An Intelligent Inhabited Environment Comfort Control Method

Dynamic comfort control is a developing direction of comfort control strategy for intelligent inhabited environment. Dynamic comfort control is a control strategy that the thermal comfort zone is divided into two parts, comfortable zone and energy-saving zone, and environment is periodically alternated between comfortable and energy-saving zone. There are many difficulties to design an optimal dynamic comfort control system, because of lack of coordinative control evaluation methods for both inflicting indices of comfort and energy-saving. And the effect of outdoor environment on indoor environment is often neglected. To solve these above problems, this paper presents an optimal dynamic comfort control method for intelligent inhabited environment.The main contents of the paper are presented as follows:(1) The evaluation index of comfort for dynamic comfort control which is based on users’preference is got. Two kinds of comfort evaluation indexes are presented. One index is the predicted index of indoor dynamic thermal comfort, and the other one is the number of complaint events which is obtained by cold/hot complaint event model. The thermal comfort can be predicted approximately by the predicted index. A cold/hot complaint event model for dynamic comfort control is derived from modified level-crossing theory. PMV (Predicted Mean Vote) time-series data and a time interval are given as the inputs of the complaint event model, then the number of cold/hot complaint events can be calculated.(2) The energy index of indoor thermal environment is got. The power consumption of air conditioner constitutes a high proportion of energy of indoor thermal environment. The power consumption of air conditioner is the nonlinear function of the temperature setting, run-time and outdoor temperature. Taking advantage of the property of neural network in non-linear functional approximation, a kind of HCMAC (Hyperball Cerebellar Model Articulation Controller) neural network is used to get the energy-consuming model of air conditioner.(3) The relevant model of indoor and outdoor environment parameters (temperature and humidity) is obtained. The performance of air conditioner is liable to be affected by outdoor environment, such as the time needed for reaching the temperature setting, dehumidifying performance, and so on. Therefore, a relevant model of indoor and outdoor environment parameters which will provide conditions for realizing the optimal control of indoor environment is proposed. According to the values of outdoor temperature, humidity and temperature setting of air conditioner, the indoor temperature and humidity can be obtained by this model. A fuzzy relation matrices model is adopted to get the relevant model. This method can avoid the complex process of building model and make it possible to analyze the system which has non-decomposable multi-outputs.(4) The optimal control method for dynamic comfort control based on improved PSO (Particle Swarm Optimization) is presented. An optimal control model is built based on comfort index and energy index of dynamic thermal comfort. And the effect of outdoor environment on indoor environment is also considered in the optimal control model. According to users’balance between comfort and energy saving, optimal setting values of parameters in dynamic comfort control system can be calculated by this model. When comfort level is equal, the energy consumption of dynamic comfort control can be reduced by 17.5% compared to steady comfort control method which is based on PMV.