Research on moisture content detection and distribution imaging of insulating paperboard based on terahertz technology
Abstract
Moisture content in oil-paper insulation systems of power transformers serves as a critical indicator for assessing their insulation condition and aging degree. Traditional testing methods suffer from issues such as sample destruction, poor resistance to interference, and an inability to obtain images of moisture distribution. We propose the use of terahertz time-domain spectroscopy (THz-TDS) technology for quantitative monitoring and imaging of moisture content in oil-immersed insulating paperboard. Mineral oil and vegetable oil-immersed paperboard samples with varying moisture contents were prepared and analyzed for their time-domain and frequency-domain characteristics in the 0.3 to 1.8 THz frequency band. Results indicate that as moisture content increases, THz pulse amplitude attenuation becomes significant and phase delay increases. The extracted absorption coefficient integral value (Sa) exhibited a linear correlation with moisture content (R2=0.97), and this quantitative relationship was consistently observed in the two representative transformer insulating oils investigated in this study, namely, mineral oil and vegetable oil. Density functional theory simulations using Gaussian 09 software revealed, at the microscopic level, that hydrogen bond resonance and cooperative vibrations between water molecules and cellulose hydroxyl groups constitute the physical mechanism driving terahertz energy absorption. Finally, this technique enabled imaging of moisture distribution in unevenly dampened cardboard. The study confirms the significant application potential of terahertz technology in the nondestructive detection of moisture content in insulating materials.
