A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process
This project assesses the possibility of a new hybrid method based on Kalman Filter and Adaptive Neural Network for the robustness improvement of Fault Detection and Identification (FDI) Process. Two classes of approaches are introduced, 1) the system identification approach and 2) the observer-base...
| Main Author: | |
|---|---|
| Format: | Research Book Profile |
| Language: | English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/36259/1/A%20new%20hybrid%20method%20based%20on%20kalman%20filter%20and%20adaptive%20neural%20network%20for%20the%20robustness%20improvement%20of%20fault%20detection%20and%20identification%20process.pdf |
| _version_ | 1796995457779499008 |
|---|---|
| author | Pebrianti, Dwi |
| author_facet | Pebrianti, Dwi |
| author_sort | Pebrianti, Dwi |
| collection | UMP |
| description | This project assesses the possibility of a new hybrid method based on Kalman Filter and Adaptive Neural Network for the robustness improvement of Fault Detection and Identification (FDI) Process. Two classes of approaches are introduced, 1) the system identification approach and 2) the observer-based approach using the Kalman filter. The Kalman filter recognizes data from the sensors of the system and indicates the fault of the system in the sensor reading. Error prediction is based on the fault magnitude and the time occurrence of fault. A representative Artificial Neural Network (ANN) model is designed and used to classify the fault class. The proposed technique is implemented on a quad-rotor Micro-Aerial Vehicle (MAV) as a complex system that has Multi Input Multi Output (MIMO). In particular, two FDI scenarios are considered: 1) the estimation of an unknown actuator fault and 2) an unknown sensor fault. The result comparison of the residual signal before filter and after filter showed that Kalman-ANN is able to identify and immediately acknowledge the system to operate in the normal state. By comparing the system performance of the FDI technique, Kalman-ANN is more effective in identifying parts of the system that experiences failure. Kalman- ANN is also able to acknowledge user on the parts of quadrotor that experience failure and provide user with the best instructions or solutions for the situation, ensuring a safe landing. |
| first_indexed | 2024-03-06T13:02:38Z |
| format | Research Book Profile |
| id | UMPir36259 |
| institution | Universiti Malaysia Pahang |
| language | English |
| last_indexed | 2024-03-06T13:02:38Z |
| publishDate | 2014 |
| record_format | dspace |
| spelling | UMPir362592023-01-04T03:23:27Z http://umpir.ump.edu.my/id/eprint/36259/ A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process Pebrianti, Dwi TK Electrical engineering. Electronics Nuclear engineering This project assesses the possibility of a new hybrid method based on Kalman Filter and Adaptive Neural Network for the robustness improvement of Fault Detection and Identification (FDI) Process. Two classes of approaches are introduced, 1) the system identification approach and 2) the observer-based approach using the Kalman filter. The Kalman filter recognizes data from the sensors of the system and indicates the fault of the system in the sensor reading. Error prediction is based on the fault magnitude and the time occurrence of fault. A representative Artificial Neural Network (ANN) model is designed and used to classify the fault class. The proposed technique is implemented on a quad-rotor Micro-Aerial Vehicle (MAV) as a complex system that has Multi Input Multi Output (MIMO). In particular, two FDI scenarios are considered: 1) the estimation of an unknown actuator fault and 2) an unknown sensor fault. The result comparison of the residual signal before filter and after filter showed that Kalman-ANN is able to identify and immediately acknowledge the system to operate in the normal state. By comparing the system performance of the FDI technique, Kalman-ANN is more effective in identifying parts of the system that experiences failure. Kalman- ANN is also able to acknowledge user on the parts of quadrotor that experience failure and provide user with the best instructions or solutions for the situation, ensuring a safe landing. 2014 Research Book Profile NonPeerReviewed pdf en http://umpir.ump.edu.my/id/eprint/36259/1/A%20new%20hybrid%20method%20based%20on%20kalman%20filter%20and%20adaptive%20neural%20network%20for%20the%20robustness%20improvement%20of%20fault%20detection%20and%20identification%20process.pdf Pebrianti, Dwi (2014) A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process. , [Research Book Profile: Research Report] (Unpublished) |
| spellingShingle | TK Electrical engineering. Electronics Nuclear engineering Pebrianti, Dwi A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process |
| title | A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process |
| title_full | A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process |
| title_fullStr | A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process |
| title_full_unstemmed | A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process |
| title_short | A new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process |
| title_sort | new hybrid method based on kalman filter and adaptive neural network for the robustness improvement of fault detection and identification process |
| topic | TK Electrical engineering. Electronics Nuclear engineering |
| url | http://umpir.ump.edu.my/id/eprint/36259/1/A%20new%20hybrid%20method%20based%20on%20kalman%20filter%20and%20adaptive%20neural%20network%20for%20the%20robustness%20improvement%20of%20fault%20detection%20and%20identification%20process.pdf |
| work_keys_str_mv | AT pebriantidwi anewhybridmethodbasedonkalmanfilterandadaptiveneuralnetworkfortherobustnessimprovementoffaultdetectionandidentificationprocess AT pebriantidwi newhybridmethodbasedonkalmanfilterandadaptiveneuralnetworkfortherobustnessimprovementoffaultdetectionandidentificationprocess |