Condition monitoring of the components of a domestic air conditioner

This work discusses the design and simulation of fuzzy control of an air conditioning system at different pressures. The first order Mamdani fuzzy inference system is utilized to model the system and create a controller design. In addition, an estimation of the heat transfer rate and water flow rate injected into or withdrawn out of the air conditioning system is determined by the fuzzy IF-THEN rules. The results show that when the pressure increases the amount of water flow rate and heat transfer rate decrease within the lower ranges of inlet temperatures. On theother hand, and as pressure increases the amount of water flow rate and heat transfer rate increases within the higher ranges of inlet temperatures. The inflection in the pressure effect trend occurs at lower temperatures as the air humidity increases. From these experiments, it is shown that the fuzzy controller can successfully track the room temperature and relative humidity set points. The fuzzy controller was insensitive to the coupling of the temperature and relative humidity. Even if the set point of one variable was changed, the other variable kept stable. Experimental results also showed the good disturbance rejection capability of the fuzzy controller. The input scaling factors andoutput gains used can significantly influence the fuzzy controller performance. Larger scaling factors and gains can increase the speed of the control action, but these can lead to oscillations and overshoots. Compared with the input variable scaling factors, the control gains at the output have a more significant effect. It was shown that fuzzy logic could be successfully applied to room temperature and relative humidity control. However, the fuzzy logic controller is built on linguistic rules given by human experts. Sometimes, there is not enough expert knowledge available or the initial fuzzy rules may even be incorrect. The fuzzy rules adopted for one system may not be applicable to other system. In these conditions, tuning of fuzzy controller is necessary.

Keywords:

Condition monitoring , domestic air conditioner , fuzzy logic , Matlab