Soft computing technique-based prediction of water quality index

Abstract

Water quality (WQ) plays a crucial role in management of water resources. Water quality index (WQI) is frequently used methods to assess of water quality for drinking purposes. WQI can be predicted using chemical analysis which might not, however, be viable for a longer period in all the country-scale rivers. Thus, in this investigation, two neural-based soft computing techniques-artificial neural network (ANN), generalized neural network (GRNN)- and one hybrid soft computing techniques- adaptive neuro-fuzzy interference system (ANFIS) with four membership function- were used to predict WQI in Khorramabad, Biranshahr and Alashtar sub-watersheds in Iran. Ten distinct physiochemical parameters were used as input variables and WQI as output. Simultaneously, a correlation plot and pairs were used to ascertain the relation of input and output variables. The soft computing techniques were compared using six fitness criteria: NSE, MAE, LMI, RMSE, MAPE, CC. Results indicated that ANN better predicted WQI than did GRNN and ANFIS. Among the different membership functions of ANFIS, ANFIS_trimf was far better than were other membership functions. Thus, it was concluded that ANN was a viable tool for the prediction of WQI.

Application of artificial intelligence-based modelling for the prediction of crop water stress index

“…The beauty of these modelling techniques is that they do not explicitly require the mathematical form of the model and still can give good results, even with minimum data. These techniques have been widely used in modelling various hydrological processes including predicting evapotranspiration (Adeloye et al 2012;Shiri et al 2020), reservoir operations (Goyal et al 2013), streamflow prediction (Muhammad Adnan et al 2019), forecasting of dam water level (Hipni et al 2013), infiltration modelling (Sihag et al 2020), water quality prediction (Abobakr Yahya et al 2019;Singh et al 2021), agricultural drought prediction (Abbasi et al 2020), rainfall-runoff modelling (Chandwani et al 2015), groundwater potential mapping (Naghibi et al 2018) and many more (ASCE 2000;Deka 2014). However, studies concerning the application of AI techniques in modelling the CWSI are very limited, despite their proven potential in numerous hydrological processes.…”