Vehicle network interaction regulation and control method and system based on charging pile data

Abstract

The invention discloses a vehicle network interaction regulation and control method and system based on charging pile data. The method comprises the steps that electric quantity data of each adjustable vehicle at the current collection moment, power supply load data of each node in a power distribution network, position data of each charging pile and charging power data, charging load data and environment data at each collection moment are collected, and the charging load of each charging pile is predicted; obtaining load prediction data of each charging pile, calculating to obtain a vehicle-network interaction suitability index of each charging pile, and determining the power supply load prediction data of each node based on the vehicle-network interaction suitability indexes of all the charging piles, the electric quantity data of all the adjustable and controllable vehicles and the determined matching degree among the power supply load prediction data of each node. And executing a vehicle network interaction regulation and control strategy corresponding to the matching result. According to the charging pile data-based vehicle network interaction regulation and control method and system provided by the embodiment of the invention, the efficiency of vehicle network interaction regulation and control is improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle-grid interaction technology, and in particular to a vehicle-grid interaction control method and system based on charging pile data. Background Art

[0002] Vehicle-grid interaction refers to the interaction between electric vehicles and the power grid, which enables electric vehicles to absorb electricity when the grid load is low and release electricity when the grid load is high. With the popularization of electric vehicles and the development of intelligent power grids, electric vehicles are used as distributed energy storage resources. Through vehicle-grid interaction control methods, they are included in the dispatching scope of virtual power plants and participate in the peak-shaving, frequency regulation and standby services of the power grid. Electric vehicles have expanded from single charging to participating in real-time control and peak-shaving auxiliary services of the power grid in both charging and discharging forms.

[0003] Through vehicle-grid interaction, it can promote the consumption of new energy, reduce the load pressure of the power grid during peak hours, and reduce the investment in the power grid, which has important practical significance for the construction of new power systems. However, in the process of vehicle-grid interaction, huge and dynamically changing data are involved, and it is difficult to analyze these data efficiently, resulting in low efficiency in the regulation of vehicle-grid interaction.

[0004] It can be seen that how to improve the efficiency of vehicle-grid interactive regulation has become a technical problem that technical personnel in this field need to solve urgently. Summary of the invention

[0005] The present invention provides a vehicle-grid interactive control method and system based on charging pile data, so as to solve the technical problem that the accuracy of the prior art in detecting anomalies in core network logs is low.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a vehicle-grid interactive control method based on charging pile data.

[0007] Collect the power data of each adjustable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time respectively;

[0008] Using a long short-term memory neural network to train a charging pile load prediction model based on all the charging load data, all the location data and all the environmental data, and using the trained charging pile load prediction model to predict the charging load of each charging pile to obtain load prediction data for each charging pile;

[0009] Obtaining the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile; processing the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time, and calculating the vehicle-grid interaction suitability index of each charging pile;

[0010] Based on the matching degree between the vehicle-grid interaction suitability index of all the charging piles, the power data of all the controllable vehicles and the determined power supply load forecast data of each of the nodes, a vehicle-grid interaction control strategy corresponding to the matching result is executed.

[0011] As one of the preferred solutions, the method of obtaining the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile includes:

[0012] Arrange all the charging power data of each charging pile in ascending order according to the time sequence of collection to obtain a charging power time series of each charging pile, and use a mutation point detection algorithm to extract all mutation points in each charging power time series;

[0013] Based on the difference characteristics of the change trends on both sides of all the mutation points in each of the charging power time series, the phase mutation points and non-stationary mutation points of each of the charging power time series are obtained;

[0014] Dividing each of the charging power time series into a plurality of charging power time subsequences based on all the stage mutation points;

[0015] The vehicle-grid interaction stability coefficient of each of the charging power time subsequences is calculated according to the deviation characteristics of all the non-stationary mutation points in each of the charging power time subsequences.

[0016] As one of the preferred solutions, the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time are processed to calculate the vehicle-grid interaction suitability index of each charging pile, including:

[0017] If the charging load data of the charging pile at the current collection time is less than or equal to the load forecast data, the vehicle-grid interaction suitability index of the charging pile is set to 0;

[0018] If the charging load data of the charging pile at the current collection time is greater than the load forecast data, the charging load data, the load forecast data, and the vehicle-grid interaction stability coefficient of the charging pile are input into the vehicle-grid interaction evaluation expression to calculate the vehicle-grid interaction suitability index of the charging pile;

[0019] The vehicle-grid interaction evaluation expression is:

[0020] VA i =Norm((CL i -YL i )×AW i )

[0021] Among them, VA i is the vehicle-grid interaction suitability index of charging pile i, Norm() is the normalization function, CL i is the charging load data of charging pile i at the current collection time, YL i is the load forecast data of charging pile i at the current collection time, AW i is the vehicle-grid interaction stability coefficient of charging pile i.

