Comparison of Bat and Fuzzy Clusterization for Identification of Suitable Locations for a Small-Scale Hydropower Plant

Abstract

Hydroelectric plants are an environmentally friendly renewable energy source, but, due to uncertainties in flow patterns, often such energy generation projects fail. Also, deliberations from displaced people and environmental activists (due to large-scale disturbances to the natural ecosystems of adjacent areas) make some highly efficient hydropower projects unfeasible. That is why the success of hydropower projects depends largely on the selection of location. Currently, the efficiency of selecting the ideal locations depends mainly on expert opinion or linear models and other decision-making methodologies where human judgment and opinion play a major role in the reliability of the selection. But, as usual, the error rate in such procedures is generally unsatisfactory. The present study tries to apply clusterization algorithms to identify ideal locations for small hydropower plants in such a way that the need for expert or opinion can be reduced. In the clusterization of a suitable hydropower location, the food foraging behavior of bats and fuzzy-logic-based theory of maximization were applied to a sample population of locations available for hydropower generation including a on where a hydropower plant had already been installed and was operating at rated capacity. The efficiency of the algorithm in identifying this location was analyzed to determine the suitability of the algorithms in estimating the ideal location for hydropower plants. The results showed that both approaches were able to identify the most suitable location, but when the time taken to make the identification was taken into account, fuzzy logic was found to perform better than the bat algorithm as the former took only one iteration to identify the location, whereas the latter needed six iterations but the sensitivity with which the algorithms identified the ideal location was better for bat than fuzzy.