American Journal of Applied Sciences

NEURAL CASCADED WITH FUZZY SCHEME FOR CONTROL OF A HYDROELECTRIC POWER PLANT

A. Selwin Mich Priyadharson, T. Ramesh Kumar, M. S. Saravanan, C. ThilipKumar and D. Dileepan

DOI : 10.3844/ajassp.2014.1659.1670

American Journal of Applied Sciences

Volume 11, Issue 9

Pages 1659-1670

Abstract

A novel design for flow and level control in a hydroelectric power plant using Programmable Logic Controller (PLC)-Human Machine Interface (HMI) and neural cascaded with fuzzy scheme is proposed. This project will focus on design and development of flow and level controller for small scale hydro generating units by implementing gate control based on PLC-HMI with the proposed scheme. The existing control schemes have so many difficulties to manage intrinsic time delay, nonlinearity due to uncertainty of the process and frequent load changes. This study presents the design of neuro controllers to regulate level, cascaded with fuzzy controller to control flow in gate valve to the turbine. A prototype model is fabricated in the laboratory as experimental setup for flow and level control and real time simulation studies were carried out using PID and neural cascaded with fuzzy scheme. The designed prototype model is fabricated with 5 levels in the upper tank and 2 levels in the lower tank. Based on the outputs of the level sensors from the upper and lower tanks, the ladder logic is actuated. This project work uses PLC of Bernecker and Rainer (B and R) Industrial Automation inbuilt with 20 digital inputs and provides 12 potential free outputs to control the miniaturized process depicted in this work. Finally, the performance of the proposed neural cascaded with fuzzy scheme is evaluated by simulation results by comparing with conventional controllers output using real time data obtained from the hydro power plant. The advantages of the proposed neural cascaded with fuzzy scheme over the existing controllers are highlighted.

Copyright

© 2014 A. Selwin Mich Priyadharson, T. Ramesh Kumar, M. S. Saravanan, C. ThilipKumar and D. Dileepan. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.