Enhanced growth optimizer algorithm with dynamic fitness-distance balance method for solution of security-constrained optimal power flow problem in the presence of stochastic wind and solar energy

In modern power system applications, security constrained optimal power flow (SCOPF), in which different contingency operating situations arise, is a complex, non-convex, and nonlinear optimization problem. With the inclusion of stochastic solar and wind energy sources in the power system, the energy efficiency is maximized. However, this increases the complexity of the SCOPF problem and makes its solution difficult. To tackle these challenges, it's essential to design an algorithm with a search behavior that aligns with the characteristics of the OPF/SCOPF problem. For this reason, in this paper, a novel dynamic fitness-distance balance based growth optimizer (dFDB-GO) algorithm was proposed to find the optimal solution of the OPF/SCOPF problem. By using the dFDB method, the guide selection in the learning stage of the GO algorithm was redesigned to improve the search capability and to provide a good balance between exploration and exploitation. To prove the performance of the dFDB-GO algorithm, the simulation study was performed on the SCOPF and benchmark problems. In the simulation study carried out on the optimization of benchmark problems using the six variations of GO and the base GO algorithms, all variations outperformed the base GO algorithm according to Friedman test results. In other simulation study, the performance of the proposed algorithm was tested on OPF/SCOPF problem including wind and solar energy sources. One of the important contribution of this study is to present a comprehensive simulation study to the literature, including twelve case studies using modified IEEE 30-bus and 57-bus systems and nine different objective functions. These case studies were solved by the dFDB-GO and eleven up-to-date MHS algorithms. The another important contribution was that a comprehensive analysis was carried out using the minimum, maximum, mean, and standard deviation values of the algorithms and the statistical analysis methods. Accordingly, the proposed dFDB-GO algorithm achieved better results than its rival in 11 of 12 case studies. Moreover, it ranked first with 1.3077 value among its competitors according to Friedman test. On the other hand, the stability of an algorithm on solving SCOPF problem was evaluated for the first time in this study. For this, while the dFDB-GO algorithm achieved 94.87% mean success rate for solving the SCOPF problem, the GO algorithm had the 12.09% mean success rate value. To sum up, all analysis results prove the superior performance of the proposed algorithm in solving the SCOPF problem against to its rivals.