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

In the case of a fault occurrence, the pattern of the fault currents obtained by the standard impulse tests contains a typical signature of the nature and the location of the insulation failure involved. This paper presents a new approach to classify the pattern of the arc discharge location as one of the important types of internal incipient faults in transformer windings. The continuous wavelet transform (CWT) has been used to calculate the most predominant frequency of each fault and its time of occurrence. The data obtained from the field tests of a 66 kV/25 MVA interleaved transformer winding and the computer simulations have been used for the classification.

Pattern classification of internal incipient faults during impulse tests using continuous wavelet analysis

The development of high temperature superconductors (HTSC) technology has reached the stage where their application in large scale powers systems is now possible. Recently a prototype fault current limiter (FCL) with a high Tc superconducting (HTS) saturable magnetic core has been produced at the University of New South Wales. This paper analyses the behavior of this type of FCL under symmetrical fault conditions based on the derived characteristics, by utilizing a computer simulation. The results indicate that this type of FCL can limit fault current effectively in transient and steady state situations. It can also limit the voltage drop on the bus bars to acceptable levels during normal operation and fault situations.

http://ieeexplore.ieee.org/iel7/6776774/6780690/06780754.pdf

The performance of a power system high T c superconducting fault current limiter with saturable core under symmetrical fault conditions

This paper presents results of PD test at elevated frequency of supply. Particular attention was given to the discharges that occur in moist pressboard layers immersed in transformer oil. The effect of moisture content on the PD activity was investigated based on the observation of the PD quantities and the characteristics of the PD distributions. Data obtained from the PD test is presented in phase-resolved analysis (PRA) and voltage different analysis (VDA).

PD patterns in oil-impregnated insulation: A comparative study

Results of PD tests in vegetable oil impregnated insulating system are discussed in this paper. The main objective of the study is to investigate the PD behaviour in the particular insulating systems at elevated temperature. Two different sample arrangements of electrical pressboard and insulating paper were used to simulate insulation defects. The sample was placed in a glass cell and was fully immersed in vegetable oil. The initiated PDs were detected and analysed using a wide band PD detector. The PD waveforms were picked up by means of a HFCT sensor and displayed using a digital oscilloscope. Investigation results show that there are some differences in intrinsic characteristics of PD pulses measured from the two different sample configurations.

Measured PD pulses in vegetable-oil-impregnated insulation system

This paper presents a study of the properties of partial discharges at elevated temperature. Particular interest was given to new and aged oil-impregnated insulation as applied in power and instrument transformers. The data analysis was focused on the PD quantities that include the peak and average PD magnitudes and the PD repetition rate. In addition to the PD quantities, an investigation of the discharge characteristics using statistical moments was also conducted. Results indicated that the PD quantities and the characteristics of their distributions are greatly affected by the temperature change. Variation in the PD activity was found to be considerable and very dependent on the insulation condition.

Trevor Blackburn

Papers

Pattern classification of internal incipient faults during impulse tests using continuous wavelet analysis

The performance of a power system high T c superconducting fault current limiter with saturable core under symmetrical fault conditions

PD patterns in oil-impregnated insulation: A comparative study

Measured PD pulses in vegetable-oil-impregnated insulation system

The statistical properties of PD distribution in oil-impregnated insulation at elevated temperature