Dynamic Economic Load Dispatch Using Hybrid genetic algorithm and the Method of Fuzzy Number Ranking

Publication Type:
Conference Proceeding
Proceedings at 2005 International Conference on Machine learning and Cybernetics, 2005, pp. 2472 - 2477
Issue Date:
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2005002788.pdf376.88 kB
Adobe PDF
This paper proposes a new economic load dispatch model that considers cost coefficients with uncertainties and the constraints of ramp rate. The uncertainties are represented by fuzzy numbers, and the model is known as fuzzy dynamic economic load dispatch model (FDELD). A novel hybrid genetic algorithm with quasi-simplex techniques is proposed to handle the FDELD problem. The algorithm creates offspring by using generic operation and quasi-simplex techniques in parallel. The quasi-simplex techniques consider two potential optimal search directions in generating prospective offspring. One direction is the worst-opposite direction, which is used in the conventional simplex techniques, and the other is the best-forward direction, which is a ray from the centroid of a polyhedron whose vertexes are all the points but the best one towards the best point of the simplex. In addition, in order to reserve more fuzzy information, the fuzzy number ranking method is used to optimize the cost function, avoiding the lost some useful information by getting /spl lambda/-level set. The experimental study shows that FDELD is more practical model; the algorithm and techniques proposed are very efficient to solve FDELD problem.
Please use this identifier to cite or link to this item: