This paper proposes an innovative genetic algorithm (GA) approach to solving the thermal unit commitment (UC) problem using a constraint satisfaction technique. Minimum up-time and down-time constraints on the generating units are embedded in the delicately designed binary strings to represent the on-off states of the units. Ramp rate constraints on the units being started up or shut down are tackled in the economic dispatch subprogram by limiting the associated maximum available capacities for generating. Violations of the other constraints are considered by integrating penalty factors into the cost function. The developed algorithm is further paralleled on an 8-processor transputer network, processors of which are arranged in master-slave and dual-direction ring structures, respectively. The proposed approach is tested on the simple 4 thermal units system and the practical Taiwan Power system of 38 thermal units. Speed-up and efficiency for each structure with different number of processors are compared to those of the sequential GA approach. The proposed approach is shown to be well amenable to parallel implementation.
[1]
Francisco D. Galiana,et al.
Unit commitment by simulated annealing
,
1990
.
[2]
Gary Boone,et al.
Optimal capacitor placement in distribution systems by genetic algorithm
,
1993
.
[3]
G. Sheblé,et al.
Genetic algorithm solution of economic dispatch with valve point loading
,
1993
.
[4]
Anil Pahwa,et al.
Optimal selection of capacitors for radial distribution systems using a genetic algorithm
,
1994
.
[5]
Allen J. Wood,et al.
Power Generation, Operation, and Control
,
1984
.
[6]
Vladimiro Miranda,et al.
Genetic algorithms in optimal multistage distribution network planning
,
1994
.