Improving the energy efficiency of coal fired power plants using deep learning with optimisation algorithm

Reducing CO2 emissions from coal-fired power plants (CFPP) is crucial for sustainable energy production. Continuous emission monitoring systems (CEMS) are commonly used but are costly and complex. This paper proposes the integration of lignite coal with biomass to reduce emissions and enhance efficiency. Additionally, optimising feed water heaters (FWH) minimises heat transients and fuel consumption. Key parameters such as reheat steam temperature, turbine extraction pressure and temperature, and combustion ratio are considered. To further optimise energy efficiency, an artificial neural network (ANN) enhanced with the improved seagull optimisation algorithm (ISOA) is employed. Compared to ANN, GA, and PSO, the proposed approach significantly reduces system errors and improves overall plant efficiency. Experimental results demonstrate an energy efficiency of 98.95%, highlighting the effectiveness of CFPP optimisation with deep learning and advanced optimisation techniques.