China
Africa
Explore chapters and articles related to this topic
Data-Mining Methods for Electricity Theft Detection
Published in Ahmed F. Zobaa, Trevor J. Bihl, Big Data Analytics in Future Power Systems, 2018
Trevor J. Bihl, Ahmed F. Zobaa
Detection involves finding a signal of interest in the presence of noise (Kelly, 1986). One variant of detection is anomaly detection which involves finding statistical anomalies, samples that are considerably different from the majority of samples (Duda, Hart, & Stork, 2012). For electricity theft detection, anomaly detection methods can be used to find load profiles which are statistically different from normal, and majority, profiles (McLaughlin, Holbert, Fawaz, Berthier, & Zonouz, 2013).
Control of the plausibility of measurements
Published in Lucien Wald, Fundamentals of Solar Radiation, 2021
Looking for anomalies implies that typical values or bounds have been established, or even a model or a pattern of values that are expected, since an anomaly is an unexpected, unexplained deviation from standard values. Consequently, the plausibility check consists first of all in establishing typical values and limits and then looking for anomalies with respect to these values and bounds.
A Review of Intrusion Detection and Prevention on Mobile Devices: The Last Decade
Published in Georgios Kambourakis, Asaf Shabtai, Constantinos Kolias, Dimitrios Damopoulos, Intrusion Detection and Prevention for Mobile Ecosystems, 2017
Weizhi Meng, Jianying Zhou, Lam-For Kwok
Anomaly-based detection is the process of comparing normal profiles against observed events to identify significant deviations. The profiles are developed by monitoring the characteristics of typical activity over a period of time. The system can use statistical methods to compare the characteristics of current activity to thresholds related to the profile.
A Review of Some Sampling and Aggregation Strategies for Basic Statistical Process Monitoring
Published in Journal of Quality Technology, 2021
Inez M. Zwetsloot, William H. Woodall
Jeske et al. (2018) provided a review of computer network surveillance methods. Generally the methods require temporal aggregation of the data or sampling at a specified frequency. As discussed by Androulidakis et al. (2006), many computer network anomaly detection techniques are based on modeling the dynamic statistical properties of typical traffic data in order to accurately and in a timely fashion detect network anomalies. Anomaly detection is based on the principle that changes from typical behavior might suggest the presence of anomalies, faults, or attacks. They pointed out that the problem of anomaly detection becomes more complex when network traffic data are sampled and compared the performance of three types of sampling plans.