Remaining Useful Life Prognostics for the Electrohydraulic Servo Actuator Using Hellinger Distance-Based Particle Filter
| Autor: | Runxia Guo, Jianfei Sui |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Series (mathematics)
Computer science 020208 electrical & electronic engineering Process (computing) Scheduling (production processes) 02 engineering and technology Nonlinear system Control theory 0202 electrical engineering electronic engineering information engineering Prognostics Electrical and Electronic Engineering Hellinger distance Particle filter Instrumentation |
| Zdroj: | IEEE Transactions on Instrumentation and Measurement. 69:1148-1158 |
| ISSN: | 1557-9662 0018-9456 |
| DOI: | 10.1109/tim.2019.2910919 |
| Popis: | The electrohydraulic servo actuator (EHSA) has been widely used in the flight control actuation system of civil aircraft. Accurate remaining useful life (RUL) prediction for EHSA is of great significance in inhibiting the unexpected failures of aircrafts, improving the scheduling ability, and reducing the maintenance costs, which has become highlight for airline operators in recent years. The particle filtering (PF) is a commonly accepted technique in RUL prediction due to its superiority in coping with nonlinear and non-Gaussian behaviors. However, the irrational selection of importance density can induce the problem of particle degeneracy and restrict the estimation performance, resulting in the decrease of RUL prediction accuracy. In this paper, an importance density choice scheme grounded on the Minimum Hellinger Distance principle, namely, MHD-PF, is proposed to fight against the degeneracy problem at the early phase. In addition, the strict mathematical derivation process of the proposed scheme is proffered. Then, the MHD-PF is used to track the degradation state and predict the RUL of EHSA. Finally, the specific performance of the MHD-PF method is performed by a series of actual experiments. The effectiveness of the proposed RUL prognostics approach is also experimentally verified. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|