ASSESSMENT OF DROUGHT FORECAST USING FUZZY LOGIC

WFL is based on the notion of decomposing predictors and predictands into frequency bands and then utilizing those bands to rebuild the predictand series. The strongest frequency bands were determined by averaging the wavelet spectra of predictors and the predictand. It was shown that WFL performed far better than a fuzzy logic model without wavelet banding when this hybrid model was applied to the state of India. WFL was shown to be more precise than both the ANN and WANN models when it came to predicting the occurrence of droughts. It's important to remember that ENSO fluctuation is not a universal indicator of upcoming drought. Because of this, considerable effort is needed to discover relevant independent predictors before using such a data-driven model in diverse locations. Long-range, precise drought forecasting has substantial engineering applications.