A Neural Network Based Energy Scheduling Method for Distributed Microgrid Groups

Abstract

The invention discloses a neural network-based distributed microgrid group energy scheduling method. The neural network model is used to optimize the internal energy distribution of the system, and a central control unit is set in the sub-microgrid to centrally control the sub-microgrid itself. The single neuron model makes a unified decision on the output power of the sub-microgrid, making it a stable output (input) unit in the microgrid group; while at the level of the microgrid group, multi-agents are used to communicate between each sub-microgrid. Construct a distributed control structure between them, consider the loss in the energy transmission process, the external power demand of the microgrid group and the power demand of each microgrid within the microgrid group, and use the BP neural network to reasonably allocate the power inside the entire microgrid group. Reasonably deploy the energy in the microgrid group, so as to ensure the reliability of the control system of the microgrid group, effectively reduce the control nodes of the microgrid group, and improve the operating efficiency of the system.

Description

technical field [0001] The invention relates to a neural network-based centralized and distributed energy dispatching method for a microgrid group, which can make decisions on its own according to system internal load changes and uncertainties of distributed power sources, and belongs to the technical field of new energy and microgrids. Background technique [0002] The microgrid group is to interconnect geographically adjacent microgrids and distributed power generation systems to form a microgrid cluster system, which represents the future development trend of microgrids. However, the research on microgrids is still in its infancy. Faced with many challenges such as the impact of a large number of distributed power sources on the stability of the power grid, and the abandonment of wind and light. The microgrid itself is a complex system that contains different types of distributed power sources and loads. While considering the power generation efficiency of different power...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problems in the prior art, the present invention provides a neural network-based energy scheduling method for distributed micro-grid groups, aiming at the large number of control nodes brought about by the access of a large number of distributed power sources and loads in the micro-grid group, and the control method cannot fully consider each In order to solve the problems of the efficiency of distributed power supply and the complex types of control units in the micro-grid group, a micro-grid group control system with self-determination, integration of collection and distribution, and strong scalability is proposed.

Images