This study describes a low-cost sensor-based scheme for online moisture measurement in transformer oil. Considering a harsh environment, moisture measurement of oil at ppm level is challenging, so the sensor should be highly sensitive and thermally and chemically stable. The conventional methods are mostly offline, have long test times and are costly. A thin-film parallel plate capacitive moisture sensor has been fabricated using inexpensive materials to measure the moisture of fresh, aged and 0.5% water content oil samples. The moisture sensing film is highly hydrophilic and is made of alumina oxide, a very stable material. Experiments were conducted to determine the response parameters of the sensor in the presence of moist nitrogen gas, the concentration of which was measured by a commercial SHAW dew point meter. Experiments were also performed to measure the moisture of different oil samples in the laboratory. Results show that the output of the sensor is sensitive (6 pF/ppm moisture), has negligible hysteresis (∼1% at 1 kHz) and has fast response and recovery times (T res = 70 s, T rec = 170 s).

https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/iet-smt.2019.0319

Highly sensitive thin-film capacitive sensor for online moisture measurement in transformer oil

The performance of oil-impregnated paper significantly affects the safety and reliability of transformers. Previous studies on water reducing the breakdown strength of oil-impregnated paper have been mainly due to increasing charge carriers and conductivity. However, the influence of water on the cellulose crystal structure of oil-impregnated paper is still not clear; in particular, its role on breakdown strength remains ambiguous. In this study, the AC breakdown strength and crystal structure of vegetable and mineral oil-impregnated paper with different moisture contents were investigated. The results show that with increasing moisture content, the cellulose crystallinity and crystal size increases linearly, while the breakdown strength decreases. The breakdown strength of vegetable oil-impregnated paper is higher than that of mineral oil-impregnated paper and is less affected by moisture. Herein, a model is proposed on the effects of water on the crystal structure of oil-impregnated paper and the correlation of breakdown strength with crystal structure is reported.

Influence of moisture content on cellulose structure and breakdown strength of vegetable oil-impregnated paper

Condition assessment of insulation is widely used to identify the state of aging of power transformers. Traditionally, condition assessment is done using dielectric response measurements conducted at low frequencies. However, due to long testing time required for these methods, attention has now been paid towards measurements at microwave (above 1 GHz) frequencies. In this paper, we establish a correlation between dielectric properties and the state of aging of transformer pressboard samples. This is useful as it paves way to develop techniques to assess the state of aging by measuring dielectric properties at high frequencies. Dielectric measurements are made by using a microstrip ring resonator. The state of aging is determined by measuring the tensile index. Tests were carried out on dry and wet oil-impregnated pressboard samples thermally aged at laboratory conditions for durations up to 45 days. The results were compared with field aged samples collected from a 33 kV sealed type distribution transformer of 18 years in service. It was found that the permittivity values had a good correlation with respective tensile index values whereas loss tangent values showed limited correlation for both dry and wet samples. The results of field aged samples matched well with the established correlation curves. It can be concluded that the permittivity of pressboard samples estimated from the ring resonator can be effectively used to evaluate the state of aging of transformer insulation.

Aging assesment of transformer pressboard insulation by micro-strip ring resonator at GHz frequencies

