Integrated approach for optimizing groundwater monitoring systems using evolutionary algorithms

This study aimed to develop a new approach to build a functioning groundwater monitoring system by detecting a reduced set of observation wells (OWs) that optimally matches the hydraulic heads measured by other OWs within the field, namely as leader wells (LWs). The optimization models used in this work are the well-known genetic algorithm (GA) and modified genetic algorithm (MGA) and a new progressive combination (PC) model. Optimization was applied to achieve three sequential selection processes: best input combinations (BICk), LWs and core leader wells (CLWs). This approach was applied to the Assiut New Barrage (ANB), a megaproject located in Assiut city, Egypt. The results show that nine LWs among 33 OWs are adequate for regular monitoring, with a reduction ratio of 72.72%. Moreover, assigning CLWs among LWs increases the accuracy of fitting to existing OWs, and helps in understanding the spatial relationships among OWs.