Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

As transformers age, their internal condition degrades, which increases the risk of failure. To prevent these failures and to maintain transformers in good operating condition is a very important issue for utilities. Traditionally, routine preventative maintenance programs combined with regular testing were used. The change to condition-based maintenance has resulted in the reduction, or even elimination, of routine time-based maintenance. Instead of doing maintenance at a regular interval, maintenance is only carried out if the condition of the equipment requires it. Hence, there is an increasing need for better nonintrusive diagnostic and monitoring tools to assess the internal condition of the transformers. If there is a problem, the transformer can then be repaired or replaced before it fails. An extensive review is given of diagnostic and monitoring tests, and equipment available that assess the condition of power transformers and provide an early warning of potential failure.

Review of condition assessment of power transformers in service

Pattern recognition

Dissolved gas analysis (DGA) is used to assess the condition of power transformers. It uses the concentrations of various gases dissolved in the transformer oil due to decomposition of the oil and paper insulation. DGA has gained worldwide acceptance as a method for the detection of incipient faults in transformers.

A review of dissolved gas analysis measurement and interpretation techniques

A concise review is given of the progress made in the field of partial discharges (PD) at DC voltage. Although ample reference will be made to work of other authors in this field, the paper will concentrate on the progress that was made at Delft University of Technology over a period of 14 years in three PhD projects. In the first project, a start was made with the analysis of the physics of partial discharges at DC voltage and different types of PD were characterized based on parameters like time interval between PD and PD magnitude. In a second project, PD analysis was applied to HVDC apparatus and different means of classification of PD at DC voltage were proposed. In the third project, PD analysis was applied to HVDC mass-impregnated cables and test specifications were proposed. In this paper the work performed in the above three Ph.D. projects is summarized with ample reference to papers of other workers in this field. Attention is given to the mechanism of PD at DC voltage as compared to AC voltage and techniques for measurement and analysis of DC PD patterns. Examples of practical application of DC PD testing are given. Finally, some thoughts on future work are presented.

/pdf/partial-discharges-at-dc-voltage-their-mechanism-detection-2ka0gekh8r.pdf

Partial discharges at DC voltage: their mechanism, detection and analysis

This paper investigates the possibility of applying a fibre optic sensor to detect the ultrasonic signals in HV power transformers. Since these signals are wide-band, the system should be able to detect the components in the low-frequency range. Descriptions of the operating principle together with the sensor construction and also the electronic circuitry necessary to process the optical signals are given. >

Optical fibre sensor for partial discharge detection and location in high-voltage power transformer

Transformer is one of the crucial equipments in electric power network. Moisture in paper insulation acts as a catalyst in transformer insulation aging. Experience has shown that humidity will affect electrical strength of paper and pressboard as well as other insulation system materials by reducing degree of polymerization and accelerated aging. Thus, it is essential to determine the extent of moisture content within insulation system before exciting the transformer and also during normal operation. In this regard, chemical tests on oil have been carried out for many years. Also, electrical tests for assessing the transformer condition have been developed and improved in recent years while most introduced methods aim to extract moisture from transformer insulation system. In this study, the results of practically applicable methods in oil sampling (KFT and capacitance sensors), paper sampling (Dean-Stark method) and also dielectric response (RVM, PDC, FDS) are examined in determining insulation system humidity. Test results through various methods are compared to analyze the sensitivity of different approaches. Crucial practical recommendations as well as advantages and disadvantages of each method are discussed. At last, to recommend an accurate and reliable method for practical implementations, practical tests have been performed simultaneously on three distribution transformers and the results of them are discussed.

Dean-Stark vs FDS and KFT methods in moisture content recognition of transformers

It is well-known that the deformation of transformer winding can produce detectable changes to the frequency response spectrum compared with a referenced past measurement. To interpret such changes for diagnostic purposes, main causes of the trace deviation need to be recognized precisely. In addition, it is useful that the interpretation of transformer frequency response is classified in a way that IoT-based techniques can be developed in the near future to analyze the transformer mechanical integrity. This paper has specifically concentrated on the inductance and capacitance variation due to the axial and radial disk deformation of transformer winding. Analytical analyses on self- and mutual-inductance variations are discussed and capacitance variation is studied in detail for symmetrical and asymmetrical transformer disk deformations. A numerical example is provided to establish the analytical approach and compare inductance and capacitance variation. The analytical approach is finally examined through the experimental study of disk deformation in a 66 kV transformer winding.

Air Core Transformer Winding Disk Deformation: A Precise Study on Mutual Inductance Variation and Its Influence on Frequency Response Spectrum

The crucial role of photoionization in the three-dimensional (3-D) formation of a streamer discharge in a void, filled with air, in an insulator, is demonstrated by performing calculations both with and without a 3-D treatment of photoionization. Secondary electron emission from the walls due to ion impact is found to be too slow to affect the initial streamer formation.

The role of photoionization in streamer discharge formation in voids

We present numerical results that provide an explanation of the finite steps of 20–50 m taken by lightning as it descends from the clouds, or as lightning rises from tall buildings towards the clouds. We also explain why positive streamers rise steadily from the ground to meet the descending lightning bolt at a distance of about 40 m above the ground, completing the discharge channel. Thus, the results define the limits of protection afforded by Franklin rods to buildings. Calculations for a 0.5 m positive point-plane gap show that the axial electric field of an ascending streamer has a remarkable structure comprised of a primary streamer head, followed by a secondary streamer head, and then a long plateau at a level greater than 30 kV cm −1 . This plateau in the electric field continues to elongate as the streamer propagates, maintaining the conductivity in the streamer channel. Once all the voltage is dropped along the channel plateau, streamer propagation must stop until heating modifies the channel properties. A limit is thus defined for the length of the streamer channel of around 20–50 m for typical lightning voltages. The step length must also reduce as the channel approaches the Earth, as observed, because of voltage dropped along the lengthening channel.

Trevor Blackburn

Papers

Optical fibre sensor for partial discharge detection and location in high-voltage power transformer

Dean-Stark vs FDS and KFT methods in moisture content recognition of transformers

Air Core Transformer Winding Disk Deformation: A Precise Study on Mutual Inductance Variation and Its Influence on Frequency Response Spectrum

The role of photoionization in streamer discharge formation in voids

The stepped nature of lightning, and the upward connecting streamer