Electronics and Telecommunications Trends (전자통신동향분석)

Electronics and Telecommunications Research Institute (한국전자통신연구원)
권호 표지
DOI QR코드

DOI QR Code

Status and Development of Physics-Informed Neural Networks in Agriculture

Physics-Informed Neural Networks 연구 동향 및 농업 분야 발전 방향

  • S.Y. Lee ;
  • H.J. Shin ;
  • D.H. Park ;
  • W.K. Choi ;
  • S.K. Jo
  • 이상연 (농축해양수산지능연구센터) ;
  • 신학종 (농축해양수산지능연구센터) ;
  • 박대헌 (농축해양수산지능연구센터) ;
  • 최원규 (농축해양수산지능연구센터) ;
  • 조성균 (농축해양수산지능연구센터)
  • Published : 2024.08.01
Download PDF
⟨ Previous Next ⟩

Abstract

Mathematical modeling is the process of representing physical phenomena using equations, and it often describes various scientific phenomena through differential equations. Numerical analysis, which is capable of approximating solutions to partial differential equations representing physical phenomena, is widely utilized. However, in high-dimensional or nonlinear systems, computational costs can substantially increase, leading to potential numerical instability or convergence issues. Recently, Physics-Informed Neural Networks (PINNs) have emerged as an alternative approach. A PINN leverages physical laws even with limited data to provide highly reliable predictive performance and can address the convergence issues and high computational costs associated with numerical analysis. This paper analyzes the weak signals, research trends, patent trends, and case studies of PINNs. On the basis of this analysis, it proposes directions for the development of PINN techniques in the agricultural field. In particular, the application of PINNs in agriculture is expected to be more effective than in other industries because of their ability to reflect real-time changes in biological processes. While the technology readiness level of PINNs remains low, the potential for model training with minimal data and real-time prediction capabilities suggests that PINNs could replace traditional numerical analysis models. It is anticipated that the research and industrial applications of PINN will develop at an increasing pace while focusing on addressing the complexity of mathematical models in agriculture, mathematical modeling and the application of various biological processes; securing key patents related to PINNs; and standardizing PINN technology in the field of agriculture.

Keywords

References

  1. E.A. Bender, An Introduction to Mathematical Modeling. Courier Corporation, 2000.
  2. 정영준, "물리 기반 신경망을 이용한 탄성체의 거동 해석," 석사학위논문, 서울대학교, 2022.
  3. S.C. Chapra and R.P. Canale, Numerical Methods for Engineers, Mcgraw-Hill, vol. 1221, 2011.
  4. J.G. Hoffer et al., "Mesh-free surrogate models for structural mechanic FEM simulation: A comparative study of approaches," Appl. Sci., vol. 11, no. 20, 2021, article no. 9411.
  5. L. Sun et al., "Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data," Comput. Methods Appl. Mech. Eng., vol. 361, 2020, article no. 112732.
  6. R. Anantharaman et al., "Accelerating simulation of stiff nonlinear systems using continuous-time echo state networks," arXiv preprint, CoRR, 2020, arXiv: 2010.04004.
  7. Y. Kim et al., "A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder," J. Comput. Phys., vol. 451, 2022, article no. 110841.
  8. Y. Zhu et al., "Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data," J. Comput. Phys., vol. 394, 2019, pp. 56-81.
  9. M. Raissi et al., "Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations," J. Comput. Phys., vol. 378, 2019, pp. 686-707.
  10. A.D. Jagtap and G.E. Karniadakis, "Extended physicsinformed neural networks (xpinns): A generalized spacetime domain decomposition based deep learning framework for nonlinear partial differential equations," Commun. Comput. Phys., vol. 28, no. 5, 2020, pp. 2002-2041.
  11. B. Moseley et al., "Finite basis physics-informed neural networks (FBPINNs): A scalable domain decomposition approach for solving differential equations," Adv. Comput. Math., vol. 49, no. 4, 2023.
  12. S. Lunderman et al., "Estimating parameters of the nonlinear cloud and rain equation from a large-eddy simulation," Physica D: Nonlinear Phenom., vol. 410, 2020, article no. 132500.
  13. A.V. Hu et al., "Predicting and Reconstructing Aerosol- Cloud-Precipitation Interactions with Physics-Informed Neural Networks," Atmosphere, vol. 14, no. 12, 2023.
  14. D. Aboelyazeed et al., "A differentiable, physics-informed ecosystem modeling and learning framework for largescale inverse problems: Demonstration with photosynthesis simulations," Biogeosciences, vol. 20, no. 13, 2023.
  15. 최수길, 김기영, 오진태, "미래 유망기술의 Weak Signal 탐지 방안," 전자통신동향분석, 제31권 제2호, 2016, pp. 18-27.
  16. R. Eckhoff et al., "Detecting innovation signals with technology-enhanced social media analysis-experiences with a hybrid approach in three branches," Int. J. Innov. Sci. Res., vol. 17, no. 1, 2015, pp. 120-130.
  17. L. Lu and Q. Cao, "Motion estimation and system identification of a moored buoy via physics-informed neural network," Appl. Ocean Res., vol. 138, 2023, article no. 103677.
  18. S. Han et al., "Approaching epidemiological dynamics of COVID-19 with physics-informed neural networks," J. Franklin Inst., 2024, article no. 106671.
  19. F. Pavirani et al., "Demand response for residential building heating: Effective monte carlo tree search control based on physics-informed neural networks," Energy Build., 2024, article no. 114161.
  20. H. Kuang et al., "A physics-informed graph learning approach for citywide electric vehicle charging demand prediction and pricing," Appl. Energy, vol. 363, 2024, article no. 123059.
  21. 최영배, 이인복, "물리 정보기반 인공신경과 전산유체역을 활용한 자연환기온의 실시간 유동 해석 시뮬레이션 모델 개발," 한국생물환경조절학회 추계학술대회, 2023.
  22. A.D. Jagtap and G.E. Karniadakis, "Extended physicsinformed neural networks(XPINNs): A generalized spacetime domain decomposition based deep learning framework for nonlinear partial differential equations," Commun. Comput. Phys., vol. 28, no. 5, 2020.
  23. K. Shukla, A.D. Jagtap, and G.E. Karniadakis, "Parallel physics-informed neural networks via domain decomposition," J. Comput. Phys., vol. 447, 2021, article no. 110683.
  24. B. Moseley et al., "Finite Basis Physics-Informed Neural Networks (FBPINNs): A scalable domain decomposition approach for solving differential equations," Adv. Comput. Math., vol. 49, no. 4, 2023.
  25. J. Cho et al., "Separable PINN: Mitigating the curse of dimensionality in physics-informed neural networks," arXiv preprint, CoRR, 2022, arXiv: 2211.08761.
  26. J. Cho et al., "Separable physics-informed neural networks," Adv. Neural Inf. Process. Syst., vol. 36, 2024.