[0022] As one of the preferred solutions, the step of obtaining the phase mutation point and the non-stationary mutation point of each charging power time series based on the difference characteristics of the change trends on both sides of all the mutation points in each charging power time series includes:

[0023] The least square method is used to perform curve fitting on the preset left neighborhood of each mutation point to obtain the left fitting parameters of each mutation point; the least square method is used to perform curve fitting on the preset right neighborhood of each mutation point to obtain the right fitting parameters of each mutation point;

[0024] The absolute value of the difference between the left fitting parameter and the corresponding right fitting parameter of each mutation point is used as the trend difference index on both sides of each mutation point; the average value of the trend difference index on both sides of all the mutation points in each charging power time series is used as the trend difference reference index of each charging power time series;

[0025] All the mutation points in each of the charging power time series that are greater than or equal to the trend difference reference index are taken as stage mutation points, and all the mutation points in each of the charging power time series that are less than the trend difference reference index are taken as non-stationary mutation points.

[0026] As one of the preferred solutions, the method of calculating the vehicle-grid interaction stability coefficient of each charging power time subsequence according to the deviation characteristics of all the non-stationary mutation points in each charging power time subsequence includes:

[0027] Taking each of the non-stationary mutation points as the center, construct a neighborhood window of each of the non-stationary mutation points, replace the element values ​​of each of the non-stationary mutation points in each of the charging power time subsequences with the average value of all element values ​​in the corresponding neighborhood window, and obtain a smooth mutation charging power sequence corresponding to each of the charging power time subsequences;

[0028] Calculating the first Hurst index of each of the charging power time subsequences and the second Hurst index of the corresponding smooth mutation charging power sequence, inputting the second Hurst index and the corresponding first Hurst index into a stable evaluation expression, and calculating the vehicle-grid interaction stable coefficient of each of the charging power time subsequences;

[0029] The stable evaluation expression is:

[0030] CarW j =exp(-(Hst j ′-Hst j ))

[0031] Among them, CarW j is the vehicle-grid interaction stability coefficient of the charging power time subsequence j, exp() is an exponential function with a natural constant as the base, Hst j is the first Hurst exponent of the charging power time subsequence j, Hst j ′ is the second Hurst exponent of the smooth mutation charging power sequence corresponding to the charging power time subsequence j.

[0032] Another embodiment of the present invention provides a vehicle-grid interactive control system based on charging pile data, the system comprising:

[0033] The vehicle-grid interaction data collection module is used to collect the power data of each adjustable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time;

[0034] A charging pile load prediction module, used to train a charging pile load prediction model based on all the charging load data, all the location data and all the environmental data using a long short-term memory neural network, and to predict the charging load of each charging pile using the trained charging pile load prediction model to obtain load prediction data for each charging pile;

[0035] A vehicle-grid interaction evaluation module is used to obtain the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile; the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time are processed to calculate the vehicle-grid interaction suitability index of each charging pile;

[0036] The vehicle-grid interaction control module is used to execute a vehicle-grid interaction control strategy corresponding to the matching result based on the matching degree between the vehicle-grid interaction suitability index of all the charging piles, the power data of all the controllable vehicles and the determined power supply load forecast data of each of the nodes.

[0037] As one of the preferred solutions, the method of obtaining the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile includes:

[0038] Arrange all the charging power data of each charging pile in ascending order according to the time sequence of collection to obtain a charging power time series of each charging pile, and use a mutation point detection algorithm to extract all mutation points in each charging power time series;

[0039] Based on the difference characteristics of the change trends on both sides of all the mutation points in each of the charging power time series, the phase mutation points and non-stationary mutation points of each of the charging power time series are obtained;

[0040] Dividing each of the charging power time series into a plurality of charging power time subsequences based on all the stage mutation points;

[0041] The vehicle-grid interaction stability coefficient of each of the charging power time subsequences is calculated according to the deviation characteristics of all the non-stationary mutation points in each of the charging power time subsequences.

[0042] As one of the preferred solutions, the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time are processed to calculate the vehicle-grid interaction suitability index of each charging pile, including:

[0043] If the charging load data of the charging pile at the current collection time is less than or equal to the load forecast data, the vehicle-grid interaction suitability index of the charging pile is set to 0;

[0044] If the charging load data of the charging pile at the current collection time is greater than the load forecast data, the charging load data, the load forecast data, and the vehicle-grid interaction stability coefficient of the charging pile are input into the vehicle-grid interaction evaluation expression to calculate the vehicle-grid interaction suitability index of the charging pile;

[0045] The vehicle-grid interaction evaluation expression is:

[0046] VA i =Norm((CL i -YL i )×AW i )

[0047] Among them, VA i is the vehicle-grid interaction suitability index of charging pile i, Norm() is the normalization function, CL i is the charging load data of charging pile i at the current collection time, YL i is the load forecast data of charging pile i at the current collection time, AW i is the vehicle-grid interaction stability coefficient of charging pile i.

