In-station control system and method for battery swap station in electric power service area

Abstract

The invention relates to the technical field of in-station control of a battery swap station, in particular to an in-station control system and method for a battery swap station in an electric power service area, and solves the technical problem that effective prediction and scheduling are difficult to carry out when multiple devices of the battery swap station in the service area are accessed in the prior art. The regional charging station coordinated dispatching system is elaborated from the aspects of system structure, load prediction, coordinated control strategy and the like, and the system starts from reality, and a layered and partitioned electric vehicle dispatching method is applied according to the structural layering characteristic of an electric vehicle charging and discharging service network; an electric vehicle charging station coordination control strategy is established, and different scheduling requirements including scheduling control over a charging and discharging station are met.

Description

Technical Field

[0001] The present invention relates to the field of in-station control technology for a battery swap station, and in particular to an in-station control system and method for a battery swap station in an electric power service area. Background Art

[0002] The power service area battery swap station involves the services of various new energy electric vehicles. With the continuous growth of the number of electric vehicles, the random and disorderly access of large-scale electric vehicles to the power grid in the power service area for charging and discharging will have a certain impact on the operation of the power grid; allowing electric vehicle users to charge disorderly will not only affect the economic efficiency of the power grid operation, but also seriously deteriorate the power quality, produce voltage deviations at access nodes, and affect the safety and stability of the power grid. If electric vehicle users can charge in an orderly manner, it will not only not burden the power grid, but also achieve peak load reduction and valley filling, and reduce the cost of purchasing electricity. Considering that charging stations are the carriers for concentrated access to the power grid for electric vehicles, a hierarchical and zoned scheduling method can be adopted. The upper layer is the regional dispatching center that issues scheduling plans for charging stations, and the lower layer realizes orderly charging of electric vehicles through charging control strategies within charging stations.

[0003] Based on the above, it is of great significance to establish an in-station control system for battery swap stations with charging station load demand prediction function, multiple scheduling objectives, network control function under peak shaving and valley filling, and orderly charging control function within the station. Summary of the invention

[0004] The present invention aims to solve the technical problem in the prior art that it is difficult to effectively predict and schedule the power consumption of multiple devices in a power exchange station in a service area, and to provide an in-station control system and method for a power exchange station in a power service area.

[0005] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0006] A control system and method for a power service area power exchange station, comprising:

[0007] The input data layer is used to provide data support for the algorithms of the business logic layer 2;

[0008] The algorithms in the business logic layer include:

[0009] Historical charging data algorithm analysis library, parameter correction library, charging facility equipment library, electric vehicle data, battery data, real-time data interface, and dispatch instruction interface;

[0010] The supporting data provided by the historical charging data algorithm analysis library for the business logic layer include: historical charging data, the historical charging data include: load forecasting learning data, load forecasting data, regional level scheduling data and station level coordinated control data, parameter correction data;

[0011] The parameter correction library includes: parameter correction strategies, correction schemes and correction results that have been applied;

[0012] Wherein, the business logic layer generates prediction results and scheduling processes for charging and discharging stations in the region;

[0013] The business logic layer includes a prediction learning module, a load prediction module, a model parameter correction module, and a regional scheduling module; and

[0014] Simulation evaluation module;

[0015] The prediction learning module continuously adjusts the load prediction module and parameters using the historical charging data algorithm analysis library;

[0016] The load forecasting data performed by the load forecasting module includes:

[0017] Station-level and regional-level forecast data;

[0018] The model parameter correction module corrects the load prediction model and the parameters in the model according to the load prediction data and the real-time data to ensure that the load prediction data is close to the real data;

[0019] The regional dispatch module decomposes the superior dispatch instructions and forms a dispatch plan for each region based on the load forecast data of each region, so as to form a coordinated control strategy within the station according to the load forecast data and charging demand within the station through station-level coordinated control;

[0020] The simulation evaluation module simulates the prediction and dispatch results by simulating different parameters and conditions, and evaluates the coordinated control strategy from the perspective of peak load shaving and valley filling and economic operation;

[0021] Afterwards, after receiving the data from the business logic layer 1, the display application layer can display the prediction results, scheduling plans and simulation evaluation results to the user in a variety of forms such as concise and clear charts, curves, tables, etc., to form an orderly charging control plan and coordinated control evaluation results within the station.

