Zhiyang Jia

Prof. Zhiyang Jia


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 June 2024 in Measurement
Verified
Journal Article
Closed access
ZXZhao Xu
Zhao Xu
ZJZhiyang Jia
Zhiyang Jia
YWYiWei Wei
YiWei Wei
SZShuyan Zhang
Shuyan Zhang
ZJZhong Jin
Zhong Jin
WDWenpei Dong
Wenpei Dong

21 October 2022
Verified
Conference Paper
Open access
ZJZhiyang Jia
Zhiyang Jia
JWJihe Wang
Jihe Wang
CDCheng Deng
Cheng Deng
The Industrial Internet of Things (IIoT) promises to provide an expanded awareness of field assets and equipment, access to data from across locations, and actionable insights for maximizing operational performance and safety of the oil and gas industry. Using automation and machine learning, with the application of predictive maintenance, efficiencies can be boosted and problems can be mitigated sooner and more effectively. The proposed system is mainly based on the data collection, processing, analysis, and modeling of an enormous number of historical and real-time data generated during the operation of the equipment on the edge side. The data-driven predictive maintenance used machine learning models and deep learning models to predict the remaining useful life (RUL). Bi-LSTM based prediction model has been trained on the cloud, and deployed onto the edge devices. The predictive maintenance process includes data acquisition, data processing, training of machine learning model, equipment health assessment, remaining useful life prediction, strategy formulation, and strategy execution. The predictive maintenance solution driven by the IIoT helps oil and gas companies make predictions before equipment failures have a significant impact on their company's safety level and profits to improve asset reliability and promote cost savings.

16 January 2025 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
YWYiwei Wei
Yiwei Wei
HZHengyang Zhou
Hengyang Zhou
SYShaozu Yuan
Shaozu Yuan
MCMeng Chen
Meng Chen
HSHaitao Shi
Haitao Shi
XHXiaodong He
Xiaodong He
Multimodal sarcasm detection (MSD) requires predicting the sarcastic sentiment by understanding diverse modalities of data (e.g., text, image). Beyond the surface-level information conveyed in the post data, understanding the underlying deep-level knowledge–such as the background and intent behind the data–is crucial for understanding the sarcastic sentiment. However, previous works have often overlooked this aspect, limiting their potential to achieve superior performance. To tackle this challenge, we propose DeepMSD, a novel framework that generates supplemental deep-level knowledge to enhance the understanding of sarcastic content. Specifically, we first devise a Deep-level Knowledge Extraction Module that leverages large vision-language models to generate deep-level information behind the text-image pairs. Additionally, we devise a Cross-knowledge Graph Reasoning Module to model how humans use prior knowledge to identify sarcastic cues in multimodal posts. This module constructs cross-knowledge graphs that connect deep-level knowledge with surface-level knowledge. As such, it enables a more profound exploration of the cues underlying sarcasm. Experiments on the public MSD dataset demonstrate that our approach significantly surpasses previous state-of-the-art methods.

1 January 2024 in IET Computers & Digital Techniques
Verified
Journal Article
Open access
DZDan Zhang
Dan Zhang
ZJZhiyang Jia
Zhiyang Jia
In the Internet era, the e‐commerce industry has risen, its development scale continues to expand, cross‐border e‐commerce (CBEC) has also been born, and it is now in the stage of sustainable development. The rapid development of CBEC also needs the strong support of logistics, the two are inseparable, and today, the development scale of CBEC is constantly expanding. The existing e‐commerce logistics (ECL) model is also gradually unable to meet the increasingly diverse needs of users, and new logistics models need to be actively explored. To change this situation, this paper carried out a specific analysis of CBEC logistics model, and applied embedded technology to ECL, which also built a logistics tracking system. At the same time, combined with the ant colony algorithm, the paper carried out experimental research on the logistics package distribution route planning problem. From the experimental results, in terms of average delivery time, the algorithm’s result was 25.95 hr, while the traditional algorithm was 32.53 hr; in terms of average distribution freight cost, the algorithm’s result was 163.3 yuan, while the traditional algorithm was 257.7 yuan; in terms of average distribution cost, this algorithm’s result was 131.53 yuan, while the traditional algorithm was 211.68 yuan. To sum up, this algorithm could effectively optimize the distribution route of logistics packages and improve the efficiency of package transportation.

26 November 2024 in IEEE Sensors Journal
Verified
Journal Article
Closed access
ZXZhao Xu
Zhao Xu
ZJZhiyang Jia
Zhiyang Jia
YWYiwei Wei
Yiwei Wei
ZGZengwei Gao
Zengwei Gao
JSJiahang Sun
Jiahang Sun
JLJiakang Liu
Jiakang Liu
Recently, multi-sensor fault diagnosis (MSFD) has become a research hotspot in the field of large machine maintenance. However, current research rarely explores in depth the spatial-temporal structural features unique to sensor signals. To address this gap, this paper proposes the ARM-STDS, a novel multi-sensor fault diagnosis model with adaptive spatial-temporal dual-scale remodeling. Sensor signals exhibit sparsity and periodicity in the temporal domain, which poses challenges for efficient processing by time-series models. To tackle this challenge, we have designed an adaptive subseries-level patches segmentation method that enhances the information capacity of short time slices in the signals. This method effectively extracts local semantic information while reducing computational complexity quadratically. Additionally, we introduce a directed graph structural modeling algorithm to better capture the dependencies between sensors. Utilizing directed graphs proves more effective in representing the coupling relationships between sensors compared to traditional undirected graphs. In comparison to other state-of-the-art MSFD models, ARM-STDS demonstrates the most effective diagnostic performance on four publicly available datasets. In the constant system, the average accuracy increased by 3-5%. Notably, in the non-constant system, ARM-STDS exhibited an accuracy improvement of over 5%.The code is available at https://github.com/XZHBUT/ARM-STDS.

25 August 2023
Verified
Conference Paper
Open access
ZXZhao Xu
Zhao Xu
ZJZhiyang Jia
Zhiyang Jia
JSJiahang Sun
Jiahang Sun
WDWenpei Dong
Wenpei Dong
ZLZhang Li
Zhang Li
Artificial intelligence techniques such as machine learning and deep learning are prevalent in the field of medical imaging, especially image segmentation. In the field of medicine, image segmentation helps to diagnose diseases accurately and hence has sparked interest. Neural-driven image segmentation methods, especially U-Net, are prominent.U-Net, although effective, suffers from severe information loss during downsampling and insufficiently detailed texture of the segmentation results, which limits its application. To solve this problem, we introduce DO-UNet, which strengthens the structure of U-Net. DO-UNet employs jump links at the downsampling layer, preserves high-resolution features, and suppresses data loss. Dense Block with Bottleneck (DBB) replaces the 3*3 convolutional block to efficiently extract spinal features and improve performance. Octave convolution extends the model during upsampling, improving accuracy by expanding the feature space and minimizing parameters. Experiments on the Verse2019 and Verse2020 datasets show that DO-UNet outperforms other models in terms of dice coefficients and accuracy. Visualization results confirm DO-UNet's superiority in accurately segmenting spinal structures. The model excels in capturing details and shows robustness and interpretability. These results highlight the practical significance and promising future of DO-UNet in spine segmentation.