Neural Network Applications in a Power Station
T. K Sai1, K. A. Reddy2
1T. K. Sai, received the B.Tech. Degree in Electronics and Instrumentation Engineering from Kakatiya Institute of Technology and Science (KITS), Warangal.
2K. Ashoka Reddy, received the B.Tech. Degree in Electronics and Instrumentation Engineering from Kakatiya Institute of Technology and Science (KITS), Warangal.
Manuscript received on January 02, 2014. | Revised Manuscript received on January 04, 2014. | Manuscript published on January 05, 2014. | PP: 112-120 | Volume-4 Issue-6, January 2014. | Retrieval Number: F2478014615/2015©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The integration of Soft Computing techniques in traditional real-time systems is a promising approach to cope with the growing complexity of real-world applications. A power station is a complicated multivariable controlled plant, which consists of boiler, turbine, generator, power network and loads. The demands being placed on Control & Instrumentation engineers include economic optimization, practical methods for adaptive and learning control, software tools that place state-ofart methods . As a result, Neural network applications are explored in Measurement and Control. In real time systems, Information plays a vital role for the efficient operation and maintenance in a power station. However there are limitations on making available information online due to instrumentation limitation, hazardous environment condition etc. The Furnace Exit Gas Temperature (FEGT) is an important design and operating parameter. The furnace of a boiler is such a zone where online measurement of temperature is difficult because of high temperature and adverse conditions. Considering the complexity of power plant operating condition and number of parameters involved, the best solution to this problem lies in adopting the Neural Networks to measure FEGT in a 500 MW Thermal Power Plant. Also, Steam temperature Control is one of the most challenging control loops in a power plant boiler because it is highly nonlinear and has a long dead time and time lag. . The Superheated temperature is to be controlled by adjusting the flow of spray water to within +/- 10 deg C during transient states and +/- 5 deg C at the steady state. A neural network based Model Predictive Control ( MPC ) is proposed in this paper
Keywords: Neural Networks, Boiler, Superheater temperature, Furnace exit gas temperature, Measurement Control, Power Plan