A novel approach for constructing robust Mamdani fuzzy system was proposed, which consisted of an efficiency robust estimator(partial robust M-regression, PRM) in the parameter learning phase of the initial fuzzy syst...A novel approach for constructing robust Mamdani fuzzy system was proposed, which consisted of an efficiency robust estimator(partial robust M-regression, PRM) in the parameter learning phase of the initial fuzzy system, and an improved subtractive clustering algorithm in the fuzzy-rule-selecting phase. The weights obtained in PRM, which gives protection against noise and outliers, were incorporated into the potential measure of the subtractive cluster algorithm to enhance the robustness of the fuzzy rule cluster process, and a compact Mamdani-type fuzzy system was established after the parameters in the consequent parts of rules were re-estimated by partial least squares(PLS). The main characteristics of the new approach were its simplicity and ability to construct fuzzy system fast and robustly. Simulation and experiment results show that the proposed approach can achieve satisfactory results in various kinds of data domains with noise and outliers. Compared with D-SVD and ARRBFN, the proposed approach yields much fewer rules and less RMSE values.展开更多
The atomic force microscope(AFM)can measure nanoscale morphology and mechanical properties and has a wide range of applications.The traditional method for measuring the mechanical properties of a sample does so for th...The atomic force microscope(AFM)can measure nanoscale morphology and mechanical properties and has a wide range of applications.The traditional method for measuring the mechanical properties of a sample does so for the longitudinal and transverse properties separately,ignoring the coupling between them.In this paper,a data processing and multidimensional mechanical information extraction algorithm for the composite mode of peak force tapping and torsional resonance is proposed.On the basis of a tip–sample interaction model for the AFM,longitudinal peak force data are used to decouple amplitude and phase data of transverse torsional resonance,accurately identify the tip–sample longitudinal contact force in each peak force cycle,and synchronously obtain the corresponding characteristic images of the transverse amplitude and phase.Experimental results show that the measured longitudinal mechanical characteristics are consistent with the transverse amplitude and phase characteristics,which verifies the effectiveness of the method.Thus,a new method is provided for the measurement of multidimensional mechanical characteristics using the AFM.展开更多
分类决策问题是人工智能领域的核心问题,因素空间针对这一问题构建了相应的粒度信息决策算法,其中较为典型的有因素分析法及差转计算算法,这两个算法的本质是:根据单一条件信息与决策信息在论域中形成的等价类包含关系诱导出概念知识,...分类决策问题是人工智能领域的核心问题,因素空间针对这一问题构建了相应的粒度信息决策算法,其中较为典型的有因素分析法及差转计算算法,这两个算法的本质是:根据单一条件信息与决策信息在论域中形成的等价类包含关系诱导出概念知识,但存在不能描述不同类型的因素在概念形成过程中的作用的局限。为解决这一问题,本文基于因素空间及商集合理论,定义了因素的析取、合取变换,在此基础上构造了数据粒度变换方法。为验证变换方法的有效性,以同为产生式推理算法的决策树算法为对比算法,在UCI经典数据集Wisconsin Breast Cancer上进行了实例验证,结果表明:数据粒度变换方法是有效的,知识挖掘形成经验推理系统的时间成本要低于决策树,经验推理系统泛化效果同决策树持平。展开更多
基金Project(61473298)supported by the National Natural Science Foundation of ChinaProject(2015QNA65)supported by Fundamental Research Funds for the Central Universities,China
文摘A novel approach for constructing robust Mamdani fuzzy system was proposed, which consisted of an efficiency robust estimator(partial robust M-regression, PRM) in the parameter learning phase of the initial fuzzy system, and an improved subtractive clustering algorithm in the fuzzy-rule-selecting phase. The weights obtained in PRM, which gives protection against noise and outliers, were incorporated into the potential measure of the subtractive cluster algorithm to enhance the robustness of the fuzzy rule cluster process, and a compact Mamdani-type fuzzy system was established after the parameters in the consequent parts of rules were re-estimated by partial least squares(PLS). The main characteristics of the new approach were its simplicity and ability to construct fuzzy system fast and robustly. Simulation and experiment results show that the proposed approach can achieve satisfactory results in various kinds of data domains with noise and outliers. Compared with D-SVD and ARRBFN, the proposed approach yields much fewer rules and less RMSE values.
基金This project is supported by the General Program of the National Natural Science Foundation of China(62073227)the National Natural Science Foundation of China(61927805 and 61903359).
文摘The atomic force microscope(AFM)can measure nanoscale morphology and mechanical properties and has a wide range of applications.The traditional method for measuring the mechanical properties of a sample does so for the longitudinal and transverse properties separately,ignoring the coupling between them.In this paper,a data processing and multidimensional mechanical information extraction algorithm for the composite mode of peak force tapping and torsional resonance is proposed.On the basis of a tip–sample interaction model for the AFM,longitudinal peak force data are used to decouple amplitude and phase data of transverse torsional resonance,accurately identify the tip–sample longitudinal contact force in each peak force cycle,and synchronously obtain the corresponding characteristic images of the transverse amplitude and phase.Experimental results show that the measured longitudinal mechanical characteristics are consistent with the transverse amplitude and phase characteristics,which verifies the effectiveness of the method.Thus,a new method is provided for the measurement of multidimensional mechanical characteristics using the AFM.
文摘分类决策问题是人工智能领域的核心问题,因素空间针对这一问题构建了相应的粒度信息决策算法,其中较为典型的有因素分析法及差转计算算法,这两个算法的本质是:根据单一条件信息与决策信息在论域中形成的等价类包含关系诱导出概念知识,但存在不能描述不同类型的因素在概念形成过程中的作用的局限。为解决这一问题,本文基于因素空间及商集合理论,定义了因素的析取、合取变换,在此基础上构造了数据粒度变换方法。为验证变换方法的有效性,以同为产生式推理算法的决策树算法为对比算法,在UCI经典数据集Wisconsin Breast Cancer上进行了实例验证,结果表明:数据粒度变换方法是有效的,知识挖掘形成经验推理系统的时间成本要低于决策树,经验推理系统泛化效果同决策树持平。
