A high precision in-bore velocity measurement system of railgun based on improved Bi-LSTM network. (February 2021)
- Record Type:
- Journal Article
- Title:
- A high precision in-bore velocity measurement system of railgun based on improved Bi-LSTM network. (February 2021)
- Main Title:
- A high precision in-bore velocity measurement system of railgun based on improved Bi-LSTM network
- Authors:
- Li, Songcheng
Lu, Junyong
Cheng, Long
Zeng, Delin - Abstract:
- Highlights: The B-dot magnetic probe position correction method is presented. An improved Bi-LSTM network for damaged B-dot probe data repair is presented. The proposed methods can improve the in-bore velocity measurement accuracy. The superiority of the proposed method is verified by simulations and experiments. Abstract: The B-dot magnetic probe array is one of the most common method to measure the in-bore velocity of electromagnetic railgun projectile, but in the exiting research, the errors caused by armature shape and the uneven distribution of current are often ignored. In order to achieve more accurate velocity measurement, this paper calculates probes induced voltage based on the Biot–Savart theorem for the projectile with C-shaped armature. Based on the analytic expression of evaluation, the positions of magnetic probes are corrected by using the cuckoo searching (CS) algorithm. In order to solve the problem that the accuracy of velocity measurement system declines under the condition of some abnormal probe signals, this paper proposes an improved Bidirectional Long Short-Term Memory (Bi-LSTM) network with Correction Head Network (CHN), and trains the whole network based on the experimental data which is under the same working condition. The algorithm proposed in this paper is verified by simulation and experiment. The results show that the position correction method of B-dot magnetic probes and the data repair method of damaged probe can effectively improve theHighlights: The B-dot magnetic probe position correction method is presented. An improved Bi-LSTM network for damaged B-dot probe data repair is presented. The proposed methods can improve the in-bore velocity measurement accuracy. The superiority of the proposed method is verified by simulations and experiments. Abstract: The B-dot magnetic probe array is one of the most common method to measure the in-bore velocity of electromagnetic railgun projectile, but in the exiting research, the errors caused by armature shape and the uneven distribution of current are often ignored. In order to achieve more accurate velocity measurement, this paper calculates probes induced voltage based on the Biot–Savart theorem for the projectile with C-shaped armature. Based on the analytic expression of evaluation, the positions of magnetic probes are corrected by using the cuckoo searching (CS) algorithm. In order to solve the problem that the accuracy of velocity measurement system declines under the condition of some abnormal probe signals, this paper proposes an improved Bidirectional Long Short-Term Memory (Bi-LSTM) network with Correction Head Network (CHN), and trains the whole network based on the experimental data which is under the same working condition. The algorithm proposed in this paper is verified by simulation and experiment. The results show that the position correction method of B-dot magnetic probes and the data repair method of damaged probe can effectively improve the accuracy and robustness of the in-bore velocity measurement system. … (more)
- Is Part Of:
- Measurement. Volume 169(2021)
- Journal:
- Measurement
- Issue:
- Volume 169(2021)
- Issue Display:
- Volume 169, Issue 2021 (2021)
- Year:
- 2021
- Volume:
- 169
- Issue:
- 2021
- Issue Sort Value:
- 2021-0169-2021-0000
- Page Start:
- Page End:
- Publication Date:
- 2021-02
- Subjects:
- Railgun -- B-dot probe -- In-bore velocity measurement -- Position correction -- Long short-term memory network
Weights and measures -- Periodicals
Measurement -- Periodicals
Measurement
Weights and measures
Periodicals
530.8 - Journal URLs:
- http://www.sciencedirect.com/science/journal/02632241 ↗
http://www.elsevier.com/journals ↗ - DOI:
- 10.1016/j.measurement.2020.108501 ↗
- Languages:
- English
- ISSNs:
- 0263-2241
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 5413.544700
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 15168.xml