Blind vibration component separation and nonlinear feature extraction applied to the nonstationary vibration signals for the gearbox multi-fault diagnosis

Fault diagnosis of gearboxes, especially the gears and bearings, is of great importance to the long-term safe operation. An unexpected damage on the gearbox may break the whole transmission line down. It is therefore crucial for engineers and researchers to monitor the health condition of the gearbox in a timely manner to eliminate the impending faults. However, useful fault detection information is often submerged in heavy background noise. Thereby, a new fault detection method for gearboxes using the blind source separation (BSS) and nonlinear feature extraction techniques is presented in this paper. The nonstationary vibration signals were analyzed to reveal the operation state of the gearbox. The kernel independent component analysis (KICA) algorithm was used hereby as the BSS approach for the mixed observation signals of the gearbox vibration to discover the characteristic vibration source associated with the gearbox faults. Then the wavelet packet transform (WPT) and empirical mode decomposition (EMD) nonlinear analysis methods were employed to deal with the nonstationary vibrations to extract the original fault feature vector. Moreover, the locally linear embedding (LLE) algorithm was performed as the nonlinear feature reduction technique to attain distinct features from the feature vector. Lastly, the fuzzy k-nearest neighbor (FKNN) was applied to the fault pattern identification of the gearbox. Two case studies were carried out to evaluate the effectiveness of the proposed diagnostic approach. One is for the gear fault diagnosis, and the other is to diagnose the rolling bearing faults of the gearbox. The nonstationary vibration data was acquired from the gear and rolling bearing fault test-beds, respectively. The experimental test results show that sensitive fault features can be extracted after the KICA processing, and the proposed diagnostic system is effective for the multi-fault diagnosis of the gears and rolling bearings. In addition, the proposed method can achieve higher performance than that without KICA processing with respect to the classification rate.     

液液萃取-GC/MS法分析地表水中的多氯联苯

建立了液液萃取-GC/MS联用分析地表水中的多氯联苯的方法,并对实际水样进行了测定。各目标物的线性回归R值均大于0.999,MDL为0.08-3.9ng/L,RSD为0.25%-6.2%,回收率为87.6%-113%。     

基于LLE降维和BP_Adaboost分类器的GIS局部放电模式识别

在气体绝缘组合电器(gas insulated switchgear,GIS)实体模型中分别放置了针-板、悬浮金属颗粒和绝缘子表面固定金属颗粒放电模型,用超声波传感器采集到其放电波形。对放电波形提取的特征向量进行局部线性嵌入(local linear embedding,LLE)算法降维处理,用降维后的向量作为输入对BP_Adaboost分类器进行训练和测试类型识别。识别结果表明,用这样方法进行GIS绝缘缺陷类型识别可以在减少计算量的同时保持较高的识别率,说明了其在局部放电模式识别应用中的有效性。     

基于近邻消息传递的自适应局部线性嵌入

针对局部线性嵌入(LLE)降维算法中邻域参数的人工设定及其全局性的缺陷,研究了聚类和降维的内在联系,提出了邻域参数的自适应选取策略,构建了一种聚类和降维的自适应局部线性嵌入(ALLE)算法,为每个样本点设计最佳的近邻搜索空间,自适应选取邻近点计算权值重建矩阵,基于类信息重新定义了重构误差函数。实验表明,新算法更能体现出数学上流形概念的局部坐标化本质,对不相关数据、冗余数据和噪声数据具有良好的鲁棒性,在实际识别问题中体现出优越的性能。     

液液萃取与固相微萃取二锅头香气成分的比较

采用液液萃取和固相微萃取结合气相色谱-质谱联用技术对清香型牛栏山二锅头的香气成分进行了定性分析,共得到42种香气成分,主要包括酯类化合物17种,酸类化合物8种,醇类化合物7种,杂环类化合物5种、烷烃类化合物4种和醛类化合物1种。液液萃取主要呈香物质有3-甲基-1-丁醇、十六酸乙酯、2-羟基-丙酸乙酯、油酸乙酯和亚油酸乙酯。固相微萃取主要呈香物质有3-甲基-1-丁醇、己酸乙酯、3-甲基-1-丁醇乙酸、辛酸乙酯和乙酸乙酯。液液萃取和固相微萃取对香气成分的分离效果存在较大差异,表明两种方法对白酒的定性分析有一定的互补性。     

