Topics in Intelligent Computing and Industry Design (ICID) 2(2) (2020) 01-04 Quick Response Code Access this article online Website: www.intelcomp-design.com DOI: 10.26480/etit.02.2020.01.04 Cite The Article: Ritesh Ranjan, Sujit Karmakar(2020).Thermodynamic Performance Analysis Of A 500mwe Supercritical Power Plant Based On Parametric And Neuro - Genetic Optimization Techniques. Topics In Intelligent Computing And Industry Design, 2(2): 01-04. ISBN: 978-1-948012-17-1 Ethics and Information Technology (ETIT) DOI: http://doi.org/10.26480/etit.02.2020.01.04 THERMODYNAMIC PERFORMANCE ANALYSIS OF A 500MWE SUPERCRITICAL POWER PLANT BASED ON PARAMETRIC AND NEURO-GENETIC OPTIMIZATION TECHNIQUES Ritesh Ranjan, Sujit Karmakar * Department of Mechanical engineering, National Institute of technology Durgapur, West Bengal, India *Corresponding Author Email: sujitkarmakar@yahoo.com This is an open access article distributed under the Creative Commons Attribution License CC BY 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ARTICLE DETAILS ABSTRACT Article History: Received 25 October 2020 Accepted 26 November 2020 Available online 03 December 2020 The thermal power plant efficiency depends on many parameters, which includes main steam pressure and temperature, intermediate turbine steam inlet (reheat) pressure, reheat temperature, deaerator pressure, low-pressure turbine steam inlet pressure to mention a few. An ANN-GA (neuro-genetics), parametric based optimization tools are performed to optimize chosen parameters of a 500 MWe super-critical high ash Indian coal fired power plant in this study for getting the best possible net plant efficiency. The net efficiency values of this base plant for random data generated between the ranges of all the chosen parameters is obtained using Fortran based commercially available software called ‘Cycle-Tempo’. Levenberg-Marquardt feed forward back propagation technique is used to train the data using Neural network modeling in MATLAB 2018a. Parametric optimization of the plant yielded 7.3% rise in the net plant energy efficiency, 7.32% rise in net exergy efficiency compared to that of base plant whereas neuro–genetic tool resulted in enhancement of energy ,exergy efficiencies by 7.8% and 7.82% respectively. Further, a comparative study is made between the results obtained through ANN-GA methodology and parametric optimization technique. KEYWORDS coal fired power plant, supercritical, exergy, artificial neural network, genetic algorithm. 1. INTRODUCTION A developing country like India needs electricity for fastening the growth and economic development. Currently the coal reserve in India is about 319 billion tonnes which is about 10% of world reserve (Ministry of Coal, 2019). In India, coal fired thermal power plant contributes about 54.2% of total installed capacity (Ministry of Power, 2019). The current scenario of sub-critical and supercritical Thermal power plant is as, around 88.65% is based on sub-critical technology and just around 11.35% is based on super-critical technology. Almost every coal based thermal power plant that uses coal consisting of high ash (HA) have net plant efficiency below 35% on HHV basis. Due to depletion of fossil fuel resources at a faster rate and increase in the emission of CO2 draws attention for the installation of more efficient thermal power plant so that emission of CO2 is less. As the number of super- critical thermal power plant is less in number and the present super- critical plant is working in low range and this allow researchers to optimise the various parameters in order to improve the net plant efficiencies as much as possible. The power plant efficiency depends on so many parameters like main steam pressure and temperature, deaerator pressure, reheat pressure and temperature and condenser pressure. Simultaneous optimization of these parameters is difficult task so the artificial intelligence tool like Artificial neural network (ANN) along with the Genetic algorithm (GA) is used to solve such complex problem. ANN is broadly applied in various field like optimization, forecasting, image and voice recognition. Suresh, Reddy and Kolar worked on neuro-genetic optimization of high ash coal-fired super- critical power plant (Suresh et al., 2011). Rajarshi, Jitendra and Tarun modelled a Thermal power plant using neural network and regression technique (Dixit et al., 2015). Genetic Algorithm is used for solving both unconstrained and constrained problems based on natural selection process that is inspired by biological evolution such as mutation, crossover and selection. Wagner, Pereira, Pius and Roberto applied genetic algorithm (GA) for optimising turbine extraction of a pressurized-water reactor (Sacco et al., 2002). In previous study, author have taken three different thermal power plants i.e. subcritical, supercritical and Ultra-supercritical based for optimisation of steam extraction pressure which will helps in enhancing the net plant efficiency along with decrease in specific fuel consumptions (Kumar et.al 2016). The study presented here used a neuro-genetic optimization technique consisting artificial neural network coupled with genetic algorithm for determining the maximum possible plant net efficiency. This power plant is based on the high ash coal-fired with super-critical steam condition. A 500 MWe super-critical power plant is used for optimization using ANN- GA technique. Also, in this study a comparison is done for energy and exergy analysis of optimised plant based on ANN-GA technique, Parametric optimization technique and further compare the base plant efficiency with parametric optimised plant and neuro-genetic optimised plant. This paper was presented at International Conference on Contemporary Issues in Computing (ICCIC-2020) - Virtual IETE Sector V, Salt Lake, Kolkata From 25th-26th July 2020