A method and system for predicting performance degradation and evaluating life of key components of a ship based on a proxy model

Abstract

This invention discloses a method and system for predicting performance degradation and assessing the lifespan of critical ship components based on a surrogate model. It addresses the technical problems of existing methods failing to meet the real-time prediction requirements for performance degradation of critical ship components in engineering projects, and the lack of accuracy in lifespan assessment. The method includes acquiring multiple sets of operating condition parameters for the critical ship component and the corresponding performance degradation data for each set of parameters; constructing a spatiotemporal dataset based on this data; using the Adam optimizer and mean squared error loss function to construct an optimal deep neural network surrogate model based on the spatiotemporal dataset; acquiring the operating condition parameters to be predicted for the component; performing performance degradation prediction through the optimal deep neural network surrogate model and outputting the prediction results; and finally, using cubic spline interpolation to conduct a lifespan assessment based on the performance degradation prediction results, outputting the final remaining service life of the critical ship component.

Images