张泽银 - 浙江大学 - 数学科学学院

个人简介

张泽银，浙江大学教授、博士（后）。 1997-至今，浙江大学，工作； 1995-1997，浙江大学，博士后； 1991-1994，武汉大学，博士。专业方向： 1、复杂系统安全监控与优化控制、目标监控与轨迹优化； 2、大数据与人工智能、小波分析应用与图象处理。先后主持或参与完成国家自然科学基金重点项目和面上项目、国家863计划、中国人民解放军总装备部基金、国家航天科工创新基金、国家知识产权局专利实施许可项目、中石化科技攻关项目、浙江省自然科学基金、浙江省科技厅国际合作等国家、国际纵向项目20余项。相关科研成果，已在International Journal of Control,Journal of the Franklin Institute,Engineering Optimization,Journal of Chemometrics,Journal of Applied Polymer Science,Applied SoftComputing等国际高水平学术期刊发表SCI/EI论文70余篇，其中SCI论文30余篇(近3年SCI论文近20篇)，获浙江省科技进步二等奖。项目介绍展示项目名称：基于相关失效理论的采煤机整机动态最优可靠性研究 Project title:Dependent failure theory based researches on optimal dynamic reliability of the whole shearer 编号：NSFC 61603336 Granted No：NSFC 61603336 作为煤矿综采工作面生产核心设备的采煤机，其可靠性不仅事关煤炭的产量，更事关工作面人员设备的安全；然而，长期以来采煤机事故在国内外屡有发生，除人为因素外，都是因为采煤机长期大载荷甚至超负荷工作，导致其有效寿命期内某些关键系统或零部件就已经濒临失效，从而导致采煤事故的发生，其本质原因在于：采煤机作为机械、电气、液压系统高度一体化的大型复杂设备，其各部件的失效具有普遍的相关性。忽略失效的相关性，进行零部件和系统的可靠性分析与计算将会导致较大的误差，甚至得出错误的结论；另外，采煤机的运行是一个典型的动态过程，传统的可靠性分析理论与方法大多针对的是系统的静态特性，而实际生产过程中，工作载荷的变化将直接影响到采煤机零部件与系统的可靠性。因此有必要在考虑相关失效的前提下，对随工作载荷而变化的采煤机动态可靠性进行研究。本项目，因此提出了基于相关失效理论的采煤机整机动态最优可靠性研究，并以采煤机运行最为重要的采煤机切削运行模式的可靠性为具体对象展开研究。首先，基于相关失效的应力-强度干涉理论、Copula函数和Sklar定理，建立了采煤机整机切削模式可靠性模型，然而研究结果表明其可靠性预报效果不理想，表明采煤机各切削模式可靠性之间的非线性十分强烈，使得基本相关失效理论的Copula-Sklar方法难以表征，混沌理论对各切削模式的非线性的研究结果也表明了其强烈的非线性特征。因此进一步引入了对非线性预报具有良好效果的非线性机器学习理论对可靠性进行研究，并进一步提出了群智能优化的采煤机切削运行可靠性OCS在线校正策略，提出了多种非线性模型参数辨识方法和动态优化寻优方法，以克服预报模型参数人为选择因素对结果的影响和采煤机切削运行过程不稳定性所导致的模型失配问题，实际生产数据运行结果（详见表1）和国际报道结果的比较研究（详见表2）都表明了所提出的动态最优可靠性预测方法的有效性。 Shearer is the core equipment of integrated mining work surface in the mining production,whose reliability is not only related to the coal production,but also related to the safety of personnel and equipment.However,shearer accidents occur frequently both at home and abroad for a long time,in addition to human factors,all because of shearer long-term big load and overload work,lead to the useful life period of some key systems or components is on the verge of failure,leading to shearer accident.The essential reason is that:as the mechanical,electrical,hydraulic system of shearer highly integration of large complex equipment,the failure of its components are of general correlation.Ignoring the correlation of failure,the reliability analysis and calculation of parts and systems will lead to large errors or even draw wrong conclusions.In addition,the operation of shearer is a typical dynamic process,the traditional reliability analysis theory and method are mostly aimed at the static characteristics of the system,but in the actual production process,the change of working load will directly affect the reliability of shearer parts and system.Therefore,it is necessary to study the dynamic reliability of the shearer which changes with the working load under the premise of considering the related failure.In this project,the dynamic optimal reliability study of the whole shearer based on the relevant failure theory is proposed,and the reliability of the cutting operation mode of the shearer,which is the most important shearer operation mode,is taken as the specific object to carry out the research.At first,based on the failure stress-strength interference theory,copulas connect function and Sklar theorem,the shearer whole cutting model reliability model is established,however,the results show that the reliability prediction effect is bad,showing that the nonlinear between shearer cutting model reliability is very strong,which makes the basic theory of copulas connect related failure-Sklar method is difficult to characterize,chaos theory study of the cutting model of the nonlinear results also prove that the strong nonlinear characteristics.Furthermore,the nonlinear prediction with good effect of the non-linear machine learning theory study is introduced to the research of shearer reliability,and further puts forward the swarm intelligence optimization of shearer cutting operation reliability of OCS online correction strategy,combining with the nonlinear model parameters identification method and the dynamic optimization method,in order to overcome the forecast model parameter selective factors on the results and the influence of shearer cutting operation instability caused by model mismatch problem,the running result of the actual production data(please see Table 1)and comparison research with report(please see Table 2)the results show that the proposed dynamic optimal reliability prediction method is effective. 表1带OCS的在线智能各切削模式可靠性预报准确率切削模式 OCS-CMPSO-WLSSVM OCS-MGSA-RVM OCS-CDA-GP 模式1 98.71% 99.20% 98.46% 模式2 97.23% 98.86% 97.42% 模式3 97.92% 98.51% 97.68% 模式4 98.03% 98.92% 97.99% 模式5 98.05% 98.13% 97.80% 模式6 98.17% 98.54% 98.03% 整体可靠性准确率 98.02% 98.69% 97.90% 表2与已人工智能方法的准确率对比方法整体可靠性准确率 EMD-NN[1](J.Vib.Meas.Diagnosis,2012) 94.40% FCM-FGOA[2](Appl.Sci.,2016) 95.40% MFOA-PNN[3](Sensors,2016) 97.50%* OCS-CDA-GP 97.90% OCS-CMPSO-WLSSVM 98.02% OCS-MGSA-RVM 98.69%** *文献报道最好结果 **本课题研究最好结果