[0048] As one of the preferred solutions, the step of obtaining the phase mutation point and the non-stationary mutation point of each charging power time series based on the difference characteristics of the change trends on both sides of all the mutation points in each charging power time series includes:

[0049] The least square method is used to perform curve fitting on the preset left neighborhood of each mutation point to obtain the left fitting parameters of each mutation point; the least square method is used to perform curve fitting on the preset right neighborhood of each mutation point to obtain the right fitting parameters of each mutation point;

[0050] The absolute value of the difference between the left fitting parameter and the corresponding right fitting parameter of each mutation point is used as the trend difference index on both sides of each mutation point; the average value of the trend difference index on both sides of all the mutation points in each charging power time series is used as the trend difference reference index of each charging power time series;

[0051] All the mutation points in each of the charging power time series that are greater than or equal to the trend difference reference index are taken as stage mutation points, and all the mutation points in each of the charging power time series that are less than the trend difference reference index are taken as non-stationary mutation points.

[0052] As one of the preferred solutions, the method of calculating the vehicle-grid interaction stability coefficient of each charging power time subsequence according to the deviation characteristics of all the non-stationary mutation points in each charging power time subsequence includes:

[0053] Taking each of the non-stationary mutation points as the center, construct a neighborhood window of each of the non-stationary mutation points, replace the element values ​​of each of the non-stationary mutation points in each of the charging power time subsequences with the average value of all element values ​​in the corresponding neighborhood window, and obtain a smooth mutation charging power sequence corresponding to each of the charging power time subsequences;

[0054] Calculating the first Hurst index of each of the charging power time subsequences and the second Hurst index of the corresponding smooth mutation charging power sequence, inputting the second Hurst index and the corresponding first Hurst index into a stable evaluation expression, and calculating the vehicle-grid interaction stable coefficient of each of the charging power time subsequences;

[0055] The stable evaluation expression is:

[0056] CarW j =exp(-(Hst j ′-Hst j ))

[0057] Among them, CarW j is the vehicle-grid interaction stability coefficient of the charging power time subsequence j, exp() is an exponential function with a natural constant as the base, Hst j is the first Hurst exponent of the charging power time subsequence j, Hst j ′ is the second Hurst exponent of the smooth mutation charging power sequence corresponding to the charging power time subsequence j.

[0058] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0059] (1) respectively collecting the power data of each controllable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time, using a long short-term memory neural network to train a charging pile load prediction model based on all charging load data, all location data and all environmental data, and using the trained charging pile load prediction model to predict the charging load of each charging pile to obtain the load prediction data of each charging pile. By combining the environment and location information of the charging pile, the reliability of the prediction of the charging load of the charging pile is improved;

[0060] (2) The vehicle-grid interaction stability coefficient of each charging pile is obtained according to the volatility characteristics of the charging power data of each charging pile. The vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time are processed to calculate the vehicle-grid interaction suitability index of each charging pile, which reflects the ability of the charging pile to interact with the power grid in energy and information. It is used to evaluate the suitability of each charging pile to participate in the vehicle-grid interaction. By analyzing the difference between the charging load data and the corresponding load forecast data, the vehicle-grid interaction ability of the charging pile is predicted. In addition, the suitability of the charging pile to participate in the vehicle-grid interaction is predicted in combination with the stability of the charging power change of the charging pile, thereby improving the reliability of the prediction of the suitability of the charging pile to participate in the vehicle-grid interaction.

[0061] (3) Based on the matching degree between the vehicle-grid interaction suitability index of all charging piles, the power data of all controllable vehicles and the power supply load forecast data of each node, the vehicle-grid interaction control strategy corresponding to the matching result is executed, the matching degree between the power supply load forecast data and the power data is comprehensively analyzed, and charging piles with higher vehicle-grid interaction suitability index are selected as the connection equipment between the electric vehicle and the power grid, so as to ensure that the energy exchange between the electric vehicle and the power grid is more efficient and stable, thereby improving the efficiency of vehicle-grid interaction control, and verifying the vehicle-grid interaction control strategy through simulation, executing the vehicle-grid interaction control strategy based on the results of the simulation verification, and further verifying the vehicle-grid interaction control strategy, thereby reducing the uncertainty and risk in the vehicle-grid interaction process and further improving the efficiency of vehicle-grid interaction control;

[0062] (4) By accurately predicting the power supply load of nodes in the distribution network and the charging load of charging piles and providing corresponding vehicle-grid interaction control strategies, real-time and accurate vehicle-grid interaction control can be performed, thereby improving the efficiency of vehicle-grid interaction control. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flow chart of a vehicle-grid interactive control method based on charging pile data in one embodiment of the present invention;

[0064] Figure 2 This is a structural block diagram of a vehicle-grid interactive control system based on charging pile data in one embodiment of the present invention. DETAILED DESCRIPTION

[0065] 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.

