实际工业过程往往具有多变量、非线性和动态性等特点,建模数据包含过多冗余信息和时序依赖特征,从而导致建模复杂度增加和模型性能下降。因此,提出一种基于非负绞杀的稀疏化有序神经元长短时记忆网络(ordered neurons long short-term m...实际工业过程往往具有多变量、非线性和动态性等特点,建模数据包含过多冗余信息和时序依赖特征,从而导致建模复杂度增加和模型性能下降。因此,提出一种基于非负绞杀的稀疏化有序神经元长短时记忆网络(ordered neurons long short-term memory,ONLSTM)用于工业软测量建模。将非负绞杀收缩系数嵌入ONLSTM输入层权重矩阵,对其进行收缩绞杀,剔除冗余输入节点的同时实现变量选择。将非负绞杀收缩系数与ONLSTM隐藏层权重矩阵相结合,根据不同隐藏神经元重要性设计权重分配规则,剔除网络隐藏层冗余节点及其对应的信息传递通路,进行网络结构稀疏优化。通过数值仿真验证了所提算法的有效性,并将其应用于某火电厂烟气脱硫过程排放净烟气SO2浓度预测。实验结果表明所提算法能有效实现变量选择,并在保证预测性能的前提下,使模型结构得到稀疏优化,展现出比较广阔的应用前景。展开更多
子痫前期(preeclampsia,PE)是妊娠期高血压疾病(hypertensive disorders of pregnancy,HDP)之一,严重损害母婴健康,影响3%~5%的妊娠。随高通量测序技术的发展和普及,肠道微生物与人类健康和疾病的关系逐渐清晰。PE患者肠道微生物群结构...子痫前期(preeclampsia,PE)是妊娠期高血压疾病(hypertensive disorders of pregnancy,HDP)之一,严重损害母婴健康,影响3%~5%的妊娠。随高通量测序技术的发展和普及,肠道微生物与人类健康和疾病的关系逐渐清晰。PE患者肠道微生物群结构变化与疾病的发生发展密切相关,失调的肠道菌群很可能会成为PE早期诊断的生物标志物或干预的靶点。本文总结了以测序技术为基础获得的正常妊娠女性和PE患者肠道微生物分布规律数据,并提出益生菌调理建议,以期为PE的科学预测、精准干预提供理论支持。展开更多
用先交联后固定法(即先用戊二醛交联使壳聚糖载体活化,后将壳聚糖载体与乳糖酶进行固定)制备固定化乳糖酶。研究其固定化最优条件为1.0 g壳聚糖载体,先用7.5 m L质量分数0.4%戊二醛溶液,于30℃条件下交联16 h,再用10 m L质量分数1.0%乳...用先交联后固定法(即先用戊二醛交联使壳聚糖载体活化,后将壳聚糖载体与乳糖酶进行固定)制备固定化乳糖酶。研究其固定化最优条件为1.0 g壳聚糖载体,先用7.5 m L质量分数0.4%戊二醛溶液,于30℃条件下交联16 h,再用10 m L质量分数1.0%乳糖酶溶液,于4℃条件下固定9 h,制备固定化乳糖酶活力为0.735 U/g。展开更多
The physical model describing the Yin-Yang balance in the tai-chi diagram via the melting and freezing processes in a rotating device presented in parts 1 and 2 is further developed for the contemporary tai-chi diagra...The physical model describing the Yin-Yang balance in the tai-chi diagram via the melting and freezing processes in a rotating device presented in parts 1 and 2 is further developed for the contemporary tai-chi diagram and in the yuan-chi diagram. The contemporary tai-chi diagram shown in Fig.1 is a simplification form of the ancient tai-chi diagram presented in Reference [2]. There are two semi-circles forming the interface curve between the yin and yang in the contemporary tai-chi diagram. By knowing the location of the interface between the yin and yang in the contemporary tai-chi diagram, the requirement for the simulation model is to find the condition to match the interface location. The simplification changes not only the structure but also the physical insight of the ancient tai-chi diagram, which will be described in the present study.The yuan-chi diagram shown in Fig.2 is the combination of the Master Chen’s tai-chi diagram presented in References [1,2] and the contemporary tai-chi diagram.展开更多
The purpose of the paper is to establish a theory of the constant heat flux ratio across the frozen layer based on the dimensional analysis of the system equations describing the freezing processes. An analytical mode...The purpose of the paper is to establish a theory of the constant heat flux ratio across the frozen layer based on the dimensional analysis of the system equations describing the freezing processes. An analytical model is then developed, utilizing this theory, for solving the planar, cylindrical and spherical freezing problems with both inward and outward freezing. As there is no exact solution available for the cylindrical and spherical freezing processes, the temperature distribution in the planar solidification obtained from the model is compared with the exact solution. They are in excellent agreement. For the cylindrical and spherical freezing, the complete inward solidification times calculated by the model are compared with those obtained from references. The results are in good agreement. The great advantage of the proposed model is its simplicity and is sufficiently accurate for most practical展开更多
