A Method and System for Predicting the Total Charging Load of Electric Vehicles

Abstract

The present invention provides a method for predicting the total amount of electric vehicle charging load, including detecting whether there are changes in the characteristics of electric vehicle charging behavior, including electricity price policy, battery technology, charging facilities and charging technology; if at least one change is detected, a preset first electric vehicle ownership determination mode is selected to obtain the ownership of various types of electric vehicles; if no change is detected, a preset second electric vehicle ownership determination mode is selected to obtain the ownership of various types of electric vehicles; according to the ownership of various types of electric vehicles, and in combination with the daily charging load of various types of electric vehicles, the power of various types of electric charging piles is obtained, and further according to the power of various types of electric charging piles, the total amount of electric vehicle charging load is predicted. The implementation of the present invention clarifies the impact of the access of electric vehicles and their charging facilities on the distribution network, simplifies the behavior simulation of a large number of electric vehicles, and overcomes the problem of too many charging time parameters involved in traditional prediction methods.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid load forecasting, and particularly to a method and system for forecasting the total load of electric vehicle charging. Background Art

[0002] For the regional distribution network planning work, power demand forecasting is one of the most core basic tasks. Its forecasting results are the basis for guiding and supporting the work of regional distribution network power balance, substation site selection and capacity determination, medium and high voltage distribution network planning, investment allocation, etc., and have a decisive impact on the power supply capacity, power supply quality, construction economy, etc. during the regional distribution network planning year. Therefore, the scientificity and rationality of the power demand forecasting method, and the accuracy and applicability of the forecasting results have extremely important guiding significance for the development of regional distribution network planning and construction work.

[0003] Under the influence of the new situation of the development of the distribution network such as the rapid popularization of electric vehicles, the energy consumption structure and load characteristics of the traditional distribution network have changed significantly. However, the existing methods for forecasting the total load of electric vehicle charging obviously cannot meet the power consumption demands of the large-scale development of electric vehicle charging facilities. In particular, for the urban distribution network of extra-large scale, once there are deviations in the rationality and applicability of the distribution network planning scheme, the adverse impacts and economic losses caused to the regional social development will be extremely huge.

[0004] For example, using historical data for model extrapolation (trend extrapolation method), under the influence of the large-scale access of electric vehicle charging facilities, the load level and load change curve of the regional distribution network have changed significantly. The historical annual data does not include such changes during the fitting process. Therefore, the trend extrapolation results based on historical data cannot reflect the development and changes of the regional electric vehicle charging load. Another example is using the load density index for load distribution forecasting. The forecasting results generally can only represent the load level of the target year and cannot predict the annual changes of the intermediate years. Moreover, this forecasting method involves a large amount of complex data calculations and is difficult to complete only by manual work. It is necessary to rely on professional computer-aided software to ensure its forecasting efficiency and accuracy. Another example is using data such as economy and population as the main parameters for forecasting. This method uses the correlation between power grid construction and social economy and population for forecasting. As a new energy supply and consumption structure in the distribution network, the development law of electric vehicle charging facilities has not been reflected in the traditional distribution network. Therefore, traditional forecasting methods are not applicable. Another example is forecasting "from bottom to top" mainly based on user installation. From the perspective of the development of electric vehicles, this method is one of the forecasting means that is most likely to accurately forecast the total level of charging load. However, its forecasting results lack the description of the load characteristics and related indicators of the regional power grid. In the actual construction and management process of the distribution network, the supporting and decision-making roles of its forecasting results are not obvious.

[0005] Therefore, there is an urgent need for a load forecasting method suitable for the new situation and background of the development of electric vehicles, which clarifies the impact of the access of electric vehicles and their charging facilities on the distribution network, simplifies the behavior simulation of a large number of electric vehicles, and overcomes the problem of excessive charging time parameters involved in traditional forecasting methods. Summary of the Invention

[0006] The technical problem to be solved by the embodiments of the present invention is to provide a method and system for forecasting the total charging load of electric vehicles, which clarifies the impact of the access of electric vehicles and their charging facilities on the distribution network, simplifies the behavior simulation of a large number of electric vehicles, and overcomes the problem of excessive charging time parameters involved in traditional forecasting methods.

