Multilayer Fuzzy Inference System for Air Conditioner

Nowadays the consumption of air conditioning devices is exponentially increased in warm countries like India. The enhancement in utilization of the cooling devices makes it essential for them to have stable energy consumption by providing steady compressor speed. Achieving an appropriate mechanism for compressor speed control requires use of Fuzzy Inference System (FIS). However, Fuzzy Inference System with defuzzification is not adaptive to other methods that can enhance the performance of the system. Fuzzy Inference System with weighted average cannot control the system perfectly. Such inference systems causes varying outcome to a large extent when used separately. This paper presents the proposed Multilayer Fuzzy Inference System, a combination of defuzzification and weighted average mechanism. The result shows that proposed approach gives output in human understandable form and predicts the compressor speed from temperature and humidity. Such multilayer approach calculates steady compressor speed to have stable energy utilization.