Simulation Analysis And Predictive Diagnosis Of Large Scale Public Buildings

At the request of energy-saving and emission-reduction in the whole society,building energy conservation has become the focus of study on energy conservation,owing to its rising proportion of the total social energy consumption.And the big total quantity and high level of energy consumption have made the large scale public building energy conservation become the priority among the priorities.Thus it is of great practical significance and social value to study the energy saving of large scale public buildings.In this thesis,the optimization of envelope structure is combined with the optimization of thermal performance design parameters.The energy-saving potential of each parameter is determined by sensitivity analysis,which could provide guidance for energy-saving design and optimization.Aim at extensive operation and management in building service stage,an energy consumption prediction and anomaly diagnosis method based on support vector machine is proposed to discover the energy waste and further provide theoretical support and implementation path for energy conservation of large scale public buildings.From the perspective of energy-saving sensitivity,a large-scale office building was taken as an example to carry out quantitative analysis of six commonly used building envelop thermal performance design parameters.The parameters included the comprehensive heat transfer coefficient of exterior wall,the comprehensive heat transfer coefficient of the roof,the comprehensive heat transfer coefficient of the exterior window,the solar heat gain coefficient of the exterior window,the ratio of window to wall and the tightness of the exterior window.The recommended range of each parameter was given as well.High-sensitivity parameters were selected as the object of scheme optimization,and orthogonal experimental design was used to arrange schemes.According to the energy consumption simulation results,the lowest energy consumption scheme was selected as the optimal solution.This method could provide comprehensive support for energy conservation of large scale public buildings.Based on historical energy consumption data series,climatic factors and time factors,11 input parameters were selected as sample characteristics,and a large scale public building energy consumption forecasting model based on support vector machine was constructed to forecast the daily energy consumption of buildings.On the basis of energy consumption forecasting,mean and maximum relative error of test set were taken as the criteria to diagnose the anomaly in energy consumption.This method was applied to the abnormal diagnosis of air conditioning system energy consumption.Through the comparison between the predicted value and the actual value of energy consumption,the unreasonable use in the operation of the air conditioning system was found successfully.It proved that this method could provide reference for building energy-saving management.