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

Data Mining Practical Machine Learning Tools and Techniques

Intermolecular And Surface Forces

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

Medium and high voltage power cables are widely used in the electrical industry with substantial growth over the last 20-30 years ago, particular in the use of XLPE insulated systems. Ageing of the cable insulation is becoming an increasing problem that requires development of reliable methods for on-line condition assessment. For insulation condition assessment of MV and HV cables, partial discharge (PD) monitoring is one of the most effective techniques. However on-site and on-line PD measurements are affected by electromagnetic interference (EMI) that makes sensitive PD detection very difficult, if not impossible. This paper describes implementation of wavelet transform techniques to reject noise from on-line partial discharge measurements on cables. A new wavelet threshold determination method is proposed with the technique. With implementation of this novel de-noising method, PD measurement sensitivity has been greatly improved. In addition, a full AC cycle data recovery can be achieved instead of focusing only on recovering individual PD pulses. Other wavelet threshold de-noising methods are discussed and examined under a noisy environment to compare their performance with the new method proposed here. The method described here has been found to be superior to the other wavelet-based methods

A novel wavelet transform technique for on-line partial discharge measurements. 1. WT de-noising algorithm

The location of a partial discharge (PD) source inside a transformer can be determined from the time differences of arrival (TDOA) between signals that are captured by an array of UHF sensors. The TDOA can be acquired from the received PD waveforms. In this paper, three different methods of acquiring the TDOA from the PD waveforms are discussed. The time difference can be calculated either by taking the first peak of the signal as the arrival instant, or from the cross-correlation of the PD waveforms, or by applying the similarity function to the plots of the PD signals cumulative energy. Computation algorithms for determining the TDOA automatically are introduced so that possible bias from human interpretation is avoided. The presence of noise and its effect on the accuracy of the PD localization will also be presented. Experimental results show the first-peak method has higher accuracy than the two other methods. The application of signal denoising further improves the localization accuracy.

Partial discharge localization in transformers using UHF detection method

Integration of renewable energy including solar energy is growing faster than ever before. Solar energy supplies more than 1.3% of global power, and it is predicted to become the largest electricity source by 2050 with about 11% of global power consumption. However, the improper installation, non-frequently scheduled maintenance, and aging effect can accelerate the deterioration of PV system components, which directly increase the possibility of arc fault occurrence. The undetected arc faults pose a severe fire hazard to residential, commercial, and utility-scaled PV systems. To deliver electricity in a safe and reliable manner, such a dangerous event must be detected at early stage. This paper presents a comprehensive review of the-state-of-art techniques for DC arc faults detection in photovoltaic systems (PV). Different methods and the features used for detection are discussed and compared in detail. This paper also emphasizes the importance of DC arc fault simulation for characteristics study and fault diagnosis purpose. Several DC arc fault models have been reviewed and compared.

A comprehensive review on DC arc faults and their diagnosis methods in photovoltaic systems

After referring briefly to the earlier developments of computer-based PD measurements in the 1970s, the paper outlines how the introduction of new technology has resulted in the design and production of a number of digital systems for use in test laboratories and on site. In order to avoid overgeneralization, the basis of the hardware required for detection and recording of the PD pulses is described by reference to a particular system, it being understood that many detailed variations are possible depending on the application. The paper describes a number of parameters that can be calculated by software and used to characterize the PD behavior. These include the IEC-270 integrated quantities, statistical moments and other fingerprints for recognizing the PD patterns within the power frequency cycle. Examples are given of the patterns obtained on equipment and samples. These are necessarily selective and are intended to indicate how the techniques may be applied. It is concluded that these new methods will become valuable in procedures for assessing the condition of electrical insulation, in particular faults or defects, but must be used in conjunction with existing techniques until improved confidence is achieved in interpreting results, especially in respect of the various phase-resolved distributions and calibration procedures. >

Development of computer-based measurements and their application to PD pattern analysis

Dissolved gas-in-oil analysis (DGA) is a sensitive and reliable technique for the detection of incipient fault condition within oil-immersed transformers. The presence of certain key gases is monitored and quantified. There are a number of methods developed for analyzing these gases and interpreting their significance: Key Gas, Rogers Ratio, Doernenburg, Logarithmic Nomograph, IEC Ratio and Duval Triangle. This paper investigates the accuracy and consistency of these methods in interpreting the transformer condition. The evaluation is carried out on DGA data obtained from the local power utilities and from published papers. The data consists of 92 different cases. The key gases considered are hydrogen, methane, ethane, ethylene and acetylene. A MATLAB program was developed to automate the evaluation of the methods.

B.T. Phung

Papers

A novel wavelet transform technique for on-line partial discharge measurements. 1. WT de-noising algorithm

Partial discharge localization in transformers using UHF detection method

A comprehensive review on DC arc faults and their diagnosis methods in photovoltaic systems

Development of computer-based measurements and their application to PD pattern analysis

Comparative Study and Analysis of DGA Methods for Transformer Mineral Oil