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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Building energy prediction and management has become increasingly importantin recent decades, driven by the growth of Internet of Things (IoT) devices andthe availability of more energy data. However, energy data is often collectedfrom multiple sources and can be incomplete or inconsistent, which can hinderaccurate predictions and management of energy systems and limit the usefulnessof the data for decision-making and research. To address this issue, paststudies have focused on imputing missing gaps in energy data, including randomand continuous gaps. One of the main challenges in this area is the lack ofvalidation on a benchmark dataset with various building and meter types, makingit difficult to accurately evaluate the performance of different imputationmethods. Another challenge is the lack of application of state-of-the-artimputation methods for missing gaps in energy data. Contemporaryimage-inpainting methods, such as Partial Convolution (PConv), have been widelyused in the computer vision domain and have demonstrated their effectiveness indealing with complex missing patterns. To study whether energy data imputationcan benefit from the image-based deep learning method, this study comparedPConv, Convolutional neural networks (CNNs), and weekly persistence methodusing one of the biggest publicly available whole building energy datasets,consisting of 1479 power meters worldwide, as the benchmark. The results showthat, compared to the CNN with the raw time series (1D-CNN) and the weeklypersistence method, neural network models with reshaped energy data with twodimensions reduced the Mean Squared Error (MSE) by 10% to 30%. The advanceddeep learning method, Partial convolution (PConv), has further reduced the MSEby 20-30% than 2D-CNN and stands out among all models.

Filling time-series gaps using image techniques: Multidimensional context autoencoder approach for building energy data imputation