Prediction of rainfall, maximum and minimum temperature in summer monsoon months using artificial neural network

Artificial neural networks (ANNs), are computer based mathematical processing systems inspired by the biological neural networks on which the animals brain works and is being used by state of the art computer technology for faster computing systems. In this paper this advanced computing system has been utilized for forecasting rainfall, maximum and minimum temperature during summer monsoon months from ANN models that has been generated by inputting 8 years(2014 to 2021) relevant and available meteorological data. Rainfall amount, maximum and minimum temperature in summer monsoon months are important inputs for crops in kharif season, and if forecasted with accuracy with substantial lead time, then it will be helpful in optimization of crop yield which subsequently will boost the economy of our country. The rainfall, maximum and minimum temperature of year 2022 & 2023 of summer monsoon months was predicted by five best fit models by using twelve meteorological parameters data of previous day and the results found were encouraging. ANN-based forecasting model formula has been created by constructing three layer networks. The data, on analysis, revealed small deviation in forecast values during the event of normal rainfall, maximum and minimum temperature and the model captures nicely the normal trend except in the event of sudden steep change of occurrence of rainfall or maximum and minimum temperature, in which case ensemble forecast of NWP and ANN can be explored.