Moisture ingress is one of the major causes of aging in oil–paper insulated transformers; the measurement of the moisture content of the insulating material is thus necessary during transformer disassembly for maintenance. One critical engineering consideration is the need for rapid and nondestructive moisture content measurement of the oil–paper insulation to minimize the power-outage time. Herein, a rapid characterization method for the moisture content of oil–paper insulation based on the dielectric response obtained using a microstrip petal-like ring resonator (MPRR) is proposed. The geometric parameters of the resonator and dielectric response model were established based on simulations and theoretical analysis. The geometry of the resonant ring was optimized by modifying the originally circular ring structure of the microstrip ring resonator (MRR) into a petal-like ring structure; this increases the number of resonant peaks from five to seven in the frequency range of 0.5–6.0 GHz for the same electrode area. The effects of the coupling mode, coupling gap, and microstrip ring size on the resonance characteristics of the MPRR were simulated and analyzed. Compared with the MRR, the MPRR has the advantages of a reduced fundamental resonant frequency and a smaller resonant-ring area that improves radiation efficiency. The impact of moisture content on the dielectric response of the pressboard obtained using the MPRR was studied. A moisture-calculation equation based on the measured dielectric response for a selected type of pressboard was established by the Lasso regression. The results of verification experiments show that the error of the proposed method is sufficient for practical applications.

This paper proposes a fast frequency domain dielectric spectroscopy (FDS) test scheme for diagnosing oil-paper insulation. The high-voltage output section introduces dynamic power supply tracking (DPST) and cascading techniques to generate high-voltage excitation signals. The DPST can increase the amplifier's voltage slew rate (SR) from 13 V per microsecond to 36.3 V per microsecond without crossover distortion. In the micro-current measurement section, the error active suppression structure of bias current and bias current under time-varying temperature conditions is proposed to improve the temperature stability of the measurement circuit. The fast testing technology shortens the test period by 65% in the low-frequency band by using the method of multi-frequency signal synthesis and partial cycle waveform fitting. The experimental results show that the total harmonic distortion and noise (THD+N) of the output voltage of the instrument are less than or equal to 0.078%. The voltage range is 780 Vpp. The standard deviation of current is less than 1.5%, and the standard deviation of tan δ is less than 1.7%, which verifies the stability of the designed instrument.

Detecting power transformer winding degradations at an early stage is very important for the safe operation of nuclear power plants. Most transformer failures are caused by insulation breakdown; the winding turn-to-turn short circuit fault is frequently encountered. Experience has shown that routine testing techniques, e.g., winding resistance, leakage inductance, and sweep frequency response analysis (SFRA), are not sensitive enough to identify minor turn-to-turn short defects. The SFRA technique is effective only if the fault is in such a condition that the flux distribution in the core is prominently distorted. This paper proposes the frequency domain reflectometry (FDR) technique for detecting and locating transformer winding defects. FDR measures the wave impedance and its change along the measured windings. The wire over a plane model is selected as the transmission line model for the transformer winding. The effectiveness is verified through lab experiments on a twist pair cable simulating the transformer winding and field testing on a real transformer. The FDR technique successfully identified and located the turn-to-turn short fault that was not detected by other testing techniques. Using FDR as a complementary tool for winding condition assessment will be beneficial.

Moisture ingression accelerates the ageing and degradation of the oil‐paper insulation. Rapid and accurate moisture assessment of power apparatus in the field is thus essential. In this paper, an innovative design of a microstrip petal‐like complementary split ring resonator (MPCSRR) is proposed for the moisture assessment of the oil‐paper insulation in the GHz range. The MPCSRR has advantages in concentrating the electromagnetic field, significantly increasing the number of resonance peaks, and achieving a sensitivity of approximately 3–4 times higher than the traditional microstrip ring resonators (MRRs). A mathematical model for analysing the dielectric response is established, combining the theoretical analysis and equivalent circuit model. The effects of microstrip width and substrate thickness on impedance matching, as well as the incidence depth and test sensitivity of the resonator, are explored using simulation. The experimental results demonstrate that the MPCSRR achieves rapid moisture detection within tens of seconds and accurately characterises the wetness of the oil‐paper insulation. A moisture assessment model based on Lasso regression achieves an accuracy of 4%, confirming the reliability and validity of the proposed method. The results also verify that the sensitivity of the MPCSRR is also higher than that of the traditional MRR, showing its potential for precise and efficient moisture detection in oil‐paper insulation.

Moisture is one of the most critical factors contributing to the inability of oil-paper insulating bushings to operate safely and stably, primarily originating from the internal capacitive core and the end screen area. This paper investigates a rough estimation method for moisture distribution in oil-paper insulated bushings based on the Polarization and Depolarization Current testing technique. First, simulation models of internal and external moisture ingress in the bushing were constructed using COMSOL, verifying the differences in moisture distribution under the two moisture ingress conditions. Subsequently, a Debye model of macroscopic charge polarization was derived, and the impact of the proton conduction migration mechanism of hydrolyzed H+ ions on the difference in positive and negative ion mobility was analyzed. Combined with experimental results from non-uniformly moistened samples, it was demonstrated that the current difference observed during polarization (depolarization) under polarity reversal originates from the disparity in macroscopic charge polarization intensities due to the differing migration strengths of the ions. Finally, a rough identification method for moisture distribution in bushings was proposed, and its effectiveness was validated through experiments using a scaled-down bushing model.