Charging load prediction method and system for electric vehicle

Abstract

The invention discloses a charging load prediction method and system for an electric vehicle. The method comprises the steps that a road traffic network matrix is constructed according to collected road information, and an origin-destination matrix is constructed based on collected trip chain information; constructing an electric vehicle dynamic travel model based on the road traffic network matrix and the origin-destination matrix, performing game processing on the electric vehicle dynamic travel model by using a Bayesian game method to obtain an electric vehicle charging strategy, and obtaining a charging demand prediction result based on the electric vehicle charging strategy; and inputting the charging demand prediction result into an electric vehicle charging load model constructed by adopting a Markov chain to obtain charging load data of the electric vehicle. According to the charging load prediction method of the electric vehicle provided by the embodiment of the invention, the technical problem of low accuracy of charging load prediction can be solved, and the effect of accurately predicting the charging load of the electric vehicle can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicles, and in particular to a charging load prediction method and system for an electric vehicle. Background Art

[0002] In the scenario of electric vehicle charging load prediction, existing technologies usually use statistical methods based on single data to predict charging load.

[0003] However, the charging behavior of electric vehicles is affected by multiple factors. The complex relationships and uncertainties among these factors make it difficult for statistical methods based on historical data to accurately capture all changes, thus limiting the accuracy of the prediction results.

[0004] It can be seen that how to accurately predict the charging load of electric vehicles has become a technical problem that needs to be urgently solved by those skilled in the art. Summary of the invention

[0005] The present invention provides a charging load prediction method and system for an electric vehicle, so as to solve the technical problem of low charging load prediction accuracy, so as to achieve the effect of accurately predicting the charging load of the electric vehicle.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a charging load prediction method for an electric vehicle, comprising:

[0007] Construct a road traffic network matrix based on the collected road information, and construct an origin-destination matrix based on the collected travel chain information;

[0008] Building a dynamic travel model for electric vehicles based on the road traffic network matrix and the origin-destination matrix, using a Bayesian game method to perform game processing on the dynamic travel model for electric vehicles to obtain an electric vehicle charging strategy, and obtaining a charging demand prediction result based on the electric vehicle charging strategy;

[0009] The charging demand prediction result is input into an electric vehicle charging load model constructed using a Markov chain to obtain the charging load data of the electric vehicle, wherein the charging load data of the electric vehicle is configured to simulate the changes in the power state of the electric vehicle at different times by establishing a state transition probability matrix to obtain the charging load data of the electric vehicle.

[0010] As one of the preferred solutions, the starting and ending point matrix is ​​constructed based on the collected travel chain information, and the construction process includes:

[0011] Acquire the travel chain information of the target electric vehicle by using the vehicle-mounted equipment, and pre-process the travel chain information of the target electric vehicle;

[0012] The preprocessed travel chain information of the target electric vehicle is segmented into a plurality of independent travel segments, and a start-end point matrix is ​​constructed based on the travel segments.

[0013] As one of the preferred solutions, the game processing process includes:

[0014] A priori probability distribution is constructed based on the electric vehicle dynamic travel model, a benefit function is determined based on the priori probability distribution, and the electric vehicle charging strategy is obtained by performing Bayesian Nash equilibrium solution according to the benefit function.

[0015] As one of the preferred solutions, the method of solving the Bayesian Nash equilibrium according to the profit function to obtain the electric vehicle charging strategy includes:

[0016] The Bayesian Nash equilibrium solution of the profit function is performed using simulated game to obtain the electric vehicle charging strategy.

[0017] As one preferred solution, after obtaining the charging load data of the electric vehicle, the charging load prediction method of the electric vehicle further includes:

[0018] The charging load data obtained by the charging pile detection is used as a benchmark, and the Spearman rank correlation coefficient is used to evaluate the predicted charging load data of the electric vehicle.

