Research On The Optimal Dispatching Decision Method Of Regional Integrated Energy System Considering The Consumption Of High Proportion Of Renewable Energy

Clean substitution and electric energy substitution in energy production and consumption are the main technological paths to achieve "carbon peaking" and "carbon neutral" strategies.Among them,as a form of renewable energy development and utilization,integrated energy system has been widely concerned.On the one hand,integrated energy system integrates multiple forms of flexible loads and energy conversion devices and realize the complementary superiority of dispatchable resources and improve the absorption capacity of renewable energy;on the other hand,the uncertainty output of renewable energy and the high fluctuation of demand on the load side have posed new challenges to the active coordination ability of the integrated energy system.To address this problem,this paper carries out the research on the optimal dispatching decision method of regional integrated energy system considering the consumption of high proportion of renewable energy.The main contents are as follows:(1)The typical architecture and operation mode of integrated energy system is analyzed.Firstly,based on the multi-energy coupling characteristics of the integrated energy system,the physical and operation control architecture of the integrated energy system are proposed,and the energy conversion characteristics of photovoltaic,wind power,hydrogen and energy storage in the integrated energy system with renewable energy intervention are analyzed.Secondly,three gridconnected operation modes of integrated energy system,namely user-oriented,regional,and urban models,are analyzed.Finally,an optimal dispatching framework for the integrated energy system is proposed to address the optimal and dispatching issues in integrated energy system.(2)The optimal dispatching strategy of regional integrated energy system considering stepped carbon trading and dual response of supply and demand is proposed.Firstly,a model of regional integrated energy system with ladder-type carbon trading is constructed for the multiplicity and dispatchability,a dual-response model of supply and demand is furtherly constructed and a response compensation mechanism is proposed.Secondly,an optimal dispatching model is established with the objective of minimizing the cost of operation,demand response compensation and carbon trading.Finally,the CPLEX toolbox is used to solve the proposed model.Through case analysis of different scenarios,the effects of the dual response strategy of supply and demand and the stepped carbon trading mechanism on the operation optimization,energy conservation and emission reduction of the regional integrated energy system are analyzed,and the simulation results verify the superiority of the proposed strategy.(3)The optimal dispatching intelligent regional integrated energy system considering sourceload uncertainty is proposed.In view of the random fluctuations of wind power,photovoltaic power and load power as well as the complexity and uncertainty of the operation status and parameters of the integrated energy system,an intelligent energy management architecture based on mechanism and data fusion is proposed.Then an optimal dispatching model for energy management of regional integrated energy system containing various energy forms such as wind,photovoltaic,hydrogen and energy storage is established.A deep learning-based source and load power prediction method is proposed to solve the dispatching scenario prediction problem;a deep reinforcement learning-based energy management optimal decision model is proposed,according to wind,photovoltaic and load prediction information,deep Q network is used to simulate the energy management strategy set of the regional integrated energy system,and the best optimal strategy is obtained through deep reinforcement learning and realize the safe and economic operation of the regional integrated energy system.Finally,compared with the energy management method based on deterministic scenario,the feasibility and effectiveness of the deep reinforcement learning method are verified.