Power transformer is one of the electrical equipment that has a central and critical role in the power system. In order to avoid power transformer failure, information system that provides the transformer condition is needed. This paper is present the information system for diagnostic transformer condition based on health index. Health Index method provides a comprehensive transformer condition assessment. This method is resulting power transformer condition that divided into several categories according to the life time prediction and degradation level of transformer components. Analysis on dissolved gas, oil, and furan (insulation paper) are subjected to detect the faults types that could be occurs in transformer. The outputs of the analysis are life time prediction, types of possible faults, and recommendations for future maintenance action. For practical application, this method is applied to the transformer data test that provided by Indonesia Power Electric Company or PT PLN. Results of the analysis found that the smaller Transformer Health Index means greater number of population failure probability. This relationship is quite strong with the value of the correlation coefficient r is -0.63.

This paper presents a comparative study on mother wavelets using a fault type classification algorithm in a power system. The study aims to evaluate the performance of the protection algorithm by implementing different mother wavelets for signal analysis and determines a suitable mother wavelet for power system protection applications. The factors that influence the fault signal, such as the fault location, fault type, and inception angle, have been considered during testing. The algorithm operates by applying the discrete wavelet transform (DWT) to the three-phase current and zero-sequence signal obtained from the experimental setup. The DWT extracts high-frequency components from the signals during both the normal and fault states. The coefficients at scales 1–3 have been decomposed using different mother wavelets, such as Daubechies (db), symlets (sym), biorthogonal (bior), and Coiflets (coif). The results reveal different coefficient values for the different mother wavelets even though the behaviors are similar. The coefficient for any mother wavelet has the same behavior but does not have the same value. Therefore, this finding has shown that the mother wavelet has a significant impact on the accuracy of the fault classification algorithm.

Under a DC stress, a wire particle within gas insulation will exhibit complex motion behaviour, leading to more challenges to detect its existence. Therefore, this work deals with an investigation of corona discharge characteristics generated from a wire particle with different particle positions in an air gap. Three different wire particles are prepared in this investigation. The particle is placed between parallel-plane electrodes with a fixed gap of 20 mm. Five different positions are evaluated to study the PD behaviour generated from the wire particle. The exponential probability distribution is used to evaluate the pulse occurrences from different positions. Pulse sequence analysis (PSA) technique is employed, and an interpretation algorithm is introduced to understand the discharge pattern. The results show that each of the particle positions creates a unique pattern, which discharge mechanism behind the pattern is explained. From the pattern observation, a wire particle has single cluster of corona type as attached to the electrode while multi clusters are found when it is floating in the gap. Moreover, it is also found that different particle lengths are also affecting the discharge pattern characteristic.

Partial discharge (PD) monitoring plays a crucial role in identifying insulation defects in high-voltage rotating machinery, where accurate classification is essential for improving the reliability and efficiency of condition-based maintenance (CBM). This work proposes hybrid Convolutional Neural Network (CNN) models to classify PRPD patterns from six different defects in a rotating machine insulation. Various hybrid models were evaluated by integrating CNN with machine learning algorithms, which include Support Vector Machines (SVM), k-Nearest Neighbors (KNN), Logistic Regression (LR), Decision Trees (DT), Random Forests (RF), and Naive Bayes (NB). The results reveal that all proposed hybrid models consistently outperform CNN in terms of computational efficiency, by achieving an average accuracy of 94.87% across all models using two optimizers, ADAM and SGDM. Notably, CNN-RF and CNN-KNN models achieve the best performance, with an accuracy exceeding 96% with lower computational time compared to CNN, which only achieves 94.44% accuracy. Thus, this work provides valuable insight into enhancing PRPD classification with lower computational cost while increasing the classification accuracy of PRPD patterns from rotating machine insulation.

This paper deals with AC particle-triggered corona discharge as a follow-up to our previous research with DC voltage to clarify the particle-triggered corona discharge process in SF₆ gas. Corona current pulses, charges associated with a corona current pulse, and corona light pulses were observed with an aluminum ellipsoidal particle suspended in a parallel plane electrode system under a SF₆ gas pressure range of 30 kPalesPles50 kPa by changing the particle position. Corona mode, phase (Phi)-charge (q) characteristics as well as corona discharge processes were discussed and the following results were obtained. AC corona mode depended on the instantaneous applied voltage, voltage gradient as well as the particle position and then, the Phi-q characteristics were also affected by those parameters. Charges flow into the floating particle due to coronas on the both tips of particle and excite field fluctuations around the opposite side of particle in addition to the applied AC field. The field fluctuation in SF₆ gas by the corona charges was about 4% of the applied field and much lower than that in air gap which was about 70%. That is, the corona development was suppressed effectively by high electron affinity of SF₆ gas even in the case of floating particle. The less effective interference between coronas on the both side of particle in SF₆ gas results in an obscure local minimum in the breakdown voltage characteristics as the particle is in the vicinity of electrode as contrasted with a drastic fall in the breakdown voltage by the particle in air.

The final objective of this research project is clarification of the particle-triggered corona mechanism in SF₆ gas-insulated electrical equipment as well as development of high precision diagnostic method of foreign metallic particle in GIS insulated by SF₆ gas using the electrical signal due to corona discharge. The following research steps are conducting to achieve our goal: 1) clarification of particle-triggered corona phenomena in air which is well-understood corona mechanism, 2) clarification of particle-triggered corona phenomena under dc and ac applied voltage in SF₆ gas at low pressure in which a stable corona occurred, and 3) clarification of the pressure effect on the particle-triggered corona mechanism. This study is the second step of our research series research, which focuses on characteristics and the mechanism of particle-triggered corona discharge in SF₆ gas on the basis of measurements of corona current pulses at a pressure range of 30 kPa to 50 kPa under dc applied voltage. The identification of positive and negative corona pulses in SF₆ gas from the waveform, amplitude and repetition rate of corona current pulses is difficult; thus, current and light pulses of coronas from the both tips of particle were measured simultaneously to discriminate the polarity of corona current pulses. Experiments showed that near the corona onset voltage, positive corona accompanied pulse current but negative corona tended to be glow-like corona. As the applied voltage was increased, negative corona changed into streamer type accompanying strong current pulse. Positive corona pulses maintained a small amplitude with an instantaneously high repetition rate, even if the field enhancement on the positive tip of particle occurred due to the positive-ion inflow to the particle produced by negative corona. On the basis of current pulse measurements, the mechanism of particle-triggered corona was discussed. Moreover, corona activity and breakdown voltage were investigated by changing the particle position. This study confirmed that the active corona development and fall in breakdown voltage were due to the effect of intermittent short-circuit arc in the minor gap between particle and electrode when the particle was in the vicinity of the electrode