[0019] Another embodiment of the present invention provides a charging load prediction system for an electric vehicle, comprising:

[0020] A construction module is used to construct a road traffic network matrix based on the collected road information and to construct an origin and destination point matrix based on the collected travel chain information;

[0021] A processing module, used for constructing a dynamic travel model of electric vehicles based on the road traffic network matrix and the origin-destination matrix, using a Bayesian game method to perform game processing on the dynamic travel model of electric vehicles to obtain an electric vehicle charging strategy, and obtaining a charging demand prediction result based on the electric vehicle charging strategy;

[0022] A prediction module is used to input the charging demand prediction result into an electric vehicle charging load model constructed using a Markov chain to obtain the charging load data of the electric vehicle, wherein the charging load data of the electric vehicle is configured to simulate the changes in the power state of the electric vehicle at different times by establishing a state transition probability matrix to obtain the charging load data of the electric vehicle.

[0023] As one of the preferred solutions, the starting and ending point matrix is ​​constructed based on the collected travel chain information, and the construction process includes:

[0024] Acquire the travel chain information of the target electric vehicle by using the vehicle-mounted equipment, and pre-process the travel chain information of the target electric vehicle;

[0025] The preprocessed travel chain information of the target electric vehicle is segmented into a plurality of independent travel segments, and a start-end point matrix is ​​constructed based on the travel segments.

[0026] As one of the preferred solutions, the game processing process includes:

[0027] A priori probability distribution is constructed based on the electric vehicle dynamic travel model, a benefit function is determined based on the priori probability distribution, and the electric vehicle charging strategy is obtained by performing Bayesian Nash equilibrium solution according to the benefit function.

[0028] As one of the preferred solutions, the method of solving the Bayesian Nash equilibrium according to the profit function to obtain the electric vehicle charging strategy includes:

[0029] The Bayesian Nash equilibrium solution of the profit function is performed using simulated game to obtain the electric vehicle charging strategy.

[0030] As one preferred solution, after obtaining the charging load data of the electric vehicle, the charging load prediction system of the electric vehicle further includes:

[0031] The evaluation module is used to evaluate the predicted charging load data of the electric vehicle using the Spearman rank correlation coefficient based on the charging load data obtained by the charging pile detection.

[0032] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0033] (1) This embodiment can capture the real travel mode of users based on the road traffic network matrix and the origin-destination matrix, avoiding inaccurate estimates based on assumptions or experience, and making the prediction of charging load closer to the actual situation;

[0034] (2) This embodiment uses the Bayesian game method to transform the electric vehicle dynamic travel model into a game theory problem, which can handle decision-making problems under incomplete information and is suitable for optimizing electric vehicle charging strategies;

[0035] (3) The charging load data of electric vehicles is obtained by using the electric vehicle charging load model constructed by Markov chain, which simulates the decision-making process of users under such uncertainty. This method takes into account the individual differences and mutual influences of users, making the prediction model more in line with the actual complex user behavior environment, thereby improving the accuracy of charging load prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic flow chart of a charging load prediction method for an electric vehicle in one embodiment of the present invention;

[0037] Figure 2 It is a schematic diagram of the structure of a charging load prediction system for an electric vehicle in one embodiment of the present invention.

[0038] Reference numerals:

[0039] Among them, 11 is a construction module; 12 is a processing module; 13 is a prediction module. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of 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.

[0041] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0042] In the description of the present application, it should be noted that, unless otherwise expressly specified and limited, the terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0043] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0044] An embodiment of the present invention provides a method for predicting the charging load of an electric vehicle. For details, see Figure 1 , Figure 1 The figure shows a flow chart of a charging load prediction method for an electric vehicle in one embodiment of the present invention, including:

[0045] S1: Construct a road traffic network matrix based on the collected road information, and construct an origin-destination matrix based on the collected travel chain information;

[0046] S2: constructing an electric vehicle dynamic travel model based on the road traffic network matrix and the origin-destination matrix, using a Bayesian game method to perform game processing on the electric vehicle dynamic travel model to obtain an electric vehicle charging strategy, and obtaining a charging demand prediction result based on the electric vehicle charging strategy;

[0047] S3: Input the charging demand prediction result into the electric vehicle charging load model constructed using Markov chain to obtain the charging load data of the electric vehicle, wherein the charging load data of the electric vehicle is configured to simulate the change of the power state of the electric vehicle at different times by establishing a state transition probability matrix to obtain the charging load data of the electric vehicle.

