资源类型

期刊论文 481

会议视频 14

年份

2024 1

2023 56

2022 53

2021 54

2020 51

2019 30

2018 28

2017 32

2016 15

2015 21

2014 12

2013 13

2012 12

2011 5

2010 4

2009 7

2008 13

2007 14

2006 9

2005 14

展开 ︾

关键词

神经网络 23

人工神经网络 5

优化 5

人工智能 4

智能制造 4

BP神经网络 3

仿真 3

微波遥感 2

控制 2

机器学习 2

模糊神经网络 2

深度神经网络 2

算法 2

系统 2

网络 2

网络安全 2

能源 2

计算机 2

预测 2

展开 ︾

检索范围:

排序: 展示方式:

Novel interpretable mechanism of neural networks based on network decoupling method

《工程管理前沿(英文)》 2021年 第8卷 第4期   页码 572-581 doi: 10.1007/s42524-021-0169-x

摘要: The lack of interpretability of the neural network algorithm has become the bottleneck of its wide application. We propose a general mathematical framework, which couples the complex structure of the system with the nonlinear activation function to explore the decoupled dimension reduction method of high-dimensional system and reveal the calculation mechanism of the neural network. We apply our framework to some network models and a real system of the whole neuron map of Caenorhabditis elegans. Result shows that a simple linear mapping relationship exists between network structure and network behavior in the neural network with high-dimensional and nonlinear characteristics. Our simulation and theoretical results fully demonstrate this interesting phenomenon. Our new interpretation mechanism provides not only the potential mathematical calculation principle of neural network but also an effective way to accurately match and predict human brain or animal activities, which can further expand and enrich the interpretable mechanism of artificial neural network in the future.

关键词: neural networks     interpretability     dynamical behavior     network decouple    

A multi-sensor relation model for recognizing and localizing faults of machines based on network analysis

《机械工程前沿(英文)》 2023年 第18卷 第2期 doi: 10.1007/s11465-022-0736-9

摘要: Recently, advanced sensing techniques ensure a large number of multivariate sensing data for intelligent fault diagnosis of machines. Given the advantage of obtaining accurate diagnosis results, multi-sensor fusion has long been studied in the fault diagnosis field. However, existing studies suffer from two weaknesses. First, the relations of multiple sensors are either neglected or calculated only to improve the diagnostic accuracy of fault types. Second, the localization for multi-source faults is seldom investigated, although locating the anomaly variable over multivariate sensing data for certain types of faults is desirable. This article attempts to overcome the above weaknesses by proposing a global method to recognize fault types and localize fault sources with the help of multi-sensor relations (MSRs). First, an MSR model is developed to learn MSRs automatically and further obtain fault recognition results. Second, centrality measures are employed to analyze the MSR graphs learned by the MSR model, and fault sources are therefore determined. The proposed method is demonstrated by experiments on an induction motor and a centrifugal pump. Results show the proposed method’s validity in diagnosing fault types and sources.

关键词: fault recognition     fault localization     multi-sensor relations     network analysis     graph neural network    

Multiscale computation on feedforward neural network and recurrent neural network

Bin LI, Xiaoying ZHUANG

《结构与土木工程前沿(英文)》 2020年 第14卷 第6期   页码 1285-1298 doi: 10.1007/s11709-020-0691-7

摘要: Homogenization methods can be used to predict the effective macroscopic properties of materials that are heterogenous at micro- or fine-scale. Among existing methods for homogenization, computational homogenization is widely used in multiscale analyses of structures and materials. Conventional computational homogenization suffers from long computing times, which substantially limits its application in analyzing engineering problems. The neural networks can be used to construct fully decoupled approaches in nonlinear multiscale methods by mapping macroscopic loading and microscopic response. Computational homogenization methods for nonlinear material and implementation of offline multiscale computation are studied to generate data set. This article intends to model the multiscale constitution using feedforward neural network (FNN) and recurrent neural network (RNN), and appropriate set of loading paths are selected to effectively predict the materials behavior along unknown paths. Applications to two-dimensional multiscale analysis are tested and discussed in detail.

关键词: multiscale method     constitutive model     feedforward neural network     recurrent neural network    

Heat, mass, and work exchange networks

Zhiyou CHEN, Jingtao WANG

《化学科学与工程前沿(英文)》 2012年 第6卷 第4期   页码 484-502 doi: 10.1007/s11705-012-1221-5

摘要: Heat (energy), water (mass), and work (pressure) are the most fundamental utilities for operation units in chemical plants. To reduce energy consumption and diminish environment hazards, various integration methods have been developed. The application of heat exchange networks (HENs), mass exchange networks (MENs), water allocation heat exchange networks (WAHENs) and work exchange networks (WENs) have resulted in the significant saving of energy and water. This review presents the main works related to each network. The similarities and differences of these networks are also discussed. Through comparing and discussing these different networks, this review inspires researchers to propose more efficient and convenient methods for the design of existing exchange networks and even new types of networks including multi-objective networks for the system integration in order to enhance the optimization and controllability of processes.