研究领域

1、复杂系统安全监控与优化控制、目标监控与轨迹优化 2、大数据与人工智能、小波分析应用与图像处理

近期论文

查看导师最新文章（温馨提示：请注意重名现象，建议点开原文通过作者单位确认）

近几年代表性SCI论文如下： [1]ZY Zhang,T Wang,X Liu.Melt Index Prediction by Aggregated RBF Neural Networks Trained with Chaotic Theory.Neurocomputing,131:368-376,2014.(SCI) [2]L Xiao,X Liu,L Ma,ZY Zhang.An effective pseudospectral method for constraint dynamic optimisation problems with characteristic times,International Journal of Control,91(3):571-581,2018.(Top SCI) [3]P Liu,X Liu,ZY Zhang,Y Wang,C Yang,W Gui.Production optimization for concentration and volume-limited fed-batch reactors in biochemical processes.Bioprocess and Biosystems Engineering,41(3):407-422,2018.(SCI) [4]Y Sun,X Liu,ZY Zhang,Quality prediction via semi-supervised Bayesian regression with application to propylene polymerization,Journal of Chemometrics,DOI:10.1002/cem.3052,2018.(SCI) [5]L Wang,X Liu,ZY Zhang.An efficient interior-point algorithm with new non-monotone line search filter method for nonlinear constrained programming.Engineering Optimization 49(2):290-310,2017.(SCI) [6]L Wang,X Liu,ZY Zhang.A new sensitivity-based adaptive control vector parameterization approach for dynamic optimization of bioprocesses.Bioprocess and Biosystems Engineering 40(2):181-189,2017.(SCI) [7]W Wang,H Chen,M Zhang,X Liu,ZY Zhang,Youxian Sun.Application of Takagi-Sugeno fuzzy model optimized with an improved free search algorithm to industrial polypropylene melt index prediction.Transactions of the Institute of Measurement and Control,39(11):1613-1622,2017.(SCI) [8]J Wu,X Liu,Y Zhu,Z Wang,Z Song,W Wang,ZY Zhang,J Zhou,Y Yu,Z Xu,T Zhang.A Novel Ship Classification Approach for High Resolution SAR Images Based on BDA-KELM Classification Model.International Journal of Remote Sensing,38(23):6457-6476,2017.(SCI) [9]Y Sun,Y Wang,X Liu,C Yang,ZY Zhang,W Gui,X Chen,B Zhu.A novel Bayesian inference soft sensor for real-time statistic learning modeling for industrial polypropylene melt index prediction.Journal of Applied Polymer Science,134(40):45384,2017.(Top SCI) [10]L Xiao,P Liu,X Liu,ZY Zhang,Y Wang,C Yang,W Gui,X Chen&B Zhu.Sensitivity-based adaptive mesh refinement collocation method for dynamic optimization of chemical and biochemical processes.Bioprocess and Biosystems Engineering,40(9),1375-1389,2017.(SCI) [11]P Liu,G Li,X Liu,ZY Zhang.Novel non-uniform adaptive grid refinement control parameterization approach for biochemical processes optimization,Biochemical Engineering Journal,111:63-74,2016.(SCI) [12]P Liu,G Li,X Liu,ZY Zhang.A novel fast dynamic optimization approach for complex multivariable chemical process systems.Canadian Journal of Chemical Engineering,94(12):2355-2363,27 2016.(SCI) [13]M Zhang,X Liu,ZY Zhang.A soft sensor for industrial melt index prediction based on evolutionary extreme learning machine,Chinese Journal of Chemical Engineering,24:1013–1019 2016.(SCI) [14]Y Sun,X Liu,ZY Zhang,Z Wang,Y Yu,T Zhang,Y Zhu,Z Song.A sparse probabilistic approach with chaotic artificial bee colony optimization for sea clutter soft computing.Applied Soft Computing,49:108–119,2016.(SCI) [15]P Zhang,H Chen,X Liu,ZY Zhang.An iterative multi-objective particle swarm optimization-based control vector parameterization for state constrained chemical and biochemical engineering problems.Biochemical Engineering Journal,103:138-151,2015.(SCI) [16]T Wang,X Liu,ZY Zhang.Characterization of chaotic multiscale features on the time series of Melt Index in industrial propylene polymerization system.Journal of the Franklin Institute,351(2):878-906,2014.(Top SCI)