Chaotic Property Analysis And Prediction Model Study For A Large Public Building Energy Consumption

Large public buildings in China consume about 22% of the total electricity consumption of cities and towns,and they are typical “large energy users”.Therefore,research on energy conservation and consumption reduction for large-scale public buildings is particularly important.Among them,energy consumption prediction is the main method to solve this problem.By predicting and analyzing the energy consumption changes in the future,it can provide scientific support for large public buildings energy allocation and avoid energy waste.The energy consumption of large public buildings is affected by various factors such as electromechanical equipment,weather,holidays,etc.,which reflects complex nonlinear characteristics,making it difficult for traditional prediction methods to achieve ideal prediction results.Chaos theory can fully exploit the nonlinear law inside the large public buildings energy consumption through phase space reconstruction.Therefore,this paper proposes a chaotic time series prediction model for large public buildings energy consumption.This paper takes an office building in Xi’an as the research object.The main research contents are as follows:(1)The chaotic characteristics of energy consumption of an office building are analyzed.The EM algorithm is used to process the energy consumption data of the collected office buildings,and then the C-C method is used to obtain the delay time and the embedded dimension of the energy consumption time series,and the two parameter values are combined,and the energy consumption data is determined by the small data amount method.The Lyapunov index shows that the energy consumption change of theoffice building has chaotic characteristics.(2)A chaotic time series prediction model for energy consumption of an office building was established.Based on the combination of neural network and chaos theory,a chaotic time series prediction model based on BP and RBF neural networks is established for the characteristics of energy consumption data,and a comparative study is carried out.The model was evaluated using the mean MAPE and RMSE.The experimental results show that the MAPE based on BP neural network is only 3.48%,and the RMSE is only 1/5 of the BP neural network prediction model.The MAPE based on RBF neural network is only 3.43%.The RMSE is only 1/4 of the RBF neural network prediction model.Therefore,compared with the neural network prediction model,the model prediction results combining chaos theory and neural network are more accurate.(3)A chaotic time series prediction software module for energy consumption of an office building was developed,which improved the function of the building energy management information system,completed the operation interface design of the module,and provided strong support for energy saving and consumption reduction of the building.This paper analyzes the chaotic characteristics of energy consumption data of an office building in Xi’an,establishes its chaotic prediction model of energy consumption,develops its corresponding functional software module,realizes the visualization function of energy management,and can effectively guide such large-scale public construction.Scientific energy and energy conservation work have guiding significance for future scientific energy conservation.