Failure of sealing materials is a principal cause to the leakage of electrical penetration assemblies (EPA) in nuclear plants, and the essence can be attributed to small deformations or defects taking place in the sealing materials (glass, epoxy, etc.) as a result of harsh environment influence (high temperature, pressure and ionizing radiation), which leads to leakage when the stress/strain exceeds the threshold value. Metal-to-glass sealing EPA has unique advantages of higher temperature and pressure endurance than organic material sealing EPA, and has been applied in the pressure vessel of High-Temperature Reactor Pebble-bed Modules (HTR-PM) at the Shidao Bay Nuclear Power Plant in China. To achieve on-line state monitoring, we proposed a new method to monitor the stress in the sealing glass by optical fiber sensing technique. Our research showed that the stress in sealing glass could be measured via embedding an optical fiber Bragg grating (FBG) sensor in glass. Optical fiber sensing technique has been widely used for stress measurement in many fields, however applications in metal-to-glass sealing EPA have not been reported in the literature yet. Taking advantage of the small size of a fiber sensor, the embedding of fiber will not affect the sealing structure. And taking advantage of the similar chemical content, fiber and glass can be fused together easily without affecting insulation. In this paper, a brief review on applications of FBG in nuclear facilities was present. The model of FBG embedded EPA was built based on finite element method. Sensitivity analysis about the impact of environment parameters including temperature and pressure on stress had been studied numerically. And the theoretical Bragg wavelength shift of the embedded sensor was derived from the strain/stress distribution. Experiments had been carried out in some main aspects, including pressure and thermal test, from which the relationship between environment parameters and Bragg wavelength shifts was obtained. This research makes an initial attempt for realizing an on-line real-time long-term state monitoring and sets a base for the life cycle diagnostics of EPA in nuclear reactors.

This content is only available via PDF.
You do not currently have access to this content.