Emergency electric logistics fleet optimization scheduling method for improving toughness of distribution network

Abstract

The invention discloses an emergency electric logistics fleet optimal scheduling method for improving the toughness of a distribution network, and the method specifically comprises the steps: detecting fault related data at a fault part of the distribution network, carrying out the post-disaster distribution network fault evaluation, and carrying out the prediction through a recurrent neural network, and obtaining the electric energy needed by the fault recovery of the distribution network, and predicting the fault recovery time; on the basis of an actual road network structure, constructing directed graph representation, and embedding influence factors such as road maintenance and traffic jam in directed edge attributes; a vehicle is used as an intelligent agent, a vehicle scheduling problem is constructed into a Markov decision process, an intelligent agent action space is constructed based on a road information network, and a vehicle scheduling result meeting a target is solved under constraints by taking minimization of user energy supply loss and scheduling cost as a target. According to the emergency logistics motorcade scheduling model constructed based on road information and post-disaster required electric energy prediction, the vehicles can be scheduled with long-term attention in the process of responding to the energy demand, and the user loss and the scheduling cost are minimized.

Description

technical field [0001] The invention relates to the technical field of power system distribution network toughness optimization, in particular to an emergency electric logistics fleet optimization scheduling method that improves distribution network toughness. Background technique [0002] In recent years, due to the frequent occurrence of natural disasters such as floods, extreme hot weather, and snowstorms, there are more and more power grid failure events. In order to reduce user losses caused by grid failure events, the power system has put forward higher and higher requirements for grid resilience and post-disaster fault repair, aiming to shorten the post-disaster fault time of the power grid and ensure the safe operation of the power grid. With the popularity of electric vehicles, the potential of electric vehicles as a mobile energy source is gradually increasing. Under the background of the increasing electrification of the urban transportation system, the dispatcha...

AI Technical Summary

Problems solved by technology

[0003] Chinese Patent Publication No. CN113675867A discloses a method and device for resilience recovery of distribution network containing electric buses, constructing a distribution network model and an electric bus model in the resilient recovery process of distribution network, and constructing a distribution network containing electric buses based on the above model A function model for grid resilience recovery scheduling, and determine the number of electric buses that supply power to the distribution network, the charging power of electric buses, or the operating load; Chinese Patent Publication No. CN113111476A provides a human-vehicle-thing Emergency resource optimal scheduling method, constructing an emergency resource scheduling framework and the first, second, and third emergency resource allocation models with the goals of minimizing power outage time, minimizing load shedding, and minimizing emergency resource scheduling costs respectively, and solving them under certain constraints And determine the optimal scheduling plan to complete the emergency repair of all the fault points; the above two methods fail to consider the transportation capacity and energy supply capacity of the urban electric logistics fleet after the disaster grid, and fail to give full play to the urban large-scale electric logistics fleet scheduling potential

Images