MOSFETs are used in several industrial applications, such as power electronics to supply switching loads, motor power supply, switching power supply, audio amplifier, inverter etc. The estimation of the RUL (Remaining Useful Life) of these devices is very interesting for the industry. Indeed, it allows gaining availability as well as an improvement of the safety of the system. The modern industry is focused on PHM (Prognostics and Health Management) to estimate the RUL and to design modern expert systems that ensure a good reliability to the industry.In this work, authors have developed a new methodology based on the deep learning ANN (Artificial Neural Network) in order to predict the RUL of MOSFETs. To validate the proposed Long Short-Term Memory (LSTM) algorithm, the data of the Ames PCOE (Prognostics Center of Excellence) was used. In addition, the evaluation was performed by comparing the errors between the real MOSFET RUL and the predicted MOSFET RUL. Besides, a comparison with different algorithms of prediction existing in the literature and using the same PCOE data set was done to present a solid evaluation of the proposed LSTM Recurrent Neural Network (LSTM-RNN).

This paper presents a Machine Learning (ML) application that involves Long Short-Term Memory (LSTM) Artificial Neural Network (ANN). The proposed LSTM-ANN. architecture addresses the prediction challenge of sleep/wake phases in order to minimize night wake stages in elderly patients. Accurate predictions guarantee a good sleep quality for elderly patients and support medication adherence. For this, Vivago® Care watch technology (IST Vivago® Oy) is used in this work. Each watch was advantaged by actimetric functionality. Watches were worn on the wrist of the considered elderly patients. Automatically, one recording was sent every minute to the base station located in the nursery room, where the recorded data are stored and backed up. Experimental results are encouraging and also they are a rationale and evidence for the allocation of investments for developing online monitoring systems for sleep/wake phase’s prediction using only previous historical data. Our proposed LSTM-ANN proposal despite the novelty of this subject and the lack of literature, highlights prototypes results. is accurate for Personal Health Management (PHM) applications.

In 1988, Fere discovered the Electrodermal activity (EDA) and it was defined originally as the property of human skins. Nowadays, it is well known as the characteristics of the human body that causes an incessant variation of the electrical skin potential. In this work, the EDA signal is used to detect the Polyneuropathy (PNP) disease. The main two steps of the proposed strategy is to extract several features via EDA signals and to classify them in two classes (Healthy case and PNP case) by using Support Vector Machine (SVM) algorithm. For this purpose, four different domains of feature extraction are investigated (morphology, time, frequency and time-frequency). The Emrirical Mode Decomposition (EMD) algorithm is used to decompose original EDA to some sub-signals ranged from high to low frequency order. Consequently, the time-frequency domain is investigated, and the EDA analyse is performed considering diffirent frequency ranges. Then, the extracted features were classified using SVM and 83.79% of accuracy was achieved. Compared to previous works, experimental results show that the proposed method is truthful for PNP detection.

In the framework of rotating machines prognosis considering naturally progressing degradations, this work proposes a new adaptive strategy for the estimation of the Remaining Useful Life (RUL) of bearings. In fact, although they are the indispensable mechanical elements of rotating machines, bearings are the most stressed part and their damage causes unexpected stops. The most of advanced fault prognosis techniques proposed in the literature are based on vibration signals because they are rich in information. However, the noisy and nonlinear natures of the raw vibratory collected signals make the prognostic task hard and more challenging. In this sense, a new strategy for bearing state of health estimation is proposed in this work. The proposed strategy is based on the application of the Extended Kalman filter (EKF) with some advanced digital processing steps. The EKF is used to approximate the nonlinear variation of the selected feature. In fact, the EKF algorithm is based on two steps of online feature extraction and evaluation. These two steps ensure the online selection of the best feature considering some mathematical characteristics. This method has been validated on vibratory data from the full-scale test bench of the University of Cincinnati, USA. Experimental results show that the proposed approach illustrates good prediction capabilities even with a long horizon and it can be applied to the Prognostic and Health Management (PHM) of several other assets.

Common mechanical failures in wind turbine generators (WTGs) result in unplanned downtime, loose of production and increase the maintenance cost. Statistical studies have shown that failures due to high-speed shaft bearing (HSSB) account for 64% of all drivetrain failures. Consequently, prognostic and health management (PHM) of WTGs aims to estimate the future state of health and predict the reaming useful life (RUL) of HSSB. This paper considers a new data-driven approach based on vibration signals. This approach extracts statistical time-domain features that reflect the behavior of the system and its degradation. Then, the extracted features are evaluated to select the most trendable condition indicators that will be considered as inputs for an Elman neural network (ENN). Moreover, this paper proposes a new ENN architecture for direct RUL estimation of HSSB validated by use of real measured data from a WTG drivetrain. The proposed method reveals accurate estimation capability even with noisy measurements and harsh conditions.

Track degradation of wind turbine high-speed shaft bearing can reduce unscheduled maintenance events, and safe power generation system. This paper proposes a particle filter-based prognostic approach for high-speed shaft bearing track degradation; this approach is validated by inspecting a real data from a wind turbine drivetrain. The particle filter-based prognostic results are compared with the standard support vector regression and Kalman smoother results. The particle filter method shows better results. For longer prediction times, the error of the proposed method is equal to or smaller than that of the regression method. The main improvement of the particle filter-based prognostic approach is its ability to produce a probabilistic result based on input parameters with uncertainties. The distributions of the input parameters propagate through the filter, and the remaining useful life is presented using a particle distribution.