Multiobjective Multi Verse Optimization Algorithm to Solve Dynamic Economic Emission Dispatch Problem with Transmission Loss Prediction by an Artificial Neural Network

This work presents a potent multiobjective multiverse algorithm to solve the highly complicated dynamic economic emission dispatch problem. Solving the dynamic economic dispatch problem with power balance equality constraint, valve point constraints, and ramp limit constraints not only minimizes the fuel cost but also reduces the environmental pollution caused by the thermal stations. A novel algorithm is proposed where the dynamic economic dispatch problem with losses is converted into a lossless dispatch by using an artificial neural network that predicts the transmission loss of the power system. A fuzzy membership-based approach is used to select the best compromise solution from the existing Pareto Optimal solution during each hour of the dispatch period. The proposed algorithm is first tested on five and ten thermal unit systems with B-loss coefficients to prove the competitive behavior of the algorithm. The proposed algorithm outperforms many of the state-of-the-art methods. Then the algorithm is tested on IEEE 30 and IEEE 118 bus system with widely varying demand patterns along with an offline trained neural network. The results reveal the algorithm is much faster and can outperform the existing algorithms in minimizing the fuel cost and reducing pollutant emission