A Q-learning based method of optimal fault diagnostic policy with imperfect tests.
Yajun LiangMingqing XiaoXi-Lang TangYawei GeXiaofei WangPublished in: J. Intell. Fuzzy Syst. (2019)
Keyphrases
- dynamic programming
- artificial intelligence
- cost function
- high accuracy
- objective function
- preprocessing
- machine learning
- pairwise
- significant improvement
- potential field
- real time
- model free
- function approximation
- test data
- segmentation method
- detection method
- cooperative
- optimal solution
- multi agent
- reinforcement learning
- neural network