Predictive Control Of Aircraft Environment On Ground Operation

This paper mainly studied in the predictive control of temperature in cabin when the aircraft parked in the transaction,the purpose is to improve the control efficiency of cabin temperature and to reduce carbon emissions and operation cost of airline companies.Through collecting and analyzing the factors data of changing the cabin temperature,the author uses the neural network to forecast the cabin temperature.In the case of obtaining the change trend of cabin temperature,the priority control of cabin temperature can reduce the overshoot in the control process and improve the efficiency of control by controlling the ground air-condition unit.During the process of the entire study, using temperature recorders and wind speed recorders with memory function to record the cabin temperature value and the corresponding wind speed in collaboration with the airport and airline companies.In different seasons,different weather, different regions, the data we respectively collected including cabin temperature,wind speed of outlet,cabin size,the usage of airport ground equipment and environmental control equipment on the aircraft.And based on these data,a complete training and validation database is established,which provides a sufficient amount of data for this study being carried out smoothly.According to the actual measurement value of cabin temperature, using interpolation method to establish the cabin temperature field, which were analyzed. And then the comparative analysis of the cabin history temperature,the external environment temperature,the outlet temperature,the heat exchange in cabin door,the heat exchange When the sun point-blank bulkhead,thermal power of the lamps and entertainment system in the cabin, the influence of human body thermal load on temperature field are made.Using the superiority of neural network to deal with nonlinear problem to establish a prediction model of the cabin thermal environment,and using Matlab to predict the cabin temperature field.Finally taking a large aircraft as a platform,we collect a large number of data for validation,to ensure the accuracy of prediction model within the error range.Through combining the neural network prediction model and the traditional predictive control,a cabin thermal environment predictive control model is built.This can make the system to quickly and efficiently control the cabin thermal environment.And on the basis of the thermal comfort in the cabin to evaluate and verify the thermal environment control,which can confirm that the predictive control can control the cabin thermal environment effectively and quickly.