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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Forecasting natural gas consumption, considering seasonality and trends, iscrucial in planning its supply and consumption and optimizing the cost ofobtaining it, mainly by industrial entities. However, in times of threats toits supply, it is also a critical element that guarantees the supply of thisraw material to meet individual consumers' needs, ensuring society's energysecurity. This article introduces a novel multistep ahead forecasting ofnatural gas consumption with change point detection integration for modelcollection selection with continual learning capabilities using data streamprocessing. The performance of the forecasting models based on the proposedapproach is evaluated in a complex real-world use case of natural gasconsumption forecasting. We employed Hoeffding tree predictors as forecastingmodels and the Pruned Exact Linear Time (PELT) algorithm for the change pointdetection procedure. The change point detection integration enables selecting adifferent model collection for successive time frames. Thus, three modelcollection selection procedures (with and without an error feedback loop) aredefined and evaluated for forecasting scenarios with various densities ofdetected change points. These models were compared with change point agnosticbaseline approaches. Our experiments show that fewer change points result in alower forecasting error regardless of the model collection selection procedureemployed. Also, simpler model collection selection procedures omittingforecasting error feedback leads to more robust forecasting models suitable forcontinual learning tasks.

A Natural Gas Consumption Forecasting System for Continual Learning Scenarios based on Hoeffding Trees with Change Point Detection Mechanism