[0007] To solve the above technical problem, an embodiment of the present invention provides a method for forecasting the total charging load of electric vehicles, and the method includes the following steps:

[0008] Detect whether there is a change in the charging behavior characteristics of electric vehicles; wherein, the charging behavior characteristics of electric vehicles include electricity price policy, battery technology, charging facilities and charging technology;

[0009] If at least one change is detected, select a preset first electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles; if no change is detected, select a preset second electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles;

[0010] According to the ownership of various types of electric vehicles, and in combination with the daily charging load of various types of electric vehicles, obtain the power of various types of electric vehicle charging piles, and further predict the total charging load of electric vehicles based on the power of various types of electric vehicle charging piles.

[0011] Among them, the step of selecting a preset first electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles is specifically:

[0012] Obtain the probability result of the charging load of electric vehicles at 24 o'clock, and calculate the ownership of various types of electric vehicles according to the probability result of the charging load of electric vehicles at 24 o'clock.

[0013] Among them, the probability result of the charging load of electric vehicles at 24 o'clock is calculated based on a pre-set charging probability occurrence model of electric vehicles at 24 o'clock; among them, the charging probability occurrence model of electric vehicles at 24 o'clock is constructed based on the statistical charging probabilities of various types of electric vehicles at 24 o'clock in battery replacement stations, residential area charging stations and public place charging stations, and three charging probability occurrence curves obtained by Monte Carlo simulation.

[0014] Among them, the steps of selecting a preset second electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles specifically include:

[0015] Obtain the typical charging load data of various types of electric vehicles to analyze the charging habits and charging load characteristics of various types of electric vehicles, and based on the charging habits and charging load characteristics of various types of electric vehicles, obtain the occurrence probabilities of the charging loads of various types of electric vehicles, and further calculate the ownership of various types of electric vehicles according to the occurrence probabilities of the charging loads of various types of electric vehicles.

[0016] Among them, the steps of obtaining the occurrence probabilities of the charging loads of various types of electric vehicles according to the charging habits and charging load characteristics of various types of electric vehicles are specifically as follows:

[0017] According to the charging habits and charging load characteristics of various types of electric vehicles, construct the charging load characteristic curves of various types of electric vehicles, and fit the charging load characteristic curves of various types of electric vehicles with Gaussian distribution to obtain the Gaussian distribution fitting curves corresponding to the charging load characteristics of various types of electric vehicles, and further solve each Gaussian distribution fitting curve to obtain the occurrence probabilities of the charging loads of various types of electric vehicles.

[0018] An embodiment of the present invention further provides an electric vehicle charging total load prediction system, including:

[0019] A detection unit for detecting whether there are changes in the charging behavior characteristics of electric vehicles; among them, the charging behavior characteristics of electric vehicles include electricity price policies, battery technologies, charging facilities, and charging technologies;

[0020] A calculation unit for, if at least one change is detected, selecting a preset first electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles; if no change is detected, selecting a preset second electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles;

[0021] A prediction unit for, according to the ownership of various types of electric vehicles and in combination with the daily charging loads of various types of electric vehicles, obtaining the power of various types of electric vehicle charging piles, and further predicting the total charging load of electric vehicles according to the power of various types of electric vehicle charging piles.

[0022] Among them, the calculation unit includes:

[0023] A first calculation module for obtaining the results of the occurrence probabilities of the charging loads of electric vehicles at 24 o'clock and calculating the ownership of various types of electric vehicles according to the results of the occurrence probabilities of the charging loads of electric vehicles at 24 o'clock;

[0024] A second calculation module, configured to obtain typical charging load data of various types of electric vehicles, analyze the charging habits and charging load characteristics of various types of electric vehicles, obtain the occurrence probabilities of the charging loads of various types of electric vehicles according to the charging habits and charging load characteristics of various types of electric vehicles, and further calculate the ownership of various types of electric vehicles according to the occurrence probabilities of the charging loads of various types of electric vehicles.

[0025] Among them, the result of the occurrence probability of the charging load of the electric vehicle at the 24th moment is calculated based on a pre-set occurrence probability model of the charging probability of the electric vehicle at the 24th moment; among them, the occurrence probability model of the charging probability of the electric vehicle at the 24th moment is constructed based on Monte Carlo simulation of three charging probability occurrence curves after statistics of the charging probabilities of various types of electric vehicles at the 24th moment in battery swapping stations, residential area charging stations, and public place charging stations.

