Research Of Boeing 737 Aircraft Cabin Energy Consumption Prediction Based On Combined Forecast

From the operating costs and safety,air quality,noise control,and even the angle of the plane congestion analysis,the use of ground specific equipment to replace aircraft APU has multiple benefits.However,the use of special equipment on the ground,the aircraft cabin thermal comfort is poor,uneven distribution of the temperature field of the cabin and other issues.In order to achieve the optimal energy consumption under the premise of cabin thermal comfort,the research on the energy consumption prediction of aircraft cabin is of great significance to the use and control of ground air conditioning.In this paper,we mainly study on the prediction method of the energy consumption of the Boeing 737 aircraft cabin:(1)The structure of energy consumption of air conditioning system and the input parameters of energy consumption prediction model are analyzed theoretically.The purpose of aircraft cabin energy consumption prediction is to provide technical support for the control of ground based air conditioning system.This paper analyzes the factors that affect the energy consumption from two aspects of the cabin and the ground air conditioning,and determines the main factors affecting the energy consumption.(2)Aiming at the problem that the energy consumption of aircraft cabin is more accurate,a new method based on particle swarm optimization and neural network is proposed.This method is based on the generalized regression neural network to establish the prediction model,and the neural network parameters are optimized by combining the particle swarm optimization algorithm and the chaotic mapping algorithm.The simulation results show that the neural network parameters can be improved effectively by the improved particle swarm optimization algorithm.(3)To solve the problem of high convergence speed for real-time prediction of aircraft cabin energy consumption,a neural network,particle swarm optimization and coral reef prediction method are proposed.Firstly,the energy consumption prediction model based on wavelet neural network is established,and then the parameters of the model are optimized by the combination of coral reef and particle swarm optimization algorithm.The simulation results show that the neural network parameters can improve the convergence speed of thecabin energy consumption effectively by the improved particle swarm optimization algorithm.(4)In order to meet the requirements of the accuracy and rapidity of the cabin energy consumption prediction in real-time control,a nonlinear combination forecasting method based on support vector machine(SVM)is proposed.First select the generalized regression neural network forecasting model and wavelet neural network forecasting model,on the basis of joining the prediction model to improve the prediction generalization ability of support vector machine;secondly the injection principle of adaptive allocation for the three single forecasting model of weight coefficients,the weight distribution after the single forecasting model as a result of the support vector machine training input,for nonlinear combination forecasting based on support vector machine.The simulation results show that the nonlinear combination forecasting method can not only meet the accuracy and rapidity of the forecast,but also improve the generalization ability.