[0022] Specifically, the load forecasting module is constructed as a load forecasting model;

[0023] The load forecasting model is a load demand forecast for the charging station, and the forecasting data information provided includes: the number and type of electric vehicles;

[0024] The load forecasting model adopts an AR (Auto-Regressive Model) model or an ARMA (AutoRegressive and Moving Average Model) model for forecasting under the main stable sequence load condition.

[0025] Specifically, when the ARMA (AutoRegressive and Moving Average Model) model is used for prediction based on the EBL (Explanation-based learning) machine learning method, the model correction process is:

[0026] Step S101, preset and read an instance;

[0027] Step S102, finding response rules and forming causal explanations;

[0028] Step S103, summarizing the overall extraction judgment control method;

[0029] Step S104, determining whether the new data parameters need to be adjusted, and when the result is N, the process ends;

[0030] In step S105, when the result is Y, the parameters are diverged in different directions and dimensions, and compared with the instance accuracy to see whether the parameters meet the requirements. If it is judged to be yes, the process ends, and if it is judged to be no, step S105 is repeated.

[0031] Specifically, it also includes: coordinated control method:

[0032] The process of the coordinated control method is:

[0033] Each day is divided into n time periods. There are two types of coordinated control decisions. One is to reduce the peak-to-valley difference of the power grid load, thereby alleviating the peak-to-valley pressure of the power grid. The other is to minimize the power purchase cost and maximize the profit of the charging station. The coordinated control decision adopts the Lemke algorithm to control the coordinated control decision.

[0034] Specifically, the execution process of the Lemke algorithm is:

[0035] S201, after initialization, find a quasi-complementary basic feasible solution;

[0036] S202, finding a new complementary basic feasible solution by the pivot elimination method;

[0037] S203, when it is determined that a non-basic variable is found, the process ends; when it is determined that no non-basic variable is found, the process executes step S202.

[0038] Specifically, it also includes: the deployment of control systems within battery swap stations in the power service area, specifically: a hierarchical deployment approach is adopted, with deployment carried out in the centralized control center and each charging and swap station. The main system of the centralized control center obtains the dispatching information through the interface with the power grid dispatching system, decomposes the dispatching plan and sends it to each charging and swap station, and then the station-level system of each charging and swap station arranges the charging of electric vehicles.

[0039] The present invention has the following beneficial effects:

[0040] The regional charging station coordination and dispatching system is elaborated in detail from the aspects of system structure, load forecasting, coordinated control strategy, etc. The system starts from reality and applies the layered and partitioned electric vehicle dispatching method according to the structural layered characteristics of the electric vehicle charging and discharging service network. A coordinated control strategy for electric vehicle charging stations is established to achieve different dispatching needs, including the dispatching control of charging and discharging stations (including discrete charging piles). BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0042] Figure 1 It is a schematic diagram of the system structure of the present invention;

[0043] Figure 2 This is an embodiment of the daily load forecast diagram of the present invention;

[0044] Figure 3 It is a parameter correction flow chart of the present invention;

[0045] Figure 4 It is the Lemke algorithm flow chart of the present invention. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention; it should be noted that in this application, for the convenience of description, the "left side" in the current view is the "first end", the "right side" is the "second end", the "upper side" is the "first end", and the "lower side" is the "second end". The purpose of such description is to clearly express the technical solution, and it should not be understood as an improper limitation on the technical solution of this application.

