The Verification, Validation, and Uncertainty Quantification Framework of the EAST Tokamak Diagnostic System
Shaohua Yuan, Haiqing Liu, Kazuaki Hanada, Mitsutaka Isobe, Y. Zhang, Ting Lan, Xiang Liu, Hui Lian, Yuqi Chu, Shouxin Wang, Zhiyong Zou, Yuan Yao, Ruiping Zhu, Yuyang Liu, Y. X. Jie
Abstract The diagnostic system in a tokamak serves as the foundation for research in plasma physics, plasma operation control, and device protection. The accuracy and reliability of diagnostic data have long been critical considerations in tokamak data analysis and processing. To achieve systematic error assessment, optimized design, and coupling with other systems or simulation modules, the development of a digital diagnostic system module, digital twin, can significantly reduce labor and time costs. Based on a digital diagnostic model, uncertainty quantification and sensitivity analysis are conducted for the diagnostic system, providing a basis for its upgrade. Meanwhile, the engineering design, system hardware, and data processing system of the actual diagnostic system are hierarchically validated to identify uncertainty sources, which are then compared with those of the digital system. This enables complementary validation between the real system and its digital twin. By analyzing uncertainty sources and conducting sensitivity analysis, high-impact uncertainty sources are prioritized for optimization, thereby enhancing the accuracy and reliability of the diagnostic system. This paper quantifies uncertainties and sensitivities of the POINT system and its digital module. The VVUQ analysis yields a total uncertainty of 6.17% and a high confidence coefficient of 0.995. The agreement between experiment and digital module exceeds 97.5% except in the boundary region (≈82.5% due to missing chords). Sensitivity analysis shows that laser frequency stability (≈30%), core channel position (≈23%), and ion mass (≈11%) are the dominant factors. Future devices should improve frequency stabilization, core chord coverage, and fuel calibration to reduce density inversion uncertainty.