摘要：Abstract. Process sensor data allow for not only the control of industrial processes but also an assessmentof plant conditions to detect fault conditions and wear by using sensor fusion and machine learning (ML). Afundamental problem is the data quality, which is limited, inter alia, by time synchronization problems. Toexamine the influence of time synchronization within a distributed sensor system on the prediction performance,a test bed for end-of-line tests, lifetime prediction, and condition monitoring of electromechanical cylindersis considered. The test bed drives the cylinder in a periodic cycle at maximum load, a 1 s period at constantdrive speed is used to predict the remaining useful lifetime (RUL). The various sensors for vibration, force,etc. integrated into the test bed are sampled at rates between 10 kHz and 1 MHz. The sensor data are usedto train a classification ML model to predict the RUL with a resolution of 1 % based on feature extraction,feature selection, and linear discriminant analysis (LDA) projection. In this contribution, artificial time shiftsof up to 50 ms between individual sensors’ cycles are introduced, and their influence on the performance ofthe RUL prediction is investigated. While the ML model achieves good results if no time shifts are introduced,we observed that applying the model trained with unmodified data only to data sets with time shifts resultsin very poor performance of the RUL prediction even for small time shifts of 0.1 ms. To achieve an acceptableperformance also for time-shifted data and thus achieve a more robust model for application, different approacheswere investigated. One approach is based on a modified feature extraction approach excluding the phase valuesafter Fourier transformation; a second is based on extending the training data set by including artificially timeshifted data. This latter approach is thus similar to data augmentation used to improve training of neural networks.
