Fault-tolerant control for high-speed trains based on neural network embedded compensation control
Kaydedildi:
| Yayımlandı: | Automatika vol. 66, no. 4 (Dec 2025), p. 841-853 |
|---|---|
| Yazar: | |
| Diğer Yazarlar: | , , , |
| Baskı/Yayın Bilgisi: |
Taylor & Francis Ltd.
|
| Konular: | |
| Online Erişim: | Citation/Abstract Full Text - PDF |
| Etiketler: |
Etiket eklenmemiş, İlk siz ekleyin!
|
MARC
| LEADER | 00000nab a2200000uu 4500 | ||
|---|---|---|---|
| 001 | 3276433972 | ||
| 003 | UK-CbPIL | ||
| 022 | |a 0005-1144 | ||
| 022 | |a 1848-3380 | ||
| 024 | 7 | |a 10.1080/00051144.2025.2569119 |2 doi | |
| 035 | |a 3276433972 | ||
| 045 | 2 | |b d20251201 |b d20251231 | |
| 100 | 1 | |a Hao, Zixu |u Transportation Collage, Jilin University , Changcun , People's Republic of China | |
| 245 | 1 | |a Fault-tolerant control for high-speed trains based on neural network embedded compensation control | |
| 260 | |b Taylor & Francis Ltd. |c Dec 2025 | ||
| 513 | |a Journal Article | ||
| 520 | 3 | |a To address the position and velocity tracking control problems of high-speed trains (HSTs), a neural network embedded fault-tolerant control (FTC) method is proposed in this paper. The unknown resistances and interactive forces between the connected carriages are taken into account. The stability of the neural networks (NNs) embedded FTC is proved by a common formal derivative of Lyapunov function, in which an NN-embedded item is integrated with a base controller which is stable for the system. On account of the system uncertainties and actuator faults, a value adaptive sliding mode control for estimating equivalent term composed of the unknown nonlinear terms and the disturbance is used and the base FTC is designed based on this method. The results of simulations show that the method of NN embedded optimization technology proposed in this paper can compensate and optimize the performance of the base FTC with only a few conditions. In the absence of actuator faults, NN-embedded FTC proposed in this paper reduces position error by about <inline-formula> <inline-graphic xlink3ahref="taut_a_2569119_ilm0001.gif"></inline-graphic> <tex-math notation="TeX"> $ 5\% $ </tex-math> 5 % </inline-formula> and velocity error by <inline-formula> <inline-graphic xlink3ahref="taut_a_2569119_ilm0002.gif"></inline-graphic> <tex-math notation="TeX"> $ 94\% $ </tex-math> 94 % </inline-formula>. In case of actuator faults, it reduces position error by about <inline-formula> <inline-graphic xlink3ahref="taut_a_2569119_ilm0003.gif"></inline-graphic> <tex-math notation="TeX"> $ 3\% $ </tex-math> 3 % </inline-formula> and velocity error by <inline-formula> <inline-graphic xlink3ahref="taut_a_2569119_ilm0004.gif"></inline-graphic> <tex-math notation="TeX"> $ 71\% $ </tex-math> 71 % </inline-formula>. | |
| 653 | |a Velocity errors | ||
| 653 | |a High speed rail | ||
| 653 | |a Velocity | ||
| 653 | |a Neural networks | ||
| 653 | |a Tracking control | ||
| 653 | |a Fault tolerance | ||
| 653 | |a Optimization | ||
| 653 | |a Sliding mode control | ||
| 653 | |a Carriages | ||
| 653 | |a Faults | ||
| 653 | |a Liapunov functions | ||
| 653 | |a Actuators | ||
| 653 | |a Position errors | ||
| 700 | 1 | |a Liu, Yumei |u Transportation Collage, Jilin University , Changcun , People's Republic of China | |
| 700 | 1 | |a Hu, Ting |u Transportation Collage, Jilin University , Changcun , People's Republic of China | |
| 700 | 1 | |a Liu, Pengcheng |u Transportation Collage, Jilin University , Changcun , People's Republic of China | |
| 700 | 1 | |a Liu, Ming |u Transportation Collage, Jilin University , Changcun , People's Republic of China | |
| 773 | 0 | |t Automatika |g vol. 66, no. 4 (Dec 2025), p. 841-853 | |
| 786 | 0 | |d ProQuest |t Research Library | |
| 856 | 4 | 1 | |3 Citation/Abstract |u https://www.proquest.com/docview/3276433972/abstract/embedded/L8HZQI7Z43R0LA5T?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text - PDF |u https://www.proquest.com/docview/3276433972/fulltextPDF/embedded/L8HZQI7Z43R0LA5T?source=fedsrch |