[0026] Implementing the embodiments of the present invention has the following beneficial effects:

[0027] 1. Based on the analysis of the charging behavior characteristics of electric vehicles, the present invention combines the existing load prediction method with the quantitative analysis results of the impact of the access of electric vehicles, constructs a load prediction parameter index system suitable for the development of electric vehicles, and finally optimizes the load prediction method for the distribution network planning, thereby optimizing and enhancing the rationality and applicability of the load prediction work of the regional distribution network, clarifying the impact of the access of electric vehicles and their charging facilities on the distribution network, simplifying the behavior simulation of a large number of electric vehicles, and overcoming the problem of too many charging time parameters involved in the traditional prediction method.

[0028] 2. Based on the typical charging load characteristic curves of various types of electric vehicles, the present invention determines the overall charging load law with time series characteristics of various types of electric vehicles in the distribution network through Gaussian curve simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, obtaining other drawings based on these drawings still belongs to the scope of the present invention.

[0030] Figure 1 It is a flowchart of a method for predicting the total charging load of electric vehicles provided by an embodiment of the present invention.

[0031] Figure 2 It is a schematic structural diagram of a system for predicting the total charging load of electric vehicles provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.

[0033] like Figure 1 As shown in the figure, a method for predicting the total charging load of an electric vehicle is proposed in an embodiment of the present invention, and the method comprises the following steps:

[0034] Step S1, detecting whether there is a change in the charging behavior characteristics of the electric vehicle; wherein the charging behavior characteristics of the electric vehicle include electricity price policy, battery technology, charging facilities and charging technology;

[0035] Step S2: if at least one change is detected, the preset first electric vehicle ownership determination mode is selected to obtain the ownership of each type of electric vehicles; if no change is detected, the preset second electric vehicle ownership determination mode is selected to obtain the ownership of each type of electric vehicles;

[0036] Step S3: According to the number of electric vehicles in each category and in combination with the daily charging load of each category of electric vehicles, the power of each category of electric charging piles is obtained, and further according to the power of each category of electric charging piles, the total charging load of the electric vehicles is predicted.

[0037] The specific process is that before step S1, a 24-hour charging probability model for electric vehicles is pre-set, and the model is constructed based on three charging probability curves obtained by Monte Carlo simulation after statistics on the 24-hour charging probabilities of various electric vehicles in battery replacement stations, residential charging stations and public charging stations.

[0038] In step S1, combined with the actual application of electric vehicles, the operating characteristics of the electric vehicle charging facilities themselves are sorted out in detail, and their main load characteristics are clarified to determine the research basis for optimizing and improving the load forecasting method. Different types of electric vehicles have certain differences in their charging behaviors. In order to study the characteristics of electric vehicle charging loads, it is necessary to analyze the characteristics of electric vehicle charging behavior, that is, it is necessary to detect whether the characteristics of electric vehicle charging behavior have changed. Among them, the characteristics of electric vehicle charging behavior include electricity price policy, battery technology, charging facilities, charging technology and other influencing factors.

[0039] It should be noted that the load characteristics of charging facilities are the result of the combined effects of multiple factors, but through sorting and summarizing, various factors can be summarized into two directions, namely, the type of electric vehicle and the charging method of electric vehicles.

[0040] In step S2, considering that the level of charging car ownership is an important factor affecting the load level of the power grid, the larger the ownership scale, the greater the possibility of increasing the maximum load level of the power grid. However, orderly charging management can effectively alleviate the impact of electric vehicle charging load on the power grid load, and scientific and reasonable management of electric vehicle charging behavior can further optimize the regional load characteristic level, narrow the load peak-valley gap, and further improve the economic efficiency of distribution network operation. Therefore, it is necessary to calculate the ownership of various types of electric vehicles based on whether the charging behavior characteristics of electric vehicles have changed, so as to obtain the power of various types of electric charging piles to predict the total charging load of electric vehicles.

[0041] Among them, the specific process of calculating the number of various types of electric vehicles is as follows:

[0042] (1) If at least one of the characteristics of the electric vehicle charging behavior is detected to have changed, the preset first electric vehicle ownership determination mode is selected to obtain the ownership of each type of electric vehicle. For example, based on the above-mentioned electric vehicle 24-hour charging probability occurrence model, the 24-hour charging load occurrence probability result of the electric vehicle is obtained, and based on the 24-hour charging load occurrence probability result of the electric vehicle, the ownership of each type of electric vehicle is calculated.

[0043] (2) If it is detected that all the characteristics of the charging behavior of the electric vehicle have not changed, the preset second electric vehicle ownership determination mode is selected to obtain the ownership of each type of electric vehicle. For example, typical charging load data of each type of electric vehicle is obtained to analyze the charging habits and charging load characteristics of each type of electric vehicle, and the probability of occurrence of charging load of each type of electric vehicle is obtained based on the charging habits and charging load characteristics of each type of electric vehicle, and further based on the probability of occurrence of charging load of each type of electric vehicle, the ownership of each type of electric vehicle is calculated.

