A New Method to Predict the Thermal Sensation of an Occupant Using a Neural Network and Its Application to the Automobile HVAC System

抄録

The purpose of this study is to develop a method for predicting the thermal sensation of an automobile occupant, and evaluate the prediction method's effectiveness in controlling an automobile heating, ventilating and air conditioning(HVAC)system. The main purpose of a HVAC system is not to keep the temperature constant, but to make people comfortable. This HVAC system was designed to control the interior environment according to the thermal sensation of an occupant, and thus has a direct effect on the occupant's comfort. The thermal sensation level was predicted by a neural network from the interior temperature of the vehicle and facial skin temperature of the occupant. The prediction of the thermal sensation using the neural network was satisfactorily accurate. The HVAC system using the neural network could control the thermal sensation of the occupant and could maintain the occupant's level of comfort even if the air temperature was constant.