Method for Predicting the Water Absorption of External Insulating Silicone Rubber

• Published in: IEEE transactions on dielectrics and electrical insulation Vol. 29; no. 4; pp. 1242 - 1250
• Main Authors: Cheng, Li, Liu, Yunfan, Chen, Rongxin, Zhang, Sida, Liao, Ruijin, Yang, Lijun, Wang, Tingting
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.08.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Absorption; Algorithms; Bonding strength; Contraction–expansion (C–E) algorithm; Curve fitting; Debonding; Diffusion coefficient; Diffusion welding; Fractures; Humidity; Hygroscopicity; Insulation; Langmuir; Moisture; Moisture effects; molecular simulation; prediction; Prediction algorithms; Prediction models; Predictive models; Rubber; Silicone resins; Silicone rubber; Silicones; Temperature; Water absorption
• ISSN: 1070-9878; 1558-4135
• DOI: 10.1109/TDEI.2022.3183653

The intrusion of moisture may induce an abnormal increase in temperature or may even cause fractures in insulation equipment. Therefore, the hygroscopicity of rubber materials must be characterized. However, the existing methods for weighing organic polymer materials, such as silicone rubber, for moisture-absorption tests are time-consuming. In this article, a practical method for testing the moisture-absorption performance of silicone rubber materials for external insulation is proposed. The basic characteristics of the Langmuir diffusion equation for the water-absorption characteristics of silicone rubber are analyzed. A Langmuir curve fitting method based on the contraction-expansion (C-E) algorithm is developed, and its fitting effect is preliminarily verified by combining the experimental errors obtained from the experiment. Results show that the fitting effect has a strong relationship with the initial range of diffusion coefficient, bonding coefficient, and debonding coefficient. The time-consuming process of directly applying the fitting method is reduced from 70 to 50 days. The initial ranges of diffusion, bonding coefficient, and debonding coefficient are determined via the results of molecular simulations and experiments combined with an analysis of the literature. A prediction model is optimized and then validated against the experimental data from the recent literature on the water-absorption characteristics of silicone rubber at different temperatures. The experimental time required to reach saturated water absorption is remarkably reduced from 70 days to 1 day. Therefore, the method developed herein is feasible for engineering applications. Moreover, the entire water-absorption curve can be possibly plotted with an error of not greater than 5% via this prediction method.