Prediction And Evaluation Study Of Power Consumption In A Large Airport Terminal Building

In recent years,with the rapid development of the aviation industry,civil airports are being built faster,and it is expected that the number of civil airports in China will exceed 270 by 2025,more than 30 more than in 2020.As one of the transportation hubs and logistics centers of cities,airport terminals consume more than two times more energy than general public buildings.The energy saving and consumption reduction of terminal buildings not only improves energy efficiency,but also helps to achieve the "double carbon" goal.Therefore,it is important to build a suitable prediction model to accurately predict power consumption and scientifically give power consumption evaluation indexes to facilitate accurate control of power consumption equipment in terminal buildings,which is also important for the development of energy saving and emission reduction in airports.In this paper,we take a large airport terminal in the northern region as the research object,and use the measured data of the terminal from January 2020 to August 2022 to predict and evaluate the power consumption of the terminal.The methods used in this paper include correlation analysis,GA-BP neural network and K-means clustering analysis to analyze the variation pattern of power consumption in the terminal building and the variation characteristics of the main mechanical and electrical systems;the influence of factors such as the number of passengers and outdoor meteorological parameters on power consumption is investigated;a power consumption prediction model is established based on the power consumption characteristics of the terminal building and the influencing factors,and the reliability of the model is tested using actual data.The reliability of the model was tested by using actual data;the evaluation index of daily power consumption per passenger was given for different passenger ranges.The main conclusions are as follows:(1)This paper studied and analyzed the measured passenger flow and power consumption data from January 2020 to August 2022,and grasped the changing characteristics of the number of passengers,total power consumption(power consumption per passenger,power consumption per unit area)and power consumption of the main mechanical and electrical systems in the terminal building.The Pearson correlation analysis and Spearman correlation analysis were also used to analyze the factors influencing the power consumption of the terminal building.(2)Based on the power consumption characteristics of the terminal,the optimal input variable type,activation function and number of layers of the hidden layer were determined by optimizing the internal parameters of the BP neural network,and the initial weight threshold of the neural network was optimized using a genetic algorithm(GA).The prediction results showed that the accuracy of the GA-BP model was higher than that of the BP model,and the MAPE decreased from 11.14% to 5.58%,and the accuracy of the optimized prediction model was significantly improved.(3)K-means clustering analysis of the number of terminal passengers and daily power consumption per unit passenger shows that the terminal power con sumption characteristics can be roughly classified into six levels according to t he number of daily passengers,i.e.,N≤10,000 passengers/day,10,000 passenge rs/day<N≤30,000 passengers/day,30,000 passengers/day <N≤50,000 passengers/day,50,000 passengers/day<N≤70,000 passengers/day,70,000 passengers/day <N≤100,000 passengers/day,100,000 passengers/day<N,and the evaluation index es of daily power consumption per passenger under six passenger number level s are given respectively.(4)According to the power consumption prediction model proposed in this paper,the power consumption of the terminal building under different annual total passenger numbers and the power consumption in different seasons were predicted and analyzed,and the evaluation indexes obtained by K-means cluster analysis were used to evaluate the daily power consumption levels under different passenger numbers,further indicating that there is some room for energy saving in the terminal building,and providing a certain reference for the subsequent proposal of optimized operation strategies for mechanical and electrical equipment.This will provide a certain reference for the subsequent optimization strategies of mechanical and electrical equipment.