English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Advanced metering infrastructure (AMI) has been widely used as an intelligentenergy consumption measurement system. Electric power was the representativeenergy source that can be collected by AMI; most existing studies to detectabnormal energy consumption have focused on a single energy source, i.e.,power. Recently, other energy sources such as water, gas, and heating have alsobeen actively collected. As a result, it is necessary to develop a unifiedmethodology for anomaly detection across multiple energy sources; however,research efforts have rarely been made to tackle this issue. The inherentdifficulty with this issue stems from the fact that anomalies are not usuallyannotated. Moreover, existing works of anomaly definition depend on onlyindividual energy sources. In this paper, we first propose a method fordefining anomalies considering not only individual energy sources but alsocorrelations between them. Then, we propose a new Correlation-drivenMulti-Level Multimodal Learning model for anomaly detection on multiple energysources. The distinguishing property of the model incorporates multiple energysources in multi-levels based on the strengths of the correlations betweenthem. Furthermore, we generalize the proposed model in order to integratearbitrary new energy sources with further performance improvement, consideringnot only correlated but also non-correlated sources. Through extensiveexperiments on real-world datasets consisting of three to five energy sources,we demonstrate that the proposed model clearly outperforms the existingmultimodal learning and recent time-series anomaly detection models, and weobserve that our model makes further the performance improvement as morecorrelated or non-correlated energy sources are integrated.

Correlation-Driven Multi-Level Multimodal Learning for Anomaly Detection on Multiple Energy Sources