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

This paper presents an intelligent model for the recognition of high-voltage direct current interference in geomagnetic observation stations. Firstly, it introduces the history and current status of geomagnetic observation in China, highlighting the issue of station interference from HVDC. Next, it discusses the application of traditional methods and deep learning techniques in the identification of geomagnetic data interference, along with related research. To address these issues, the paper emphasizes the proposed model framework, which includes four main components: data preprocessing, model training, interference recognition, and visualization. Data preprocessing is carried out to eliminate dimensional differences between data by using standardization and data augmentation techniques, increasing the diversity and robustness of training data. Model training involves the use of an LSTM network, which learns temporal patterns and relevant features in the input data, implicitly performing feature extraction and representation learning. In the interference recognition stage, the concept of anomaly scores is introduced, and the anomaly score for each data point is calculated using mean and covariance to determine if the point is an anomaly. Finally, the results of interference recognition are presented through visualization. In the experimental section, the paper conducts a comprehensive evaluation of four different models (LSTM, RNN&#x005F;TANH, RNN&#x005F;RELU, and GRU) when used as the training network for the proposed model. The evaluation focuses on three aspects: model performance, computational cost, and Friedman&#x2019;s test. The experimental results demonstrate that selecting LSTM as the training network with a time step of 3 achieves optimal performance in all three aspects, showcasing strong generalization capabilities.

Multi-step Ahead prediction for Anomaly Detection of Geomagnetic Observation in HVDC transmission