检索结果-内蒙古大学图书馆

Deep learning-based artistic style transformation algorithm in visual communication

INTERNATIONAL JOURNAL OF SYSTEM ASSURANCE ENGINEERING AND MANAGEMENT 2024年 1-10页

作者： zheng, shuwen Shanghai Inst Commerce & Foreign Languages Shanghai 201399 Peoples R China

Artistic style transition is a popular visual communication technique, the subject matter of one image with the aesthetics of another to create stunning new works. Deep learning (DL) revolutionized computers by solving algorithm-impractical issues, including voice recognition, machine translation, vision, etc. We collect content and style images for our algorithm in this study. Gaussian filter preprocessing is a technique for improving data. This smoothing method removes visual noise and inconsistencies to let the neural network focus on important information and design aspects. We process preprocessed images using a Dense Convolutional Network (DenseNet) to acquire feature representations. Our technique classifies style and content using Long Short-Term Memory (LSTM) *** networks enhance visual communication by effectively capturing and processing sequential visual data, improving the accuracy of tasks like video analysis and image *** can represent the sequential character of creative styles using LSTM networks since they function well with sequence *** altered images produce an intriguing and creative visual result, which is aesthetically pleasing while retaining the semantic and structural coherence of the source *** use the Python program recommended for visual communication. The findings showed that in terms of accuracy (96%), precision (93%), and runtime (80%)Compared with other DL-based models we examined, the average performance of our proposed method was *** approach successfully generates visually appealing and semantically coherent images, demonstrating the potential of advanced neural networks in creative applications.

关键词： Deep learning Artistic style Visual Images Communication Aesthetics

来源：评论

学校读者我要写书评

暂无评论

BiGraphormer: A Bidirectional Graphormer on Directed Causal Graph for Fault Detection in Complex Systems

引用

IEEE TRANSACTIONS ON RELIABILITY 2024年

作者： zheng, shuwen Liu, Jie Chen, Yunxia Beihang Univ Sch Reliabil & Syst Engn Beijing 100191 Peoples R China

Effective fault detection of complex systems can significantly enhance safety, availability, and maintainability. Recently, graph neural networks (GNNs), which leverage spatial structures among variables, have gained substantial attention due to advancements over previous data-driven methods. However, existing GNN-based fault detection models primarily focus on adjacent neighborhood for feature fusion, neglecting the long-range dependencies in graphs. Moreover, these models often utilize undirected correlational graphs, potentially limiting their applicability and modeling efficiency for target systems. In this work, a causal graph-based bidirectional Graphormer (BiGraphormer) is proposed for complex systems fault detection. The causal relationships among monitoring variables are mined and represented as a directed acyclic causal graph, in which nodes denote variables and directed edges indicate causal influence. Then, the dependencies including the global spatial structure, the node and edge information are encoded and fused using the proposed BiGraphormer. By incorporating both ancestral cause and descendant effect nodes along the directed causal graphs, comprehensive representations are constructed for system fault detection. To validate the effectiveness of the proposed framework, a case study concerning real monitoring data of high-speed train braking systems is conducted, with the results showing efficacy of the proposed method.

关键词： Fault detection Monitoring Feature extraction Complex systems Encoding Transformers Cause effect analysis Voltage Adaptation models Streams Causal discovery complex systems fault detection graph neural network high-speed train transformer

来源：评论

学校读者我要写书评

暂无评论

A Study of Adding Psychological Feeling into the Model of Synaesthesia at the Morpheme Level in Modern Chinese 24th

A Study of Adding Psychological Feeling into the Model of Sy...

引用

24th Chinese Lexical Semantics Workshop (CLSW)

作者： Wen, Han zheng, shuwen Zhejiang Int Studies Univ 299 Liuhe Rd Hangzhou 310023 Peoples R China Beijing Language & Culture Univ 15 Xueyuan Rd Beijing 100083 Peoples R China

ISBN: (纸本)9789819705825;9789819705832

The synaesthetic mapping pattern of modern Chinese lexicon have long been a focal point in both national and international linguistic studies. However, issues at the morpheme level are scarce, and there is rarely incorporating psychological sensations into mapping models. This paper meticulously analyzes 279 synaesthetic cases, derived from 190 adjectival sensory words in the Modern Chinese Dictionary (7th Edition). Through this analysis, 27 synaesthesia patterns are identified. It is discovered that the phenomenon of synaesthesia in modern Chinese significantly differs at the morpheme level and lexicon level. Notably, psychological feeling in Chinese synaesthesia is frequently mapped as destination, becoming a distinctive feature of these metaphors. This finding provides new perspectives and methods for research in lexicology, etymology, and machine translation. Studying sensory metaphors within words in the Chinese lexicon and comparing them with those within phrases revealed the characteristics of sensory word formation in modern Chinese, offering fresh insights and examples for research in etymology and semantic evolution.

