There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

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

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Data Mining - Concepts and Techniques.

Data Mining Practical Machine Learning Tools and Techniques

This paper presents a new robust method for probabilistically estimating the fault location during a voltage sag using a set of arbitrarily accurate selection of monitors located throughout a distribution network The method uses statistical analysis to solve the fault location equations and find the most likely location for the fault location The method draws on and advances existing impedance based fault location algorithms The presented method is practical, using erroneous measurement data which is both available now, and expected to be available in future smart distribution networks

Probabilistic fault location using erroneous measurement devices

We address the problem of predicting the transient stability status of a power system as quickly as possible in real time subject to probabilistic risk constraints. The goal is to minimise the average time taken after a fault to make the prediction, and the method is based on ideas from statistical sequential analysis. The proposed approach combines probabilistic neural networks with dynamic programming. Simulation results show an approximately three-fold increase in prediction speed when compared to the use of pre-committed (fixed) prediction times.

/pdf/application-of-sequential-testing-problem-to-online-2fbjzq2w7n.pdf

Application of sequential testing problem to online detection of transient stability status for power systems

The paper presents a method for the detection of areas affected by voltage unbalance in distribution networks and presents the results graphically by “Heat Map”. Voltage unbalance causes overheating of equipment, increased losses and general reduction in efficiency of power system equipment and user devices. This results in additional economic costs for both Distribution Network Operators (DNOs) and consumers. This paper uses three-phase full Newton-Raphson load flow method and Monte Carlo simulations to probabilistically estimate the voltage unbalance level at all buses in the network under unbalanced loading conditions. The usage of “Heat map” graphical representation of the results facilitates easy identification of the areas of the network that are affected the most by unbalance. Proposed methodology is demonstrated on a real 24 bus network with actual network parameters. (5 pages)

/pdf/probabilistic-estimation-of-voltage-unbalance-in-1uhffzt7nx.pdf

Probabilistic estimation of voltage unbalance in distribution networks with asymmetrical loads

The paper investigates the risk of subsynchronous resonance (SSR) in meshed power networks with compensated AC lines operating in parallel to voltage source converter (VSC) HVDC lines. Different compensation levels of AC line, different power transfer through VSC HVDC line and various contingencies are considered to assess the risk of mechanical torques amplification in turbine shaft due to SSR. Both, active and passive compensation of AC lines are considered.

Study of subsynchronous resonance in meshed compensated AC/DC network

This paper investigates the effect of demand side management (DSM) program in distribution network on network losses and loading margin. The composite load model comprising constant impedance, constant current, constant power and induction motor loads is used to model the loads in the network. This enables a more accurate analysis of the changes in load flow and network loadability limit after applying a DSM program across all the buses in a distribution network with distributed generation. The applied DSM programme aims at maintaining as close to self-sufficient operation of the network as possible, by adjusting flexible demand in the network to the volatile distributed generation, and minimizing its reliance on the upstream network. The analysis shows how shifting different load components (namely constant impedance load or induction motors), and reconnecting them at different times following a DSM action (the so called load payback) affect the network performance. The network used in this paper is a modified IEEE 33 bus distribution network

Jovica V. Milanovic

Papers

Probabilistic fault location using erroneous measurement devices

Application of sequential testing problem to online detection of transient stability status for power systems

Probabilistic estimation of voltage unbalance in distribution networks with asymmetrical loads

Study of subsynchronous resonance in meshed compensated AC/DC network

The Effect of Load-follow-generation Motivated DSM Programme on Losses and Loadability of a Distribution Network with Renewable Generation