Background: A large number of nature-based optimization methods have been proposed to use as efficient tools in scientific studies. Genetic Algorithm (GA), which operates based on human genetical evolution, has been an outstanding mostly used solver in a wide range of applications. This algorithm reflects the process of natural selection where the fittest individuals are selected for reproduction in order to produce offspring of the next generation. Initialization, selection, crossover, and mutation are the main parts of the GA population-based method which enables GA to have a prominent explorative feature. On the other hand, the Teaching Learning Based Optimization algorithm (TLBO) is of great performance during searching for the optimum solution among individuals. Therefore, it is expected that the combination of both algorithms in a certain logical way improves the optimization time.
Objective: The study intends to determine ways of improving the performance of the TLBO algorithm to solve a complex non-linear problem. Power system studies are one of the most complex problems for analysis. Therefore, a powerful heuristic optimization procedure would have a valuable contribution to solving such problems. In addition, the proposed heuristic algorithm will help scientists to apply the technique to their problems.
Methodology: According to the aforementioned explanation, a new efficient optimization approach is proposed which optimizes the parameters of multi-machine power system stabilizers (PSSs). The TLBO algorithm includes two different stages in its main structure, which are aptly called teacher and student stages. The student stage of TLBO is replaced by the genetic algorithm in order to improve the explorative feature of the main TLBO. The PSS parameters are obtained for four PSSs which are connected to four generators.
Results: The performance of the proposed stabilizer is compared with other formerly designed stabilizers reported in the literature consisting of multi-band PSSs for two areas four-machine power system. Simulation results demonstrate the effectiveness and robustness of the proposed PSS in damping local and inter-area oscillation modes under various disturbances and confirm its superiority in comparison with other types of PSSs.
Conclusion: A search heuristic method like the genetic algorithm can dramatically improve the performance of meta-heuristic optimization technique. In actuality, the TLBO as a meta-heuristic optimization technique suffers from a direct search of random solutions in its student stage. Then, the TLBO relinquishes some parts of search space which may restrict the algorithm to find absolute maximums or minimums. In this condition, the GA with a great ability in searching the whole search space effectively improves the TLBO. According to the obtained results, the proposed algorithm, named Genetic-TLBO, obviates the conventional TLBO flaws successfully.