State Evaluation of Large Ships Diesel Engine Based on SOM Neural Network

There is still no fixed evaluation method for the state evaluation of large-scale ships diesel engines. Here, the SOM self-organizing map neural network model is applied to judge and evaluate the state of the marine diesel engine. Firstly, according to the basic functions and structure of the diesel engine, the selection and establishment of scientific indicators should be carried out on the principle of evaluating indicators. Then, after the selected state indicator parameters are normalized, the data is input to the SOM neural network for training, and finally use the trained network to evaluate the state of the input parameters. The trained SOM neural network can effectively improve the evaluation efficiency and reduce the influence of some subjective factors, and the experimental results show that the model is feasible in the state evaluation of Marine diesel engines.