Research On Key Technologies Of Power Load Forecasting And Dispatching Optimization In Port Micro Grid

With the popularization of shore-to-ship power technology in the Yangtze River inner port,the stability and environmental protection requirements of shore-to-ship power are gradually improved,which also puts forward higher requirements for the management of port power grid.Due to the addition of shore-to-ship power load,the peak capacity of port power load is increased,and some cold chain and chemical transport ship loads require high power stability.Therefore,in this paper,for the port with shore-to-ship power supply system,the micro grid system of port with wind and solar distributed power supply and energy storage is designed,and the scheduling optimization method of micro grid is proposed.The main contents of this paper are as follows:1)This paper will analyze the demand of shore-to-ship power in the port,and the shore-to-ship power load is included in the port power grid system.The micro grid system of wind and solar distributed power supply and energy storage is designed,and the mathematical model and simulation modeling method of each component of the micro grid system are given.2)Power load forecasting is one of the most important tasks in micro grid scheduling optimization.This paper will focus on port power load forecasting.In order to solve the problem of port power load forecasting,the time series forecasting problem of load forecasting is transformed into supervised learning problem by machine learning technology.Based on the Long and short term memory neural network(LSTM),a load forecasting model is proposed.Aiming at the structural parameter selection of LSTM,simulated annealing algorithm(SA)is proposed to optimize the super parameters of LSTM.According to the historical load data,the future load data is predicted,and the load prediction data is used in the scheduling optimization of the micro grid.3)In the study of micro grid scheduling,the scheduling model of distributed generation and energy storage system power output is established with the minimum cost of micro grid as the scheduling goal,and the scheduling plan of each distributed generation is solved by genetic algorithm(GA).In the coding design of genetic algorithm,two-dimensional real genetic code is designed,and the crossover and mutation operation of individuals is improved.After each operation,the genetic code is adjusted to ensure that the scheduling scheme meets the micro grid operation constraints.In the optimization of genetic algorithm,an improved genetic algorithm based on iterative local search(ILS)is proposed to improve the local search ability of the algorithm.Finally,the simulation results show that the improved genetic algorithm based on iterative local search(ILS-GA)can effectively reduce the power cost of port micro grid.