English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Prognostics or Remaining Useful Life (RUL) Estimation from multi-sensor timeseries data is useful to enable condition-based maintenance and ensure highoperational availability of equipment. We propose a novel deep learning basedapproach for Prognostics with Uncertainty Quantification that is useful inscenarios where: (i) access to labeled failure data is scarce due to rarity offailures (ii) future operational conditions are unobserved and (iii) inherentnoise is present in the sensor readings. All three scenarios mentioned areunavoidable sources of uncertainty in the RUL estimation process oftenresulting in unreliable RUL estimates. To address (i), we formulate RULestimation as an Ordinal Regression (OR) problem, and propose LSTM-OR: deepLong Short Term Memory (LSTM) network based approach to learn the OR function.We show that LSTM-OR naturally allows for incorporation of censored operationalinstances in training along with the failed instances, leading to more robustlearning. To address (ii), we propose a simple yet effective approach toquantify predictive uncertainty in the RUL estimation models by training anensemble of LSTM-OR models. Through empirical evaluation on C-MAPSS turbofanengine benchmark datasets, we demonstrate that LSTM-OR is significantly betterthan the commonly used deep metric regression based approaches for RULestimation, especially when failed training instances are scarce. Further, ouruncertainty quantification approach yields high quality predictive uncertaintyestimates while also leading to improved RUL estimates compared to single bestLSTM-OR models.

Data-driven Prognostics with Predictive Uncertainty Estimation using Ensemble of Deep Ordinal Regression Models