Urban distributed source network load storage scheduling optimization method based on LSTM algorithm

Abstract

The invention discloses an urban distributed source network load storage scheduling optimization method based on an LSTM (Long Short Term Memory) algorithm in the field of load regulation. The method comprises the following steps: 1, acquiring historical load data of a medium-voltage line through a power system; 2, an available historical load data sequence is cleaned and reserved, and a line load prediction model is established based on an LSTM algorithm; 3, inputting the real-time load of the line at the current moment into the line load prediction model, and performing real-time prediction to obtain a line load value at the next moment; step 4, obtaining an adjustable load interval of the charging station corresponding to the current line; 5, according to the line load value obtained through prediction in the step 3, the optimal load adjustment method of the charging station in the adjustable load interval at the next moment is obtained based on the particle swarm optimization; and 6, the charging station executes the load adjustment method in the step 5, and the steps 3-6 are repeated. According to the invention, the operation efficiency of the distribution network line after large-scale access of various distributed source network load storage can be effectively improved, the network loss is reduced, and the load fluctuation is stabilized.

Description

technical field [0001] The invention relates to the field of load regulation, in particular to an LSTM algorithm-based scheduling optimization method for urban distributed source network load storage. Background technique [0002] At present, new energy has clearly defined its dominant position in the future power system. In this context, new energy sources such as wind power, photovoltaics, and energy storage will all achieve explosive growth, and electric vehicles and related industries will also enter a stage of doubling. [0003] With the continuous expansion of various distributed energy sources, energy storage and charging piles connected to the distribution network, different types of energy units have different power consumption characteristics, which may cause power reverse transmission, large fluctuations in load levels, and energy waste. The problem. For example, weather factors have a great impact on photovoltaic power generation, which will lead to randomness ...

AI Technical Summary

Problems solved by technology

[0003] With the continuous expansion of various distributed energy sources, energy storage and charging piles connected to the distribution network, different types of energy units have different power consumption characteristics, which may cause power reverse transmission, large fluctuations in load levels, and energy waste. The problem

For example, weather factors have a great impact on photovoltaic power generation, which will cause the load of distribution network lines with photovoltaics to be prone to randomness and fluctuation, so that the line load is unbalanced

The superposition of electric vehicle charging time or the charging behavior during peak load hours will increase the burden on the distribution network

However, if only energy storage is used to balance the volatility of renewable energy such as wind and light, it will be costly and difficult to achieve.

[0004] Therefore, the large-scale access of photovoltaic stations, energy storage and charging piles under the distribution network has a greater impact on the load level of the distribution network, which greatly increases the difficulty of the economic operation and safety management of the power system, and restricts the distribution network. Further integration and consumption of renewable energy, and currently there is no effective method for coordinated scheduling of various distributed energy sources, electric vehicle charging and energy storage

[0005] To sum up, the applicant proposes an improvement plan for the problem that the distributed source-network-load-storage coordination and optimization scheduling for photovoltaic power stations and electric vehicle charging stations under the distribution network line is relatively small.

Images