[0066] 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.

[0067] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are only for illustrative purposes, 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.

[0068] 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.

[0069] An embodiment of the present invention provides a vehicle-grid interactive control method based on charging pile data. For details, see Figure 1 , Figure 1 The figure shows a flow chart of a vehicle-grid interactive control method based on charging pile data in one embodiment of the present invention.

[0070] A flow chart of a vehicle-grid interactive control method based on charging pile data in one embodiment of the present invention includes the following steps S1-S5, which are as follows:

[0071] Step S1: respectively collect the power data of each adjustable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time.

[0072] It should be noted that electric vehicles, as energy storage units of the power grid, can output electric energy to the power grid when the power grid load is too high, thus playing a role in peak load reduction; when the power grid load is too low, electric vehicles enter charging mode, storing excess power in the power grid, thus playing a role in valley filling. The realization of bidirectional charging and discharging depends on bidirectional charging piles to support the two functions of electric vehicles discharging to the power grid and charging from the power grid. Therefore, by comprehensively analyzing the power data of electric vehicles, the power supply load data of nodes in the distribution network, and the charging data of charging piles, interactive regulation of the vehicle and the grid can be achieved.

[0073] Specifically, the power data of each controllable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time are collected respectively, and the collected data are processed for missing values.

[0074] It should be noted that environmental data includes temperature, humidity, etc. The charging load data of the charging pile is the product of the rated voltage and the operating current.

[0075] Step S2: Using a long short-term memory neural network to train a charging pile load prediction model based on all charging load data, all location data, and all environmental data, the charging load of each charging pile is predicted using the trained charging pile load prediction model to obtain load prediction data for each charging pile.

[0076] Specifically, the location data of each charging pile, all charging load data and all environmental data are input into the long short-term memory neural network, and the mean square error is used as the loss function for prediction and training to obtain a trained charging pile load prediction model, and the trained charging pile load prediction model is used to predict the charging load of each charging pile to obtain the load prediction data of each charging pile.

[0077] It should be noted that the long short-term memory neural network can dynamically control the retention and forgetting of information by introducing three gating mechanisms: forget gate, input gate and output gate, solving the problem of gradient vanishing or gradient exploding of traditional RNN when processing long sequence data, so that the network can learn long-term dependencies. The training of long short-term memory neural network is a well-known technology and will not be elaborated in this application.

[0078] Step S3: Obtain the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile.

[0079] It should be noted that if the charging power on the left side of a mutation point shows a steady upward trend, while the right side suddenly drops to a new stable level, then the mutation point can be classified as a stage mutation point. If the charging power on both sides of a mutation point shows a continuous fluctuation state without obvious stage characteristics, then the mutation point can be classified as a non-stationary mutation point.

[0080] Specifically, all charging power data of each charging pile are arranged in ascending order according to the time sequence of collection to obtain the charging power time series of each charging pile, and a mutation point detection algorithm is used to extract all mutation points in each charging power time series.

[0081] It should be noted that mutation point detection refers to the process of identifying mutation points in a signal or data sequence. These mutation points usually represent sudden changes in the signal, such as fault occurrence, state switching, or event triggering. By performing mutation point detection on the charging power time series, the moment when the charging power time series fluctuates can be identified. The mutation point detection algorithm is a well-known technology and will not be described in detail in this application.

[0082] In this step, the specific method for obtaining the stage mutation point and non-stationary mutation point of each charging power time series based on the difference characteristics of the change trends on both sides of all mutation points in each charging power time series is:

[0083] The least square method is used to perform curve fitting on the preset left neighborhood of each mutation point to obtain the left fitting parameters of each mutation point, and the least square method is used to perform curve fitting on the preset right neighborhood of each mutation point to obtain the right fitting parameters of each mutation point;

[0084] The absolute value of the difference between the left fitting parameter of each mutation point and the corresponding right fitting parameter is used as the trend difference index on both sides of each mutation point, and the average value of the trend difference index on both sides of all mutation points in each charging power time series is used as the trend difference reference index of each charging power time series;

[0085] All mutation points in each charging power time series that are greater than or equal to the trend difference reference index are regarded as stage mutation points, and all mutation points in each charging power time series that are less than the trend difference reference index are regarded as non-stationary mutation points.

