Research On Optimal Dispatch Of Building Integrated Energy System Based On Multiple Loads Forecasting

With the proposal of the "dual carbon" goal,the integrated energy system has gradually become the focus of research in the field of energy.The integrated energy system contains electricity,heat,cooling,gas and other energy sources,and the mutual conversion between various energy sources is realized through the internal energy coupling equipment,which greatly improves the energy utilization rate of the system.At the same time,due to the in-depth research on power-to-gas and cogeneration technologies,the conversion between multiple energy sources in the system is also more efficient,which improves the new energy absorption capacity and system stability,and also reduces the carbon emissions of the system.Therefore,it is very important to carry out multi-load forecasting on the energy consumption side of the integrated energy system and optimize the dispatch of the energy supply side.The main research contents of this paper are as follows:Firstly,the load forecasting and optimal dispatching theory of integrated energy system are analyzed.The characteristics of short-term load forecasting and multi-energy load characteristics in integrated energy systems are expounded.The data preprocessing method applied to load forecasting and the model prediction error evaluation index to judge the accuracy of the forecast model are introduced.Then,according to the internal structure of the integrated energy system,the mathematical model of clean energy power generation equipment,the mathematical model of energy storage equipment and the mathematical model of energy coupling equipment within the system were analyzed in terms of source,grid and storage.Then,the multi-load forecasting method of integrated energy system is studied.In order to improve the accuracy of the prediction model,a multi-load forecasting method based on error compensation is proposed.Based on the long-short-term memory neural network,this method uses the gated recurrent unit as the error compensation network,and reconstructs the results of the two into the final load prediction results.The grey correlation analysis method is used to select the factors affecting load forecasting,and a two-layer prediction network model is established,which verifies that the multi-load forecasting method proposed in this paper has higher accuracy through in-depth comparative analysis of sliding window width,sampling time,error compensation model and different network models.Finally,the optimal dispatching strategy of building integrated energy system is studied.In order to eliminate the influence of the two-sided uncertainty of source and load on the optimal scheduling of the system,the optimization scheduling of the building integrated energy system considering the uncertainty is proposed.The mathematical model of heat demand,grid model and dynamic model of natural gas grid for building users in the integrated energy system are introduced and analyzed.Aiming at the uncertainty of source-side wind power and photovoltaic output and the uncertainty of load-side load forecasting,the opportunity constraint planning method is used to transform the uncertainty constraint in the system into a deterministic constraint.Then,taking the minimum operating cost of the system as the objective function,and taking the operating conditions of the equipment in the system,the energy grid model,the user heat demand model and the opportunity constraint planning as the constraints,the optimal scheduling model of the integrated energy system is established.By comparing the scheduling results under different scenarios,the reliability of the optimal scheduling strategy proposed in this paper is verified.