A5101762609- Fault Detection and Control Systems
- Mineral Processing and Grinding
- Iron and Steelmaking Processes
- Spectroscopy and Chemometric Analyses
- Pluripotent Stem Cells Research
- Congenital heart defects research
- Anomaly Detection Techniques and Applications
- Cardiovascular, Neuropeptides, and Oxidative Stress Research
- MicroRNA in disease regulation
- Geoscience and Mining Technology
- Control Systems and Identification
- Industrial Vision Systems and Defect Detection
- Cardiovascular Disease and Adiposity
- Metallurgical Processes and Thermodynamics
- Neural Networks and Applications
- CRISPR and Genetic Engineering
- Advanced Computational Techniques and Applications
- Advanced Control Systems Optimization
- Cardiac Fibrosis and Remodeling
- Apelin-related biomedical research
- Advanced Algorithms and Applications
- Simulation and Modeling Applications
- Machine Learning and ELM
- Cancer-related molecular mechanisms research
- Advanced Sensor and Control Systems
Zhejiang University of Technology
2008-2025
State Key Laboratory of Industrial Control Technology
2011-2025
Zhejiang University
2008-2025
Xi’an University of Posts and Telecommunications
2009-2024
The First Affiliated Hospital, Sun Yat-sen University
2013-2023
Zhongshan Hospital
2013-2023
Fudan University
2013-2023
Sun Yat-sen University
2013-2023
Shanxi University
2023
Wuhan Institute of Virology
2022
Sinter ore is the main raw material of blast furnace, and burn-through point (BTP) has a direct influence on yield, quality, energy consumption ironmaking process. Since iron sintering very complex industrial process with strong nonlinearity, multivariable coupling, random noises, time variation, traditional soft-sensor models are hard to learn knowledge In this article, multistep prediction model, called denoising spatial–temporal encoder–decoder, developed predict BTP in advance. First,...
Fault detection technique is essential for improving overall equipment efficiency of semiconductor manufacturing industry. It has been recognized that fault based on k nearest neighbor rule (kNN) can effectively deal with some characteristics processes, such as multimode batch trajectories and nonlinearity. However, the computation complexity storage space involved in neighbors searching NN prevent it from online monitoring, especially high dimensional cases. To this difficulty, principal...
Abstract Chemical modifications are important for RNA function and metabolism. N4-acetylcytidine (ac4C) is critical the translation stability of mRNA. Although ac4C found in viruses, detailed mechanisms through which affects viral replication unclear. Here, we reported that 5′ untranslated region enterovirus 71 (EV71) genome was modified by host acetyltransferase NAT10. Inhibition NAT10 mutation sites within internal ribosomal entry site (IRES) suppressed EV71 replication. enhanced via...
In the blast furnace iron-making process (BFIP), there still has been a significant push to maintain stable and ensure maximum efficiency. Although some control systems can compensate for multiple types of disturbances, faults always require precise human intervention avoid safety hazards. Therefore, it is crucial develop an efficient diagnosis system identify these quickly. This paper focuses on novel approach called deep stationary kernel learning support vector machine (DSKL-SVM) BFIP...
Background: Diabetic heart dysfunction is a common complication of diabetes. Cell death core event that leads to diabetic dysfunction. However, the time sequence cell pathways and precise intervene particular type remain largely unknown in heart. This study aims identify responsible for propose promising therapeutic strategy by intervening pathway. Methods: Type 2 diabetes models were established using db/db leptin receptor–deficient mice high-fat diet/streptozotocin–induced mice. The 1...
As a key thermal-state indicator of the iron ore sintering process, content ferrous oxide (FeO) in finished sinter is directly related to product quality. Based on massive data data-driven soft sensor model provides good choice for real-time FeO detection. However, complex characteristics data, including dynamics, nonlinearity, and multisource heterogeneity, are still main obstacles improving modeling accuracy. To solve this problem, article, information fusion autoformer (MIF-Autoformer)...
Predicting burn-through point (BTP) in advance is a quite critical task for the sintering process. However, complex physicochemical reaction process, and strong spatial–temporal correlations of data make multistep prediction very challenging. The previous BTP model only extracts spatial features high-level layers, leaving low-level layers not learned. Specifically, considers relationships between process variables BTP, ignoring coupling relations among variables. Further, existing loss...
The shortage of computation methods and storage devices has largely limited the development multi-objective optimization in industrial processes. To improve operational levels process industries, we propose a framework based on cloud services distribution system. Real-time data from manufacturing procedures are first temporarily stored local database, then transferred to relational database cloud. Next, system with elastic compute power is set up for framework. Finally, model deep learning...
