Experimental investigation of cylinder liners and performance optimization using ANN

This research investigates the structural integrity and performance optimization of cylinder liners in internal combustion engines, focusing on material selection and thermomechanical stress response. The study aims to enhance efficiency and durability by evaluating the performance of cast iron, nickel-chromium iron alloy, and aluminum alloy under real-world engine conditions. Using Finite Element Analysis, material properties are experimentally assessed to determine thermomechanical stresses, wear characteristics, and heat dissipation behavior. Additionally, Artificial Neural Networks are employed to optimize performance parameters by predicting material behavior under varying thermal and load conditions. Results indicates that•Nickel-Chromium iron alloy exhibits the lowest combined stress, making it the most suitable material due to superior resistance to thermal stress.•Unlike cast iron, where thermal-induced stress is too much to bear, aluminum alloy is lightweight but suffers from deformation when it undergoes thermal expansion.•However, the ANN-based optimization was successful in lowering stress values and allows us to consider future applications as a prediction modelling tool in material selection. The ANN optimization will reduce thermal stress by up to 20 % and is thus a successful method to use with predictive maintenance and improved performance of cylinder liners. This study shows that using both computational and experimental approaches can allow for optimized performance for engine components and provide future direction for cylinder liner designs that are both stronger and have better performance.