An enhanced real coded approach for the optimization of the Unit Commitment Problem

In the present study, a novel enhanced real-coded approach is proposed for the simultaneous optimization of Unit Commitment and Economic Dispatch. In the proposed method, the objective variables represent the power generated by each unit at each hour. By utilizing the operational characteristics of each unit, its states and generation levels during the scheduling period are optimized concurrently. Moreover, a repair mechanism, which stochastically adjusts the generation of the units, is implemented to ensure the feasibility of the derived operation schedules. The solver employed for the optimization of the Unit Commitment Problem using the proposed method is L-SHADE, which is a state-of-the-art adaptive algorithm based on Differential Evolution. The real-coded approach is tested on a set of power systems consisting of up to 100 thermal units, available in the literature. The experimental results indicate that the proposed method exhibits more robust solution distributions and lower minimum operation cost for large scale systems compared to other state-of-the-art methods.