[0047] The present invention aims to solve the technical problem in the prior art that it is difficult to effectively predict and dispatch the power consumption of multiple devices in the service area battery swap station. It provides an in-station control system and method for the power swap station in the power service area. The regional charging station coordination control system serves as a tool for dispatchers to assist in decision-making, which can realize the prediction of the load demand and spare power of the charging stations in the region. According to the two goals of smoothing the load curve and minimizing the cost of purchasing electricity, the electric vehicle charging and swapping service network is structurally layered, different control measures are taken for charging stations of different levels, and scheduling plans for each charging station at multiple time and space scales in the region are given to realize the coordinated scheduling of charging stations within the region. The system structure is divided into a basic data layer, a business logic layer and a display application layer, as shown in the attached figure. Figure 1 As shown, the layers are interconnected, including: input data layer 1, which is used to provide data support for the algorithms of business logic layer 2;

[0048] The algorithms in the business logic layer include:

[0049] Historical charging data algorithm analysis library, parameter correction library, charging facility equipment library, electric vehicle data, battery data, real-time data interface, and dispatch instruction interface;

[0050] The supporting data provided by the historical charging data algorithm analysis library for the business logic layer 2 include: historical charging data, the historical charging data include: load forecasting learning data, load forecasting data, regional level scheduling data and station level coordinated control data, parameter correction data;

[0051] The parameter correction library includes: parameter correction strategies, correction schemes and correction results that have been applied;

[0052] Wherein, the business logic layer generates prediction results and scheduling processes for charging and discharging stations in the region;

[0053] The business logic layer includes a prediction learning module, a load prediction module, a model parameter correction module, and a regional scheduling module; and

[0054] Simulation evaluation module;

[0055] The prediction learning module continuously adjusts the load prediction module and parameters using the historical charging data algorithm analysis library;

[0056] The load forecasting data performed by the load forecasting module includes:

[0057] Station-level and regional-level forecast data;

[0058] The model parameter correction module corrects the load prediction model and the parameters in the model according to the load prediction data and the real-time data to ensure that the load prediction data is close to the real data;

[0059] The regional dispatch module decomposes the superior dispatch instructions and forms a dispatch plan for each region based on the load forecast data of each region, so as to form a coordinated control strategy within the station according to the load forecast data and charging demand within the station through station-level coordinated control;

[0060] The simulation evaluation module simulates the prediction and dispatch results by simulating different parameters and conditions, and evaluates the coordinated control strategy from the perspective of peak load shaving and valley filling and economic operation;

[0061] Afterwards, after receiving the data from the business logic layer, the application layer can display the prediction results, scheduling plans and simulation evaluation results to the user in a variety of forms such as concise and clear charts, curves, tables, etc., to form an orderly charging control plan and coordinated control evaluation results within the station.

[0062] In a specific embodiment, the load forecasting module is constructed as a load forecasting model;

[0063] The load forecasting model is a load demand forecast for the charging station, and the forecasting data information provided includes: the number and type of electric vehicles;

[0064] The load forecasting model adopts an AR (Auto-Regressive Model) model or an ARMA (AutoRegressive and Moving Average Model) model for forecasting under the main stable sequence load condition.

[0065] In a specific embodiment, when the ARMA (AutoRegressive and Moving Average Model) model is used for prediction based on the EBL (Explanation-based learning) machine learning method, the model correction process is:

[0066] Step S101, preset and read an instance;

[0067] Step S102, finding response rules and forming causal explanations;

[0068] Step S103, summarizing the overall extraction judgment control method;

[0069] Step S104, determining whether the new data parameters need to be adjusted, and when the result is N, the process ends;

[0070] In step S105, when the result is Y, the parameters are diverged in different directions and dimensions, and compared with the instance accuracy to see whether the parameters meet the requirements. If it is judged to be yes, the process ends, and if it is judged to be no, step S105 is repeated.

[0071] Specifically, after the order and parameters of the ARMA model for predicting the load demand of the charging station are determined, the machine learning method of EBL is selected to modify the parameters of the ARMA model. Machine learning is to establish a computational model or cognitive model similar to the human learning process by imitating the human learning mechanism. As a type of machine learning, EBL divides the entire learning process into two steps. First, an instance is provided to the system, and the system knowledge base searches for rules related to the instance, and searches and records the causal relationship explanation between the rules related to the instance and the instance. Then, the system summarizes the causal relationship explanation, retains the main influencing factors, omits the secondary influencing factors, and appropriately combines the main factors to finally form a judgment rule or control method.