[0048] In step S1, road information can be collected in a variety of ways, such as measuring the length, width, slope and other geometric attributes of the road on the spot, using traffic sensors to obtain dynamic information such as real-time traffic flow and vehicle speed of the road, and extracting the topological structure of the road from the geographic information system (GIS), where the topological structure includes the connection relationship between nodes and road sections.

[0049] This information is integrated to construct a road traffic network matrix, where the rows and columns of the road traffic network matrix correspond to the nodes in the road network, such as intersections, charging station locations, etc., and the elements represent the connection relationships between nodes, such as distance, travel time, traffic flow weight, etc.

[0050] The travel chain information of the target electric vehicle is obtained by using the on-board equipment and the geographic information system (GIS), and the travel chain information is preprocessed. The preprocessing steps include removing duplicate records, processing erroneous data, and integrating multi-source data.

[0051] The preprocessed travel chain information of the target electric vehicle is segmented into several independent travel segments, and a start-end point matrix is ​​constructed based on the travel segments. The rows of the start-end point matrix represent the starting area of ​​the trip, and the columns represent the end area. The matrix elements represent the travel flow from a specific starting area to the end area.

[0052] In step S2, a dynamic travel model of electric vehicles is constructed using traffic simulation software in combination with the road traffic network matrix and the origin and destination matrix. The dynamic travel model takes into account the dynamic changes of the traffic network and the driving characteristics of electric vehicles.

[0053] Based on the dynamic travel model of electric vehicles, the Bayesian game method is used to conduct game processing on the dynamic travel model of electric vehicles to obtain the electric vehicle charging strategy.

[0054] It should be noted that Bayesian game is developed on the basis of incomplete information game. In the electric vehicle charging scenario, electric vehicle users cannot fully understand the actions of other participants and all the information of the environment, such as the real-time usage of charging stations, the charging needs of other users, etc. Each electric vehicle user has different user preferences, such as sensitivity to charging prices, tolerance for waiting time, etc., and this type is based on a certain probability distribution.

[0055] Therefore, a priori probability distribution is constructed based on the dynamic travel model of electric vehicles, and the profit function is determined based on the prior probability distribution. The Bayesian Nash equilibrium is solved according to the profit function to obtain the electric vehicle charging strategy.

[0056] Specifically, Bayesian Nash equilibrium is a stable strategy combination. In this state, the strategy chosen by each electric vehicle user is optimal given the probability distribution of his own type and the strategies of other electric vehicle users. The Bayesian Nash equilibrium can be solved through mathematical programming methods or simulation methods.

[0057] In addition to using the Bayesian game method to perform game processing on the electric vehicle dynamic travel model to obtain the electric vehicle charging strategy, the method based on multi-agent system can also be used. In the method based on multi-agent system, electric vehicles and charging stations are regarded as agents, and each agent has its own goals and behavior rules.

[0058] The goal of the electric vehicle agent is to meet its own charging needs, and the goal of the charging station agent is to reasonably allocate charging resources to improve efficiency. The electric vehicle agent sends a charging request to the charging station agent, and the charging station agent responds according to its own resource conditions and the requests of other electric vehicles, thereby obtaining the electric vehicle charging strategy.

[0059] In game theory, the payoff function is a mathematical representation of the rewards that participants can obtain based on a specific strategy combination. For the prediction of electric vehicle charging load, the payoff function includes multiple aspects such as charging cost, time cost, battery health, and grid load impact.

[0060] The electric vehicle charging scenario is a complex dynamic system involving many factors such as the distribution of charging stations, fluctuations in electricity prices, and vehicle travel plans. Simulation games can effectively simulate this complex real-world environment and reflect the actual situation by setting different parameters and rules.

[0061] Through multiple simulations of the game process, each electric vehicle user can adjust his or her strategy according to the profit function. As the number of simulations increases, the strategies of the participants will gradually converge to a relatively stable state, which is the Bayesian Nash equilibrium state, and the corresponding strategy is the electric vehicle charging strategy.

[0062] Since the Bayesian game takes into account the uncertainty of the participant types, the simulation game handles this uncertainty by setting the probability distribution of different types of users. In each simulation, different types of users are randomly generated according to the probability distribution, and then the users are allowed to play games in the simulation environment to find the electric vehicle charging strategy under uncertain conditions, and obtain the charging demand forecast results based on this electric vehicle charging strategy.

