A Digital-Twin Framework for Predicting the Remaining Useful Life of Piezoelectric Vibration Sensors with Sensitivity Degradation Modeling

Abstract

Piezoelectric vibration sensors (PVSs) are widely applied to vibration detection in aerospace engines due to their small size, high sensitivity, and high-temperature resistance. The precise prediction of their remaining useful life (RUL) under high temperatures is crucial for their maintenance. Notably, digital twins (DTs) provide enormous data from both physical structures and virtual models, which have potential in RUL predictions. Therefore, this work establishes a DT framework containing six modules for sensitivity degradation detection and assessment on the foundation of a five-dimensional DT model. In line with the sensitivity degradation mechanism at high temperatures, a DT-based RUL prediction was performed. Specifically, the PVS sensitivity degradation was described by the Wiener–Arrhenius accelerated degradation model based on the acceleration factor constant principle. Next, an error correction method for the degradation model was proposed using real-time data. Moreover, parameter updates were conducted using a Bayesian method, based on which the RUL was predicted using the first hitting time. Extensive experiments on distinguishing PVS samples demonstrate that our model achieves satisfying performance, which significantly reduces the prediction error to 8 h. A case study was also conducted to provide high RUL prediction accuracy, which further validates the effectiveness of our model in practical use.

RUL Prediction for Piezoelectric Vibration Sensors Based on Digital-Twin and LSTM Network

“…Each module of the framework is described in detail below. Aiming to predict the RUL of PVS by the LSTM network, an RUL prediction DT framework based on DT data and LSTM is established based on previous research [17]. As shown in Figure 8, the main improvement is in the content of the virtual model management and application module, which adjusts the content of each module from data acquisition to processing and application according to the needs of LSTM.…”

“…Predicting the RUL of a product requires a quantitative characterization of the degradation process. The degradation process of PVSs mainly manifests in the degradation of sensitivity [17], so sensitivity is a necessary feature. According to the analysis in Section 2.1, the sensitivity varies according to the vibration frequency and acceleration, which should be considered.…”

“…Modeling using historical degradation data and then model correction using realtime degradation data is a popular strategy in RUL prediction. In a previous study [17], sensitivity degradation is modeled, and the time at which the degradation first reached the failure threshold was predicted as a method of RUL prediction. This approach assumes that the time at which sensitivity degradation reaches a threshold is the time to failure.…”

“…Our previous study used a statistical model-based data-driven approach and Digital-Twin (DT) to predict the RUL of PVS [17]. But there are still some problems.…”