| With the rapid development of the Belt and Road Initiative,the number of airport terminal has increased rapidly.Intelligent regulation of the heating system in the airport terminal is an important means to improve the energy saving and indoor thermal comfort of the airport.Accurate heat load predicting provides a theoretical basis for optimal dispatching of energy stations.Due to the large number of influencing factors,strong randomness,and difficulty in predicting the heat load,it is particularly important to accurately predict the heat load of the airport terminal.At the same time,the airport heating system is mainly consists of three heat sources: municipal heating,mid-deep geothermal and gas boilers.The implementation of scientific and reasonable optimization scheduling strategies is related to economic and social benefits.The thesis conducts a detailed study on the load predicting of the airport heating system and the optimal scheduling of multiple heat sources.The main contents are as follows:Firstly,establish a heat load prediction model for the airport terminal.Use the Pearson correlation coefficient method to filter out the variables with high heat load factors,and use the variational mode decomposition(VMD)method to decompose the variables.Variables are used as the input of the gated recurrent unit(GRU)prediction model for training to form the VMD-GRU heat load prediction model.Secondly,a multi-heat source optimal scheduling strategy is proposed.The multi-heat source optimal scheduling model is studied.With environmental scheduling and economic scheduling as the goal,the constraints of heat load balance and flow balance are established on the basis of predicting heat load,and the bare-bones particle swarm optimization algorithm is used to solve the scheduling model,and then proposed.An optimized dispatch strategy that takes into account the economy and the environment.Finally,the airport heating supervision system based on the Internet of Things was developed,and the architecture design,hardware selection and point layout of the supervision system were completed,and the airport heating supervision system database was established,and functions such as heat load predicting and heat source optimization dispatching were added to it.The functions of load predicting and energy consumption management of the airport heating supervision system are realized.The comparison of simulation experiments shows that compared with BP,GRU and VMD-BP,the average absolute error of the established heat load prediction model is reduced by 1428674,1318.424 and 630.927,respectively.Compared with the conventional scheduling method,the proposed multi-heat source optimal scheduling strategy reduces the operating cost by an average of 13.94% and the comprehensive emissions of pollutants by an average of 14.96%.The airport terminal heating supervision system solves the problems of inaccurate heat load predicting of the airport terminal,realizes the comprehensive management of the heating system,and provides guidance and reference for other types of building heat source scheduling and energy-saving management. |
