A Physics-Informed Training Approach for Data-Driven Method in Remaining Useful Life Estimation.
Yuxuan HeHuai SuEnrico ZioLin FanJinjun ZhangPublished in: ICSRS (2022)
Keyphrases
- data driven
- synthetic data
- neural network
- high accuracy
- experimental evaluation
- high precision
- significant improvement
- pairwise
- objective function
- training process
- training algorithm
- parametric models
- similarity measure
- support vector machine svm
- parameter estimation
- training procedure
- segmentation method
- training examples
- maximum likelihood estimation
- optimization algorithm
- maximum likelihood
- motion estimation
- computational cost
- dynamic programming
- hidden markov models
- prior knowledge
- preprocessing
- computer science
- feature selection
- artificial intelligence