关键词： Synaesthesia Morpheme Mapping Patterns Psychological Feeling

来源：评论

学校读者我要写书评

暂无评论

Empirical study on fine-tuning pre-trained large language models for fault diagnosis of complex systems

引用

RELIABILITY ENGINEERING & SYSTEM SAFETY 2024年 252卷

作者： zheng, shuwen Pan, Kai Liu, Jie Chen, Yunxia Beihang Univ Sch Reliabil & Syst Engn Beijing 100191 Peoples R China

With increasing complexity and interconnectivity of the modern industrial systems, effectively diagnosing faults is a core step to Prognostic and Health Management (PHM). Language models, particularly Large Language Models (LLMs) that pre-trained on massive corpora, have demonstrated remarkable capabilities in Natural Language Processing (NLP) and its related downstream tasks. However, how to leverage these models for facilitating system fault diagnosis and verify its effectiveness is rarely explored. This paper incorporates the potential comprehension ability of pre-trained LLMs, and investigates the efficacy of fine-tuning LLMs for realizing efficient system fault diagnosis. The experiments conducted in this study involve both open-source and closed-source models, and utilize a simulation and a real fault diagnosis dataset. We find that these models consistently achieve high performance across various metrics compared to the baselines. Additionally, qualitative and quantitative analysis is performed to investigate several aspects of the approach, such as the impact of dataset size, data normalization, missing values and explainability of the diagnosis, further showcasing the potential as well as limitations of the approach.

关键词： prognosis and health management fault diagnosis large language model complex systems high-speed train chemical plant

来源：评论

学校读者我要写书评

暂无评论

Fault detection in complex mechatronic systems by a hierarchical graph convolution attention network based on causal paths

引用

RELIABILITY ENGINEERING & SYSTEM SAFETY 2024年 243卷

作者： zheng, shuwen Wang, Chong Zio, Enrico Liu, Jie Beihang Univ Sch Reliabil & Syst Engn Beijing 100191 Peoples R China PSL Res Univ Ctr Rech Risques & Crises MINES ParisTech Sophia Antipolis France Politecn Milan Energy Dept Via Masa 34 I-20156 Milan Italy Kyung Hee Univ Dept Nucl Engn 1732 Deogyeong daero Yongin 17104 South Korea

Fault detection in mechatronic systems is crucial for their maintainability and safety. However, the systems monitoring variables are often abundant, with intricate connections. It is difficult to characterize their re-lationships and to extract effective features. In this paper, a hierarchical graph convolution attention network based on causal paths (HGCAN) is developed to improve the performance and interpretability of data-driven fault detection models in complex mechatronic systems. A hybrid causal discovery algorithm is introduced to discover the inherit causality among monitoring variables. The causal paths that sequentially connect the cause-effect variables serve as the reception fields to extract features using multiscale convolution. Different levels of the features are aggregated based on a hierarchical attention mechanism, which assigns adaptive weights consid-ering the varied feature importance. To verify the effectiveness of the proposed method, a dataset of real high -speed train braking systems is considered. Experimental results demonstrate promising performance improve-ment of the proposed method, and analysis on interpretability indicates its potential to facilitate practical de-cision-making.

关键词： Mechatronic systems Fault detection Causal path Graph convolutional network Attention mechanism High -speed train

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning for Ventilation System Control in Pig Buildings

Deep Reinforcement Learning for Ventilation System Control i...