[0063] In step S3, the charging demand prediction result is introduced into the electric vehicle charging load model constructed based on the Markov chain as key input information.

[0064] When using Markov chains to build a model, the different power levels of electric vehicles are defined as a series of discrete states, such as low power, medium power, high power, full power, etc. The transition probabilities between these states describe the changes in the power level of the electric vehicle at different time points.

[0065] By establishing a state transition probability matrix to simulate the changes in the power state of electric vehicles at different times, and combining the charging demand prediction results with the state transition probability matrix, the charging behavior of electric vehicles at different time points can be simulated.

[0066] According to the initial charge state and state transition probability of the electric vehicle, the charge state of the electric vehicle at the next time point can be calculated. The Markov chain model simulates the charging behavior of the electric vehicle in an iterative manner. In each iteration, the charge state of the electric vehicle is updated according to the current state and state transition probability matrix, and the corresponding charging load is calculated, thereby predicting the charging load data of the electric vehicle.

[0067] After obtaining the charging load data of the electric vehicle, the charging load data obtained by the charging pile detection is used as a benchmark, and the Spearman rank correlation coefficient is used to evaluate the predicted charging load data of the electric vehicle.

[0068] The charging load data obtained from the charging pile detection system is the load record generated by the electric vehicle during the actual charging process. The actual charging load data and the predicted charging load data are sorted to obtain their respective ranks. If there are repeated values ​​in the data, they are assigned the same average rank.

[0069] The correlation between the actual charging load data and the predicted charging load data is calculated using the formula of the Spearman rank correlation coefficient.

[0070] If the calculated coefficient is close to 1, it means that there is a strong positive correlation between the predicted charging load data and the actual charging load data, that is, the predicted value increases with the increase of the actual value; if the calculated coefficient is close to -1, it means that there is a strong positive correlation between the predicted charging load data and the actual charging load data, that is, the predicted value increases with the increase of the actual value; if the calculated coefficient is close to 0, it means that there is no obvious correlation between the predicted charging load data and the actual charging load data.

[0071] The embodiment of the present invention provides a charging load prediction system for electric vehicles. For details, see Figure 2 , Figure 2 The figure shows a schematic diagram of the structure of a charging load prediction system for an electric vehicle in one embodiment of the present invention, comprising:

[0072] A construction module 11 is used to construct a road traffic network matrix based on the collected road information, and to construct an origin and destination matrix based on the collected travel chain information;

[0073] A processing module 12 is used to construct an electric vehicle dynamic travel model based on the road traffic network matrix and the origin and destination matrix, use a Bayesian game method to perform game processing on the electric vehicle dynamic travel model to obtain an electric vehicle charging strategy, and obtain a charging demand prediction result based on the electric vehicle charging strategy;

[0074] The prediction module 13 is used to input the charging demand prediction result into the electric vehicle charging load model constructed using the Markov chain to obtain the charging load data of the electric vehicle, wherein the charging load data of the electric vehicle is configured to simulate the changes in the power state of the electric vehicle at different times by establishing a state transition probability matrix to obtain the charging load data of the electric vehicle.

[0075] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0076] (1) This embodiment can capture the real travel mode of users based on the road traffic network matrix and the origin-destination matrix, avoiding inaccurate estimates based on assumptions or experience, and making the prediction of charging load closer to the actual situation;

[0077] (2) This embodiment uses the Bayesian game method to transform the electric vehicle dynamic travel model into a game theory problem, which can handle decision-making problems under incomplete information and is suitable for optimizing electric vehicle charging strategies;

[0078] (3) The charging load data of electric vehicles is obtained by using the electric vehicle charging load model constructed by Markov chain, which simulates the decision-making process of users under such uncertainty. This method takes into account the individual differences and mutual influences of users, making the prediction model more in line with the actual complex user behavior environment, thereby improving the accuracy of charging load prediction.

