Research On Energy Management Of Multi-Energy Systems Based On Distributed Neurodynamic Algorithms

With advanced communication technology,information processing technology and control technology linking multiple energy networks such as electricity,heat and gas,multi-energy systems are seen as an effective way to achieve energy conservation and reduce emissions and improve energy utilization.In the context of today’s large number of distributed energy resources and smart devices connected to energy networks,distributed optimization algorithms are widely used to solve the energy management problems of energy systems.Nevertheless,the existing distributed algorithms have some problems that need further research:(1)the loads in multi-energy networks are often highly random and volatile,requiring high convergence speed of the algorithms;(2)the effective implementation of some algorithms makes too many assumptions on the energy management problem itself,which is difficult to be applied in practical scenarios.In this thesis,energy management problems and efficient distributed algorithms in different scenarios are investigated in the context of multi-energy systems.The main research work is as follows:1.The energy management problems in the hybrid energy system under the coupling of electricity and heat networks are studied.In this thesis,the hybrid energy system is divided into several regional energy systems.Considering that thermal energy is limited by transmission distance and transmission mode,the heat supply network is only applicable in the regional energy network,and each regional energy network can trade electrical energy and communicate information with each other.Given the volatility of load,this thesis considers it as a time-varying variable and proposes a model of time-varying energy management problem for hybrid energy systems.For this problem,a distributed neurodynamic algorithm with fast convergence speed is proposed in this thesis.Following the fixed-time stability theory and the contraction analysis theory,the stability proof of the algorithm is given,and the convergence speed and tracking error of the algorithm are clearly defined.Finally,the effectiveness and real-time performance of the algorithm are demonstrated by simulation experiments on an IEEE-39 bus system.2.The energy management problem under the influence of time-varying loads is studied in the scenario of an integrated energy system where electricity,heat and gas networks exist simultaneously.With the objective of minimizing the overall operating cost of the system,the operating constraints of each energy supply unit are fully considered,and the time-varying energy management problem of the integrated energy system is modeled under the premise of satisfying the supply-demand balance and line congestion of each energy network.In order to accelerate the convergence speed of the distributed algorithm,the idea of Nesterov accelerated gradient descent method is embedded into the algorithm design of distributed neurodynamics to provide stable and real-time reference power values for each energy unit and load in the integrated energy system.In addition,the optimality and stability proof of the algorithm are proved by using Lyapunov stability theory,and the effectiveness and fast convergence of the algorithm are verified by simulation experiments under different condition settings.