Simulated annealing with local search-a hybrid algorithm for unit commitment

Abstract

A hybrid algorithm has been developed for the solution of the unit commitment problem. This hybrid technique uses simulated annealing as the main algorithm. At each temperature, fresh solutions are generated randomly, and with a high likelihood of being feasible. Local search is made in the neighborhood of the best solution, using a heuristic decommitment technique. The hybrid algorithm is robust and has improved convergence, compared with earlier algorithms. The results obtained in system studies indicate its potential for solving the unit commitment problem.