引用

作者： zheng, shuwen Leiden University

学位级别：硕士

Indoor thermal environment plays an important role in pigs' health, welfare, production, and reproduction, and ventilation system focuses on ensuring the comfort of animals by using exhaust fans and air inlets com- monly in livestock buildings. Model-free method Deep Reinforcement Learn- ing(DRL), well known for the performance in game-playing and robotics control, recently has applied in buildings heating, ventilation, and air con- ditioning (HVAC) systems control. This study explored the effectiveness of a DRL algorithm, Deep Q-Network(DQN), in ventilation system con- trol for pig buildings. The results showed that the DQN agent managed to maintain the room temperature within the comfortable range in 99.13%, 90.1%, 92.01% of test days in winter, spring and summer, respectively. The DQN agent outperformed the baseline method with the saving of power con- sumption by 17.66%, 30.04%, 6.89% in the test days of winter, spring and summer, respectively. The DQN algorithm applied the same neural network architecture and hyperparameter settings and was trained and tested in dif- ferent periods of time, indicating the generalization capability of the DQN algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Electronic structures and optical properties of Hf-, Zr-and Nb-doped 2D (β-Ga2O3

引用

PHYSICA B-CONDENSED MATTER 2025年 703卷

作者： Chen, Nan Xi, Zhihao Ma, Xiaochen Li, Shuti zheng, shuwen South China Normal Univ Sch Elect Sci & Engn Sch Microelect Foshan 528200 Peoples R China South China Normal Univ Guangdong Engn Res Ctr Optoelect Funct Mat & Devic Guangzhou 510006 Guangdong Peoples R China

The electronic structures and optical properties of 2D (1-Ga2O3 and Hf-, Zr-, Nb-doped 2D (1-Ga2O3 were systematically studied by GGA + U method based on density functional theory. Hf- and Zr-doped 2D (1-Ga2O3 tend to form under O-poor conditions while Nb-doped 2D (1-Ga2O3 tends to form under O-rich condition. Although Hf-, Zrand Nb-doped 2D (1-Ga2O3 can obtain high electron concentrations, Nb-doped 2D (1-Ga2O3 has deep donor levels and a large electron effective mass, which is not conducive to electron migration. In contrast, Hf- and Zrdoped 2D (1-Ga2O3 have good electrical conductivity and small work functions, which is easy to form ohmic contact with the metals. Besides, the optical absorption edges of Hf- and Zr-doped 2D (1-Ga2O3 are larger compared with 2D (1-Ga2O3 and Nb-doped 2D (1-Ga2O3, so Hf- and Zr-doped 2D (1-Ga2O3 are suitable as TCO materials for DUV optoelectronic devices. But Nb-doped 2D (1-Ga2O3 exhibits ferrimagnetic with a magnetic moment of 2 mu B and can be used as a spin material.

关键词： First-principles Doping Electronic structure Optical property

来源：评论

学校读者我要写书评

暂无评论

Strength and microscopic properties of S95 ground granulated blast furnace slag - Calcium carbide slag -Phosphogypsum carbonation and standard-cured waste soil

引用

CONSTRUCTION AND BUILDING MATERIALS 2025年 465卷

作者： Ma, Wene Tong, Liyuan Sun, Dafang Zhang, Tiantian zheng, shuwen Southeast Univ Sch Transportat Nanjing 211189 Peoples R China Nanjing Metro Town Dev & Construct Grp Co Ltd Nanjing 210000 Peoples R China

Resource utilization of solid waste and waste soil is crucial for addressing resource and environmental challenges and reducing carbon dioxide emission. This paper studies the strength and microscopic properties of waste soil (silty clay) solidified by industrial solid wastes(S95 ground granulated blast furnace slag (GGBS), calcium carbide slag (CS), semi-water phosphogypsum (PG)) at various mixture ratios under carbonation and standard curing conditions. Tests for compressive strength, flexural strength, pH, X-ray diffraction (XRD), and scanning electron microscopy (SEM) were conducted. Results indicate that under standard curing, the strength of GGBS-CS-PG solidified soil increased with higher GGBS-CS ratios and curing time, with the highest 7-d strength and the lowest pH value at 3:1 GGBS-CS ratio and 4 % PG. Because excessive Ca(OH)2 crystals from CS weakened the system's water resistance and reduced strength, and a small amount of PG has the sulphate excitation effect and SO4 2- can react with C-A-H to form ettringite (AFt) to improve strength. However, specimens cured with GGBS-CSPG under carbonation conditions exhibited lower strengths and pH values compared to those cured under standard conditions due to the decomposition of AFt. PG is not suitable as a cementitious material for carbonation curing. For GGBS-CS cured soil under carbonation conditions, the higher relative dosage of CS helped the carbonation reaction to continue, at the dosage ratios of 1:3 and 2:3, the compressive strength increased with age due to the accumulation of the reaction product CaCO3. And the strength at 1:3 dosage ratio is significantly lower than that at 2:3 due to excessive Ca(OH)2 crystals from CS. In addition, owing to the brittleness of CaCO3, the over prolonged carbonation curing will reduce the flexural strength, which increased in 1-4 days and decreased in 4-7 days. Consequently, it is reasonable to take 2:3 mixture ratio of GGBS-CS and 2-4 days of carbonation curing age.

