| Paper Details: | Downloads: 1096 |
| Serial Number: | P1121340294
|
| Title: | Unit Commitment Using a Hybrid Differential Evolution with Triangular Distribution Factor for Adaptive Crossover
|
| Authors: | N. Malla Reddy and K. Ramesh Reddy and N. V. Ramana
|
| Abstract: | In the present day power scenario Unit Commitment (UC) is one of the complex challenging tasks for power system operators. UC is a nonlinear, non-convex, large scale, mixed integer problem. To mitigate this complex problem in this paper a hybrid Differential Evolution with local search technique and an adaptive Crossover using triangular distribution factor (DE-TCR) is presented. The salient features of the proposed DE-TCR are: An intelligent chromosome representation is used which is independent of number of units present in UC problem thereby reducing the length of chromosome. It is able to interlink the cross over probability in conjunction with the non-separable and decision variable dependency of UC problems. Local search using Sequential Quadratic Programming, which has proved in improving the performance of the classical DE algorithm. Initially, the proposed DE-TCR is used to determine an optimal generation schedule for each hourly demand. Later, SQP is utilized to find the optimal dispatch strategy to minimize the fuel cost. The effectiveness of the proposed algorithm is tested on standard 4 units, 8 hour and 10 units, 24 hour UC systems. Results demonstrate that the proposed algorithm can perform better and produce global optimal solutions compared to that of other reported methods.
|
| Keywords: | Unit Commitment, Differential Evaluation, Sequential Quadratic Programming, Adaptive cross over and Triangular Distribution factor
|
| Journal/Conference: | International Journal of Artificial Intelligence and Machine Learning
|
| Volume: | 14
|
| Issue: | 1
|
| Submission Date: | 9/28/2013 12:00:00 AM
|
| Review Date: | 2/20/2014 12:00:00 AM
|
| Publishing Date: | 3/6/2014 12:00:00 AM
|
| Article Downloads: | 1096
|
| Download: |
|