Novel direct remaining useful life estimation of aero-engines with randomly assigned hidden nodes

Aircraft engines Remaining Useful Life prediction with an adaptive denoising online sequential Extreme Learning Machine

“…For instance, Bai et al [11] introduced a new fast training approach for RUL prediction based on a "single-batch" ELM after an appropriate features selection and noise filtering for accurate estimation. Bektas et al [10] proposed multi-regime normalization approaches followed by filtering methods for best features representation and noise reduction before feeding the feed-foreword neural network for RUL prediction.…”

“…Bektas et al [10] proposed multi-regime normalization approaches followed by filtering methods for best features representation and noise reduction before feeding the feed-foreword neural network for RUL prediction. Lu et al [11] improved the recursive learning of the OS-ELM by reducing the noise produced from Recursive Least Squares (RLS) linear approximation with Kalman filter objective function and proposed a new weights adaptation method to accurately predict the RUL in aircraft engines.…”

“…DDFF is proposed based on [6]  Calculate the initial output weights βAE(k+1) and βos(k+1) of the DAE and DOS-ELM as given in equation ( 10) and (11), respectively;…”

“…The best solution for this problem or time-varying data problems, in general, is to integrate a dynamic tracking and selection strategy in the data-driven models to resist against any variation in the new coming chunks [6], [7], [8], [9]. However, such a solution is feasible under the constraint that the training data are not contaminated with noise with unknown source or behavior [10], [11].…”

Prognostics uncertainty reduction by right-time prediction of remaining useful life based on hidden Markov model and proportional hazard model

“…With the increasing requirement of safety and economy of equipment operation, more and more attention is paid to the prediction of remaining useful life(RUL). Scholars in various fields have conducted extensive research on this issue, including aerospace field [1,17] , electronics [28], mechanical industrial [16,22] , and many other fields [7,20]. Most researchers devoted to improve the accuracy of the prediction results, and have investigated many effective methods for RUL prediction, including various neural networks (NN) [41], support vector regression (SVR) [34], stochastic process [13,36], and other methods [3,21].…”