Genetic algorithm unit commitment program
Date
1997
Authors
Kondragunta, Sridhar
Major Professor
Advisor
Lamont, John W.
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The deregulation of the electric power market gives the problem of unit commitment a different dimension. The goal of the problem is maximizing the profit and management of spinning reserves instead of the traditional cost minimization of production. Among the traditional methods, dynamic programming and linear programming have been very popular until a more recent method called Lagrangian relaxation came into picture. But these methods are either computationally inefficient and suffer from the curse of dimensionality, require the problem to be convex, or yield suboptimal solutions because of the discrete nature of the problem. On the other hand, the genetic algorithm approach is based on the philosophy of proliferation of more fit candidates. It is a robust technique and has an inherent advantage of providing multiple solutions with different operating scenarios but with approximately equal objective values. An effort has been made to solve unit commitment by the genetic algorithm approach with all its limits and real time constraints by taking the external market transactions at each hour into account.
Series Number
Journal Issue
Is Version Of
Versions
Series
Academic or Administrative Unit
Type
thesis