[0044] It should be noted that the total amount of charging load of various types of electric vehicles changes over time and is highly similar to the result of a Gaussian distribution curve or the superposition of multiple Gaussian curves. Based on this research result, the typical charging load characteristic curves of various types of vehicles can be used as a basis, and the overall charging load law with time series characteristics of various types of electric vehicles in the distribution network can be determined through Gaussian curve simulation. Therefore, according to the charging habits and charging load characteristics of various types of electric vehicles, the charging load characteristic curves of various types of electric vehicles can be constructed, and the charging load characteristic curves of various types of electric vehicles are fitted with Gaussian distribution to obtain the Gaussian distribution fitting curves corresponding to the charging load characteristics of various types of electric vehicles, and further solve the Gaussian distribution fitting curves to obtain the probability of occurrence of charging loads of various types of electric vehicles.

[0045] In step S3, the number of electric vehicles of each type is multiplied by the daily charging load of each type of electric vehicle, and the products obtained are the power of each type of electric charging pile. The power of each type of electric charging pile is accumulated, and the sum obtained is the predicted value of the total charging load of electric vehicles, that is, the grid load.

[0046] It can be understood that the above steps realize the 24-hour charging load prediction of electric vehicles. Based on the 24-hour charging load prediction of electric vehicles, it can be combined with the annual load prediction result to determine the annual load prediction result of the region in the context of electric vehicle development.

[0047] like Figure 2 As shown, an electric vehicle charging total load prediction system is provided in an embodiment of the present invention, comprising:

[0048] The detection unit 110 is used to detect whether there is a change in the charging behavior characteristics of the electric vehicle; wherein the charging behavior characteristics of the electric vehicle include electricity price policy, battery technology, charging facilities and charging technology;

[0049] The calculation unit 120 is configured to select a preset first electric vehicle ownership determination mode to obtain the ownership of each type of electric vehicles if at least one change is detected; and select a preset second electric vehicle ownership determination mode to obtain the ownership of each type of electric vehicles if no change is detected;

[0050] The prediction unit 130 is used to obtain the power of various electric charging piles according to the number of various electric vehicles and the daily charging load of various electric vehicles, and further predict the total charging load of electric vehicles according to the power of various electric charging piles.

[0051] Wherein, the calculation unit 120 includes:

[0052] The first calculation module is used to obtain the probability result of the charging load of the electric vehicle at 24 hours, and calculate the number of various types of electric vehicles according to the probability result of the charging load of the electric vehicle at 24 hours;

[0053] The second calculation module is used to obtain typical charging load data of various electric vehicles to analyze the charging habits and charging load characteristics of various electric vehicles, and obtain the probability of occurrence of charging loads of various electric vehicles based on the charging habits and charging load characteristics of various electric vehicles, and further calculate the number of electric vehicles in use based on the probability of occurrence of charging loads of various electric vehicles.

[0054] Among them, the probability result of the charging load of the electric vehicle at 24 o'clock is calculated based on a pre-set charging probability occurrence model of the electric vehicle at 24 o'clock; among them, the charging probability occurrence model of the electric vehicle at 24 o'clock is constructed by Monte Carlo simulation based on the charging probability statistics of various electric vehicles at 24 o'clock in battery swapping stations, residential area charging stations and public charging stations, and three charging probability occurrence curves are obtained.

[0055] Implementing the embodiments of the present invention has the following beneficial effects:

[0056] 1. Based on the analysis of the charging behavior characteristics of electric vehicles, the present invention combines the existing load forecasting method with the quantitative analysis results of the impact of electric vehicle access, constructs a load forecasting parameter index system suitable for the development of electric vehicles, and finally optimizes the load forecasting method for distribution network planning, so as to optimize and improve the rationality and applicability of the load forecasting work of the regional distribution network, clarify the impact of the access of electric vehicles and their charging facilities on the distribution network, simplify the behavior simulation of a large number of electric vehicles, and overcome the problem of too many charging time parameters involved in traditional forecasting methods;

[0057] 2. Based on the typical charging load characteristic curves of various electric vehicles, the present invention determines the overall charging load law with time series characteristics of various electric vehicles in the distribution network through Gaussian curve simulation.

