Journal of the Korean Society of Marine Environment & Safety (해양환경안전학회지)

The Korean Society of Marine Environment and safety (해양환경안전학회)
권호 표지
DOI QR코드

DOI QR Code

Study on the Performance Improvement of Marine Engine Generator Exciter Control using Neural Network Controller

신경망 회로 제어기를 이용한 선박 엔진 발전기의 여자기 제어 성능 개선에 관한 연구

  • HeeMoon Kim (Eco-Friendly Propulsion Team, Korea Marine Equipment Research Institute.) ;
  • JongSu Kim (Division of Marine System Engineering, Korea Maritime & Ocean University) ;
  • SeongWan Kim (Division of Marine AI & Cyber Security, Korea Maritime & Ocean University) ;
  • HyeonMin Jeon (Division of Marine System Engineering, Korea Maritime & Ocean University)
  • 김희문 ((재)한국조선해양기자재연구원 친환경추진기술팀) ;
  • 김종수 (한국해양대학교 기관시스템공학부) ;
  • 김성완 (한국해양대학교 해사인공지능.보안학부) ;
  • 전현민 (한국해양대학교 기관시스템공학부)
  • Received : 2023.08.21
  • Accepted : 2023.10.27
  • Published : 2023.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

The exciter of a ship generator adjusts the magnetic flux through excitation current control to maintain the output terminal voltage constant. The voltage controller inside the exciter typically uses a proportional integral control method. however, the response characteristics determined by the gain and time constant produce unwanted output owing to an inappropriate setting value that can reduce the quality and stability of power within the ship. In this study, a neural network circuit is learned using stable input/output data that can be obtained through the AC4A type exciter model provided by IEEE, and the simulation is performed by replacing the existing proportional integral control type voltage controller with the learned neural network circuit controller. Consequently, overshooting was improved by up to 9.63% compared with that of the previous model, and excellence in stable response characteristics was confirmed.

선박 발전기의 여자기는 출력 단자 전압을 일정하게 유지하기 위하여 여자전류 제어를 통해 자속을 조정한다. 여자기 내부에 있는 전압제어기는 통상적으로 비례 적분 제어방식이 사용되는데 게인과 시정수에 의해 결정되는 응답 특성은 적절치 못한 설정값에 의해 원하지 않는 출력을 내며 이로 인해 선내 전력의 품질과 안정성을 떨어뜨릴 수 있다. 본 논문에서는 IEEE에서 제공하는 AC4A 타입의 여자기 모델을 통해 얻을 수 있는 안정적인 입출력 데이터를 활용하여 신경망 회로를 학습시킨 후 기존의 비례 적분 제어방식의 전압제어기를 학습된 신경망 회로 제어기로 대체하여 시뮬레이션을 수행하였다. 그 결과 기존 대비 최대 9.63%까지 오버슈팅이 개선되었으며, 안정적인 응답 특성에 대한 우수성을 확인하였다.

Keywords

Acknowledgement

이 연구는 2023년 해양수산부 재원으로 해양수산과학기술진흥원(KIMST)의 지원을 받아 수행된 연구임(No. 20210369, 전기복합 추진어선 핵심 기자재 기술개발).

References

  1. Al_Issa, H. A., M. Drechny, I. Trrad, M. Qawaqzeh, V. Kuchanskyy, O. Rubanenko, S. Kudria, P. Vasko, O. Miroshnyk, and T. Shchur(2022), Assessment of the Effect of Corona Discharge on Synchronous Generator Self-Excitation, Energies, Vol. 15, No. 6.
  2. Bento, M. E. C.(2023), Physics-Guided Neural Network for Load Margin Assessment of Power Systems, IEEE Transactions on Power Systems.
  3. Da Silva Filho, J. I., R. A. Bispo de Oliveira, M. C. Rodrigues, H. M. Cortes, A. Rocco, M. C. Mario, D. V. Garcia, J. M. Abe, C. R. Torres, V. B. D. Ricciotti, A. C. D. Ricciotti, A. de Carvalho, Jr. and G. Lambert-Torres(2023), Predictive Controller Based on Paraconsistent Annotated Logic for Synchronous Generator Excitation Control, Energies, Vol. 16, No. 4.
  4. Hu, Z., Y. Ren, Q. Meng, P. Yun, C. Fang, and Y. Pan(2023), Improvement of Frequency Support for a DFIG Using a Virtual Synchronous Generator Strategy at Large Power Angles, Energies, Vol. 16, No. 2.
  5. IEEE(2016), IEEE Recommended Practice for Excitation System Models for Power System Models for Power System Stability Studies, IEEE Power and Energy Society. IEEE Std 421.5-2016 (Revision of IEEE Std 421.5 [2003]): 1-207.
  6. Kumar, R. S., P. Saravanan, and R. Tiwari(2023), Impact of Artificial Intelligence Techniques in Distributed Smart Grid Monitoring System, Smart Energy and Electric Power Systems. pp. 79-103.
  7. Olcay, K. and N. Cetinkaya(2023), Analysis of the Electric Vehicle Charging Stations Effects on the Electricity Network with Artificial Neural Network, Energies, Vol. 16, No. 3.
  8. Vanting, N. B., Z. Ma, and N. Jorgensen(2021), A scoping review of deep neural networks for electric load forecasting, Energy Informatics. Academy Conference Asia 2021, Vol. 4, No. 49.
  9. Wu, J., K. Zhao, Y. Jiang, L. Cheng, Q. Liu, Y. Xue, and K. Peng(2018), Maximum Power Point Tracking Algorithm for Laser Power Beaming Based on Neural Networks, 2018 IEEE 3rd International Conference on Cloud Computing and Internet of Things (CCIOT).
  10. Wu, T., Z. Wang, B. Ozpineci, M. Chinthavali, and S. Campbell(2019), Automated Heatsink Optimization for Air-Cooled Power Semiconductor Modules, IEEE Transactions on Power Electronics, Vol. 34, No. 6, pp. 5027-5031. https://doi.org/10.1109/TPEL.2018.2881454
  11. Yu, X.(2020), The Correlation of Network Topology and Power System Resilience by Using Neural Network Analysis, 2020 IEEE 11th International Symposium on Power Electronics for Distributed Generation Systems (PEDG).
  12. Zhan, X., W. Wang, and H. Chung(2018), A Neural-Network-Based Color Control Method for Multi-Color LED Systems, IEEE Transactions on Power Electronics, Vol. 34, No. 8, pp. 7900-7913. https://doi.org/10.1109/TPEL.2018.2880876
  13. Zhao, S., F. Blaabjerg, and H. Wang(2021), Interactions of the Wakes of Two Spheres Placed Side by Side, IEEE Transactions on Power Electronics, Vol. 36, No. 4, pp. 4633-4658.  https://doi.org/10.1109/TPEL.2020.3024914