光伏故障检测对光伏电站智能运维具有重要意义。针对光伏组件红外图像中热斑目标小、难检测的问题,研究了基于改进Faster R CNN的光伏组件红外热斑故障检测模型。将Swin Transformer作为Faster R CNN模型中的特征提取模块,捕获图像的全...光伏故障检测对光伏电站智能运维具有重要意义。针对光伏组件红外图像中热斑目标小、难检测的问题,研究了基于改进Faster R CNN的光伏组件红外热斑故障检测模型。将Swin Transformer作为Faster R CNN模型中的特征提取模块,捕获图像的全局信息,建立特征之间的依赖关系,提高模型的建模能力;进一步利用BiFPN进行特征融合,改善了热斑故障由于目标小和特征不明显容易被模型忽略掉的问题;同时为了抑制光伏红外图像中背景和噪声的干扰,加入轻量级注意力模块CBAM,使模型更加关注重要通道和关键区域,提高对热斑故障检测精度。在自建光伏组件图像数据集上进行实验,热斑故障检测精度高达915,验证了本文模型对光伏组件热斑故障检测的有效性。展开更多
针对RGB(Red Green Blue)模态与热度模态信息表征形式不一致,特征信息无法有效挖掘、融合问题,提出了一种新的联合注意力强化网络-FCNet(Feature Sharpening and Cross-modal Feature Fusion Net)。首先,通过双维度注意力机制提升图像...针对RGB(Red Green Blue)模态与热度模态信息表征形式不一致,特征信息无法有效挖掘、融合问题,提出了一种新的联合注意力强化网络-FCNet(Feature Sharpening and Cross-modal Feature Fusion Net)。首先,通过双维度注意力机制提升图像特征映射能力;然后,利用跨模态特征融合机制捕获目标区域;最后,利用逐层解码结构消除背景干扰,优化检测目标。实验结果表明,该优化改进算法运算参数更少、运算时间更短,且模型整体检测性能均优于现有多模态检测模型性能。展开更多
Inflammatory bowel disease(IBD)is chronic inflammation of the gastrointestinal tract that has a high epidemiological prevalence worldwide.The increasing disease burden worldwide,lack of response to current biologic th...Inflammatory bowel disease(IBD)is chronic inflammation of the gastrointestinal tract that has a high epidemiological prevalence worldwide.The increasing disease burden worldwide,lack of response to current biologic therapeutics,and treatment-related immunogenicity have led to major concerns regarding the clinical management of IBD patients and treatment efficacy.Understanding disease pathogenesis and disease-related molecular mechanisms is the most important goal in developing new and effective therapeutics.Sphingosine-1-phosphate(S1P)receptor(S1PR)modulators form a class of oral small molecule drugs currently in clinical development for IBD have shown promising effects on disease improvement.S1P is a sphingosine-derived phospholipid that acts by binding to its receptor S1PR and is involved in the regulation of several biological processes including cell survival,differentiation,migration,proliferation,immune response,and lymphocyte trafficking.T lymphocytes play an important role in regulating inflammatory responses.In inflamed IBD tissue,an imbalance between T helper(Th)and regulatory T lymphocytes and Th cytokine levels was found.The S1P/S1PR signaling axis and metabolism have been linked to inflammatory responses in IBD.S1P modulators targeting S1PRs and S1P metabolism have been developed and shown to regulate inflammatory responses by affecting lymphocyte trafficking,lymphocyte number,lymphocyte activity,cytokine production,and contributing to gut barrier function.展开更多
This work presents the design of hollow fiber T-type zeolite membrane modules with different geometric configurations. The module performances were evaluated by pervaporation dehydration of ethanol/water mixtures. Str...This work presents the design of hollow fiber T-type zeolite membrane modules with different geometric configurations. The module performances were evaluated by pervaporation dehydration of ethanol/water mixtures. Strong concentration polarization was found for the modules with big membrane bundles. The concentration polarization was enhanced at high temperature due to the higher water permeation flux. The increase of feed flow could improve water permeation flux for the membrane modules with small membrane bundle.Computational fluid dynamics was used to visualize the flow field distribution inside of the modules with different configurations. The membrane module with seven bundles exhibited highest separation efficiency due to the uniform distribution of flow rate. The packing density could be 10 times higher than that of the tubular membrane module. The hollow fiber membrane module exhibited good stability for ethanol dehydration.展开更多
