Predicting efficiency of flat-plate solar collector using a fuzzy inference system

With the increasing world population, industrialization and comfort need of humans, energy usage is growing considerably. Nowadays most of the energy needs are obtained from fossil stratum, and hydroelectric or nuclear power plants. However, usage of these fuels has two important problems: (1) These fuels may not exist in the near future and (2) Extensive use of these fuels causes serious environmental pollution. Alternatively, solar energy that is renewable and environmentally friendly energy can be utilized. In design of such renewable energy systems, efficiency is crucial including solar water heating systems. Experiments were conducted to evaluate a flat-plate solar collector performance of thermosiphon solar water heating system during the summer season in Nicosia, North Cyprus. A fuzzy inference system was developed to predict the efficiency of the solar collector. In our fuzzy inference system, we only utilize ambiance temperature, input and output temperature of the solar heating system. The predicted values were found to be in close agreement with the experimental counterparts with 0.9469,3.13, 6.96 coefficient of determination, root mean square error and average forecasting error respectively. It was noted that the proposed fuzzy inference system can provide high accuracy and reliability for predicting the performance of a solar collector. The advantages of this approach as compared to the testing methods are speed, simplicity, and the use of expert knowledge for prediction. Hence, the soft computing approach can be a potential tool for predicting efficiency of a flat-plate solar collector.