This paper proposes an improved firefly (FF) algorithm with multiple workers for solving the unit commitment (UC) problem of power systems. The UC problem is a combinatorial optimization problem that can be posed as minimizing a quadratic objective function under system and unit constraints. Nowadays, highly developed computer systems are available in plenty, and proper utilization of these systems will reduce the time and complexity of combinatorial optimization problems with large numbers of generating units. Here, multiple workers are assigned to solve a UC problem as well as the subproblem, namely economic dispatch (ED) in distributed memory models. The proposed method incorporates a group search in a FF algorithm and thereby a global search is attained through the local search performed by the individual workers, which fine tune the search space in achieving the final solution. The execution time taken by the processor and the solution obtained with respect to the number of processors in a cluster are thoroughly discussed for different test systems. The methodology is validated on a 100 unit system, an IEEE 118 bus system, and a practical Taiwan 38 bus power system and the results are compared with the available literature.
Distributed memory models, economic dispatch problem, firefly algorithm, multiple workers, parallel computing, unit commitment problem
KOODALSAMY, BANUMALAR; VEERAYAN, MANIKANDAN BAIRAVAN; KOODALSAMY, CHANDRASEKARAN; and SIMON, SISHAJ PULIKOTTIL
"Firefly algorithm with multiple workers for the power system unit commitment problem,"
Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 24:
6, Article 18.
Available at: https://journals.tubitak.gov.tr/elektrik/vol24/iss6/18