101. Studies on a PC-based Fuzzy and Non-fuzzy On-Off Controller for a Temperature Process
- Author
-
S. Pal, S. K. Sen, G. Sarkar, and Anish Deb
- Subjects
Artificial neural network ,Computer science ,business.industry ,Competitive learning ,Computer Science::Neural and Evolutionary Computation ,Supervised learning ,Power factor ,AC power ,Backpropagation ,Euclidean distance ,Electric power system ,Control theory ,Artificial intelligence ,business ,Transformer (machine learning model) - Abstract
This paper presents a research work on artificial neural networks (ANN) to examine whether the Power Systems is secured under steady-state operating conditions. ANN determines the minimum bus voltages and maximum ratio of line-flow to permissible line- flow. Detailed load flow study is avoided if the values supplied by ANN satisfy these operating constraints. For training, fast decoupled load flow data are used. The Artificial Neural Networks used are Multilayer feed forward Network with error Backpropagation Algorithm and a Counterpropagation Neural Network. The CPNN is a network which can obtain a mapping from inputs to outputs by competitive learning and supervised learning. Extensive studies have been made by varying the network parameters of both the networks. The hidden neurons are also varied to fix the optimum architecture for the problem to be solved. The input variables to the network are the active Power of the load buses, Power factor of the loads and the net generated Power of the generating buses. The generating bus voltages and tap settings of the transformer are assumed to be constant (fixed). The algorithms are tested for security assessment on an IEEE-14 bus Systems and the test results are presented. Results of the both the ANN closely agrees with that obtained by fast decoupled load flow. The computation time of both the ANN is much smaller than that by fast decoupled load-flow. However, comparing the training time and suitability for online application in Power Systems, CPNN is best suited due to its fast learning based on Euclidean distance calculations.
- Published
- 2007
- Full Text
- View/download PDF