Designing an isolation valve system (IVS) is a critical task to improve resilience toward unexpected failures. During isolation, the hydraulics (nodal pressure, flow direction, and velocity) of the system will change. Unlike nodal pressure, flow direction and velocity changes may detach accumulated materials and induce unexpected water quality failure (WQF). In general, failure induced by pressure changes has been considered for IVS design, but WQF has not been considered yet. This study proposes a new IVS design approach to prevent unexpected WQF induced by isolation. For this, we proposed a Flow-characteristic change Ratio (FCR) which accounts for flow direction and velocity changes before and after isolation. FRC will have a higher value if the pipe did not experience either flow direction changes or high velocity (self-cleaning velocity; 0.35 m/s here) before isolation but experienced after isolation. With these considerations, the optimal design model was developed using a genetic algorithm considering reliability as an objective function and FCR as constraints. The model has been applied to a synthetic grid network and real systems and the results are compared with the traditional approach (without FCR as constraints). The results have been compared by robustness of hydraulics and hydraulic geodesic index (HGI) which is a graph theory measure quantifying hydraulic connectivity. Preliminary results showed proposed approach can prevent potential WQF threats while the robustness of hydraulics and HGI outperform compared the traditional approach. The proposed approach is expected to be a useful alternative approach to preventing unexpected consequences of traditional approaches.
