Research And Application On Public Building Energy Consumption Data Prediction And Amount Of Energy Saving Assessment Method

Currently,China's building energy consumption accounts for 33% of the total energy consumption,which can be converted to 1.1 billion tons of standard coal,of which,the energy consumption of public buildings accounts for about 22% of the total energy consumption,therefore,public buildings have considerable potential for energy conservation.In recent years,the public building energy consumption measurement system is applied in the whole country,a large amount of energy consumption data can be uploaded to the data center platform in real time.Energy consumption data is the foundation of energy saving work,thus,accurate prediction and nodal energy calculation on public building energy consumption can carry out the work of building energy efficiency assessment.However,large number of energy consumption data have brought "data disaster",there are a lot of problem data during the operation of the building,and it is difficult for the construction manager to find out the value information.As there are a lot of energy consumption prediction and energy evaluation methods have been proposed,and energy consumption simulation is generally used to predict the energy consumption of new buildings,however,the prediction method used in existing buildings is still in the exploratory stage,these methods do not take into account the characteristics of energy consumption of public buildings,lack of universal and real-time features.In this paper,energy consumption and energy consumption characteristics of different types of public buildings,energy consumption prediction methods and energy saving evaluation methods are studied.Energy consumption prediction model of public buildings based on RBF neural network is studied and optimized for predicting the energy consumption of public buildings in the future.Energy evaluation method for public building section based on full building verification is proposed to conduct evaluation of energy conservation of public buildings involved in energy conservation.Energy consumption prediction model predicts the energy consumption of the target time by establishing RBF neural network,and adopts PSO algorithm to optimize the energy consumption prediction model of public buildings based on RBF neural network to achieve better prediction.MATLAB simulation is performed,and the results have showed RBF neural network has a good fitting ability in energy consumption prediction,but there is a certain error,the error is reduced by using PSO-RBF neural network,its forecast effect is better.The energy consumption evaluation method mainly adopts the multiple linear regression method to establish the model of energy consumption,and data mining software SPSS is used to establish the regression model between energy consumption and energy consumption,by comparing the reference energy consumption and actual energy consumption after energy saving transformation,energy efficiency of public buildings is evaluated.Finally,VS2013 and MATLAB mixed programming method is also adopted to develop energy consumption prediction and energy saving analysis system for public buildings in this paper,to make public building energy consumption prediction model and energy consumption benchmark model realized in this system.Public building managers can upload data collected to the system to conduct energy consumption prediction and energy saving analysis,and managers are not required to perform complex operations and have related expertise.