甲基丙烯酸甲酯-对二甲苯-水体系的盐效应

在４０℃的条件下，研究了甲基丙烯酸甲酯－对二甲苯－水体系的盐效应。试验结果表明硫酸氢铵对该体系的ＭＭＡ产生盐析效应。用Ｅｉｓｅｎ－Ｊｏｆｆｅ方程对该含盐体系的ＬＬＥ数据进行关联，关联的值与试验值符合得较好。     

甲基环己烷－苯－DMF－水液液平衡的研究

采用作者设计的液液平衡釜，于常压下测定了甲基环己烷（１）－ＤＭＦ（２）二元系（２０．０～４７．９℃）的互溶度数据，以及常压、２０．Ｏ℃和３５．Ｏ℃下甲基环己烷（１）－苯（２）－ＤＭＦ（３）三元系和甲基环己烷（１）－苯（２）－ＤＭＦ（３）－水（４）（ＤＭＦ／Ｈ２Ｏ＝４∶１，ｍａｓｓ比）四元系的ＬＬＥ数据。     

LLE算法及其应用

LE算法针对非线性降维问题,利用线性重构的局部对称性找出高维数据空间中的非线性结构.并在保持各数据点临近位置关系情况下,把高维空间数据点映射为低维空间对应的数据点.其计算步骤包括:计算、寻找数据点或邻居数据点、构造数据点及计算权值矩阵,并通过权值矩阵计算低维向量.     

Molecular thermodynamics approach on phase equilibria of dendritic polymer systems

We suggest a molecular thermodynamic framework to describe the phase behavior of dendritic polymer systems. The proposed model, which is based on the lattice cluster theory, contains correlations of molecular structure and specific interactions such as hydrogen bonding to the phase equilibria of branch-structured polymer systems. We examine liquid-liquid equilibria (LLE) of hyperbranched polymer solutions and vapor-liquid equilibria (VLE) of dendrimer solutions in the viewpoints of effects of a branched structure and specific interaction formations among endgroups of dendritic polymer and solvent molecules. We investigate VLE of dendrimer/solvent (Benzyl Ether Dendrimer/Toluene) systems by the combination of a new lattice-based model and atomistic simulation technique. The interaction energy parameters are obtained by the pairs method [Baschnagel et al., 1991] including Monte Carlo simulation with excluded volume constraint. In the pairs method [Baschnagel et al., 1991], we do not simulate the whole molecule as in molecular dynamics or molecular mechanics, but only monomer segments interacting with solvent molecules. The proposed model shows improvements in prediction for both phase equilibria (VLE and LLE) due to the branched structure and specific interaction due to endgroups at periphery of dendritic polymer molecule. Atomic simulation technique gives good result in prediction without fitting variables. Our results show that the specific interactions between the endgroup and the solvent molecule play an important role in phase behavior of the given systems.     

一种新的有监督流形学习方法

提出了一种新的有监督流形学习方法,目的是提供将流形学习降维方法高效应用于有监督学习问题的全新策略.算法的核心思想是集成流形学习方法对高维流形结构数据的降维有效性与支撑向量机(SVM)在中小规模分类数据集上的优良特性实现高效有监督流形学习.算法具体实现步骤为:首先利用SVM在流形学习降维数据中选出对分类决策最重要的数据集,即支撑向量集;按标号返回可得到原空间的支撑向量集;在这个集合上再次使用SVM即可得到原空间的分类决策,从而完成有监督流形学习.在一系列人工与实际数据集上的实验验证了方法的有效性.