Load profiling and data mining techniques in electricity deregulated market

Nizar, A.H.; Dong, Zhaoyang; Zhao, Junhua

doi:10.1109/pes.2006.1709335

Cited by 44 publications

(17 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Hierarchical clustering algorithms, like the family of agglomerative algorithms and the COBWEB (Tsekouras et al, 2007;Chicco et al, 2004;Bidaki et al, 2010;Gerbec et al, 2002aGerbec et al, , 2002bGerbec et al, , 2003Nizar et al, 2006;Hino et al, 2013;Notaristefano et al, 2013;Kwac et al, 2014).…”

Section: Literature Survey and Contributionsmentioning

confidence: 99%

See 1 more Smart Citation

Evaluation of the performance of clustering algorithms for a high voltage industrial consumer

Panapakidis

Alexiadis

Papagiannis

2015

Engineering Applications of Artificial Intelligence

View full text Add to dashboard Cite

Section: Literature Survey and Contributionsmentioning

confidence: 99%

“…Algorithms that do not belong to the above categories, like the Expectation Maximization (EM), Support Vector Clustering (SVC), Renyi entropy (Nizar et al, 2006;Chicco and Akilimali, 2009;Chicco and Ilie, 2009), Ant Colony Clustering and a set of Subspace and Projection Clustering Methods (Piao et al, 2014).…”

Section: Literature Survey and Contributionsmentioning

confidence: 99%

Evaluation of the performance of clustering algorithms for a high voltage industrial consumer

Panapakidis

Alexiadis

Papagiannis

2015

Engineering Applications of Artificial Intelligence

View full text Add to dashboard Cite

“…This would be of particular relevance to the Australian context, and would be a likely occurrence in the evening, when demand is at a maximum [18], and solar irradiation is low.…”

Section: ) Further Workmentioning

confidence: 99%

A Sensitivity Analysis Toolkit for the Simplification of MV Distribution Network Voltage Management

Tamp

Ciufo

2014

IEEE Trans. Smart Grid

View full text Add to dashboard Cite

As distribution networks become increasingly variable and generation becomes increasingly decentralised, voltage standards are becoming more liable to violation. Accordingly, intelligent voltage management strategies are required to enable standards compliance without unduly increasing network maintenance or infrastructure costs. The relationship between network power and voltages, however, is complex, non-linear and interdependent, and thus is difficult to conceptualise and use for decision-making and control purposes. This paper introduces a software toolkit that simplifies the development of voltage management strategies by the application of sensitivity analysis. Sensitivity analysis reduces complex network relationships to simple linear equations, and thus enables easy and comprehensive conceptualisation of the effect of network modifications. Sensitivity data is obtained through a unique 'perturb-and-observe' algorithm built on top of an open-source simulation package. The toolkit enables the rapid development of sensitivity-driven network experimentation and is highly extensible. A number of applications that demonstrate the usefulness of these techniques are presented, including the development of a reactive power control algorithm for the mitigation of inverter-based, distributed generation (DG)-induced voltage rise, and the verification of voltage-supportcapacitor-bank placement for a real Australian semi-rural network. Finally, this paper presents some suggestions for possible future applications of data-driven, simulator-augmented sensitivity analysis techniques. Abstract-As distribution networks become increasingly variable and generation becomes increasingly decentralised, voltage standards are becoming more liable to violation. Accordingly, intelligent voltage management strategies are required to enable standards compliance without unduly increasing network maintenance or infrastructure costs. The relationship between network power and voltages, however, is complex, non-linear and interdependent, and thus is difficult to conceptualise and use for decision-making and control purposes. This paper introduces a software toolkit that simplifies the development of voltage management strategies by the application of sensitivity analysis. Sensitivity analysis reduces complex network P -Q -|V | relationships to simple linear equations, and thus enables easy and comprehensive conceptualisation of the effect of network modifications. Sensitivity data is obtained through a unique 'perturb-and-observe' algorithm built on top of an open-source simulation package. The toolkit enables the rapid development of sensitivity-driven network experimentation and is highly extensible.A number of applications that demonstrate the usefulness of these techniques are presented, including the development of a reactive power control algorithm for the mitigation of inverterbased, distributed generation (DG)-induced voltage rise, and the verification of voltage-support-capacitor-bank placement for a real Australian semi...

show abstract

“…FS is normally used together with classification techniques, encompassing the schemes used to find which variables are the most useful to identify the presence of NTLs [48,54]. The reduction of the number of inputs used in classification can improve the performance and ease interpretation.…”

mentioning

confidence: 99%

“…The techniques can be used to divide the classification problem in a set of multiple easier problems or to directly identify a group of customers 525 presenting irregular behavior, which may be a strong indicator of NTLs [89]. Finding load profiles representing normal consumption patterns is used compare to new consumption curves, uncovering outliers that can indicate the presence of illicit behavior or faults in metering [48].…”

mentioning

confidence: 99%

Solutions for detection of non-technical losses in the electricity grid: A review

Viegas

Esteves

Melício

et al. 2017

Renewable and Sustainable Energy Reviews

144

View full text Add to dashboard Cite

This paper is a review of literature with an analysis on a selection of scientific studies for detection of non-technical losses. Non-technical losses occurring in the electric grid at level of transmission or of distribution have negative impact on economies, affecting utilities, paying consumers and states. The paper is concerned with the lines of research pursued, the main techniques used and the limitations on current solutions. Also, a typology for the categorization of solutions for detection of non-technical losses is proposed and the sources and possible attack/vulnerability points are identified. The selected literature covers a wide range of solutions associated with non-technical losses. Of the 103 selected studies, 6 are theoretical, 25 propose hardware solutions and 72 propose non-hardware solutions. Data based classification models and data from consumption with high resolution are respectively required in about 47% and 35% of the reported solutions. Available solutions cover a wide range of cases, with the main limitation found being the lack of an unified solution, which enables the detection of all kinds of non-technical losses.

show abstract

Load profiling and data mining techniques in electricity deregulated market

Cited by 44 publications

References 21 publications

Evaluation of the performance of clustering algorithms for a high voltage industrial consumer

Evaluation of the performance of clustering algorithms for a high voltage industrial consumer

A Sensitivity Analysis Toolkit for the Simplification of MV Distribution Network Voltage Management

Solutions for detection of non-technical losses in the electricity grid: A review

Contact Info

Product

Resources

About