Mutation particle swarm optimization algorithm for solving the optimal operation model of thermal power plants

To solve joint optimization scheduling problem of thermal power plants based on the electricity market, the constraints and characteristics of optimal operation of thermal power plants are analyzed. From the perspective of economics and environmental science, we propose and establish an electricity market based joint optimization scheduling model of thermal power plants. Mutation particle swarm optimization algorithm (MPSO) is proposed to solve the model. Using an authentication instance, we compare and analyze the performances of the MPSO method and the mixed integer programming (MIP) method in solving the model. Further, considering a commonly used system, we show the efficiency of the MPSO method in solving the model; we apply two case studies to illustrate the performance of the MPSO method in solving the thermal unit commitment (UC) problem. To demonstrate the effectiveness of the MPSO method in solving the UC problem, we use MPSO method, PSO-LR (Lagrangian relaxation) H. H. Balci and J. Valenzuela, Int. J. Appl. Math. Comput. Sci. 14(3), 411–421 (1989), improved PSO B. Zhao et al., Elect. Power Energy Syst. 28, 482–490 (2006), mimetic algorithm J. Valenzuela and A. Smith, J. Heuristics 8(2), 173–195 (2002), adaptive LR W. Ongsakul and N. Petcharaks, IEEE Trans. Power Syst. 19(1), 620–628 (2004), stochastic priority list (PL) T. Senjyu et al., Electr. Power Syst. Res. 76, 283–292 (2006), PL based evolutionary algorithm D. Srinivasan and J. Chazelas, in International Conference on Power System Technology (2004), Vol. 90, pp. 1746–1751, enhanced adaptive LR W. Ongsakul and N. Petcharaks, IEEE Trans. Power Syst. 19(1), 620–628 (2004), mixed-integer linear programming (MILP) M. Carrin and J. M. Arroyo, IEEE Trans. Power Syst. 21(1), 321–332 (2006), MILP G. Morales-Espana et al., IEEE Trans. Power Syst. 28, 4897 (2013), NBD T. Niknam et al., Appl. Energy 86, 1667–1674 (2009), generalized new benders decomposition approach (NBD) S. Rahimi et al., World Appl. Sci. J. 6(12), 1665–1672 (2009), LR and genetic algorithmC. P. Cheng et al., IEEE Trans. Power Syst. 15(2), 707–714 (2000), and seeded memetic algorithm (SMA) J. Valenzuela and A. Smith, J. Heuristics 8(2), 173–195 (2002) to solve the UC problem, simulation results are compared with the ones obtained by other methods. It is shown that the optimal operation model is right, the MPSO method is better than MIP method in solving the model, and they have universality. The effectiveness of the MPSO method in solving the UC problem is better than other methods. The MPSO method has better solutions in comparison with other methods, especially for systems with a large number of units.