Optimal Operation of Multi-reservoir System Utilizing DEA, AIDE Algorithm and Flood Control Assessment by MCDM Approach

A multi-reservoir system's operation during floods is a complex problem, because it is dynamic and nonlinear nature, finding the global or near-global best solution is a difficult task. The Adaptive Immune Differential Evolution (AIDE) algorithm is one of the Evolutionary Algorithms (EA) used to solve the multi-reservoir system. It improves the Differential Evolution (DE) algorithm's exploitation and exploration capabilities. A multi-criteria decision-making (MCDM) approach is also implemented for managing flood control operations in reservoirs, aiming to handle correlations among different criteria. To eliminate correlation, principal component Analysis (PCA) is used and coupled with a weight vector as the input to the TOPSIS method, WASPAS method, and MOORA method by which the alternatives are to be determined. The results show that the dimensionality of the criterion system is lowered while simultaneously eliminating the correlation between criteria, and the ranking order of the alternatives is fair. From the results, it is clear that the AIDE algorithm would have faster convergence and a powerful global ability than the DE algorithm. The control parameters used in DE and AIDE enable these algorithms to effectively navigate complex search spaces and identify optimal or near-optimal solutions. The above optimization method is recommended for complex, large-scale reservoir operations and evaluation of ranking by MCDM approach is helpful for flood control operation of multi-reservoir systems. However, the optimal alternative sequence (3, 5, 2, 4, and 1) is suitable to manage the flood event operation by multi-reservoir.