Developing an Optimization Algorithm for Solving Economic Dispatch Problem in Microgrids

In the power system, one of the remarkably popular and fundamental optimization problems is economic power dispatch. Economic dispatch in a classical form contains only thermal generators without considering network security constraints. However, other forms of this problem like economic emission dispatch are gaining more importance since minimization of emission is predominant from the environmental point of view. Moreover, integrating renewable sources comes with challenges due to the stochastic nature of them. In this study, a multi-objective algorithm is developed to deal with the problem of economic emission power dispatch integrated with solar, wind, and small-hydro unit. Lognormal, Weibull and Gumbel distribution for predicting the accessible power of solar, small hydro, and wind power is utilized respectively. For the goal of the study, some of the traditional generators are replaced in the structure of the IEEE 30-bus network, with different renewable units of energy. Voltage limitation, capacities of the transmission line, prohibited areas of operation for the thermal generator plants, and restriction of the system are also considered. Multi-objective real coded non-dominated sorting genetic algorithm II is enforced to the problem of this study while incorporating a decent procedure of handling system restrictions, constraint domination, to meet the system limitations. Results are looked over in parts and discussed. The proposed method was then compared with two previous methods from another study to exhibit the robustness of the suggested technique. Results have been found to be significant. R-NSGA-II can reduce the cost up to $4,853.04 a year compared with the SMODE-SF method and $1,795.8 a year compared to the MOEA/D-SF method.