Title : An integrated mathematical modelling and ANFIS framework for groundwater quality assessment and contaminant transport
Abstract:
To protect and sustainably manage water resources, water quality assessment and reliable contaminant transport prediction are crucial. This keynote introduces an integrated framework of mathematical modelling, numerical simulation, artificial intelligence (AI) and field-based water quality assessment. For the derivation of the analytical solution of the one-dimensional advection–dispersion equation (ADE), the Laplace Transform Technique (LTT) is used and for obtaining its numerical solution, the Finite Difference Technique (FDT) is used. An Adaptive Neuro-Fuzzy Inference System (ANFIS) is also improved to be a computationally efficient surrogate model for predicting contaminant transport.
The applicability of the framework is illustrated using a field-based water quality assessment of Dhurwa and Kanke Dams in Ranchi, India.A field based water quality evaluation of the Dhurwa and Kanke Dams, Ranchi, India is used to illustrate the applicability of the framework. The standard method prescribed under IS 3025 was used to analyze the key physico-chemical parameters and the Water Quality Index (WQI), statistical analysis, and correlation analysis were used to assess the water quality and to find relationships among the various parameters analyzed. The findings show that some parameters are found to be beyond the allowable limit as prescribed by Bureau of Indian Standards (BIS). The water quality of Kanke Dam was comparatively poor which could be related to agricultural runoff, industrial activities and poor waste management.
The modelling results show that the contaminant concentration builds up over time and diminishes with transport distance. RMSE, MAE, MAPE and R² values showed the excellent agreement of the analytical, numerical and ANFIS solutions. Additionally, the AI-based prediction framework is validated by uncertainty analysis, ensuring its robustness and reliability. Overall, the integrated approach can be a reliable and computationally efficient tool for the contaminant transport prediction and water quality evaluation and can be applied into the sustainable management of water resources.
Keywords: Mathematical Modeling; ANFIS; Water Quality; Contaminant Transport; Groundwater.
