| In order to achieve the goal of "carbon peak" and "carbon neutrality",integrated energy system(IES)based on the coupling and complementarity of cold,thermal and electrical energy is regarded as an important approach to promote the low-carbon development of energy industry.IES can meet the requirement of multiple types of energy,improve the consumption proportion of renewable energy and energy utilization efficiency.However,IES is faced with the uncertainties of renewable energy generation on the supply side and multiple load demand on the demand side,as well as the spatio-temporal coupling characteristics between different energy conversion devices,which make energy scheduling and planning of the system difficult.This paper takes solid oxide fuel cell-based integrated tri-generation energy system(ITES)as the research object,and research on two-stage optimization method of ITES operation scheduling and capacity allocation based on multiple load and photovoltaic power prediction is carried out.The main research contents are as follows:(1)In view of the multiple load prediction for ITES,long and short-term memory(LSTM)neural network method is proposed.This method calculates the correlation degree between multiple load and weather factors,and considers the coupling characteristics between multiple load.Historical weather information and load are taken as the input variables,and cooling,heat and electric load are taken as the output variables,thus a joint prediction model of multiple load is established.Aiming at the photovoltaic power prediction,Gaussian process regression and LSTM are combined to describe the probability interval of photovoltaic power uncertainty.The test results verify the effectiveness of the proposed methods.(2)For the day-ahead scheduling optimization of solid oxide fuel cell-based ITES,dynamic programming method is proposed.The steady-state operation mathematical models of the system with combined energy storage devices are established and total scheduling economic cost is taken as the optimization target.Comprehensively considering the system energy balance constraints,operation and safety constraints,a day-ahead scheduling optimization model of ITES with given equipment capacity is developed under reasonable assumptions.Inverse recursive algorithm is used to solve the dynamic programming model to obtain the day-ahead output scheduling of each equipment under the goal of optimal economic benefit.By the comparative study of three working conditions including system only with electrical energy storage unit,system only with thermal energy storage unit,and system with combined energy storage equipment,it is found that the total scheduling economic cost can be reduced by 2.09% by the incorporation of electrical energy storage and thermal energy storage equipment respectively.(3)In view of the ITES operation scheduling and planning design,a two-stage optimization method of ITES operation scheduling and capacity allocation is proposed.This method comprehensively considers the initial investment cost and scheduling cost of the system,and develops a two-stage optimization model of ITES operation scheduling and capacity configuration.At the upper optimization stage,the equipment capacity are taken as the design variables,and the equipment investment cost is taken as the optimization objective.The solution space of design variables is searched by genetic algorithm.At the lower optimization stage,power output of system devices are taken as the decision variables,and total scheduling economic cost is taken as the optimization objective.The sub-optimization problems corresponding to the design variables,that is,optimal scheduling scheme,can be obtained by dynamic programming.The research of this paper can provide theoretical support and methodological guidance for the IES state forecasting,operation scheduling,capacity configuration design and scale promotion and application. |
PDF Full Text Request