Research On Indoor Thermal Comfort Control System Based On Smith Predictor And Feed-forward Decoupling

With the development of Chinese economy and the improvement of people’s living standards, the contradiction between people’s increasing requirement for indoor environment comfort and the energy shorting is becoming more and more serious, which force us to research on the best energy-saving schemes that can meet the high thermal comfort of indoor environment.Air-conditioner is an integral part of improving indoor thermal comfort conditions. However, there exists hysteresis in air-conditioner temperature and humidity control system. Besides, the coupling between temperature and humidity control loops is severe and the traditional controller is inefficient, which no doubt increase the energy consumption of air-conditioning system. Moreover, more and more people suffer from “air-conditioning disease”. The constant temperature and humidity environment created by air-conditioner has a lot of disadvantages. Therefore, researching the rule of indoor dynamic thermal comfort, decoupling the temperature from humidity, improving the control algorithm of air-conditioning system and building a dynamic energy saving and comfortable indoor environment is of great significant.Detailed research works have been done on the above issues. The thesis establishes the simulation model of temperature and humidity control system, designs a feed-forward decoupling device with Smith predictor, improves the control algorithm and adopts a dynamic thermal control method. The establishment of simulation model for temperature and humidity control system takes thermal comfort research laboratory in Anhui key laboratory of intelligent building as the physical model, while temperature and humidity collection platform is built based on Zig Bee Wireless Sensor Network and the characteristic parameters of the model is acquired by using method of least square on the basis of mechanism. To design a decoupling device, a feed-forward compensator is adopted between control object and the controller and the Smith predictor method is used to solve the problem of time lag. Moreover, a thermal comfort controller is applied based on the Fuzzy-PID by which parameters can be configured online. In order to validate the performance of the decoupling controller, a simulation model of the thermal comfort system is established in Smiulink environment. The results show that the thermal comfort control system has faster dynamic response, smaller overshoot and rise time with the decoupling method and Fuzzy-PID controller. Finally, a dynamic thermal comfort method is adopted to control the air-conditioner runs alternately between energy saving zone and comfort zone. The experimental results demonstrate that this method not only can meet the requirement of thermal comfort for users but also reduce the energy consumption.