CN-122000967-A - Energy management method and system suitable for electric vehicle charging station

CN122000967ACN 122000967 ACN122000967 ACN 122000967ACN-122000967-A

Abstract

The application relates to an energy management method and system suitable for an electric automobile charging station, wherein the method comprises the steps of acquiring relevant electric parameters of the charging station and energy management data of the charging station; the method comprises the steps of building an energy management and learning network of the charging station, training the energy management and learning network to obtain a well-trained energy management and learning network, and managing the energy of the charging station through the well-trained energy management and learning network. According to the application, multi-parameter collaborative accurate scheduling is realized, the operation cost of the charging station can be reduced, the voltage fluctuation of the direct current bus is controlled, the SOC value of the super energy storage system is maintained in an optimal interval, the energy utilization efficiency and the stability of the power distribution network are improved, and the large-scale and intelligent operation requirements of the charging station are adapted.

Inventors

WANG XI
CHEN YANXIA
YU KAIAN
Mei Kuang
Yin Zhengsheng
ZHANG TING
WANG YUANHAO
LIU JIN
YE XIAOHAI

Assignees

武汉市充换电技术有限公司

Dates

Publication Date: 20260508
Application Date: 20251231

Claims (7)

1. An energy management method suitable for an electric vehicle charging station, comprising: Acquiring relevant electrical parameters of a charging station and energy management data of the charging station; building an energy management and learning network of the charging station; training the energy management learning network to obtain a well-trained energy management learning network; The energy of the charging station is managed through a well-trained energy management learning network.
2. The energy management method for an electric vehicle charging station of claim 1, wherein the obtained charging station-related electrical parameters include charging station dc bus voltage SOC value of super energy storage system Side charging load power of electric automobile Price of electricity And sampling time node , Wherein, the method comprises the steps of, Is the total sampling time; The energy management instructions of the charging station comprise the output power of the power distribution network Output power of super energy storage system Charging power of electric automobile , Wherein, the method comprises the steps of, Is the total number of sampling times.
3. The energy management method for an electric vehicle charging station of claim 2, wherein the energy management instructions for the charging station are determined by charging station related electrical parameters as a function of charging peak and valley charging demand.
4. The energy management method suitable for electric vehicle charging stations according to claim 1, characterized in that the calculation process of the energy management network of the charging station is set up as follows: building an energy management learning network input matrix by : (1) Wherein, the The transpose operation of the matrix is represented, The number of neurons being the input layer; constructing an energy management learning network target output matrix by : (2) Wherein, the Is the number of neurons of the hidden layer; Related parameters of initial learning network are enveloped and input into weight matrix ; Calculating an output matrix of the hidden layer by : (3) By solving the following Estimating an output weight matrix : (4) Wherein, the The norm of l 2 is indicated, Representing the adaptive weight coefficient of the model, Representing the adaptive control coefficient; The calculation can be performed by the following formula, namely: (5) Wherein, the Is about Is the process coefficient of (1), namely: (6) Output weight matrix After being estimated, a well-trained energy management learning network is obtained.
5. The energy management method for an electric vehicle charging station of claim 4, wherein the optimal output weight matrix is obtained by solving equation (5) by an adaptive optimization algorithm The process of (2) is as follows: Initializing relevant parameters, including optimizing the size of the population Maximum number of iterations ; Defining iterative tags Setting up =1; Initializing each optimized individual at Space vector position at multiple iterations , ; Calculating the presence of each optimized individual by the formula (5) Adaptation value at time , ; Calculating the position of each optimized individual Information concentration at multiple iterations The method comprises the following steps: (7) Wherein, the And Respectively the first Maximum and minimum values of the adaptation values for the multiple iterations, Is the deviation coefficient; Calculating the attraction degree of other optimized individuals to the nth individual by the following method : (8) Wherein, the For the initial degree of attraction, the suction force, =1; The space vector position distance between the nth individual and the p th individual is that: (9) judging the current iteration times Whether or not it is greater than the maximum number of iterations If not, then If yes, jumping to step 9; updating each individual by Space vector at the next time, namely: (10) jumping to step 4; output weight matrix with optimal output 。
6. The energy management method for electric vehicle charging stations according to claim 1, wherein the charging station energy management process through a well-trained energy management learning network is: obtaining charging station related electrical parameters including charging station DC bus voltage SOC value of super energy storage system Side charging load power of electric automobile Price of electricity And sampling time node Constructing an input matrix ; Calculating an output matrix of the hidden layer by : (11) Wherein the output matrix of the output layer is calculated by : (12) By passing through Obtaining current energy management data of charging stations, i.e. output power of distribution network Output power of super energy storage system And electric automobile charging power 。
7. An energy management system for an electric vehicle charging station, comprising: the charging station parameter acquisition module is used for acquiring relevant electrical parameters of the charging station and energy management data of the charging station; The energy management and learning network module is used for building an energy management and learning network of the charging station; the network training module is used for training the energy management learning network to obtain an energy management learning network with good training; And the energy management control module is used for managing the energy of the charging station through a well-trained energy management learning network.

