Energy-saving Optimization Model And Application Based On Building Energy Consumption Data

With the rapid economic and social development of various countries,highly developed modern industries are creating huge social wealth and continuously meeting the increasingly rich material needs of the people,while also producing a large number of pollutants,making the global ecological environment deteriorating day by day.In the total energy consumption of buildings in the proportion of energy consumption,the European Union and the United States building energy consumption accounted for 40%of the total energy consumption.By 2015,the proportion of building energy consumption in China's total social energy consumption has been close to 20%,and has been on the rise year by year.Based on the above problems,this paper tries to apply the data analysis and mining technology to the analysis of building energy consumption data,establish the energy consumption data correlation model according to the characteristics of building energy consumption,and analyze building energy consumption from various aspects:1?Study on energy consumption factors of regional buildings.STIRPAT(Stochastic Regression by Regression on Population,Affluence,and Technology)model was used to predict the future energy consumption of civil buildings in Guangzhou from a macro perspective,and the factors affecting building energy consumption were analyzed.According to the model,the energy consumption of civil buildings in Guangzhou is expected to be2651.87 to 3236.16 million tons of standard coal by 2025,with an increase of 20.01%to35.92%compared with that of 2020.In addition,it is found that the gross regional product,per capita living area and household consumption expenditure have a greater positive impact on building energy consumption,while the proportion of tertiary industry gross domestic product has a greater negative impact on building energy consumption.2?Research on clustering algorithm of office building.This paper analyzes the current situation of energy utilization of office buildings in Guangzhou.According to the clustering model,office buildings in Guangzhou are divided into buildings with low energy consumption of less than 78k Wh/(m~2·a),buildings with low energy consumption of78-122kwh/(m~2·a),buildings with high energy consumption of 122-176kwh/(m~2·a),and buildings with high energy consumption of more than 176k Wh/(m~2·a).3?Research on logistic regression model of office building.Building information and energy consumption information of some office buildings were collected to establish logistic regression model and analyze the factors affecting energy consumption of office buildings.The logistic regression model shows that five factors including building height,air conditioning area ratio,indoor temperature,operating hours and building area have influence on energy consumption of office buildings.4?Energy consumption prediction model of LSTM neural network for building air-conditioning system.The LSTM neural network model was established to predict the air conditioning energy consumption of buildings with different building functions using the LSTM model.LSTM model is compared with ARIMA model and BP neural network model,and the results show that LSTM model not only has higher accuracy,but also has higher applicability than other models.This paper analyzes the building energy consumption data from multiple dimensions,makes full use of the collected energy consumption data,and deeply explores the value behind the building energy consumption data,so as to provide a basis for the further implementation of energy-saving operation and management,and guide the energy-saving renovation of buildings and the optimized operation and management of the building energy use system from multiple angles.