ZZZhiqiang Zhou

Zhiqiang Zhou

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Zhiqiang Zhou has a total of 56 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    GraphSubDetector: Time Series Subsequence Anomaly Detection via Density-Aware Adaptive Graph Neural Network

    by Weiqi Chen, Zhiqiang Zhou, Qingsong Wen, Liang Sun

    Time series subsequence anomaly detection is an important task in a large variety of real-world applications ranging from health monitoring to AIOps, and is challenging due to the following reasons: 1) how to effectively learn complex dynamics and dependencies in time series; 2) diverse and complicated anomalous subsequences as well as the inherent variance and noise of normal patterns; 3) how to determine the proper subsequence length for effective detection, which is a required parameter for many existing algorithms. In this paper, we present a novel approach to subsequence anomaly detection, namely GraphSubDetector. First, it adaptively learns the appropriate subsequence length with a length selection mechanism that highlights the characteristics of both normal and anomalous patterns. Second, we propose a density-aware adaptive graph neural network (DAGNN), which can generate further robust representations against variance of normal data for anomaly detection by message passing between subsequences. The experimental results demonstrate the effectiveness of the proposed algorithm, which achieves superior performance on multiple time series anomaly benchmark datasets compared to state-of-the-art algorithms.

  • Functionalized Double‐Gradient Janus Filters for Full‐Scale Capture and Inactivation of Multiscale Aqueous Bioaerosols

    by Zhiqiang Zhou, Di Wang, Tianle You, Zhengyuan Pan, Guilong Xu, Yun Liang, Min Tang

    Exposure to bioaerosols can result in severe infection and even death, prompting the development of numerous air filters to capture and inactivate bioaerosols. Although the wettability of filter media determines its bioaerosol‐capture efficiency, the dynamic collision behavior of multiscale aqueous bioaerosols on most filter media is poorly understood. A double‐gradient Janus paper (Janus paper) with a pore‐size gradient, asymmetric wettability, and biocidal and bioadhesion properties is fabricated in this study using an industrially available papermaking process. The dynamic collision behavior at the micrometer and millimeter scales indicates that the Janus paper enables the full‐scale blocking and/or capture of aqueous aerosols by a two‐step mechanism where the water‐repellent top layer blocks large‐size aqueous aerosols, while the hydrophilic bottom layer permits the adherence of small‐size aerosols to its fibers. The Janus paper exhibits a higher capture and inactivation capacity (inactivation efficiency: 98.3%), better dust‐holding capacity (> 21 g m−2), and a longer lifetime than commercial filter media. Commercial air purifiers comprising the Janus paper exhibit high potential for large‐scale commercial applications owing to the facile, low‐cost, and scalable fabrication method. This study enables future research on efficient and practicable Janus‐paper filter media.


  • Trichome‐Like Biomimetic Air Filters via Templated Silicone Nanofilaments

    by Zhiqiang Zhou, Tianle You, Zhengyuan Pan, Di Wang, Hao Wang, Lingyun Wang, Guilong Xu, Yun Liang, Jian Hu, Min Tang

    Air pollution threats to human health have increased awareness of the role of filter units in air cleaning applications. As an ideal energy‐saving strategy for air filters, the slip effect on nanofiber surfaces can potentially overcome the trade‐off between filtration efficiency and pressure drop. However, the potential of the slip effect in nanofibrous structures is significantly limited by the tight nanofiber stacks. In this study, trichome‐like biomimetic (TLB) air filters with 3D‐templated silicone nanofilaments (average diameter: ≈74 nm) are prepared based on an in situ chemical vapor deposition (CVD) method inspired by plant purification. Theoretical modeling and experimental results indicate that TLB air filters make significant use of the slip effect to overcome the efficiency‐resistance tradeoff. The selectable filter class (up to U15, ≈99.9995%) allows TLB air filters to meet various requirements, and their integral filtration performance surpasses that of most commodity air filters, including melt‐blown cloth, ePTFE membranes, electrospun mats, and glass fiber paper. The proposed strategy directly transforms commercial filter media and filters into TLB air filters using a bottom‐up, one‐step approach. As a proof‐of‐concept, reusable N95 respirators and air purifiers equipped with TLB air filters are fabricated, overcoming the limitations of existing filter designs and fabrication methods.


  • Bioinspired Structures Made of Silicone Nanofilaments for Upcycling Waste Masks to Reusable N95 Respirators

    by Zhiqiang Zhou, Di Wang, Zhengyuan Pan, Tianle You, Guilong Xu, Yun Liang, Min Tang

    The increasing demand for personal protective equipment such as single-use masks has led to large amounts of nondegradable plastic waste, aggravating economic and environmental burdens. This study reports a simple and scalable approach for upcycling waste masks via a chemical vapor deposition technique, realizing a trichome-like biomimetic (TLB) N95 respirator with superhydrophobicity (water contact angle >= 150 degrees), N95-level protection, and reusability. The TLB N95 respirator comprising templated silicone nanofilaments with an average diameter of similar to 150 nm offers N95-level protection and breathability comparable to those of commercial N95 respirators. The TLB N95 respirator can still maintain its N95-level protection against particulate matter and viruses after 10 disinfection treatment cycles (i.e., ultraviolet irradiation, microwave irradiation, dry heating, and autoclaving), demonstrating durable reusability. The proposed strategy provides new insight into upcycle waste masks, breaking the existing design and preparation concept of reusable masks.