关键词： Industrial waste Carbonation curing Standard curing Strength test Microtest Waste soil Phosphogypsum

来源：评论

学校读者我要写书评

暂无评论

Causal graph-based spatial-temporal attention network for RUL prediction of complex systems

引用

COMPUTERS & INDUSTRIAL ENGINEERING 2025年 201卷

作者： zheng, shuwen Liu, Jie Chen, Yunxia Fan, Yu Xu, Dan Beihang Univ Sch Reliabil & Syst Engn Beijing Peoples R China Beihang Univ Sch Energy & Power Engn Beijing Peoples R China

With diverse sensors in complex systems, Remaining Useful Life (RUL) prediction involves fusing multivariate time-series data. Incorporating features from spatial-and-temporal dimensions is a trending and challenging topic. Existing state-of-the-art researches typically fuse spatial-temporal features according to the correlation analysis, limiting their ability to capture representative features for RUL prediction. Moreover, the dynamic feature importance in system degradation is often ignored. To address the above limitations, this work proposes a Causal graph-based Spatial-Temporal Attention Network (CaSTAN) for RUL prediction of complex systems. Prior knowledge is first combined with causal discovery algorithms to explore cause-effect relationships among variables. Subsequently, strategies are introduced for adaptively extracting the spatial, temporal and causal features during the degradation process. The extracted representations are then aggregated using a gating mechanism, allowing to selectively fuse features for RUL prediction. Case studies concerning aircraft turbofan engines are conducted, along with extensive comparative and experimental analyses. The results underscore the method's effectiveness in enhancing predictive maintenance of complex systems.

关键词： Remaining useful life Spatial-temporal features Causality Complex systems Prognostics

来源：评论

学校读者我要写书评

暂无评论

Lagrange tracking-based long-term drift trajectory prediction method for Autonomous Underwater Vehicle

引用

MATHEMATICAL BIOSCIENCES AND ENGINEERING 2023年第12期20卷 21075-21097页

作者： zheng, shuwen Zhang, Mingjun Zhang, Jing Li, Jitao Harbin Engn Univ Coll Mech & Elect Engn Harbin 150001 Peoples R China Univ Jinan Sch Informat Sci & Engn Jinan 250022 Peoples R China

Autonomous Underwater Vehicle (AUV) works autonomously in complex marine environments. After a severe accident, an AUV will lose its power and rely on its small buoyancy to ascend at a slow speed. If the reserved buoyancy is insufficient, when reaching the thermocline, the buoyancy will rapidly decrease to zero. Consequently, the AUV will experience prolonged lateral drift within the thermocline. This study focuses on developing a prediction method for the drift trajectory of an AUV after a long-term power loss accident. The aim is to forecast the potential resurfacing location, providing technical support for surface search and salvage operations of the disabled AUV. To the best of our knowledge, currently, there is no mature and effective method for predicting longterm AUV underwater drift trajectories. In response to this issue, based on real AUV catastrophes, this paper studies the prediction of long-term AUV underwater drift trajectories in the cases of power loss. We propose a three-dimensional trajectory prediction method based on the Lagrange tracking approach. This method takes the AUV's longitudinal velocity, the time taken to reach different depths, and ocean current data at various depths into account. The reason for the AUV's failure to ascend to sea surface lies that the remaining buoyancy is too small to overcome the thermocline. As a result, AUV drifts long time within the thermocline. To address this issue, a method for estimating thermocline currents is proposed, which can be used to predict the lateral drift trajectory of the AUV within the thermocline. Simulation is conducted to compare the results obtained by the proposed method and that in a real accident. The results demonstrate that the proposed approach exhibits small directional and positional errors. This validates the effectiveness of the proposed method.

关键词： AUV underwater drift lagrange tracking method track prediction

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：