Introducing improved atom search optimization (IASO) algorithm: Application to optimal operation of multi-reservoir systems

Abstract

Nowadays, optimal reservoir operation is regarded as one of the most important issues in water resources management. In this regard, several optimization algorithms have been used to evaluate the issue of optimal reservoir operation. To increase the accuracy and proximity of the solutions to the absolute optimal solution, the improved atom search optimization (IASO) algorithm was introduced with a mutation strategy to improve its global search capability, as an important factor in preventing trapping in a locally optimal solution. This article implemented the atom search optimization (ASO) algorithm and its improved model to improve the results of the objective function in multi-reservoir systems. Comparing the IASO algorithm with the basic ASO algorithm on the mathematical benchmark functions revealed good accuracy and the superiority of the IASO algorithm. Then, sensitivity analysis was performed on the effective coefficients of both algorithms, in which alpha and beta, as the depth and weight coefficients used in the calculation of atoms acceleration, were considered 50 and 0.5, respectively. The results showed the relatively good accuracy of the ASO algorithm in the case of multi-reservoir systems, but implementing and comparing the results of the improved algorithm, showed improvement so that the best IASO results were improved from 95.33% to 99.99% as close to the global four-reservoir system response of 308.29 and from 89.67 to 94.36% with the solution of the ten-reservoir system in the amount of 1194.44.