[0086] It should be noted that the least squares method is a mathematical optimization technique, which is used to find the best function match for data by minimizing the sum of squares of errors. In the context of curve fitting, the least squares method attempts to find a curve that minimizes the sum of squares of errors between this curve and given data points. The least squares method is a well-known technique and is not described in detail in this application. When the trend difference on both sides of the mutation point is greater, it means that the two sides of the mutation point are more likely to belong to different charging stages, the mutation point is more likely to be the dividing point of the two stages, and the fluctuation of the mutation point is more likely to be caused by normal changes in the charging stage; when the trends on both sides of the mutation point are roughly the same, it means that the two sides of the mutation point are more likely to belong to the same charging stage, and the fluctuation of the mutation point is more likely to be caused by the instability of the charging pile.

[0087] Furthermore, the stage mutation points and non-stationary mutation points of each charging power time series are obtained based on the difference characteristics of the change trends on both sides of all mutation points in each charging power time series, and then each charging power time series is divided into multiple charging power time sub-series according to all stage mutation points. Then, the vehicle-grid interaction stability coefficient of each charging power time sub-sequence is calculated according to the deviation characteristics of all non-stationary mutation points in each charging power time sub-sequence.

[0088] In this step, the specific method for calculating the vehicle-grid interaction stability coefficient of each charging power time subsequence according to the deviation characteristics of all non-stationary mutation points in each charging power time subsequence is:

[0089] Taking each non-stationary mutation point as the center, a neighborhood window of each non-stationary mutation point is constructed, and the element value of each non-stationary mutation point in each charging power time subsequence is replaced by the average value of all element values ​​in the corresponding neighborhood window to obtain a smooth mutation charging power sequence corresponding to each charging power time subsequence;

[0090] Calculate the first Hurst index of each charging power time subsequence and the second Hurst index of the corresponding smooth mutation charging power sequence, input the second Hurst index and the corresponding first Hurst index into the stable evaluation expression, and calculate the vehicle-grid interaction stable coefficient of each charging power time subsequence;

[0091] The stable evaluation expression is:

[0092] CarW j =exp(-(Hst j ′-Hst j ))

[0093] Among them, CarW j is the vehicle-grid interaction stability coefficient of the charging power time subsequence j, exp() is an exponential function with a natural constant as the base, Hst jis the first Hurst exponent of the charging power time subsequence j, Hst j ′ is the second Hurst exponent of the smooth mutation charging power sequence corresponding to the charging power time subsequence j.

[0094] It should be noted that the Hurst index is a statistical indicator used to describe the long-term behavior characteristics of time series data, which can help distinguish whether the data has long-term memory or anti-persistence. The influence of non-stationary mutation points on the long-term correlation of the charging power time subseries is smoothed by replacing the element value of the non-stationary mutation point with the average value of the neighborhood. If the difference in the Hurst index of the charging power time subseries before and after the removal is small, it indicates that the charging power time subseries maintains good stability before and after the removal of the mutation point; if the difference in the Hurst index of the charging power time subseries before and after the removal is large, it indicates that the charging power time subseries has a significant change in stability before and after the removal of the mutation point, indicating that the non-stationary mutation point has a greater impact on the long-term correlation of the charging power time subseries, and the stability of the charging power time subseries is lower.

[0095] Step S4: Process the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data, and the charging load data of each charging pile at the current collection time to calculate the vehicle-grid interaction suitability index of each charging pile.

[0096] Specifically, if the charging load data of the charging pile at the current collection moment is less than or equal to the load forecast data, the vehicle-grid interaction suitability index of the charging pile is set to 0.

[0097] If the charging load data of the charging pile at the current collection time is greater than the load forecast data, the charging load data, load forecast data, and the vehicle-grid interaction stability coefficient of the charging pile are input into the vehicle-grid interaction evaluation expression to calculate the vehicle-grid interaction suitability index of the charging pile.

[0098] The evaluation expression of vehicle-grid interaction is:

[0099] VA i =Norm((CL i -YL i )×AW i )

[0100] Among them, VA i is the vehicle-grid interaction suitability index of charging pile i, Norm() is the normalization function, CL i is the charging load data of charging pile i at the current collection time, YL i is the load forecast data of charging pile i at the current collection time, AW i is the vehicle-grid interaction stability coefficient of charging pile i.

[0101] It should be noted that the greater the vehicle-grid interaction stability coefficient is, and the greater the difference between the charging load data and the load forecast data is, the more stable the charging power of the charging pile is, and the more likely it is to be able to supply electricity to the power grid while meeting the charging load for a period of time in the future, and the more suitable it is for vehicle-grid interaction regulation.

[0102] Step S5: Based on the matching degree between the vehicle-grid interaction suitability index of all charging piles, the power data of all controllable vehicles and the determined power supply load forecast data of each node, execute the vehicle-grid interaction control strategy corresponding to the matching result.

