The Study of Unit Commitment Based on Adaptive Mixed-Variable Particle Swarm Optimization Algorithm

The unit commitment (UC) problem is a crucial type of economic dispatch problem in power systems, and its mathematical model is a complex, large-scale mixed-integer nonlinear problem, making it difficult to obtain a theoretical optimal solution. To address this, an adaptive mixed-variable particle swarm optimization (PSO) algorithm is proposed in this paper, considering the remaining cycle count of the energy storage battery. This approach not only enhances the traditional PSO, which tends to fall into local optima, but also handles continuous, binary, and integer variables effectively. To evaluate the impact of transfer functions on the algorithm, eight different transfer functions are compared. Results from two case studies demonstrate that the proposed algorithm can efficiently solve the UC problem with nonlinear mixed variables, while also being robust. This makes it a feasible method to mitigate the unpredictability and stochastic nature of sustainable distributed generators and other intermittent energy sources.