The particle paths of the Lagrangian flow field between two cylinders simulate well the spiral arms of Galaxy M51 image [1] and the interface curve of the Yin-Yang balance in the ancient Tai-Cbi diagram [2]. The parti...The particle paths of the Lagrangian flow field between two cylinders simulate well the spiral arms of Galaxy M51 image [1] and the interface curve of the Yin-Yang balance in the ancient Tai-Cbi diagram [2]. The particle paths of the Lagrangian flow field involve four parameters. The normalization of the system of equations significantly simplifies the formulation of the flow process and reduces the original four parameters to only one parameter. Furthermore it provides the similarity between the formulation of the spiral arms of Galaxy M51 and that of the interface curve of the Yin-Yang balance in the ancient Tai-Chi diagram.展开更多
The integration of Dynamic Graph Neural Networks(DGNNs)with Smart Manufacturing is crucial as it enables real-time,adaptive analysis of complex data,leading to enhanced predictive accuracy and operational efficiency i...The integration of Dynamic Graph Neural Networks(DGNNs)with Smart Manufacturing is crucial as it enables real-time,adaptive analysis of complex data,leading to enhanced predictive accuracy and operational efficiency in industrial environments.To address the problem of poor combination effect and low prediction accuracy of current dynamic graph neural networks in spatial and temporal domains,and over-smoothing caused by traditional graph neural networks,a dynamic graph prediction method based on spatiotemporal binary-domain recurrent-like architecture is proposed:Binary Domain Graph Neural Network(BDGNN).The proposed model begins by utilizing a modified Graph Convolutional Network(GCN)without an activation function to extract meaningful graph topology information,ensuring non-redundant embeddings.In the temporal domain,Recurrent Neural Network(RNN)and residual systems are employed to facilitate the transfer of dynamic graph node information between learner weights,aiming to mitigate the impact of noise within the graph sequence.In the spatial domain,the AdaBoost(Adaptive Boosting)algorithm is applied to replace the traditional approach of stacking layers in a graph neural network.This allows for the utilization of multiple independent graph learners,enabling the extraction of higher-order neighborhood information and alleviating the issue of over-smoothing.The efficacy of BDGNN is evaluated through a series of experiments,with performance metrics including Mean Average Precision(MAP)and Mean Reciprocal Rank(MRR)for link prediction tasks,as well as metrics for traffic speed regression tasks across diverse test sets.Compared with other models,the better experiments results demonstrate that BDGNN model can not only better integrate the connection between time and space information,but also extract higher-order neighbor information to alleviate the over-smoothing phenomenon of the original GCN.展开更多
文摘实际工业过程往往具有多变量、非线性和动态性等特点,建模数据包含过多冗余信息和时序依赖特征,从而导致建模复杂度增加和模型性能下降。因此,提出一种基于非负绞杀的稀疏化有序神经元长短时记忆网络(ordered neurons long short-term memory,ONLSTM)用于工业软测量建模。将非负绞杀收缩系数嵌入ONLSTM输入层权重矩阵,对其进行收缩绞杀,剔除冗余输入节点的同时实现变量选择。将非负绞杀收缩系数与ONLSTM隐藏层权重矩阵相结合,根据不同隐藏神经元重要性设计权重分配规则,剔除网络隐藏层冗余节点及其对应的信息传递通路,进行网络结构稀疏优化。通过数值仿真验证了所提算法的有效性,并将其应用于某火电厂烟气脱硫过程排放净烟气SO2浓度预测。实验结果表明所提算法能有效实现变量选择,并在保证预测性能的前提下,使模型结构得到稀疏优化,展现出比较广阔的应用前景。
文摘子痫前期(preeclampsia,PE)是妊娠期高血压疾病(hypertensive disorders of pregnancy,HDP)之一,严重损害母婴健康,影响3%~5%的妊娠。随高通量测序技术的发展和普及,肠道微生物与人类健康和疾病的关系逐渐清晰。PE患者肠道微生物群结构变化与疾病的发生发展密切相关,失调的肠道菌群很可能会成为PE早期诊断的生物标志物或干预的靶点。本文总结了以测序技术为基础获得的正常妊娠女性和PE患者肠道微生物分布规律数据,并提出益生菌调理建议,以期为PE的科学预测、精准干预提供理论支持。
文摘用先交联后固定法(即先用戊二醛交联使壳聚糖载体活化,后将壳聚糖载体与乳糖酶进行固定)制备固定化乳糖酶。研究其固定化最优条件为1.0 g壳聚糖载体,先用7.5 m L质量分数0.4%戊二醛溶液,于30℃条件下交联16 h,再用10 m L质量分数1.0%乳糖酶溶液,于4℃条件下固定9 h,制备固定化乳糖酶活力为0.735 U/g。
基金The present work is being supported by the Natural Sciences and Engineering Research Council of Canada under Grant No. OGP0007929.