Blast furnaces are the most crucial equipment in ironmaking processes. Stable operation of blast furnace is a prerequisite for personnel safety and production efficiency. Therefore, early detection abnormalities an important task However, owing to large fluctuations quality raw materials, dynamic operating conditions, as well impact hot stoves switches, measurements show severe nonstationary characteristics. All these factors make monitoring challenging task. In this article, process method...
Soft sensor of silicon content in hot metal is challenging due to the nonlinearity, dynamics, and non-Euclidean coupling relationship between process variables. The inter-variable widely exists ironmaking process. Nevertheless, they are generally overlooked modeling. To solve this issue, paper proposes a novel temporal graph convolutional network with supervised graphs (TGCN-S). Different from traditional soft methods, TGCN-S explicitly models inherent irregular relationships First,...
Data-driven soft sensing modeling is becoming a powerful tool in the ironmaking process due to rapid development of machine learning and data mining. Although various techniques have been successfully used both sintering blast furnace, they not comprehensively reviewed. In this work, we provide an overview recent advances on process, with special focus data-driven techniques. First, present general framework based mechanism analysis characteristics. Second, detailed taxonomy current methods...
Cardiac microvascular endothelial cells (CMECs) are important angiogenic components and injured rapidly after cardiac ischaemia anoxia. Cardioprotective effects of Qiliqiangxin (QL), a traditional Chinese medicine, have been displayed recently. This study aims to investigate whether QL could protect CMECs against anoxic injury explore related signalling mechanisms. were successfully cultured from Sprague-Dawley rats exposed anoxia for 12 hrs in the absence presence QL. Cell migration assay...
Burn-through point (BTP) is a very key factor in maintaining the normal operation of sintering process, which guarantees yield and quality sinter ore. Due to characteristics time-varying multivariable coupling actual it difficult for traditional soft-sensor models extract spatial-temporal features reduce multistep prediction error accumulation. To address these issues, this study, we propose probabilistic aware network, called BTPNet, used feature accurate BTP prediction. The BTPNet model...
Anomaly detection is a cornerstone of industrial safety, enabling real-time monitoring process operations by identifying deviations from normal conditions through statistical analysis. In real-world scenarios, the nonstationary properties multivariate time-series data present common and substantial challenge. Existing methods for extracting stationary sources (SSs) mainly rely on weak stationarity (i.e., mean variance), but their performance limited long-tailed distributions in datasets....
Due to the existence of multiple operating modes, traditional fault detection techniques are ill-suited for complex industrial processes. Although there more and literature studies concerning this problem, only a few them based on hidden Markov model (HMM). However, is no exploration concern unknown mode in process it. This article proposes novel monitoring approach moving window HMM (MVHMM) real-time multimode with mode. First, built by training set. Instead just considering posterior...
Few study has been done to evaluate the stability and superiority of normalizers for serum microRNA (miRNA) in cardiovascular disease. Therefore, aim this is assess suitability several common (miR-16, SNOU6, 5S, miR-19b, miR-24, miR-15b, let 7i) disease.We evaluated seven circulating miRNAs as reference genes blood samples from patients with disease [heart failure (HF) hypertension] healthy people. Stability was quantified by combining BestKeeper, NormFinder comparative delta Cq analysis.A...
For the actual blast furnace ironmaking process (BFIP), sophisticated dynamic, nonlinear, and nonstationary characteristics make it hard to be modeled accurately with conventional monitoring methods. In this article, local dynamic broad kernel stationary subspace analysis (local-DBKSSA) is developed improve performance. Faced complex nonlinear characteristics, a single model considered unable for accurate representation. Thus, features established by time shift multikernel projection are...
Nowadays, data-driven soft sensors have become mainstream for the key performance indicators prediction, which guarantees safety and stability of industrial process. The typical autoencoder (AE) has been widely used to extract potential features through unsupervised pretraining supervised fine-tuning. However, most existing studies fail consider both time-varying process differences in contributions hidden target variable. Therefore, this article, a stacked spatial–temporal (S <sup...
Blast furnace iron-making process (BFIP) is one of the most critical procedures in iron and steel industry where timely detection accurate classification faults have always been core focus. However, coupling effects system's nonlinear nonstationary characteristics often cause consistent underlying information to be buried, allowing extraction a significant challenge. This also complicates development BFIP fault diagnosis model. Therefore, we propose novel data-driven joint strategy that...
Blast furnace iron-making process (BFIP) holds immense significance in iron and steel industry. However, due to its nonlinear dynamic properties, establishing an effective monitoring model has remained a major challenge. This paper proposes novel method called Joint Sparse Constrained Data-Dependent Kernel Canonical Variate Analysis (JS-DDKCVA) enhance performance. Initially, data-dependent feature extraction framework is developed by Isolation (IK). Unlike conventional kernels, IK directly...