computer device. The computer readable storage medium provides a storage space, which stores an operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that the computer readable storage medium herein can be a high-speed RAM memory, or a non-volatile memory such as at least one disk memory. One or more instructions stored in the computer readable storage medium can be loaded and executed by the processor to implement the steps of the charging station planning method of fusing vehicle behavior and power grid operation data in the above embodiment.

[0039] Those skilled in the art will appreciate that embodiments of the application can be supplied as a method, a system, or a computer program product. Therefore, the application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer usable program code.

[0040] The application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks. Figure 1 The device that implements the functions specified in one or more flows and / or blocks.

[0041] These computer program instructions can also be stored in a computer readable memory that can guide the computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer readable memory produce a product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks.Figure 1 the function specified in the one or more blocks.

[0042] These computer program instructions can also be loaded into computer or other programmable data processing devices, so that a series of operation steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, thus the instructions executed on the computer or other programmable data processing devices provide a process for implementing the flow Figure 1 the flow or flows and / or blocks Figure 1 the steps of the function specified in the one or more blocks.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application rather than limit the same, and although the present application has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered within the protection scope of the claims of the present application.

Claims

1. A charging station planning method that integrates vehicle behavior and power grid operation data, characterized in that, The method includes: Acquire fused data on vehicle behavior and power grid operation in segmented regions; The vehicle behavior and grid operation data of the segmented regions are fused as input to the pre-trained XGBoost regression model to obtain the charging amount of the segmented regions output by the pre-trained XGBoost regression model. According to the charging volume of the segmented area, a charging station with a corresponding capacity is configured in the segmented area.

2. The method as described in claim 1, characterized in that, The segmented regions are obtained by dividing the target region using the Uber h3 algorithm.

3. The method as described in claim 1, characterized in that, The vehicle behavior and power grid operation fusion data includes at least one of the following: area ID, number of potential customers in the area, number of potential customers in surrounding areas, number of individual customers in the area, number of individual customers in surrounding areas, electricity demand of potential customers in the area, electricity demand of potential customers in surrounding areas, electricity demand of individual customers in the area, electricity demand of individual customers in surrounding areas, number of existing DC charging piles in the area, number of existing DC charging piles in surrounding areas, number of existing AC charging piles in the area, number of existing AC charging piles in surrounding areas, number of large shopping malls in the area, number of large shopping malls in surrounding areas, number of residential buildings in the area, number of residential buildings in surrounding areas, number of parking lots in the area, and number of parking lots in surrounding areas.

4. The method as described in claim 3, characterized in that, The term "potential customers" refers to users whose non-charging stay records within the area are no less than 15 days, with an average stay of more than 30 minutes each time, and whose vehicles do not have private charging stations in the area. The term "sporadic customers" refers to users whose non-charging stay records within the area are greater than 1 day but less than 15 days, with an average stay of more than 30 minutes each time, and whose vehicles do not have private charging stations in the area.

5. The method as described in claim 3, characterized in that, The surrounding area consists of six hexagonal subdivisions adjacent to the subdivision area.

6. The method as described in claim 1, characterized in that, The training process of the pre-trained XGBoost regression model includes: Training data is constructed using historical vehicle behavior and power grid operation data from segmented regions and their corresponding charging data. The initial XGBoost regression model is trained using the training data to obtain the pre-trained XGBoost regression model.

7. The method as described in claim 6, characterized in that, The evaluation metrics during the training process of the pre-trained XGBoost regression model include at least one of the following: root mean square error metric and coefficient of determination metric.

8. An apparatus for a charging station planning method based on the fusion of vehicle behavior and power grid operation data as described in any one of claims 1-7, characterized in that, The device includes: The acquisition module is used to acquire fused data on vehicle behavior and power grid operation in segmented regions; The analysis module is used to take the vehicle behavior and power grid operation fusion data of the segmented area as input to the pre-trained XGBoost regression model and obtain the charging amount of the segmented area output by the pre-trained XGBoost regression model. The planning module is used to configure charging stations of corresponding capacity within the segmented areas according to the charging volume of the segmented areas.

9. A computer device, characterized in that, include: One or more processors; The processor is used to execute one or more programs; When the one or more programs are executed by the one or more processors, the charging station planning method that integrates vehicle behavior and power grid operation data as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that, It contains a computer program that, when executed, implements the charging station planning method that integrates vehicle behavior and power grid operation data as described in any one of claims 1 to 7.

Images