[0058] It should be noted that in the above system embodiments, the included units are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0059] Those of ordinary skill in the art can understand that all or part of the steps in implementing the above method embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium, such as ROM / RAM, disk, optical disc, etc.

[0060] The above-disclosed are only the preferred embodiments of the present invention, and of course cannot be used to limit the scope of the rights of the present invention. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A method for predicting the total charging load of electric vehicles, characterized in that, the method comprises the following steps: Detect whether there are changes in the charging behavior characteristics of electric vehicles; wherein, the charging behavior characteristics of electric vehicles include electricity price policies, battery technologies, charging facilities and charging technologies; If at least one change is detected, select a preset first electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles; if it is detected that all have not changed, select a preset second electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles; According to the ownership of various types of electric vehicles, and in combination with the daily charging load of various types of electric vehicles, obtain the power of various types of electric vehicle charging piles, and further predict the total charging load of electric vehicles based on the power of various types of electric vehicle charging piles; Among them, the step of selecting the preset first electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles is specifically: Obtain the charging load occurrence probability results of electric vehicles at 24 moments, and calculate the ownership of various types of electric vehicles according to the charging load occurrence probability results of electric vehicles at 24 moments; Among them, the step of selecting the preset second electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles specifically includes: Obtain the typical charging load data of various types of electric vehicles to analyze the charging habits and charging load characteristics of various types of electric vehicles, and obtain the charging load occurrence probabilities of various types of electric vehicles according to the charging habits and charging load characteristics of various types of electric vehicles, and further calculate the ownership of various types of electric vehicles according to the charging load occurrence probabilities of various types of electric vehicles.

2. The method for predicting the total charging load of electric vehicles according to claim 1, characterized in that, the charging load occurrence probability results of electric vehicles at 24 moments are calculated based on a preset charging probability occurrence model of electric vehicles at 24 moments; wherein, the charging probability occurrence model of electric vehicles at 24 moments is constructed by Monte Carlo simulation of three charging probability occurrence curves based on the charging probability statistics of various types of electric vehicles at 24 moments in battery replacement stations, residential area charging stations and public area charging stations.

3. The method for predicting the total charging load of electric vehicles according to claim 2, characterized in that, the step of obtaining the charging load occurrence probabilities of various types of electric vehicles according to the charging habits and charging load characteristics of various types of electric vehicles is specifically: Construct charging load characteristic curves for various types of electric vehicles according to the charging habits and charging load characteristics of various types of electric vehicles, and fit the charging load characteristic curves of various types of electric vehicles with Gaussian distribution to obtain Gaussian distribution fitting curves corresponding to the charging load characteristics of various types of electric vehicles, and further solve each Gaussian distribution fitting curve to obtain the charging load occurrence probabilities of various types of electric vehicles.

4. An electric vehicle total charging load prediction system, characterized in that, comprises; A detection unit for detecting whether there are changes in the charging behavior characteristics of electric vehicles; wherein, the charging behavior characteristics of electric vehicles include electricity price policies, battery technologies, charging facilities and charging technologies; A calculation unit, configured to select a preset first electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles if at least one change is detected; and select a preset second electric vehicle ownership determination mode to obtain the ownership of various types of electric vehicles if no change is detected. A prediction unit, configured to obtain the power of various types of electric vehicle charging piles according to the ownership of various types of electric vehicles and in combination with the daily charging load of various types of electric vehicles, and further predict the total charging load of electric vehicles according to the power of various types of electric vehicle charging piles. Wherein, the calculation unit includes: A first calculation module, configured to obtain the probability result of the charging load of electric vehicles at 24 hours, and calculate the ownership of various types of electric vehicles according to the probability result of the charging load of electric vehicles at 24 hours. A second calculation module, configured to obtain the typical charging load data of various types of electric vehicles to analyze the charging habits and charging load characteristics of various types of electric vehicles, obtain the probability of the charging load of various types of electric vehicles according to the charging habits and charging load characteristics of various types of electric vehicles, and further calculate the ownership of various types of electric vehicles according to the probability of the charging load of various types of electric vehicles.

5. The electric vehicle total charging load prediction system according to claim 4, characterized in that the probability result of the charging load of electric vehicles at 24 hours is calculated based on a preset charging probability occurrence model of electric vehicles at 24 hours; wherein, the charging probability occurrence model of electric vehicles at 24 hours is constructed by Monte Carlo simulation based on the charging probability statistics of various types of electric vehicles at 24 hours in battery replacement stations, residential area charging stations and public area charging stations.

Images