CN-122026461-A - Multi-time scale elastic response and rolling optimization regulation and control method for electric automobile

CN122026461ACN 122026461 ACN122026461 ACN 122026461ACN-122026461-A

Abstract

The invention discloses an electric automobile multi-time scale elastic response and rolling optimization regulation and control method, and relates to the technical field of electric automobiles. The method comprises the steps of obtaining historical travel data of an electric vehicle based on a Monte Carlo random sampling method to generate a historical travel data set, generating a wind-light power prediction result by adopting a wind-light power prediction model, establishing a double-target model based on micro-grid operation cost and vehicle owner charging cost according to the historical travel data set, obtaining a basic regulation strategy by combining the wind-light power prediction result, setting a rolling time window, dynamically adjusting an energy storage charge-discharge power boundary and an electric vehicle charge-discharge interval based on ultra-short-term load prediction data, correcting the basic regulation strategy in real time, solving a corrected optimization problem by adopting a parallel optimization algorithm, and forming a multi-time-scale regulation strategy by minimizing deviation caused by prediction errors. The method can reduce the influence of prediction errors and improve the running stability and the wind-solar energy absorption rate of the micro-grid.

Inventors

ZHANG LIANG
LV LING
WANG DIANBIN
JIANG LIYANG
Dong Zhangkuan
ZHENG ZHI
LI WENHUI
YANG MENGWEI

Assignees

东北电力大学

Dates

Publication Date: 20260512
Application Date: 20260130

Claims (8)