  • Conformal Build‐Up of Functionalized Air Filters with Improved Air Cleaning and Bioprotective Traps

    by Zhiqiang Zhou, Tianle You, Di Wang, Zhengyuan Pan, Guilong Xu, Yun Liang, Min Tang

    Air filtration is vital for passive pollution control toward advanced manufacturing industries and human health protection. However, large‐scale air‐filter utilization is limited by high energy costs. Therefore, it is crucial to develop filter media that exhibit a high removal efficiency, low pressure drop, and stable biocidal effects. This study reports an efficient and scalable bottom‐up approach for fabricating shellac‐based glass fiber matrices with bioprotective traps (GFMs‐traps) with tunable conformal micro/nanostructures, superhydrophobicity, and biocidal functions. The synthesis, which involves freeze‐drying‐free ice‐shellac double‐template formation, perfluorosilane‐based grafting, and filtration, eliminates capillary‐force‐induced micro/nanofibrous cluster formation. Theoretical modeling and experimental results indicate that the air cleaning and biocidal properties of the GFMs‐traps can be attributed to their uniformly distributed micro/nanofibrous structures and stable nanoparticle (NP) coatings (AgNPs, CuNPs, and TiO2NPs). The GFMs‐traps exhibit integrated air cleaning, with excellent filtration performance at the most penetrating particle size (99.97%, 234 Pa), superhydrophobicity (≥150°), a high virus‐filtration efficiency (99.995%), and good antibacterial and antiviral activity. Additionally, a commercial air purifier comprising a GFMs‐trap filter exhibits high potential for commercial applications. The facile strategy for fabricating high‐performance bioprotective filter media reported here may facilitate advanced large‐scale air filtration applications in the future.


  • Open Access

    AHPA: Adaptive Horizontal Pod Autoscaling Systems on Alibaba Cloud Container Service for Kubernetes

    by Zhiqiang Zhou, Chaoli Zhang, Lingna Ma, Jing Gu, Huajie Qian, Qingsong Wen, Liang Sun, Peng Li, Zhimin Tang

    The existing resource allocation policy for application instances in Kubernetes cannot dynamically adjust according to the requirement of business, which would cause an enormous waste of resources during fluctuations. Moreover, the emergence of new cloud services puts higher resource management requirements. This paper discusses horizontal POD resources management in Alibaba Cloud Container Services with a newly deployed AI algorithm framework named AHPA - the adaptive horizontal pod auto-scaling system. Based on a robust decomposition forecasting algorithm and performance training model, AHPA offers an optimal pod number adjustment plan that could reduce POD resources and maintain business stability. Since being deployed in April 2021, this system has expanded to multiple customer scenarios, including logistics, social networks, AI audio and video, e-commerce, etc. Compared with the previous algorithms, AHPA solves the elastic lag problem, increasing CPU usage by 10% and reducing resource cost by more than 20%. In addition, AHPA can automatically perform flexible planning according to the predicted business volume without manual intervention, significantly saving operation and maintenance costs.


  • Open Access

    AHPA: Adaptive Horizontal Pod Autoscaling Systems on Alibaba Cloud Container Service for Kubernetes

    by Zhiqiang Zhou, Chaoli Zhang, Lingna Ma, Jing Gu, Huajie Qian, Qingsong Wen, Liang Sun, Peng Li, Zhimin Tang

    The existing resource allocation policy for application instances in Kubernetes cannot dynamically adjust according to the requirement of business, which would cause an enormous waste of resources during fluctuations. Moreover, the emergence of new cloud services puts higher resource management requirements. This paper discusses horizontal POD resources management in Alibaba Cloud Container Services with a newly deployed AI algorithm framework named AHPA -- the adaptive horizontal pod auto-scaling system. Based on a robust decomposition forecasting algorithm and performance training model, AHPA offers an optimal pod number adjustment plan that could reduce POD resources and maintain business stability. Since being deployed in April 2021, this system has expanded to multiple customer scenarios, including logistics, social networks, AI audio and video, e-commerce, etc. Compared with the previous algorithms, AHPA solves the elastic lag problem, increasing CPU usage by 10% and reducing resource cost by more than 20%. In addition, AHPA can automatically perform flexible planning according to the predicted business volume without manual intervention, significantly saving operation and maintenance costs.

  • Open Access

    Towards Out-of-Distribution Sequential Event Prediction: A Causal Treatment

    by Chenxiao Yang, Qitian Wu, Qingsong Wen, Zhiqiang Zhou, Liang Sun, Junchi Yan

    The goal of sequential event prediction is to estimate the next event based on a sequence of historical events, with applications to sequential recommendation, user behavior analysis and clinical treatment. In practice, the next-event prediction models are trained with sequential data collected at one time and need to generalize to newly arrived sequences in remote future, which requires models to handle temporal distribution shift from training to testing. In this paper, we first take a data-generating perspective to reveal a negative result that existing approaches with maximum likelihood estimation would fail for distribution shift due to the latent context confounder, i.e., the common cause for the historical events and the next event. Then we devise a new learning objective based on backdoor adjustment and further harness variational inference to make it tractable for sequence learning problems. On top of that, we propose a framework with hierarchical branching structures for learning context-specific representations. Comprehensive experiments on diverse tasks (e.g., sequential recommendation) demonstrate the effectiveness, applicability and scalability of our method with various off-the-shelf models as backbones.

  • Research on power quality comprehensive control device under the situation of medium voltage and high power based on H-bridge cascade converter

    by Zhiqiang Zhou, Yu Shen, Chengkuan Wan, Wei Hu, Tao Wang, Hao Tang

    This paper proposes a power quality control device based on H-Bridge cascade converter, to solve the problems in current decentralised power quality control, such as huge investment in equipment, e...


  • Towards an optimized gallium gradient for Cu(In,Ga)Se2 thin film via an improved constant low-temperature deposition process

    by Yunxiang Zhang, Shuping Lin, Zhaojing Hu, Shiqing Cheng, Zhichao He, Zhiqiang Zhou, Shulong Sun, Wei Liu, Yun Sun