Design and implementation of heavy metal prediction in acid mine drainage using multi-output adaptive neuro-fuzzy inference systems (ANFIS) - a case study

This paper reports an attempt to show how acid mine drainage (AMD), as well as other heavy metals, pollute the environment and how this problem can be resolved. AMD is considered to be the main source of environmental pollution in areas where mining operations are undertaking. Since AMD and the factors that control it are of prime importance regarding the environmental preservation activities, this study investigates the presence of heavy metal pollutants in AMD. To achieve this goal, we implemented the ANFIS method to predict the presence of heavy metals (Zn, Mn, Fe, and Cu), taking into account pH, as well as SO4 and Mg concentrations.  Having used the ANFIS method, the comparison of predicted concentration with calculated data resulted in correlation coefficients of 0.999, 0.999, 0.999, and 0.999 for Cu, Fe, Mn, and Zn, respectively. The employed procedure proved to be easy to use and cost-effective to foresee the presence of heavy metals in AMD.

Keywords:

ANFIS , Heavy metals , Acid mine drainages , Shur River , Sarcheshmeh copper mine