Description

Energy management method and system suitable for electric vehicle charging station Technical Field The invention relates to the technical field of energy management of electric vehicle charging stations, in particular to an energy management method and system suitable for electric vehicle charging stations. Background With the rapid increase of the conservation amount of electric vehicles, the construction scale and the use frequency of charging stations are continuously improved, and the scientificity and the high efficiency of energy management become core problems of industry attention. Currently, the traditional charging station energy management is dependent on a fixed control strategy or simple logic judgment, and has obvious limitations that on one hand, the energy distribution lacks flexibility due to incapability of dynamically adapting to multidimensional electrical parameters such as charging station direct current bus voltage fluctuation, super energy storage system SOC value change, electric vehicle charging load power difference and the like, and on the other hand, the energy distribution is optimized without fully combining peak-valley electricity price difference, so that the operation cost of the charging station is increased, and unstable impact is possibly brought to a power distribution network. In addition, the energy management model capable of adaptively learning the multi-parameter association relation is lacking in the prior art, and the matching relation among the output power of the power distribution network, the charging and discharging power of the super energy storage system and the charging power of the electric automobile is difficult to dynamically adjust according to real-time working conditions, so that the problems of low energy storage resource utilization rate, insufficient charging efficiency and the like are caused, and the requirements of large-scale and intelligent operation of the charging station cannot be met. Therefore, a charging station energy management scheme capable of precisely integrating multiple source parameters and adaptively optimizing energy distribution is needed. Disclosure of Invention Aiming at the defects or improvement demands of the prior art, the energy management method and the system suitable for the electric automobile charging station are provided, so that accurate energy scheduling under multi-parameter cooperation is realized, the operation efficiency of the charging station is improved, the cost is reduced, and the stable operation of a power distribution network is ensured. In order to achieve the above purpose, the present application provides the following technical solutions: In a first aspect, an embodiment of the present application provides an energy management method suitable for an electric vehicle charging station, including: Acquiring relevant electrical parameters of a charging station and energy management data of the charging station; building an energy management and learning network of the charging station; training the energy management learning network to obtain a well-trained energy management learning network; The energy of the charging station is managed through a well-trained energy management learning network. Acquired charging station-related electrical parameters, including charging station DC bus voltageSOC value of super energy storage systemSide charging load power of electric automobilePrice of electricityAnd sampling time node,Wherein, the method comprises the steps of,Is the total sampling time; The energy management instructions of the charging station comprise the output power of the power distribution network Output power of super energy storage systemCharging power of electric automobile,Wherein, the method comprises the steps of,Is the total number of sampling times. The energy management instructions of the charging station are determined by charging station related electrical parameters according to charging peak-time and valley-time charging requirements. The calculation process of the energy management and learning network for constructing the charging station is as follows: building an energy management learning network input matrix by : (1) Wherein, the The transpose operation of the matrix is represented,The number of neurons being the input layer; constructing an energy management learning network target output matrix by : (2) Wherein, the Is the number of neurons of the hidden layer; Related parameters of initial learning network are enveloped and input into weight matrix ; Calculating an output matrix of the hidden layer by: (3) By solving the followingEstimating an output weight matrix: (4) Wherein, the The norm of l 2 is indicated,Representing the adaptive weight coefficient of the model,Representing the adaptive control coefficient; The calculation can be performed by the following formula, namely: (5) Wherein, the Is aboutIs the process coefficient of (1), namely: (6) Output weight m