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

Power Quality is one of the critical issues when operating contemporary distribution networks. The increased interest in PQ is due to the increased employment of sensitive equipment loads and renewable generation technologies. Currently, the evaluation of PQ network performance is based on evaluating different phenomena separately, and a standardised way to evaluate the PQ as a whole for a bus or a network is yet to be applied or widely accepted. The paper presents a methodology for combining the performances of different PQ phenomena and expressing the PQ performance of a bus using a single index. The proposed methodology adopts an Analytic Hierarchy Process (AHP) model to combine the harmonics, unbalance and voltage sag performances in one proposed index; i.e. Compound Bus PQ Index (CBPQI). The separate and cumulative PQ performances of a 295-bus generic distribution network were evaluated, compared and demonstrated using heat maps.

/pdf/global-assessment-of-power-quality-performance-of-networks-1tr8qa3yci.pdf

Global assessment of power quality performance of networks using the analytic hierarchy process model

This paper analyses the operation and interaction of a STATCOM and a VSC HVDC link located in close electrical proximity in an AC system, taking into consideration of various controller configurations and operating conditions for both components. It is identified that the dynamic behaviour of one component can be substantially affected by a parametric event in the control of the other component whereas little interaction is shown between the two components for an AC system short circuit event. Results also indicate that the operating point of the DC link can deteriorate the transient performance of both the STATCOM and VSC. The study is performed based on a multi-machine dynamic GB system developed from a real reference load flow case to achieve higher fidelity.

Interactions between STATCOM and VSC HVDC in dynamic GB system

CIGRE WG C4.065 Recommendations on Measurement Based and Component Based Load Modelling Practice

This paper proposes a methodology for demand profiling, namely load decomposition, of aggregated residential load based on smart meter (SM) data. The methodology is applicable to both active and reactive load, following an assumption that SMs can monitor real-time active power consumption of individual appliances. Only a number of households in the aggregation are equipped with SMs in this study. The non-monitored users' load is decomposed using artificial neural network (ANN) trained with the available SM data. Information about load composition, in terms of load categories or load controllability, can be highly beneficial for various demand response (DR) applications. Different levels of SM coverage are considered in the study to illustrate the effect of the level of SM coverage on the accuracy of total aggregated load decomposition. The results show that the consumption of some load categories can be estimated with high confidence, even at lower levels of SM coverage.

/pdf/application-of-data-analytics-for-advanced-demand-profiling-3e9xsrusq8.pdf

Application of data analytics for advanced demand profiling of residential load using smart meter data

In modern power systems, the penetration of deregulated market structures, together with the integration of renewable energy source-based generations and non-conventional loads can exhibit inherent stochastic and intermittent behaviour. Hence, bringing uncertainties to the system generation and loading profile, affects the power system's dynamic behaviour. The identification and ranking of critical system uncertain parameters is important for the efficient operation of modern power systems since they can enable better system management with less monitoring. In this paper, 6 sensitivity analysis (SA) methods have been employed for the priority ranking of uncertain parameters in a network with renewable generations from the perspective of their influence on power system voltage stability. The performances of the 6 SA methods are evaluated and their advantages and disadvantages are discussed. The modified version of the 68 bus NETS-NYPS has been used as the test system.

/pdf/identification-of-critical-parameters-affecting-voltage-1m1bxxrlkp.pdf

Jovica V. Milanovic

Papers

Global assessment of power quality performance of networks using the analytic hierarchy process model

Interactions between STATCOM and VSC HVDC in dynamic GB system

CIGRE WG C4.065 Recommendations on Measurement Based and Component Based Load Modelling Practice

Application of data analytics for advanced demand profiling of residential load using smart meter data

Identification of critical parameters affecting voltage stability in networks with renewable generations using sensitivity analysis methods