[0072] In a specific embodiment, it also includes: a coordinated control method:

[0073] The process of the coordinated control method is:

[0074] Each day is divided into n time periods. There are two types of coordinated control decisions. One is to reduce the peak-to-valley difference of the power grid load, thereby alleviating the peak-to-valley pressure of the power grid. The other is to minimize the power purchase cost and maximize the profit of the charging station. The coordinated control decision adopts the Lemke algorithm to control the coordinated control decision.

[0075] In a specific embodiment, the execution process of the Lemke algorithm is:

[0076] S201, after initialization, find a quasi-complementary basic feasible solution;

[0077] S202, finding a new complementary basic feasible solution by the pivot elimination method;

[0078] S203, when it is determined that a non-basic variable is found, the process ends; when it is determined that no non-basic variable is found, the process executes step S202.

[0079] Most of the research on the optimization strategy of orderly access of electric vehicles is based on the direct dispatch of electric vehicles by the dispatch center. Considering that this top-to-bottom dispatch mode has high requirements on the data processing and communication capabilities of the dispatch center, the system adopts a hierarchical and partitioned dispatch mode after obtaining the load forecast data, dividing a day into n time periods. There are two decision models in the system. One is to reduce the peak-to-valley difference of the power grid load with the goal of shaving the peak and filling the valley, thereby alleviating the peak-to-valley pressure of the power grid. The other is to minimize the cost of purchasing electricity and maximize the profit of the charging station. Both dispatch models can obtain dispatch instructions for the charging station. The dispatch instructions for each time period of the charging stations in each region are issued, and then the charging optimization of the single charging station is carried out in the station, and finally the charging plan for the charging station in multiple time periods is obtained. Since the dispatch instruction problem issued to the charging station is a quadratic programming problem involving equality and inequality constraints after modeling, the Lemke algorithm is used to solve the model. After obtaining the dispatch instructions issued to the charging station in each time period, the next step is to carry out orderly charging in the charging station. Since the dispatching system is mainly for lithium battery electric vehicles, it is assumed that each type of charging pile in the charging station maintains constant power within a unit time period, that is, the charging power of the charging station in a charging period is a discrete value, so the charging plan of the charging station will deviate from the dispatching order issued by the dispatching center. On the one hand, the charging plan in the charging station should be close to the dispatching order issued by the dispatching center. On the other hand, considering the operation time of charging and replacing batteries in the charging station and the labor cost of the staff, it is hoped that the number of changes in the charging piles in the charging station will be small. Therefore, for the orderly charging in the charging station, the goal is to minimize the sum of the changes in the number of charging piles in adjacent time periods under a small deviation from the dispatching order. A multi-objective dispatching model is established. Within the allowable change range of the first-level goal, the number of changes in the charging piles is minimized as much as possible to save costs and labor time.

[0080] In a specific embodiment, it also includes: system deployment, and the system deployment is: a hierarchical deployment method is adopted, and deployment is carried out in the centralized control center and each charging and swapping station. The main system of the centralized control center obtains the dispatching information through the interface with the power grid dispatching system, decomposes the dispatching plan and sends it to each charging and swapping station, and then the station-level system of each charging and swapping station arranges the charging of electric vehicles. The coordinated control system of the charging and swapping stations in the region operates stably and reliably, and can complete the load forecast and load demand forecast of the charging and swapping stations on a single day every day. The prediction accuracy meets the dispatching requirements, and can realize the dispatching instructions of the charging and swapping stations under different strategies of peak shaving and valley filling economic operation, complete the orderly charging in the charging and swapping stations, and provide coordinated control with peak shaving and valley filling and economic operation as the goals. For the above two goals, the system is simulated and verified. The technical scheme elaborates on the coordinated dispatching system of regional charging stations in terms of system structure, load forecasting, coordinated control strategy, etc. The system starts from reality, according to the structural layered characteristics of the electric vehicle charging and discharging service network, applies the layered and partitioned electric vehicle dispatching method, establishes a coordinated control strategy for electric vehicle charging stations, and realizes the dispatching control of charging and discharging stations (including discrete charging piles) under different dispatching requirements.