As a result of the deregulations in the power system networks, diverse beneficial operations have been competing to optimize their operational costs and improve the consistency of their electrical infrastructure. Having certain and comprehensive state assessment of the electrical equipment helps the assortment of the suitable maintenance plan. Hence, the insulation condition monitoring and diagnostic techniques for the reliable and economic transformers are necessary to accomplish a comprehensive and proficient transformer condition assessment.,The main intent of this paper is to develop a new prediction model for the aging assessment of power transformer insulation oil. The data pertaining to power transformer insulation oil have been already collected using 20 working power transformers of 16-20 MVA operated at various substations in Punjab, India. It includes various parameters associated with the transformer such as breakdown voltage, moisture, resistivity, tan δ, interfacial tension and flashpoint. These data are given as input for predicting the age of the insulation oil. The proposed aging assessment model deploys a hybrid classifier model by merging the neural network (NN) and deep belief network (DBN). As the main contribution of this paper, the training algorithm of both NN and DBN is replaced by the modified lion algorithm (LA) named as a randomly modified lion algorithm (RM-LA) to reduce the error difference between the predicted and actual outcomes. Finally, the comparative analysis of different prediction models with respect to error measures proves the efficiency of the proposed model.,For the Transformer 2, root mean square error (RMSE) of the developed RM-LA-NN + DBN was 83.2, 92.5, 40.4, 57.4, 93.9 and 72 per cent improved than NN + DBN, PSO, FF, CSA, PS-CSA and LA-NN + DBN, respectively. Moreover, the RMSE of the suggested RM-LA-NN + DBN was 97.4 per cent superior to DBN + NN, 96.9 per cent superior to PSO, 81.4 per cent superior to FF, 93.2 per cent superior to CSA, 49.6 per cent superior to PS-CSA and 36.6 per cent superior to LA-based NN + DBN, respectively, for the Transformer 13.,This paper presents a new model for the aging assessment of transformer insulation oil using RM-LA-based DBN + NN. This is the first work uses RM-LA-based optimization for aging assessment in power transformation insulation oil.

Enhanced optimal trained hybrid classifiers for aging assessment of power transformer insulation oil

Here, a new device is presented to supervise the ageing process of insulating oils used in power transformers. It is based on the use of a microstrip antenna whose electrical properties are very sensitive to the permittivity and conductivity of the medium. As many works showed that the drift of permittivity due to transformers aging is very small, the achieved sensor must exhibit a great sensitivity. In order to assess the aging of the insulation oil, the authors will track the resonant frequency of the antenna which is directly related to the permittivity of the oil, and the reflection coefficient level of the antenna at the resonance frequency which is linked to the conductivity. The experiments have been made on vegetable oil and results revealed to be enthusiastic. The sensor is proposed to be used in an embedded system in order to achieve a low cost and safe preventive diagnosis.

https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/iet-smt.2018.5489

Microwave-based sensor for power transformers’ insulation oil monitoring

Increased moisture content (MC) is considered as one of the main factors for the failure of transformer insulation. This paper presents a method of estimating the MC of a transformer pressboard insulation based on the relative permittivity and loss tangent measurements performed at GHz using a multi-layer micro-strip ring resonator. It was fabricated on a material of known dielectric properties and the sample to be tested (i.e. pressboard) was placed on the ring structure. The dielectric properties, namely loss tangent and permittivity, of the sample were then estimated by measuring the S 21 transmission characteristics of the ring structure using a vector network analyzer. The technique was validated using a test sample of known dielectric properties. Additionally the results were compared with traditional frequency domain dielectric spectroscopy (FDS) measurements conducted at 1 mHz −1 kHz. The measurements were performed on oil impregnated and non-impregnated pressboard samples having different MCs conditioned by drying and/or wetting processes. The actual MC of the pressboard samples were obtained by weight measurements. It was found that the resonance frequency increased when the MC was reduced. Furthermore, a positive non-linear relation could be observed between the MC and both the relative permittivity and the loss tangent. It can be concluded that the proposed method could be used to estimate the MC of pressboards.

Moisture estimation of transformer pressboard by micro-strip ring resonator at GHz frequencies

Measurement of dielectric properties of copper clad printed circuit boards using microstrip ring resonators is a well-studied problem widely in use. The problem addressed here is different from this as the sample under test is difficult to copper clad. The sample is placed on a ring resonator fabricated on a printed circuit board of known dielectric properties and the properties of the test sample are obtained from the measurement of the resulting multilayer stack. Although the measurement of dielectric constant does not pose problems, the calculation of the loss tangent needs the radiation and conduction losses. Out of the two losses, estimation of radiation loss is a difficult task, firstly because the problem is multi-layer and, secondly because the dielectric properties of one of the layers (test sample) are unknown. As a result it is usually assumed that the radiation losses are negligible. This paper examines the effects of this assumption using accurate electromagnetic simulations.

A. C. M. Ahamed

Papers

Moisture estimation of transformer pressboard by micro-strip ring resonator at GHz frequencies

The effects of radiation losses on the measurement of loss tangent using microstrip ring resonators