Phosphorene is a promising candidate for gas sensing materials. This letter describes our study of the adsorption of SF 6 and SF 6 decomposition gases (SO 2 and H 2 S) on phosphorene. We used first principles calculations to explore phosphorene's potential applications as gas sensor to diagnose the state of online gas insulated switchgear (GIS). The calculation results showed that only the adsorption of SO 2 induced a moderate adsorption energy and apparent charge transfer. We further investigated the current-voltage (I-V) relationships before and after gas absorption through the non-equilibrium Green's function method. It was found that only SO 2 induced a dramatic change in the I-V relationships. Therefore, phosphorene appears to be a promising candidate for highly sensitive and selective SF 6 decomposition gas sensors for online GIS diagnosis.
19 November 2021 in IEEE Transactions on Smart Grid
High-impedance faults (HIFs) pose the greatest challenge for distribution system protection, especially for microgrids and distribution networks with distributed generators (DGs) that have flexible operation modes. This paper analyzes the faulty features of HIFs and proposes a HIF detection method that uses empirical wavelet transform (EWT) and differential faulty energy. The proposed method is as follows. First, the various time-frequency components are obtained by utilizing the EWT to decompose the differential faulty energy and adaptively select the feature component with the largest permutation entropy. Second, the permutation variance index is constructed based on the sample point number and feature component energy, and then it is employed to detect HIFs. Finally, low voltage microgrid simulation tests, medium voltage distribution system integrated by DG simulation tests, and field tests show that the proposed method can correctly distinguish HIFs from normal disturbances, including operation mode switches, load switches, capacitor switches, and DG switches. The advantages of the proposed method are also elaborated in detail, from signal preprocessing and feature extraction.
21 July 2020 in IEEE Transactions on Industrial Electronics
Recent years have witnessed the prominent advancements of deep learning (DL) in the arsenal of prognostics and health management. However, the prognostic uncertainty problem extensively existed in industrial devices is not addressed by most DL approaches. This article formulates a novel Bayesian Deep Learning (BDL) framework to characterize the prognostic uncertainties. A distinguished advantage of the framework is its capacity of capturing the comprehensive effects of two critical uncertainties: 1) epistemic uncertainty, accounting for the uncertainty in the model, and 2) aleatoric uncertainty, representing the impact of random disturbance, such as measurement errors. The former arises from the variability of the model weights, and the latter is characterized by selected lifetime distributions. We integrate both uncertainties by defining BDL as priors of lifetime parameters. A sequential Bayesian boosting algorithm is executed to improve the estimation accuracy and compress the credible intervals. The superior prediction performance of our framework is validated by a real-world dataset collected from hydraulic mechanisms of circuit breakers.