[0081] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A control system and method for a power exchange station in a power service area, characterized in that: include: The input data layer is used to provide data support for the algorithms in the business logic layer; The algorithms in the business logic layer include: Historical charging data algorithm analysis library, parameter correction library, charging facility equipment library, electric vehicle data, battery data, real-time data interface, and dispatch instruction interface; The supporting data provided by the historical charging data algorithm analysis library for the business logic layer include: historical charging data, the historical charging data include: load forecasting learning data, load forecasting data, regional level scheduling data and station level coordinated control data, parameter correction data; The parameter correction library includes: parameter correction strategies, correction schemes and correction results that have been applied; Wherein, the business logic layer generates prediction results and scheduling processes for charging and discharging stations in the region; The business logic layer includes a prediction learning module, a load prediction module, a model parameter correction module, and a regional scheduling module; and Simulation evaluation module; The prediction learning module continuously adjusts the load prediction module and parameters using the historical charging data algorithm analysis library; The load forecasting data performed by the load forecasting module includes: Station-level and regional-level forecast data; The model parameter correction module corrects the load prediction model and the parameters in the model according to the load prediction data and the real-time data to ensure that the load prediction data is close to the real data; The regional dispatch module decomposes the superior dispatch instructions and forms a dispatch plan for each region based on the load forecast data of each region, so as to form a coordinated control strategy within the station according to the load forecast data and charging demand within the station through station-level coordinated control; The simulation evaluation module simulates the prediction and dispatch results by simulating different parameters and conditions, and evaluates the coordinated control strategy from the perspective of peak load shaving and valley filling and economic operation; Afterwards, after receiving the data from the business logic layer, the application layer can display the prediction results, scheduling plans and simulation evaluation results to the user in a variety of forms such as concise and clear charts, curves, tables, etc., to form an orderly charging control plan and coordinated control evaluation results within the station.

2. The control system and method for the power service area power exchange station according to claim 1, characterized in that: The load forecasting module is constructed as a load forecasting model; The load forecasting model is a load demand forecast for the charging station, and the forecasting data information provided includes: the number and type of electric vehicles; The load forecasting model adopts AR model or ARMA model for forecasting under the main stable sequence load condition.

3. The control system and method for the power service area power exchange station according to claim 2, characterized in that: When the ARMA model is used for prediction based on the EBL machine learning method, the model modification process is: Step S101, preset and read an instance; Step S102, finding response rules and forming causal explanations; Step S103, summarizing the overall extraction judgment control method; Step S104, determining whether the new data parameters need to be adjusted, and when the result is N, the process ends; In step S105, when the result is Y, the parameters are diverged in different directions and dimensions, and compared with the instance accuracy to see whether the parameters meet the requirements. If it is judged to be yes, the process ends, and if it is judged to be no, step S105 is repeated.

4. The control system and method for the power service area power exchange station according to claim 3, characterized in that: Also includes: Coordinated control method: The process of the coordinated control method is: Each day is divided into n time periods. There are two types of coordinated control decisions. One is to reduce the peak-to-valley difference of the power grid load, thereby alleviating the peak-to-valley pressure of the power grid. The other is to minimize the power purchase cost and maximize the profit of the charging station. The coordinated control decision adopts the Lemke algorithm to control the coordinated control decision.

5. The control system and method for the power service area power exchange station according to claim 3, characterized in that: The execution process of the Lemke algorithm is: S201, after initialization, find a quasi-complementary basic feasible solution; S202, finding a new complementary basic feasible solution by the pivot elimination method; S203, when it is determined that a non-basic variable is found, the process ends; when it is determined that no non-basic variable is found, the process executes step S202.

6. The control system and method for the power service area power exchange station according to claim 3, characterized in that: It also includes: the deployment of control systems within battery swap stations in the power service area, specifically: a hierarchical deployment approach is adopted, with deployment carried out in the centralized control center and each charging and swap station. The main system of the centralized control center obtains dispatch information through the interface with the power grid dispatch system, decomposes the dispatch plan and sends it to each charging and swap station, and then the station-level system of each charging and swap station arranges the charging of electric vehicles.

Images