关键词: process system engineering     integration methods     heat exchange network     mass exchange network     work exchange network    

Deep convolutional tree-inspired network: a decision-tree-structured neural network for hierarchical

《机械工程前沿(英文)》 2021年 第16卷 第4期   页码 814-828 doi: 10.1007/s11465-021-0650-6

摘要: The fault diagnosis of bearings is crucial in ensuring the reliability of rotating machinery. Deep neural networks have provided unprecedented opportunities to condition monitoring from a new perspective due to the powerful ability in learning fault-related knowledge. However, the inexplicability and low generalization ability of fault diagnosis models still bar them from the application. To address this issue, this paper explores a decision-tree-structured neural network, that is, the deep convolutional tree-inspired network (DCTN), for the hierarchical fault diagnosis of bearings. The proposed model effectively integrates the advantages of convolutional neural network (CNN) and decision tree methods by rebuilding the output decision layer of CNN according to the hierarchical structural characteristics of the decision tree, which is by no means a simple combination of the two models. The proposed DCTN model has unique advantages in 1) the hierarchical structure that can support more accuracy and comprehensive fault diagnosis, 2) the better interpretability of the model output with hierarchical decision making, and 3) more powerful generalization capabilities for the samples across fault severities. The multiclass fault diagnosis case and cross-severity fault diagnosis case are executed on a multicondition aeronautical bearing test rig. Experimental results can fully demonstrate the feasibility and superiority of the proposed method.

关键词: bearing     cross-severity fault diagnosis     hierarchical fault diagnosis     convolutional neural network     decision tree    

Identifying spreading influence nodes for social networks

《工程管理前沿(英文)》   页码 520-549 doi: 10.1007/s42524-022-0190-8

摘要: The identification of spreading influence nodes in social networks, which studies how to detect important individuals in human society, has attracted increasing attention from physical and computer science, social science and economics communities. The identification algorithms of spreading influence nodes can be used to evaluate the spreading influence, describe the node’s position, and identify interaction centralities. This review summarizes the recent progress about the identification algorithms of spreading influence nodes from the viewpoint of social networks, emphasizing the contributions from physical perspectives and approaches, including the microstructure-based algorithms, community structure-based algorithms, macrostructure-based algorithms, and machine learning-based algorithms. We introduce diffusion models and performance evaluation metrics, and outline future challenges of the identification of spreading influence nodes.

关键词: complex network     network science     spreading influence     machine learning    

信息网络——现代信息工程学的前沿

钟义信

《中国工程科学》 1999年 第1卷 第1期   页码 24-29

摘要:

信息网络正在各地迅猛崛起,并以史所罕见的规模和速度生长成为世界性社会基础结构,深刻地改变着人们的生产方式、工作方式、学习方式、交往方式、生活方式和思维方式,成为工程学界以至整个社会普遍关注的集点、热点和前沿。文章旨在从理论上廓清信息网络的概念,阐明为什么信息网络对于科学技术的进步、对于世界经济和人类社会的发展能够产生如此巨大和深远的作用与影响。在此基础上,论述信息网络在现代工程学中的作用与地位,以及信息网络工程学在当前的主要研究内容和方向。

关键词: 信息网络     智能化社会生产工具     网络时代     信息网络工程学    

Diffusion of municipal wastewater treatment technologies in China: a collaboration network perspective

Yang Li, Lei Shi, Yi Qian, Jie Tang

《环境科学与工程前沿(英文)》 2017年 第11卷 第1期 doi: 10.1007/s11783-017-0903-0

摘要: Real wastewater treatment technology diffusion process was investigated. The research is based on a dataset of 3136 municipal WWTPs and 4634 organizations. A new metric was proposed to measure the importance of a project in diffusion. Important projects usually involve central organizations in collaboration. Organizations become more central by participating in less important projects. The diffusion of municipal wastewater treatment technology is vital for urban environment in developing countries. China has built more than 3000 municipal wastewater treatment plants in the past three decades, which is a good chance to understand how technologies diffused in reality. We used a data-driven approach to explore the relationship between the diffusion of wastewater treatment technologies and collaborations between organizations. A database of 3136 municipal wastewater treatment plants and 4634 collaborating organizations was built and transformed into networks for analysis. We have found that: 1) the diffusion networks are assortative, and the patterns of diffusion vary across technologies; while the collaboration networks are fragmented, and have an assortativity around zero since the 2000s. 2) Important projects in technology diffusion usually involve central organizations in collaboration networks, but organizations become more central in collaboration by doing circumstantial projects in diffusion. 3) The importance of projects in diffusion can be predicted with a Random Forest model at a good accuracy and precision level. Our findings provide a quantitative understanding of the technology diffusion processes, which could be used for water-relevant policy-making and business decisions.