1. The multi-time scale elastic response and rolling optimization regulation and control method for the electric automobile is characterized by comprising the following steps of: acquiring historical trip data of the electric automobile based on a Monte Carlo random sampling method, and generating a historical trip data set; generating a wind-light power prediction result by adopting a wind-light power prediction model; According to the historical trip data set, a double-target model based on the running cost of the micro-grid and the charging cost of the vehicle owner is established, and a basic regulation strategy is obtained by combining a wind-light power prediction result; Setting a rolling time window, dynamically adjusting an energy storage charging and discharging power boundary and an electric vehicle charging and discharging interval based on ultra-short-term load prediction data, and correcting the basic regulation strategy in real time; and solving the corrected optimization problem by adopting a parallel optimization algorithm, and forming a multi-time scale regulation strategy by minimizing deviation caused by a prediction error.
2. The method for regulating and controlling multi-time scale elastic response and rolling optimization of the electric automobile according to claim 1, wherein a time sequence decomposition algorithm and an improved LSTM neural network wind-light power prediction model are fused; collecting historical wind-solar power data, generating a prediction basic data set, and preprocessing the prediction basic data set; Decomposing the preprocessed wind-solar power data into trend terms, period terms and random terms through empirical mode decomposition, and separating fluctuation features of different scales; respectively predicting wind-light power sequences of trend items, period items and random items; and adopting a weighted fusion strategy, distributing weights according to the reciprocal of each component prediction error, and fusing each component prediction result to obtain a wind-solar power prediction curve.
3. The method for regulating and controlling the multi-time scale elastic response and rolling optimization of the electric automobile according to claim 1, wherein the objective function of the running cost of the micro-grid is constructed as follows: ; Wherein, the For the wind-solar energy power-off cost, the calculation formula is as follows: ; the cost is lost in terms of unit waste electricity, The power is predicted for the photovoltaic at time t, The power is predicted for the wind power at the moment t, The micro-grid can absorb power for the moment t; For the energy storage charge and discharge loss cost, the calculation formula is as follows: ; For the unit loss cost of the battery, The energy storage and charging power is stored at the moment t, The energy storage and discharge power is the time t, In order to achieve the energy storage and charging efficiency, The energy storage and discharge efficiency is achieved; For the interactive electricity price cost of the power grid, the calculation formula is as follows: ; the power is purchased for the power grid at the moment t, For the electricity price of electricity purchase at the moment t, The electric power is sold for the power grid at the moment t, Electricity price is sold at time t; For spare capacity cost, the calculation formula is: ; Is the cost per unit of standby, And (t) is the standby power of the micro-grid at the moment t.
4. The method for optimizing and regulating multi-time scale elastic response and rolling of an electric automobile according to claim 3, wherein constructing the objective function of the charging cost of the automobile owner comprises: ; Wherein, the For charging electricity charge, the calculation formula is as follows: ; N is the number of electric vehicles participating in regulation and control, (T) is the charging power of the ith vehicle at the moment t, (T) is the charging electricity price at the moment t, Is a time interval; For battery loss cost, based on charge and discharge depth, cycle life and initial cost conversion of the battery, the calculation formula is as follows: ; for the battery cycle life of the ith vehicle after the interactive discharge of the vehicle network, For battery cycle life for which the i-th vehicle is traveling only, For the initial cost of the battery for the ith vehicle, The discharge power of the ith vehicle at the time t, The discharge time length of the ith vehicle; for the equivalent cost of the charging waiting time, the calculation formula is as follows: ; for the charge waiting period of the i-th vehicle, Is the time value coefficient of the ith vehicle owner.
5. The method for multi-time scale elastic response and rolling optimization regulation and control of an electric automobile according to claim 4, wherein the method for solving the optimization problem by adopting a genetic algorithm based on an objective function and constraint conditions to generate a basic regulation and control strategy comprises the following steps: setting algorithm parameters including population scale, iteration times, crossover probability and variation probability; adopting real number coding, wherein each chromosome corresponds to a group of electric vehicle charging and discharging power and energy storage charging and discharging power regulation scheme; taking the inverse of the converted single objective function value as an fitness function; Adopting a roulette selection method, wherein the probability of being selected by an individual with higher fitness is higher; a single-point crossover is adopted, and crossover points are randomly selected to exchange partial genes of two individuals; Carrying out random disturbance on individual genes by adopting Gaussian variation; and after iteration convergence, outputting an optimal regulation scheme as a basic regulation strategy.
6. The method for regulating and controlling multi-time scale elastic response and rolling optimization of the electric automobile according to claim 1, wherein a gradient lifting tree algorithm is adopted for ultra-short-term load prediction; Selecting historical load data, real-time load data, meteorological data, date type and time period characteristics as input characteristics, and analyzing and eliminating redundant characteristics through pearson correlation coefficients; Optimizing model super parameters by adopting a grid search method; Outputting a load predicted value in the rolling window according to the predicted period; Model parameters are updated by using actual load data feedback, and the prediction effect is optimized; and dynamically adjusting the energy storage charging and discharging power boundary and the charging and discharging interval of the electric automobile according to the deviation of the ultra-short-term load prediction data and the day-ahead prediction data.
7. The method for regulating and controlling multi-time scale elastic response and rolling optimization of an electric automobile according to claim 1, wherein the method for regulating and controlling multi-time scale by solving the corrected optimization problem by adopting a parallel optimization algorithm and forming a multi-time scale regulation and control strategy by minimizing deviation caused by prediction error comprises the following steps: decomposing the corrected optimization problem into two sub-problems of a micro-grid side and an electric vehicle user side; Adopting an improved particle swarm algorithm, and introducing an inertia weight self-adaptive adjustment strategy to solve the micro-grid side; solving the user side of the electric automobile through the self-adaptive cross variation probability by adopting a genetic algorithm; After iteration is finished, fusing optimization results at two sides to obtain a global optimal solution; and integrating the basic regulation strategy and the dynamic optimization result to obtain the multi-time scale regulation strategy.
8. The multi-time scale elastic response and rolling optimization regulation method of the electric automobile according to claim 7 is characterized in that the micro-grid side sub-problem aims at minimizing running cost and prediction deviation, and the energy storage charge and discharge power and the power grid interaction power are optimized; the problem of the user side of the electric automobile aims at minimizing the charging cost of the automobile owner and the satisfaction of travel constraint, and the charging and discharging time period and the power of a single automobile are optimized.

