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

The renewable energy source based generating technologies and flexible demand and storage devices exhibit significant temporal and spatial uncertainties in generating and loading profiles and introduce additional level of uncertainty in network operation. The dynamic behaviors of such a network can be affected and the stable operation may be compromised. This paper proposes a probabilistic analysis approach for the evaluation of the effect of uncertain parameters on power system voltage and angular stability. Load margin, the damping of critical eigenvalues and the transient stability index have been chosen as the relevant stability indices for voltage stability, small-disturbance stability, and transient stability analysis, respectively. The Morris screening sensitivity analysis method coupled with a multivariate Gaussian copula to account for parameter correlations is used for the priority ranking of uncertain parameters. The approach is illustrated on a number of case studies using modified IEEE 68-bus NETS-NYPS test system. The results obtained in this paper reveal that the critical parameters appear as groups if the input dataset is correlated, and, hence even a parameter (which may be uninfluential individually) can have a significant impact on system dynamic behavior due to its correlation with other influential parameters.

/pdf/identification-of-critical-parameters-affecting-voltage-and-1fsnns2dff.pdf

Identification of Critical Parameters Affecting Voltage and Angular Stability Considering Load-Renewable Generation Correlations

This paper presents an approach to determine strategic buses for harmonic filter placement by correlating the sensitivity index based on the inherent structure theory network (ISTN) and harmonic current injection at buses of the distribution network/feeder. The strategic buses for the harmonic filter are characterized as either the bus with maximum total harmonic voltage distortion (THDV) in a distribution feeder, or the bus where the placement of the harmonic filter would result in maximum reduction in the overall average THDV in the distribution feeder. The advantage of using this approach is that it provides insight into the interaction of the network structure with harmonic current sources in the distribution network in regards to the performance of harmonic filters. The assessment of the performance of a strategically placed passive harmonic filter based on the proposed approach is evaluated for peak and offpeak load scenarios on a radial distribution network. The results indicate that passive harmonic filters that are placed based on harmonic current emission levels perform better than harmonic filters placed based on ISTN alone. The strategically placed passive harmonic filters are highly effective in controlling harmonic propagation in distribution networks

Planning Approaches for the Strategic Placement of Passive Harmonic Filters in Radial Distribution Networks

This paper describes a methodology to estimate post process failure cost of downtime (COD). COD can be estimated using either historical or direct online costs (direct and restart costs) as input to the model. Application of the proposed COD estimation model is illustrated on an actual industrial process that experienced several downtime events due to voltage sags in its electric supply. Raw financial data obtained from industrial collaborators are normalized and used as an input to developed software for COD estimation. Several case studies, including influence of simultaneous manufacturing of different product variants and optimization (minimization) of COD are presented and discussed in detail.

/pdf/estimation-of-cost-of-downtime-of-industrial-process-due-to-58p3j16j7y.pdf

Estimation of Cost of Downtime of Industrial Process Due to Voltage Sags

In this paper a distributed control strategy for coordinating multiple battery energy storage systems to support frequency regulation in power systems with high penetration of renewable generation is proposed. The approach is based on an online convex optimisation framework that considers both the operating costs of storage systems and the frequency regulation requirements. The proposed framework uses a dynamic regret based on the last available information up to the current point in time of renewable generation and demand, which are uncertainty sources. The approach provides more accurate control than the currently employed prediction-based algorithms and it can be implemented in a distributed manner using a multi-agent system framework that facilitates plug-and-play functionality. The effectiveness and scalability of the proposed strategy is assessed through several case studies.

/pdf/distributed-control-of-battery-energy-storage-systems-for-3q5itme0tu.pdf

Distributed Control of Battery Energy Storage Systems for Improved Frequency Regulation

The authors discuss an important, but often neglected voltage sag characteristic: the phase shift. Firstly, a critical review of existing methodologies and approaches is presented. Then, theoretical and experimental results on the ranges and relationships of phase shift values are reviewed. A numerical example and several recorded 'non-conventional' voltage sag waveforms are used to highlight some of the problems encountered in calculating a phase shift. Equipment and procedures to test single-phase and three-phase equipment sensitivity to voltage sags are also discussed from a phase shift point of view.

Jovica V. Milanovic

Papers

Identification of Critical Parameters Affecting Voltage and Angular Stability Considering Load-Renewable Generation Correlations

Planning Approaches for the Strategic Placement of Passive Harmonic Filters in Radial Distribution Networks

Estimation of Cost of Downtime of Industrial Process Due to Voltage Sags

Distributed Control of Battery Energy Storage Systems for Improved Frequency Regulation

Advanced voltage sag characterisation. Part I: Phase shift