[0103] It should be noted that the vehicle-grid interaction suitability index of the charging pile reflects the ability of the charging pile to interact with the power grid in terms of energy and information. The power supply load forecast data of the node reflects the load conditions of each node in the power grid in the future. The matching degree of the power supply load forecast data and the power data is comprehensively analyzed, and charging piles with a higher vehicle-grid interaction suitability index are selected as the connection equipment between electric vehicles and the power grid to ensure that the energy exchange between electric vehicles and the power grid is more efficient and stable.

[0104] Specifically, based on the matching degree between the vehicle-grid interaction suitability index of all charging piles, the power data of all controllable vehicles and the power supply load forecast data of each node, a vehicle-grid interaction control strategy corresponding to the matching result is executed.

[0105] As an embodiment of the present application, before executing the vehicle-grid interaction control strategy corresponding to the matching result, the vehicle-grid interaction control strategy is simulated and verified to reduce uncertainty and risk in the vehicle-grid interaction process.

[0106] The vehicle-grid interactive control method based on charging pile data provided by the embodiment of the present invention has the following beneficial effects compared with the prior art:

[0107] (1) respectively collecting the power data of each controllable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time, using a long short-term memory neural network to train a charging pile load prediction model based on all charging load data, all location data and all environmental data, and using the trained charging pile load prediction model to predict the charging load of each charging pile to obtain the load prediction data of each charging pile. By combining the environment and location information of the charging pile, the reliability of the prediction of the charging load of the charging pile is improved;

[0108] (2) The vehicle-grid interaction stability coefficient of each charging pile is obtained according to the volatility characteristics of the charging power data of each charging pile. The vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time are processed to calculate the vehicle-grid interaction suitability index of each charging pile, which is used to evaluate the suitability of each charging pile to participate in the vehicle-grid interaction. By analyzing the difference between the charging load data and the corresponding load forecast data, the vehicle-grid interaction capability of the charging pile is predicted. In addition, the suitability of the charging pile to participate in the vehicle-grid interaction is predicted in combination with the stability of the charging power change of the charging pile, thereby improving the reliability of the prediction of the suitability of the charging pile to participate in the vehicle-grid interaction.

[0109] (3) Based on the matching degree between the vehicle-grid interaction suitability index of all charging piles, the power data of all controllable vehicles and the power supply load forecast data of each node, the vehicle-grid interaction control strategy corresponding to the matching result is executed, the matching degree between the power supply load forecast data and the power data is comprehensively analyzed, and charging piles with higher vehicle-grid interaction suitability index are selected as the connection equipment between the electric vehicle and the power grid, so as to ensure that the energy exchange between the electric vehicle and the power grid is more efficient and stable, thereby improving the efficiency of vehicle-grid interaction control, and verifying the vehicle-grid interaction control strategy through simulation, executing the vehicle-grid interaction control strategy based on the results of the simulation verification, and further verifying the vehicle-grid interaction control strategy, thereby reducing the uncertainty and risk in the vehicle-grid interaction process and further improving the efficiency of vehicle-grid interaction control;

[0110] (4) By accurately predicting the power supply load of nodes in the distribution network and the charging load of charging piles and providing corresponding vehicle-grid interaction control strategies, real-time and accurate vehicle-grid interaction control can be performed, thereby improving the efficiency of vehicle-grid interaction control.

[0111] Another embodiment of the present invention provides a vehicle-grid interactive control system based on charging pile data. For details, see Figure 2 , Figure 2 The following is a structural block diagram of a vehicle-grid interactive control system based on charging pile data in one embodiment of the present invention. The system includes:

[0112] The vehicle-grid interaction data collection module 11 is used to collect the power data of each adjustable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time;

[0113] The charging pile load prediction module 12 is used to train a charging pile load prediction model based on all charging load data, all location data and all environmental data using a long short-term memory neural network, and predict the charging load of each charging pile using the trained charging pile load prediction model to obtain load prediction data for each charging pile;

[0114] The vehicle-grid interaction evaluation module 13 is used to obtain the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile; the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time are processed to calculate the vehicle-grid interaction suitability index of each charging pile;

[0115] The vehicle-grid interaction control module 14 is used to execute a vehicle-grid interaction control strategy corresponding to the matching result based on the matching degree between the vehicle-grid interaction suitability index of all charging piles, the power data of all controllable vehicles and the power supply load prediction data of each node.

[0116] 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 vehicle-grid interactive control method based on charging pile data, characterized in that: The method comprises: Collect the power data of each adjustable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time respectively; Using a long short-term memory neural network to train a charging pile load prediction model based on all the charging load data, all the location data and all the environmental data, and using the trained charging pile load prediction model to predict the charging load of each charging pile to obtain load prediction data for each charging pile; Obtaining the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile; processing the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time, and calculating the vehicle-grid interaction suitability index of each charging pile; Based on the matching degree between the vehicle-grid interaction suitability index of all the charging piles, the power data of all the controllable vehicles and the determined power supply load forecast data of each of the nodes, a vehicle-grid interaction control strategy corresponding to the matching result is executed.

