Two-phase artificial intelligent-based technique for online optimal economic dispatch

A novel two-phase technique is proposed in this paper for solving the Economic Load Dispatch (ELD) problem in the online mode. It is sometimes necessary to find a fast optimal solution especially for the short-term highly-constrained ELD. This arises when considering power-generation-business profit, where fast computation simulation time scheduling of ELD is valuable. Since the solution of the ELD is a repetitive process that results in a remarkable cost saving, it is vital to find a practical methodology for obtaining a fast but accurate solution for this problem. In this paper, a two-phase Artificial Intelligent-(AI) based framework is developed to solve this problem in the online mode. The idea is to acquire database by solving the ELD problem under different load demands including extreme cases. Then, the results are generalized in the second phase using intelligent techniques to obtain solutions for new situations avoiding carrying out the classical solution algorithms over again. This approach offers the flexibility of the online modification of the operating circumstances according to the observed variations in the load demands. Results obtained from this research at different operating conditions confirm the high accuracy of the proposed generalization technique and its flexibility when dealing with the ELD problem.