Fuzzy and improved fuzzy-wavelet approach in modeling municipal residential water consumption estimation using climatic variables

Abstract

This work highlights the importance of fuzzy-wavelet denoise and fuzzy-wavelet compress in modeling the municipal residential water consumption estimation. To begin, fuzzy logic is used with different rules, membership criteria and fuzzy set. Based on accuracy of the developed model, optimum number of rules and best membership function were selected. To improve the accuracy of the single fuzzy model, wavelets technique (denoise and compress approach) was coupled with fuzzy logic and results were compared to single fuzzy technique. To map the input and output functions, the present research work includes Mamdani fuzzy inference approach based on various climatic input variables like rainfall, maximum temperature, minimum temperature and relative humidity. The models were trained based on climatic data to a certain period, and corresponding estimated models were tested for the same period. Result highlights that models with denoise and compress approach have better accuracy compared to single fuzzy model. © 2020, Springer-Verlag GmbH Germany, part of Springer Nature.