Application of Quasi-oppositional driving training-based optimization for feasible Optimal power flow solution of renewable power system with unified power flow controller

Tushnik Sarkar ¹ Chandan Paul ¹ Susanta Dutta ¹ Provas Roy ² Ghanshyam Tejani ^3,4 Seyed Jalaleddin Mousavirad ^5*

• ¹ Dr B C Roy Engineering College, Durgapur, Durgapur, West Bengal, India
• ² Kalyani Government Engineering College (KGEC ), Kalyani, West Bengal, India
• ³ Yuan Ze University, Zhongli District, Taiwan
• ⁴ Applied Science Private University, Amman, Amman, Jordan
• ⁵ Mid Sweden University, Sundsvall, Västernorrland, Sweden

The current study’s objective is to reveal the best possible solution for an optimal power flow (OPF) problem. The driving training-based optimization (DTBO) technique has been applied in this work to achieve the goal where quasi-oppositional based learning (QOBL) has been integrated with DTBO and referred to as quasi-oppositional driving training-based optimization (QODTBO). The experiments have been executed on IEEE 57 & 118 bus systems. Four different test scenarios have been considered here. The first one is the traditional IEEE 57 bus network; IEEE 57 bus with renewable energy sources (RESs) (i.e. Solar and Wind units) is chosen in the second one, the third one considers IEEE 57 bus with RESs & unified power flow controller (UPFC) & finaly IEEE 118 bus network with RESs and UPFC. In each test scenario, there are four objective functions, among which one is single objective & three of them are multi-objective. The acquired test outcomes by QODTBO, have been contrasted with the outcomes found by the use of DTBO, backtracking search optimization algorithm (BSA) & sine cosine algorithm (SCA). The effect of inherent uncertainties within RESs is gauged in the current study by the choice of appropriate probability density functions (PDF). Based on experimental outcomes using different optimization techniques over thirty trials a statistical report has been prepared which ascertains that QODTBO is the most robust optimization scheme among the optimization tools taken into consideration in this study. To represent the statistical analysis, pictorially box plots and error-bar plots are provided. One way ANOVA tests have also been conducted on test outcomes to enhance degree of reliability of the inferences made based on statistical results. From this work, it is also explored that integrating RESs & UPFC with traditional IEEE-57 bus system can improve the overall execution of the test system. If the performance of the conventional system, RESs based System, RESs and UPFC based system are observed, it can be noticed that RESs based system gives better result by 1.364790635% and RESs and UPFC based system gives 2.175247484 % better result as compared to the conventional system when cost reduction is concerned.