文摘The physical model describing the Yin-Yang balance in the tai-chi diagram via the melting and freezing processes in a rotating device presented in parts 1 and 2 is further developed for the contemporary tai-chi diagram and in the yuan-chi diagram. The contemporary tai-chi diagram shown in Fig.1 is a simplification form of the ancient tai-chi diagram presented in Reference [2]. There are two semi-circles forming the interface curve between the yin and yang in the contemporary tai-chi diagram. By knowing the location of the interface between the yin and yang in the contemporary tai-chi diagram, the requirement for the simulation model is to find the condition to match the interface location. The simplification changes not only the structure but also the physical insight of the ancient tai-chi diagram, which will be described in the present study.The yuan-chi diagram shown in Fig.2 is the combination of the Master Chen’s tai-chi diagram presented in References [1,2] and the contemporary tai-chi diagram.
文摘The purpose of the paper is to establish a theory of the constant heat flux ratio across the frozen layer based on the dimensional analysis of the system equations describing the freezing processes. An analytical model is then developed, utilizing this theory, for solving the planar, cylindrical and spherical freezing problems with both inward and outward freezing. As there is no exact solution available for the cylindrical and spherical freezing processes, the temperature distribution in the planar solidification obtained from the model is compared with the exact solution. They are in excellent agreement. For the cylindrical and spherical freezing, the complete inward solidification times calculated by the model are compared with those obtained from references. The results are in good agreement. The great advantage of the proposed model is its simplicity and is sufficiently accurate for most practical
基金sponsored by the Natural SciencesEngineering Research Council of Canada
文摘The particle paths of the Lagrangian flow field between two cylinders simulate well the spiral arms of Galaxy M51 image [1] and the interface curve of the Yin-Yang balance in the ancient Tai-Cbi diagram [2]. The particle paths of the Lagrangian flow field involve four parameters. The normalization of the system of equations significantly simplifies the formulation of the flow process and reduces the original four parameters to only one parameter. Furthermore it provides the similarity between the formulation of the spiral arms of Galaxy M51 and that of the interface curve of the Yin-Yang balance in the ancient Tai-Chi diagram.
基金funded by Guangdong Provincial Natural Science Foundation of China under Grant No.2021A1515011243Guangdong Provincial Science and Technology Plan Project under Grant No.2019B010139001Guangzhou Science and Technology Plan Project under Grant No.201902020016.
文摘The integration of Dynamic Graph Neural Networks(DGNNs)with Smart Manufacturing is crucial as it enables real-time,adaptive analysis of complex data,leading to enhanced predictive accuracy and operational efficiency in industrial environments.To address the problem of poor combination effect and low prediction accuracy of current dynamic graph neural networks in spatial and temporal domains,and over-smoothing caused by traditional graph neural networks,a dynamic graph prediction method based on spatiotemporal binary-domain recurrent-like architecture is proposed:Binary Domain Graph Neural Network(BDGNN).The proposed model begins by utilizing a modified Graph Convolutional Network(GCN)without an activation function to extract meaningful graph topology information,ensuring non-redundant embeddings.In the temporal domain,Recurrent Neural Network(RNN)and residual systems are employed to facilitate the transfer of dynamic graph node information between learner weights,aiming to mitigate the impact of noise within the graph sequence.In the spatial domain,the AdaBoost(Adaptive Boosting)algorithm is applied to replace the traditional approach of stacking layers in a graph neural network.This allows for the utilization of multiple independent graph learners,enabling the extraction of higher-order neighborhood information and alleviating the issue of over-smoothing.The efficacy of BDGNN is evaluated through a series of experiments,with performance metrics including Mean Average Precision(MAP)and Mean Reciprocal Rank(MRR)for link prediction tasks,as well as metrics for traffic speed regression tasks across diverse test sets.Compared with other models,the better experiments results demonstrate that BDGNN model can not only better integrate the connection between time and space information,but also extract higher-order neighbor information to alleviate the over-smoothing phenomenon of the original GCN.
