Esp et al. have used typical analysis methods relating to oil, which includes dissolved gas analysis (DGA), colour, moisture level, acidity, breakdown voltage and furfuraldehyde content together with unsupervised neural networks to unearth further information [Esp1].

Ashkezari et al. through the multivariate analysis method was able to provide correlation among the relevant transformer oil tests consisting of moisture content, acidity, dielectric dissipation factor, resistivity, breakdown voltage and 2-Furfural [Ashkezari1].

Chen and Lin displayed that diagnosing fault conditions from the combination of several simple fuzzy approaches are much better than traditional methods especially for transformers, which have gas-in-oil conditions around the crisp norms [Chen1].

Guardado et al. presented a comparative study of neural network efficiency for the detection of incipient faults in power transformers according to the Dornenburg, modified Rogers, Rogers, IEC and California State University of Sacramento (CSUS) criteria [Guardado1]. This study showed that the neural network rate of successful diagnosis is dependent on the criterion under consideration, with values in the range of 87-100%.

Ding et al. studied the relationship between transformer faults of power transformers and gases dissolved in insulating oil where preliminary simulation results showed a 95% correct diagnosis rate for the artificial neural network (ANN) based method with a modest amount of training data [Ding1].

Zhang et al. presented a two-step ANN method to detect faults with or without cellulose involved where good diagnostic accuracy is obtained with the proposed approach [Zhang1].

Chengbiao et al. used principal component analysis to extract information on the main characteristics of transformer fault diagnosis [Chengbiao1]. However, the results of this study indicated a loss of useful information.

Several authors have used DP as a diagnostic tool to determine the condition of transformers [Moser1, Shroff1, Allan1, Oommen2]. The following studies propose that for a DP of 150, the strength is reduced to 20% of its initial DP value and below this strength will be diminished to almost insignificant tensile strength [Fallou1, Shroff1, Hernadi1, Skubala1, Tamura1]. However recent investigations within the Cigre working group WG47 have provided little evidence that failure of transformers with low DP values are as a result of mechanical weakness of the paper [Duval5]. The DP used as a condition monitoring tool is usually measured from furan samples of the oil. This gives an indication of the general state of the cellulose insulation as the paper fibres are distributed within the oil. The disadvantage of this is that localised degradation is not easily identified and may be masked by the high volumes of oil.

Experiments in the laboratory to identify aging characteristics were performed at temperatures above 100 °C due to the shorter reaction times under these conditions [Kachler1]. The aging under normal operating conditions as yet has not been fully investigated. Emsley concluded that at most a ±20% estimated life of transformer paper insulation at normal operating temperatures can be calculated from laboratory experiments with associated statistical errors [Emsley2].

Abu-Siada and Islam introduced a new method using Gene Expression Programming (GEP) to standardise DGA interpretation techniques thus alleviating the need of expert knowledge for fault diagnosis [Abu-Siada1].

Ma et al. introduced the statistical learning techniques and their application for condition assessment of power transformers [Ma1]. This algorithm when compared to the conventional ratio based DGA interpretation methods are able to indicate several fault conditions that may occur simultaneously within a transformer or if the deterioration is due to normal operating conditions.

Jadav et al. presented findings from the analysis of polarisation and depolarisation current measurements and comparison to dissolved gas analysis [Jadav1]. It was found that although dissolved gas analysis provided useful information about the insulation condition and the presence of the fault within a transformer it appears difficult to quantify the amount of insulation degradation taking place.

It was found that most of the new methods available are usually variations and combinations of the fundamental diagnostic methods as discussed above.