2. The vehicle-grid interactive control method based on charging pile data according to claim 1 is characterized in that: The obtaining of the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile includes: Arrange all the charging power data of each charging pile in ascending order according to the time sequence of collection to obtain a charging power time series of each charging pile, and use a mutation point detection algorithm to extract all mutation points in each charging power time series; Based on the difference characteristics of the change trends on both sides of all the mutation points in each of the charging power time series, the phase mutation points and non-stationary mutation points of each of the charging power time series are obtained; Dividing each of the charging power time series into a plurality of charging power time subsequences based on all the stage mutation points; The vehicle-grid interaction stability coefficient of each of the charging power time subsequences is calculated according to the deviation characteristics of all the non-stationary mutation points in each of the charging power time subsequences.

3. The vehicle-grid interactive control method based on charging pile data according to claim 1 is characterized in that: The processing of the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data, and the charging load data of each charging pile at the current collection time to calculate the vehicle-grid interaction suitability index of each charging pile includes: If the charging load data of the charging pile at the current collection time is less than or equal to the load forecast data, the vehicle-grid interaction suitability index of the charging pile is set to 0; If the charging load data of the charging pile at the current collection time is greater than the load forecast data, the charging load data, the load forecast data, and the vehicle-grid interaction stability coefficient of the charging pile are input into the vehicle-grid interaction evaluation expression to calculate the vehicle-grid interaction suitability index of the charging pile; The vehicle-grid interaction evaluation expression is: VA i =Norm((CL i -YL i )×AW i ) Among them, VA i is the vehicle-grid interaction suitability index of charging pile i, Norm() is the normalization function, CL i is the charging load data of charging pile i at the current collection time, YL i is the load forecast data of charging pile i at the current collection time, AW i is the vehicle-grid interaction stability coefficient of charging pile i.

4. The vehicle-grid interactive control method based on charging pile data according to claim 2 is characterized in that: The step of obtaining the phase mutation point and the non-stationary mutation point of each charging power time series based on the difference characteristics of the change trends on both sides of all the mutation points in each charging power time series includes: The least square method is used to perform curve fitting on the preset left neighborhood of each mutation point to obtain the left fitting parameters of each mutation point; the least square method is used to perform curve fitting on the preset right neighborhood of each mutation point to obtain the right fitting parameters of each mutation point; The absolute value of the difference between the left fitting parameter and the corresponding right fitting parameter of each mutation point is used as the trend difference index on both sides of each mutation point; the average value of the trend difference index on both sides of all the mutation points in each charging power time series is used as the trend difference reference index of each charging power time series; All the mutation points in each of the charging power time series that are greater than or equal to the trend difference reference index are taken as stage mutation points, and all the mutation points in each of the charging power time series that are less than the trend difference reference index are taken as non-stationary mutation points.

5. The vehicle-grid interactive control method based on charging pile data according to claim 2 is characterized in that: The calculating the vehicle-grid interaction stability coefficient of each charging power time subsequence according to the deviation characteristics of all the non-stationary mutation points in each charging power time subsequence includes: Taking each of the non-stationary mutation points as the center, construct a neighborhood window of each of the non-stationary mutation points, replace the element values ​​of each of the non-stationary mutation points in each of the charging power time subsequences with the average value of all element values ​​in the corresponding neighborhood window, and obtain a smooth mutation charging power sequence corresponding to each of the charging power time subsequences; Calculating the first Hurst index of each of the charging power time subsequences and the second Hurst index of the corresponding smooth mutation charging power sequence, inputting the second Hurst index and the corresponding first Hurst index into a stable evaluation expression, and calculating the vehicle-grid interaction stable coefficient of each of the charging power time subsequences; The stable evaluation expression is: CarW j =exp(-(Hst j ′-Hst j )) Among them, CarW j is the vehicle-grid interaction stability coefficient of the charging power time subsequence j, exp() is an exponential function with a natural constant as the base, Hst j is the first Hurst exponent of the charging power time subsequence j, Hst j ′ is the second Hurst exponent of the smooth mutation charging power sequence corresponding to the charging power time subsequence j.