[0079] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for predicting charging load of an electric vehicle, characterized in that: include: Construct a road traffic network matrix based on the collected road information, and construct an origin-destination matrix based on the collected travel chain information; Building a dynamic travel model for electric vehicles based on the road traffic network matrix and the origin-destination matrix, using a Bayesian game method to perform game processing on the dynamic travel model for electric vehicles to obtain an electric vehicle charging strategy, and obtaining a charging demand prediction result based on the electric vehicle charging strategy; The charging demand prediction result is input into an electric vehicle charging load model constructed using a Markov chain to obtain the charging load data of the electric vehicle, wherein the charging load data of the electric vehicle is configured to simulate the changes in the power state of the electric vehicle at different times by establishing a state transition probability matrix to obtain the charging load data of the electric vehicle.

2. The method for predicting the charging load of an electric vehicle according to claim 1, characterized in that: The origin and destination point matrix is ​​constructed based on the collected travel chain information, and the construction process includes: Acquire the travel chain information of the target electric vehicle by using the vehicle-mounted equipment, and pre-process the travel chain information of the target electric vehicle; The preprocessed travel chain information of the target electric vehicle is segmented into a plurality of independent travel segments, and a start-end point matrix is ​​constructed based on the travel segments.

3. The method for predicting the charging load of an electric vehicle according to claim 1, characterized in that: The game processing process includes: A priori probability distribution is constructed based on the electric vehicle dynamic travel model, a benefit function is determined based on the priori probability distribution, and the electric vehicle charging strategy is obtained by performing Bayesian Nash equilibrium solution according to the benefit function.

4. The method for predicting the charging load of an electric vehicle according to claim 3, characterized in that: The method of performing Bayesian Nash equilibrium solution according to the profit function to obtain the electric vehicle charging strategy includes: The Bayesian Nash equilibrium solution of the profit function is performed using simulated game to obtain the electric vehicle charging strategy.

5. The method for predicting charging load of an electric vehicle according to claim 1, characterized in that: After obtaining the charging load data of the electric vehicle, the charging load prediction method of the electric vehicle further includes: The charging load data obtained by the charging pile detection is used as a benchmark, and the Spearman rank correlation coefficient is used to evaluate the predicted charging load data of the electric vehicle.

6. A charging load prediction system for electric vehicles, characterized in that: include: A construction module is used to construct a road traffic network matrix based on the collected road information and to construct an origin and destination matrix based on the collected travel chain information; A processing module, used for constructing a dynamic travel model of electric vehicles based on the road traffic network matrix and the origin-destination matrix, using a Bayesian game method to perform game processing on the dynamic travel model of electric vehicles to obtain an electric vehicle charging strategy, and obtaining a charging demand prediction result based on the electric vehicle charging strategy; A prediction module is used to input the charging demand prediction result into an electric vehicle charging load model constructed using a Markov chain to obtain the charging load data of the electric vehicle, wherein the charging load data of the electric vehicle is configured to simulate the changes in the power state of the electric vehicle at different times by establishing a state transition probability matrix to obtain the charging load data of the electric vehicle.

7. The charging load prediction system for electric vehicles according to claim 6, characterized in that: The origin and destination point matrix is ​​constructed based on the collected travel chain information, and the construction process includes: Acquire the travel chain information of the target electric vehicle by using the vehicle-mounted equipment, and pre-process the travel chain information of the target electric vehicle; The preprocessed travel chain information of the target electric vehicle is segmented into a plurality of independent travel segments, and a start-end point matrix is ​​constructed based on the travel segments.

8. The charging load prediction system for electric vehicles according to claim 6, characterized in that: The game processing process includes: A priori probability distribution is constructed based on the electric vehicle dynamic travel model, a benefit function is determined based on the priori probability distribution, and the electric vehicle charging strategy is obtained by performing Bayesian Nash equilibrium solution according to the benefit function.

9. The charging load prediction system for electric vehicles according to claim 8, characterized in that: The method of performing Bayesian Nash equilibrium solution according to the profit function to obtain the electric vehicle charging strategy includes: The Bayesian Nash equilibrium solution of the profit function is performed using simulated game to obtain the electric vehicle charging strategy.

10. The charging load prediction system for electric vehicles according to claim 6, characterized in that: After obtaining the charging load data of the electric vehicle, the charging load prediction system of the electric vehicle further includes: The evaluation module is used to evaluate the predicted charging load data of the electric vehicle using the Spearman rank correlation coefficient based on the charging load data obtained by the charging pile detection.

Images