基于NSGA-Ⅲ的船用柴油机模型自适应修正方法研究

The performance degradation caused by faults and aging of components in the long-term use of marine diesel engines can cause mismatch between the diesel engine model and the actual diesel engine performance, which seriously affects the performance optimization control of diesel engines. To solve this problem, a diesel engine model adaptive correction method is proposed. Firstly, a second-order response surface model is established based on the data obtained from the diesel engine prediction model built in GT-POWER software, Then, the third generation non dominated sorting genetic algorithm (NSGA Ⅲ) is used to modify the input parameters of the second-order response surface model, so that the output performance of the second-order response surface model can match the output performance of the diesel engine prediction model after state changes. The validation results indicate that the NSGA - III algorithm can not only accurately identify and correct the input parameter changes of the second-order response surface model, but also adaptively correct the optimal solution with performance parameter offset errors within 0.9%. The adaptive correction of the second-order response surface model has been achieved, demonstrating the feasibility of the proposed correction method.