6. A vehicle-grid interactive control system based on charging pile data, characterized in that: The system comprises: The vehicle-grid interaction data collection module is used to collect the power data of each adjustable vehicle at the current collection time, the power supply load data of each node in the distribution network, the location data of each charging pile, and the charging power data, charging load data and environmental data at each collection time; A charging pile load prediction module, used to train a charging pile load prediction model based on all the charging load data, all the location data and all the environmental data using a long short-term memory neural network, and to predict the charging load of each charging pile using the trained charging pile load prediction model to obtain load prediction data for each charging pile; A vehicle-grid interaction evaluation module is used to obtain the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile; the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data and the charging load data of each charging pile at the current collection time are processed to calculate the vehicle-grid interaction suitability index of each charging pile; The vehicle-grid interaction control module is used to execute a vehicle-grid interaction control strategy corresponding to the matching result based on the matching degree between the vehicle-grid interaction suitability index of all the charging piles, the power data of all the controllable vehicles and the determined power supply load forecast data of each of the nodes.

7. The vehicle-grid interactive control system based on charging pile data according to claim 6 is characterized in that: The obtaining of the vehicle-grid interaction stability coefficient of each charging pile according to the volatility characteristics of the charging power data of each charging pile includes: Arrange all the charging power data of each charging pile in ascending order according to the time sequence of collection to obtain a charging power time series of each charging pile, and use a mutation point detection algorithm to extract all mutation points in each charging power time series; Based on the difference characteristics of the change trends on both sides of all the mutation points in each of the charging power time series, the phase mutation points and non-stationary mutation points of each of the charging power time series are obtained; Dividing each of the charging power time series into a plurality of charging power time subsequences based on all the stage mutation points; The vehicle-grid interaction stability coefficient of each of the charging power time subsequences is calculated according to the deviation characteristics of all the non-stationary mutation points in each of the charging power time subsequences.

8. The vehicle-grid interactive control system based on charging pile data according to claim 6 is characterized in that: The processing of the vehicle-grid interaction stability coefficient of each charging pile, the load forecast data, and the charging load data of each charging pile at the current collection time to calculate the vehicle-grid interaction suitability index of each charging pile includes: If the charging load data of the charging pile at the current collection time is less than or equal to the load forecast data, the vehicle-grid interaction suitability index of the charging pile is set to 0; If the charging load data of the charging pile at the current collection time is greater than the load forecast data, the charging load data, the load forecast data, and the vehicle-grid interaction stability coefficient of the charging pile are input into the vehicle-grid interaction evaluation expression to calculate the vehicle-grid interaction suitability index of the charging pile; The vehicle-grid interaction evaluation expression is: VA i =Norm((CL i -YL i )×AW i ) Among them, VA i is the vehicle-grid interaction suitability index of charging pile i, Norm() is the normalization function, CL i is the charging load data of charging pile i at the current collection time, YL i is the load forecast data of charging pile i at the current collection time, AW i is the vehicle-grid interaction stability coefficient of charging pile i.

9. The vehicle-grid interactive control system based on charging pile data according to claim 7 is characterized in that: The step of obtaining the phase mutation point and the non-stationary mutation point of each charging power time series based on the difference characteristics of the change trends on both sides of all the mutation points in each charging power time series includes: The least square method is used to perform curve fitting on the preset left neighborhood of each mutation point to obtain the left fitting parameters of each mutation point; the least square method is used to perform curve fitting on the preset right neighborhood of each mutation point to obtain the right fitting parameters of each mutation point; The absolute value of the difference between the left fitting parameter and the corresponding right fitting parameter of each mutation point is used as the trend difference index on both sides of each mutation point; the average value of the trend difference index on both sides of all the mutation points in each charging power time series is used as the trend difference reference index of each charging power time series; All the mutation points in each of the charging power time series that are greater than or equal to the trend difference reference index are taken as stage mutation points, and all the mutation points in each of the charging power time series that are less than the trend difference reference index are taken as non-stationary mutation points.

10. The vehicle-grid interactive control system based on charging pile data according to claim 7, characterized in that: The calculating the vehicle-grid interaction stability coefficient of each charging power time subsequence according to the deviation characteristics of all the non-stationary mutation points in each charging power time subsequence includes: Taking each of the non-stationary mutation points as the center, construct a neighborhood window of each of the non-stationary mutation points, replace the element values ​​of each of the non-stationary mutation points in each of the charging power time subsequences with the average value of all element values ​​in the corresponding neighborhood window, and obtain a smooth mutation charging power sequence corresponding to each of the charging power time subsequences; Calculating the first Hurst index of each of the charging power time subsequences and the second Hurst index of the corresponding smooth mutation charging power sequence, inputting the second Hurst index and the corresponding first Hurst index into a stable evaluation expression, and calculating the vehicle-grid interaction stable coefficient of each of the charging power time subsequences; The stable evaluation expression is: CarW j =exp(-(Hst j ′-Hst j )) Among them, CarW j is the vehicle-grid interaction stability coefficient of the charging power time subsequence j, exp() is an exponential function with a natural constant as the base, Hst j is the first Hurst exponent of the charging power time subsequence j, Hst j ′ is the second Hurst exponent of the smooth mutation charging power sequence corresponding to the charging power time subsequence j.

Images