References

[Abu-Siada1]	Abu-Siada, A., Islam, S. “A New Approach to Identify Power Transformer Criticality and Asset Management Decision Based on Dissolved Gas-in-oil Analysis,” IEEE Transactions on Dielectrics and Electrical Insulation, Vol 19, No. 3, pages 1007-1012, June 2012
[Allan1]	Allan, D., Jones, C., Sharp, B., “Studies of the Condition of Insulation in Aged Power Transformers. Part 1: Insulation Condition and Remnant Life Assessments for In-service Units,” IEEE Proceedings of the 3rd International Conference on Properties and Applications of Dielectric Materials, Tokyo, Japan, pages 1116-1119, 8-12 July 1991
[Ashkezari1]	Ashkerzari, A. D., Ma, Hui., Ekanayake, C., Saha, T. K., “Multivariate Analysis for Correlations among Different Transformer Oil Parameters to Determine Transformer Health Index,” IEEE Power and Energy Society Meeting General Meeting, 2012
[Chen1]	Chen, A. P., Lin, C. C., “Fuzzy Approaches for Fault Diagnosis of Transformers,” Fuzzy Sets and Systems, 118(1), pages 139-151, 2001
[Chengbiao1]	Chengbiao, Z., Yunbai, L., Xishan, W., “Study on Fault Diagnosis of Transformer Based on Principal Component Analysis of Dissolved Gas,” High Voltage Engineering, Vol. 8, pages 9-11, 2005
[Ding1]	Ding, X., Yao, E., Liu, Y., Griffin, P. J., “ANN Based Transformer Fault Diagnosis Using Gas-in-oil Analysis,” PROC AM Power Conference, Vol. 57-2, pages 1096-1100, 1995
[Duval5]	Duval, M., “New Frontiers of DGA Interpretations for Power Transformers and Their Accessories,” Techcon Canada, Montreal 27-28 September 2012
[Emsley2]	Emsley, A. M., Stevens, G. C., “Review of Chemical Indicators of Degradation of Cellulosic Electric Paper Insulation in Oil-filled Transformers,” IEE Proceedings – Science, Measurement and Technology, Vol. 141, No. 5, pages 324-334, 1994
[Esp1]	Esp, D. G., Carrillo, M., McGrail, A. J., “Data Mining Applied to Transformer Oil Analysis Data,” IEEE International Symposium on Electrical Insulation, Arlington, Virginia, USA, pages 12-15, 7-10 June 1998
[Fallou1]	Fallou, B., “Synthesis of Work Carried Out at LCIE on Paper Degradation,” Rev. Gen. Elect. 79, page 645-661, 1970
[Guardado1]	Guardado, J. L., Naredo, J. L., Moreno, P., Fuerte, C. R., “A Comparative Study of Neural Network Efficiency in Power Transformers Diagnosis Using Dissolved Gas Analysis,” IEEE Transactions on Power Delivery, Vol. 16, No. 4, pages 643-647, 2001
[Hernadi1]	Hernadi, A., “Correlation of Strength and Thermal Degradation of Paper Made from Cellulose Fibres,” Cell. Chem. Technology, Vol. 16, No. 1, pages 27-34, 1982
[Jadav1]	Jadav, R. B., Saha, T. K., Ekanayake, C., “Transformer Diagnostics Using Dissolved Gas Analysis and Polarisation and Depolarisation Current Measurements – A Case Study,” Universities Power Engineering Conference (AUPEC), 2011 21st Australasian , Vol., No., Pages 1-6, 25-28 September 2011
[Kachler1]	Kachler, A. J., Hohlein, I., “Ageing of Cellulose at Transformer Service Temperatures. Part1: Influence of Type of Oil and Air on the Degree of Polymerisation of Pressboard, Dissolved Gases and Furanic Compounds in oil,” IEEE Electrical Insulation Magazine, Vol. 21, No. 2, pages 15-21, 2005
[Ma1]	Ma, H., Saha, T. K., Ekanayake, C., “Statistical Learning Techniques and Their Applications for Condition Assessment of Power Transformer,” IEEE Transactions on Dielectric and Electrical Insulation, Vol. 19, No. 2, pages 481-489, April 2012
[Moser1]	Moser, H., Dahinden, V., “Application of Cellulosic and Non-cellulosic Materials in Power Transformers,” International Conference Large High Voltage Electrical Systems, CIGRE Proceeding 31st Session, page 12, 1986
[Oommen2]	Oommen, T. V., Arnold, L. N., “Cellulose Insulation Materials Evaluated by Degree of Polymerization Measurements,” IEEE Proceedings of the 15th Electrical/Electronics Insulation Conference, pages 19-22, Chicago, IL, USA, October 1981
[Shroff1]	Shroff, D. H., Stannett, A. W., “A Review of Paper Ageing in Power Transformers,” IEE Proceedings, Vol. 132, No. 6, pages 312-319, November 1985
[Skubala1]	Skubala, W., “Investigation of Ageing of Transformer Insulation With and Without the Effects of an Electric Field,” Prz. Papierniczy, Vol. 5, No. 188, pages 40-58, 1974
[Tamura1]	Tamura, R., Anetai, H., Iskii, T., Kawawmura, T., “Diagnosis of Ageing Deterioration of Insulating Paper,” JIEE Proc. Pub. A, Vol. 101, pages 30-36, 1981
[Zhang1]	Zhang, Y., Ding, X., Liu, Y., Griffin, P. J., “An Artificial Neural Network Approach to Transformer Fault Diagnosis,” IEEE Transactions on Power Delivery, Vol. 11, No. 4, pages 1836-1841, October 1996