Research Of Fuzzy Control On Inverter Air Conditioners Based On Genetic Algorithm

The air-conditioning of frequency conversion can construct more comfortable room environment because of its energy saving and high efficiency and low noise, so it gradually becomes the direction of family air-condition industry. Certainly, the control system of the variable-frequency air-condition is more complex than ordinary air-condition, thus the research of this thesis is just on the intelligent control system of the variable-frequency air-condition.In recently years, an algorithm based on fuzzy logic control is adopted, which overcome the shortcoming of traditional PID algorithm and improve adjustment property of system greatly. However, in the algorithm membership function and control rules obtained are only depended on expert experience, so it can’t achieve real intelligence as unintimacy appears when environment changes.According to above issue, in the thesis we consider comfort of air conditioning system as optimization objective function, make use of genetic algorithm to control membership function and control rules of fuzzy control system, finally reach the purpose of intelligent control.The mian work of the thesis are as follows:1. Firstly the development process of air conditioning is elaborated in the angle of the pursuit of comfort, and analysis key technology of frequency conversion air conditioning in four aspects.2. According to the defects of the genetic algorithm, in the thesis choose operation of genetic algorithm is improved, and adaptive genetic algorithm is combined. The Improved algorithm changes crossover probability and mutation probability, besides, genetic algorithm is provided with adaptability and some self-learning function.3. In the thesis mathematical model of air conditioning system is established, then intelligence frequency conversion air conditioning system based on genetic fuzzy is designed, of which fitness function and fuzzy rules coding are confirmed. Finally in terms of principle of construction, genetic algorithm subsystem of software control part is carried on formation of software program.4. In the last blur intelligent air conditioning system based on genetic algorithm is simulated in Matlab, the process of simulation shows that genetic fuzzy algorithm need not accurate mathematical model of controlled object, and its control rules and membership functions are not only dependent on the experience of the experts. Finally we can get an effective control effect which is superior to PID algorithm and the traditional fuzzy control algorithm.