文摘光伏故障检测对光伏电站智能运维具有重要意义。针对光伏组件红外图像中热斑目标小、难检测的问题,研究了基于改进Faster R CNN的光伏组件红外热斑故障检测模型。将Swin Transformer作为Faster R CNN模型中的特征提取模块,捕获图像的全局信息,建立特征之间的依赖关系,提高模型的建模能力;进一步利用BiFPN进行特征融合,改善了热斑故障由于目标小和特征不明显容易被模型忽略掉的问题;同时为了抑制光伏红外图像中背景和噪声的干扰,加入轻量级注意力模块CBAM,使模型更加关注重要通道和关键区域,提高对热斑故障检测精度。在自建光伏组件图像数据集上进行实验,热斑故障检测精度高达915,验证了本文模型对光伏组件热斑故障检测的有效性。
文摘针对RGB(Red Green Blue)模态与热度模态信息表征形式不一致,特征信息无法有效挖掘、融合问题,提出了一种新的联合注意力强化网络-FCNet(Feature Sharpening and Cross-modal Feature Fusion Net)。首先,通过双维度注意力机制提升图像特征映射能力;然后,利用跨模态特征融合机制捕获目标区域;最后,利用逐层解码结构消除背景干扰,优化检测目标。实验结果表明,该优化改进算法运算参数更少、运算时间更短,且模型整体检测性能均优于现有多模态检测模型性能。
文摘Inflammatory bowel disease(IBD)is chronic inflammation of the gastrointestinal tract that has a high epidemiological prevalence worldwide.The increasing disease burden worldwide,lack of response to current biologic therapeutics,and treatment-related immunogenicity have led to major concerns regarding the clinical management of IBD patients and treatment efficacy.Understanding disease pathogenesis and disease-related molecular mechanisms is the most important goal in developing new and effective therapeutics.Sphingosine-1-phosphate(S1P)receptor(S1PR)modulators form a class of oral small molecule drugs currently in clinical development for IBD have shown promising effects on disease improvement.S1P is a sphingosine-derived phospholipid that acts by binding to its receptor S1PR and is involved in the regulation of several biological processes including cell survival,differentiation,migration,proliferation,immune response,and lymphocyte trafficking.T lymphocytes play an important role in regulating inflammatory responses.In inflamed IBD tissue,an imbalance between T helper(Th)and regulatory T lymphocytes and Th cytokine levels was found.The S1P/S1PR signaling axis and metabolism have been linked to inflammatory responses in IBD.S1P modulators targeting S1PRs and S1P metabolism have been developed and shown to regulate inflammatory responses by affecting lymphocyte trafficking,lymphocyte number,lymphocyte activity,cytokine production,and contributing to gut barrier function.
基金Supported by the National Natural Science Foundation of China(21222602,21490585 and 21176117)National High-tech R&D Program of China(2015AA03A602)+4 种基金the Key Project of Chinese Ministry of Education(212060)the Outstanding Young Fund of Jiangsu Province(BK2012040)Young Fund of Jiangsu Province(BK20130915)the "Six Top Talents" and "333 Talent Project" of Jiangsu Province,a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions(PAPD)the Innovation Project for Graduate Student of Jiangsu Province(KYLX15-0790)
文摘This work presents the design of hollow fiber T-type zeolite membrane modules with different geometric configurations. The module performances were evaluated by pervaporation dehydration of ethanol/water mixtures. Strong concentration polarization was found for the modules with big membrane bundles. The concentration polarization was enhanced at high temperature due to the higher water permeation flux. The increase of feed flow could improve water permeation flux for the membrane modules with small membrane bundle.Computational fluid dynamics was used to visualize the flow field distribution inside of the modules with different configurations. The membrane module with seven bundles exhibited highest separation efficiency due to the uniform distribution of flow rate. The packing density could be 10 times higher than that of the tubular membrane module. The hollow fiber membrane module exhibited good stability for ethanol dehydration.
