$\theta$-Multiobjective Teaching–Learning-Based Optimization for Dynamic Economic Emission Dispatch

This paper presents a -multiobjective-teaching-learning-based optimization algorithm to solve the dynamic economic emission dispatch problem. Teaching-learning-based-optimization (TLBO) algorithm works on the effect of a teacher on learners. This paper proposes several modifications on the basic TLBO. In the suggested method, the optimization process is done based on the phase angles, instead of the design variables themselves whereby the nonlinear characteristics of the problem are considered more efficiently. To avoid entrapping into local optima, a new learning method is proposed. Moreover, several metaheuristic techniques are applied to make a satisfactory multiobjective optimization method. A new approach is employed to select the population; thereby uniformly distributed Pareto-optimal front as well as the extreme points of the tradeoff surface can be achieved. Furthermore, a niching mechanism is applied to direct the individuals to seek the lesser explored regions. A fuzzy clustering approach is utilized to handle the size of the repository and obtain profitable solutions from the decision maker's point of view. To involve the decision maker's favor through the search process perfectly, a min-max approach is developed to select the best candidate solutions for the next generation. The applicability of the method is validated on three test systems, including 5-unit, 10-unit, and 120-unit test systems.