Economic dispatch for power generation using artificial neural network ICPE'07 conference in Daegu, Korea
This paper presents an optimal economic dispatch of electrical power plants by using back-propagation neural networks. The method of economic dispatch for generating units at different loads must have total fuel cost at the minimum point. There are many conventional methods that can use to solve eco...
Saved in:
Main Authors: | , |
---|---|
Format: | Conference Proceeding |
Published: |
2018
|
Subjects: | |
Online Access: | https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=58149087899&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/60319 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Chiang Mai University |
Summary: | This paper presents an optimal economic dispatch of electrical power plants by using back-propagation neural networks. The method of economic dispatch for generating units at different loads must have total fuel cost at the minimum point. There are many conventional methods that can use to solve economic dispatch problem such as Lagrange multiplier method, Lamda iteration method and Newton-Raphson method. However, an obstacle in optimal economic dispatch of conventional methods is the changed load. They are necessary to find the optimal economic dispatch from time to time. Moreover, they need a lot of time to repeat calculation for a new solution again. This paper presents back-propagation neural networks model to carry out instead the conventional Lamda iteration method. It is compared with the experimental results of electrical power system of 3 and 10 generating units respectively. The testing results of the back-propagation neural networks are compared with the Lamda iteration method by testing the teaching data and non-teaching data. It shows clearly that the back-propagation neural networks can find out the solutions accurately and use time to calculate less than other systems that are tested. Error of prediction will be increased slightly by the number of generating units in electrical power plants because it needs to learn a lot of input and output data in the neural network dramatically. © 2008 IEEE. |
---|