Wear assessment model for cylinder liner of internal combustion engine under fuzzy uncertainty

The wear of the piston ring-cylinder system is inevitable in operation internal combustion engines (ICEs). If exceeds maximum, will be failure. A novel assessment model proposed based on support vector regression, and fuzzy uncertainty modeled to describe random behavior under small sample. To verify model, sample data cylinder liner applied. For best results, particle swarm optimization (PSO) algorithm used optimize parameters. back propagation neural network (BPNN) employed effectiveness model. results show that regression has better prediction accuracy than other methods for this paper, can evaluate ICEs effectively. work provides a technical evaluating service performance reference regular maintenance ships.