Research On The Evaluate And Control Methods Of The Building Intelligent Light Environment

The leap-forward development of electronic and information technology has created more possibilities for China to achieve the goal of achieving overall social energy efficiency by strengthening the control of building energy consumption.This paper relies on science and technology project of ?the development and research of building energy consumption analysis and energy efficiency evaluation system‘,whichfromHousing and Urban-rural Construction ofShanxi province.It mainly studies the realization scheme of energy consumption monitoring system platform of public building and the energy consumption analysis algorithm based on datamining.The main contents of the thesis are as follows:(1)Analyze the factors that affect the energy consumption of public buildings,through the field research summary summed up the large commercial buildings and large-scale office buildingenergy consumption characteristics,determines the public building energy consumption acquisition system data collection indicators,builds energy consumption classification sub-meter monitoring data model,Research on two kinds of energy dissipation algorithm,which can be used for non-direct use of measuring instruments to collect energy consumption data in the electrical branch of the calculation of sub-energy consumption(the mixed branch of the relatively stable load-sharing equipment and the mixed branch of load-real-time changing power consumption equipment),select a unit of energy consumption and a series of practical energy consumption analysis indicators.(2)Through the functional requirements analysis of public building energy consumption monitoring system,based on the purpose of openness and scalability,design the architecture of the energy consumption monitoring system,complete the system module function,database structure and interface prototype design.(3)Based on the historical energy consumption data recorded by the monitoring system,this paper studies the abnormal energy consumption of the outlier value discovery algorithm(LOF)by using the local outliers factor.Through the empirical analysis of the daily energy consumption data of a shopping mall in 2015,and compared with the clustering algorithm based on clustering,The results show that the results of LOF algorithm are more robust and reliable.(4)Using the random forest algorithm under high dimensional data to predict the average annual energy consumption of the building,through the data from the US Department of Energy CBECS database samples for empirical analysis,using the average decline in accuracy to measure the importance of independent variables,analysishow building properties of office buildings affect energy consumption.