关键词: Innovation diffusion     Collaboration network     Wastewater treatment plant     Complex network     Data driven    

Negative weights in network time model

Zoltán A. VATTAI, Levente MÁLYUSZ

《工程管理前沿(英文)》 2022年 第9卷 第2期   页码 268-280 doi: 10.1007/s42524-020-0109-1

摘要: Time does not go backward. A negative duration, such as “time period” at first sight is difficult to interpret. Previous network techniques (CPM/PERT/PDM) did not support negative parameters and/or loops (potentially necessitating recursive calculations) in the model because of the limited computing and data storage capabilities of early computers. Monsieur Roy and John Fondahl implicitly introduced negative weights into network techniques to represent activities with fixed or estimated durations (MPM/PDM). Subsequently, the introduction of negative lead and/or lag times by software developers (IBM) apparently overcome the limitation of not allowing negative time parameters in time model. Referring to general digraph (Event on Node) representation where activities are represented by pairs of nodes and pairwise relative time restrictions are represented by weighted arrows, we can release most restraints in constructing the graph structure (incorporating the dynamic model of the inner logic of time plan), and a surprisingly flexible and handy network model can be developed that provides all the advantages of the abovementioned techniques. This paper aims to review the theoretical possibilities and technical interpretations (and use) of negative weights in network time models and discuss approximately 20 types of time-based restrictions among the activities of construction projects. We focus on pure relative time models, without considering other restrictions (such as calendar data, time-cost trade-off, resource allocation or other constraints).

关键词: graph technique     network technique     construction management     scheduling    

Online recognition of drainage type based on UV-vis spectra and derivative neural network algorithm

《环境科学与工程前沿(英文)》 2021年 第15卷 第6期 doi: 10.1007/s11783-021-1430-6

摘要:

• UV-vis absorption analyzer was applied in drainage type online recognition.

关键词: Drainage online recognition     UV-vis spectra     Derivative spectrum     Convolutional neural network    

A neural network-based production process modeling and variable importance analysis approach in corn

《化学科学与工程前沿(英文)》 2023年 第17卷 第3期   页码 358-371 doi: 10.1007/s11705-022-2190-y

摘要: Corn to sugar process has long faced the risks of high energy consumption and thin profits. However, it’s hard to upgrade or optimize the process based on mechanism unit operation models due to the high complexity of the related processes. Big data technology provides a promising solution as its ability to turn huge amounts of data into insights for operational decisions. In this paper, a neural network-based production process modeling and variable importance analysis approach is proposed for corn to sugar processes, which contains data preprocessing, dimensionality reduction, multilayer perceptron/convolutional neural network/recurrent neural network based modeling and extended weights connection method. In the established model, dextrose equivalent value is selected as the output, and 654 sites from the DCS system are selected as the inputs. LASSO analysis is first applied to reduce the data dimension to 155, then the inputs are dimensionalized to 50 by means of genetic algorithm optimization. Ultimately, variable importance analysis is carried out by the extended weight connection method, and 20 of the most important sites are selected for each neural network. The results indicate that the multilayer perceptron and recurrent neural network models have a relative error of less than 0.1%, which have a better prediction result than other models, and the 20 most important sites selected have better explicable performance. The major contributions derived from this work are of significant aid in process simulation model with high accuracy and process optimization based on the selected most important sites to maintain high quality and stable production for corn to sugar processes.

