Study On Operation Optimization Regional Buildings With Combined Heat And Power Based On Electric Load Forecasting And Heat Loss Calculation

As an application scenario of the combined heat and power system,the operation optimization of the regional building needs further study.First of all,short-term power load prediction provides the basis for energy dispatching in cogeneration regional buildings,and the accuracy of the prediction results directly affects the progress in other areas of the power system.Therefore,in this paper,the deep learning method is adopted to establish a short-term power load prediction model based on the stacked denoising autoencoder,and to predict the power load inside the system.Compared with the traditional BP neural network and autoencoder,the prediction error decreased from 3.66%and 6.16%to 2.88%,respectively.Therefore,compared with the traditional forecasting methods,the model based on SDAE has a better forecasting effect and can be used as a new short-term urban power load forecasting method.Secondly,the paper estimates the heat loss in the optimal operation strategy of the regional buildings with combined heat and power supply.The data of power load prediction will be used as the basis for the follow-up research.The novel feature of this study is that a method for estimating building heat loss based on hot water circulation is proposed and the calculation formula is derived.Heat loss calculation method for data mining and data analysis,the establishment of simplified nonlinear heat loss mathematical model.Then the model parameters were optimized,and the results show that the accuracy of the optimized model is improved from 10.22%to 5.39%.Finally,this chapter carries out energy coordination scheduling for the electric energy and heat energy of the regional buildings with cogeneration of heat and power.Combined with electricity price,natural gas price,electric load and heat demand,the corresponding operation optimization strategy is formulated.Combined with specific application examples,the corresponding optimization models are compared.