Yoshinobu Mizutani

Detection of deformation and displacement of transformer winding by frequency response analysis

Frequency response analysis (FRA) is a diagnosis method to detect troubles of a transformer, which are deformation and displacement of winging and so, by detecting change of a transfer function compared with that measured when the transformer is normal. However, identification of the troubles detected by FRA is not clear. In this paper, firstly, identification method of the troubles is investigated using a model transformer. Secondly, the proposed identification method is applied to identify a trouble of a real transformer caused by short-circuit tests.

Abnormality diagnosis of transformer winding by frequency response analysis (FRA) using circuit model

A circuit model of a real transformer is established to reproduce the transfer functions of the transformer. This model is based on the configuration and dimensions of the real transformer. Using this circuit model, the sensitivities of the circuit elements to the resonant frequencies are analyzed. In frequency response analysis (FRA), this sensitivity analysis is helpful in the diagnosis of abnormalities by extracting the circuit elements having relatively high sensitivity to the shift of the resonant frequency. Furthermore, the ability to detect abnormalities in a transformer winding is discussed quantitatively by relating the shift of the resonant frequency and the degree of abnormality via variations of circuit elements on the basis of electromagnetic theory. The accumulation of such data for transformers with different classes of rated voltage and capacity, connections of phases, winding configurations, and so will contribute construction of diagnosis criteria for transformer windings by FRA.

Relationship between remaining antioxidant content and radiation-thermal degradation in crosslinked polyethylene

To clarify the deterioration mechanism of cables installed in nuclear power plants, three kinds of XLPE sheets with various antioxidant contents were aged by 100 Gy/h γ-ray irradiation at room temperature, 60, and 100°C. The remaining antioxidant content estimated from an oxidation induction time of a differential calorimetric thermogram as well as the elongation at break decreases faster as the initial antioxidant content decreases. On the other hand, an evaluation parameter called oxidation degree, which is the absorbance ratio of the carbonyl group to the methylene group obtained by infrared spectroscopy, increases with aging time. The oxidation degree increases with an increase in the elongation at break, and the relationship of these two parameters can be plotted around a sigmoid curve. Furthermore, the elongation at break and oxidation degree show reasonably stable transitions until the antioxidant content decreases to a certain value. After such a period, the elongation at break starts to change drastically and becomes sufficiently low value. It is clarified from these experimental results that there exists a certain antioxidant content that effectively inhibits oxidation.

Proposal of objective criterion in diagnosis of abnormalities of power-transformer winding by Frequency Response Analysis

Frequency Response Analysis (FRA) is a powerful technique to detect internal abnormalities within power transformers. However, diagnosis criteria of power transformers by FRA have not been fully established yet. In this paper, an indicator of identification degree of the two transfer functions is examined using data of real transformers and problems in practical application are discussed. Based on this discussion, an improved indicator of degree of identification between two transfer functions is proposed. The proposed indicator is applied to transfer functions of real transformers with or without some abnormalities. Based on this discussion, an objective criterion for diagnosis of abnormalities of transformer windings by FRA is proposed.

Degree of polymerization estimation of insulation papers in power transformers based on load and temperature histories in Java-Bali Region of Indonesia National Electric company

Degree of polymerization (DP) is thought to be a good index to determined deterioration level of insulation paper in power transformers. In practical conditon, using this method is complicated and expensive due to the necessity of disasembling the transformer. Therefore an alternative method has been expected. In this paper a new method to estimate DP from load and temperature histories is proposed. The exsistence of moisture also considered. Mechanical strength of insulation paper of power transformers in Java Bali Region is assesed using the estimated DP based on the real load and temperature data. Furthermore failure probability of those transformers is considered based on the design stresses in a transformer winding generated by fault currents. Finally the estimated DP values are compared with those from furfural test.