关键词: big data     corn to sugar factory     neural network     variable importance analysis    

网络系统行为效用计算——概念与原理 Article

胡昌振

《工程(英文)》 2018年 第4卷 第1期   页码 78-84 doi: 10.1016/j.eng.2018.02.010

摘要:
网络的服务与应用是一个面向业务与任务的行为过程,其度量与评估目前尚处于粗略比较状态。本文把网络行为场景刻画为微分流形,根据光滑微分流形的同胚变换,给出了网络行为的数学定义,提出了网络行为路径及行为效用的数学描述。基于微分几何原理,提出了网络行为效用的计算原理与方法,并构建了基于行为效用的网络攻防对抗评估计算框架,从而奠定了网络行为定量度量和评估的数学基础。

关键词: 网络度量评估     微分流形     网络行为效用     网络攻防对抗    

PID neural network control of a membrane structure inflation system

Qiushuang LIU, Xiaoli XU

《机械工程前沿(英文)》 2010年 第5卷 第4期   页码 418-422 doi: 10.1007/s11465-010-0117-7

摘要: Because it is difficult for the traditional PID algorithm for nonlinear time-variant control objects to obtain satisfactory control results, this paper studies a neuron PID controller. The neuron PID controller makes use of neuron self-learning ability, complies with certain optimum indicators, and automatically adjusts the parameters of the PID controller and makes them adapt to changes in the controlled object and the input reference signals. The PID controller is used to control a nonlinear time-variant membrane structure inflation system. Results show that the neural network PID controller can adapt to the changes in system structure parameters and fast track the changes in the input signal with high control precision.

关键词: PID     neural network     membrane structure    

网络时代的三大基本定律与骨干网容量的演进

韦乐平

《中国工程科学》 2001年 第3卷 第5期   页码 12-16

摘要:

文章首先分析了新时期的巨大挑战及其对网络的深远影响,特别指出了网络面临的巨大容量压力。接下来分别就主宰网络时代的三个基本定律:摩尔定律、光纤定律和迈特卡尔夫定律的内涵、影响和技术极限进行了论述。最后探讨了骨干网传输链路、传送节点和业务节点的容量演进策略。

关键词: 网络     容量     骨干网    

A new automatic convolutional neural network based on deep reinforcement learning for fault diagnosis

《机械工程前沿(英文)》 2022年 第17卷 第2期 doi: 10.1007/s11465-022-0673-7

摘要: Convolutional neural network (CNN) has achieved remarkable applications in fault diagnosis. However, the tuning aiming at obtaining the well-trained CNN model is mainly manual search. Tuning requires considerable experiences on the knowledge on CNN training and fault diagnosis, and is always time consuming and labor intensive, making the automatic hyper parameter optimization (HPO) of CNN models essential. To solve this problem, this paper proposes a novel automatic CNN (ACNN) for fault diagnosis, which can automatically tune its three key hyper parameters, namely, learning rate, batch size, and L2-regulation. First, a new deep reinforcement learning (DRL) is developed, and it constructs an agent aiming at controlling these three hyper parameters along with the training of CNN models online. Second, a new structure of DRL is designed by combining deep deterministic policy gradient and long short-term memory, which takes the training loss of CNN models as its input and can output the adjustment on these three hyper parameters. Third, a new training method for ACNN is designed to enhance its stability. Two famous bearing datasets are selected to evaluate the performance of ACNN. It is compared with four commonly used HPO methods, namely, random search, Bayesian optimization, tree Parzen estimator, and sequential model-based algorithm configuration. ACNN is also compared with other published machine learning (ML) and deep learning (DL) methods. The results show that ACNN outperforms these HPO and ML/DL methods, validating its potential in fault diagnosis.

关键词: deep reinforcement learning     hyper parameter optimization     convolutional neural network     fault diagnosis    

标题 作者 时间 类型 操作

Novel interpretable mechanism of neural networks based on network decoupling method

期刊论文

A multi-sensor relation model for recognizing and localizing faults of machines based on network analysis

期刊论文

Multiscale computation on feedforward neural network and recurrent neural network

Bin LI, Xiaoying ZHUANG

期刊论文

Heat, mass, and work exchange networks

Zhiyou CHEN, Jingtao WANG

期刊论文

Deep convolutional tree-inspired network: a decision-tree-structured neural network for hierarchical

期刊论文

Identifying spreading influence nodes for social networks

期刊论文

信息网络——现代信息工程学的前沿

钟义信

期刊论文

Diffusion of municipal wastewater treatment technologies in China: a collaboration network perspective

Yang Li, Lei Shi, Yi Qian, Jie Tang

期刊论文

Negative weights in network time model

Zoltán A. VATTAI, Levente MÁLYUSZ

期刊论文

Online recognition of drainage type based on UV-vis spectra and derivative neural network algorithm

期刊论文

A neural network-based production process modeling and variable importance analysis approach in corn

期刊论文

网络系统行为效用计算——概念与原理

胡昌振

期刊论文

PID neural network control of a membrane structure inflation system

Qiushuang LIU, Xiaoli XU

期刊论文

网络时代的三大基本定律与骨干网容量的演进

韦乐平

期刊论文

A new automatic convolutional neural network based on deep reinforcement learning for fault diagnosis

期刊论文