Description

Multi-time scale elastic response and rolling optimization regulation and control method for electric automobile Technical Field The invention relates to the technical field of electric automobiles, in particular to a multi-time scale elastic response and rolling optimization regulation method of an electric automobile. Background With the rapid development of new energy automobile industry, the electric automobile conservation amount continuously rises, and a large amount of adjustable resources are provided for micro-grid dispatching. The electric automobile has charge and discharge flexibility, participates in micro-grid regulation and control through a vehicle-network interaction technology, can effectively stabilize wind-light power generation fluctuation, and improves the running stability and economy of the micro-grid. However, the existing electric automobile regulation and control method still has a plurality of defects: Firstly, the travel behavior of the electric automobile has the characteristics of randomness and volatility, the travel data is processed by adopting a simple statistical method in the existing method, and the representativeness of a sampling sample is insufficient, so that the regulation strategy is disjointed from the actual travel demand, and the travel rights and interests of the automobile owners are difficult to guarantee. Secondly, wind-solar power generation power is obviously influenced by meteorological factors, prediction errors are large, and a single prediction model is difficult to meet the precision requirement of a basic regulation strategy, so that the regulation effect is influenced. Thirdly, the existing regulation and control strategies mostly adopt single time scale scheduling, lack of dynamic correction mechanisms, cannot effectively offset the influence caused by short-term prediction errors and load fluctuation, and are poor in regulation and control flexibility and adaptability. Fourth, the regulation and control target multi-emphasis micro-grid operation economy, neglecting benefit demands such as the charging cost of the vehicle owners and the battery loss, and leading the participation enthusiasm of the vehicle owners to be low, and being difficult to realize multi-party benefit balance. Therefore, aiming at the defects of the prior art, how to provide a multi-time scale elastic response and rolling optimization regulation method for an electric automobile is a problem to be solved by the technicians in the field. Disclosure of Invention In view of the above, the invention provides a multi-time scale elastic response and rolling optimization regulation method for an electric automobile, which has the advantages of balancing data reliability, prediction precision, multi-time scale suitability and multiparty interests, reducing the influence of prediction errors and improving the running stability and wind-solar absorption rate of a micro-grid. In order to achieve the purpose, the technical scheme adopted by the invention is that the multi-time scale elastic response and rolling optimization regulation method of the electric automobile comprises the following steps: acquiring historical trip data of the electric automobile based on a Monte Carlo random sampling method, and generating a historical trip data set; generating a wind-light power prediction result by adopting a wind-light power prediction model; According to the historical trip data set, a double-target model based on the running cost of the micro-grid and the charging cost of the vehicle owner is established, and a basic regulation strategy is obtained by combining a wind-light power prediction result; Setting a rolling time window, dynamically adjusting an energy storage charging and discharging power boundary and an electric vehicle charging and discharging interval based on ultra-short-term load prediction data, and correcting the basic regulation strategy in real time; and solving the corrected optimization problem by adopting a parallel optimization algorithm, and forming a multi-time scale regulation strategy by minimizing deviation caused by a prediction error. Preferably, a time sequence decomposition algorithm is fused with an improved LSTM neural network wind-light power prediction model; collecting historical wind-solar power data, generating a prediction basic data set, and preprocessing the prediction basic data set; Decomposing the preprocessed wind-solar power data into trend terms, period terms and random terms through empirical mode decomposition, and separating fluctuation features of different scales; respectively predicting wind-light power sequences of trend items, period items and random items; and adopting a weighted fusion strategy, distributing weights according to the reciprocal of each component prediction error, and fusing each component prediction result to obtain a wind-solar power prediction curve. Preferably, a microgrid operation cost objective function is cons