Optimizing Waste Heat Recovery in Marine Power Plants

Abstract

In today’s world, the demand for marine industries to improve energy efficiency and reduce environmental impact from marine power plants through waste heat recovery (WHR) is increasing daily. The energy efficiency of the marine engine is about 50% with a large amount of energy removed as waste heat, especially in exhaust gases. The present research paper proposes the optimization of the WHR by adopting the artificial neural network (ANN) model, exploring real-world data extracted from marine power plants on board to investigate the optimization of waste heat recovery. The ANN model, which was trained with operational parameters including exhaust gas temperature, engine load and ambient conditions, showed a very good predictive capability producing an R2 of 0.95, RMSE of 3.85kW, and MAE of 2.68kW. The ANN model proved to be more accurate and reliable than conventional thermodynamic methods considering fluctuating operational conditions. As indicated by the sensitivity analysis, exhaust gas temperature and engine load were found to produce the greatest impact on the potential of WHR. The results have a tremendous impact on the maritime sector, enabling ship operators and owners to improve efficiency and sustainability while reducing exhaust gas emissions.