Remaining service life prediction method meeting cross-working-condition scene application

The invention discloses a remaining service life prediction method meeting cross-working-condition scene application. The remaining service life prediction method comprises the following steps: 1) giving degradation data of an active domain and a target domain; 2) carrying out data normalization on the collected data; 3), extracting degradation features of the source domain and the target domain, and separating the degradation features into private features and common features; 4) aligning common features of the source domain and the target domain from two angles of global distribution and local distribution; 5) maximizing mutual information of the input data and the extracted features by adopting an InfoNCE loss function in contrast learning, and ensuring that the extracted features have representativeness; and 6) regression prediction: performing RUL prediction on the test data by using the model trained in the steps, thereby greatly improving the prediction precision. 本发明公开了一种满足跨工况场景应用的剩余使用寿命预测方法，包括以下步骤：1)