基于振动频谱的磨矿过程球磨机负荷参数集成建模方法

以磨矿过程的湿式球磨机为背景, 针对传统磨机负荷(ML)检测方法只能依靠灵敏度较低的轴承振动、筒体振声和磨机功率等信号监督判断ML状态, 难以检测磨机内部负荷参数的问题, 提出了一种基于高灵敏度的筒体振动频谱的集成建模方法. 首先, 依据磨矿过程的研磨机理, 将振动频谱采用波峰聚类方法自动划分具有不同物理意义的分频段; 然后利用核偏最小二乘(KPLS)算法分别建立各分频段的ML参数子模型; 最后, 依据子模型训练数据预测误差的信息熵获得初始权重, 加权得到最终的ML参数集成预测模型; 在线使用中则根据子模型预测误差的变化进行权值的在线自适应更新. 仿真结果证明了该方法的有效性.     

Feature extraction based on the fractional Fourier transform for vibration signals with application to measuring the load of a tumbling mill

Feature extraction plays a major role in data preprocessing and soft sensing. In this paper, a feature extraction method based on the fractional Fourier transform (FrFT) is proposed to estimate the load of a tubular ball mill. The FrFT, a generalised form of FFT, can draw intermediate time–frequency representations for a signal. Vibration signals are first transformed into a frequency spectrum by using the FrFT with multiple-orders, yielding multi-order frequency spectrums (MFS). Then, according to the sensitivity thresholds under different wear conditions, multi-order characteristic frequencies (MCF) are selected from the MFS, and some statistical characteristics are extracted from MCF as initial features. Only informative features can significantly benefit the forecasting accuracy. Therefore, some candidate features used to build the soft-sensing model are chosen from statistical characteristics. Finally, the load of a tumbling ball mill can be obtained by fusing the results of multi-order regression models based on an adaptive-weighting coefficient. Compared with some feature extraction methods and the performance of the proposed method in the Fourier domain, experimental results show that the feature extraction method proposed is effective in modelling the load parameters of a tubular ball mill.     

基于多尺度振动和振声频谱特征自适应提取与选择的磨机负荷参数软测量

针对磨机筒体振动/振声多尺度频谱与磨机负荷参数间的模糊特性、多源多尺度频谱间的冗余性与互补性,以及现有文献中潜结构选择性集成模型难以模拟运行专家“听音”推理识别磨机负荷参数等问题,提出一种基于多尺度振动/振声频谱特征自适应提取与选择的磨机负荷参数软测量模型.该方法首先进行多尺度频谱获取,然后基于核潜结构映射和互信息进行多尺度振动和振声频谱特征的自适应提取和选择,最后采用同步聚类算法、Madani模糊模型以及基于分支定界和自适应加权融合算法的选择性集成学习策略构建磨机负荷参数软测量模型.通过实验球磨机验证了所构建的软测量模型能够模拟运行专家的模糊推理机制,具有较好的建模精度.     

基于ASOS-ELM的球磨机负荷参数软测量系统设计

球磨机负荷参数是决定磨矿工作质量和效率的重要指标,为了提升球磨机负荷参数软测量系统的运行性能,降低综合测量误差,利用ASOS-ELM算法,从硬件和软件两个方面,设计球磨机负荷参数软测量系统。加设球磨机工作数据采集器,改装处理器与输出显示屏,扩大存储器空间,利用电源电路连接硬件设备,完成硬件系统的设计。根据球磨机的组成结构和工作原理,构建球磨机数学模型,在该模型下,采集球磨机不同工况下的振动信号。利用ASOS-ELM算法,提取振动信号特征,得出球磨机球负荷、料负荷、水负荷参数的软测量结果,实现球磨机负荷参数软测量系统设计。通过系统测试实验结果表明,设计系统负荷参数的综合测量误差降低了约0.064kg,且系统运行性能得到明显提升。     

Feature extraction and selection from acoustic emission signals with an application in grinding wheel condition monitoring

Feature extraction and feature selection are two important issues in sensor-based condition monitoring of any engineering systems. In this study, acoustic emission signals were first collected during grinding operations, next processed by autoregressive modeling or discrete wavelet decomposition for feature extraction, and then the best feature subsets are found by three different feature selection methods, including two proposed ant colony optimization (ACO)-based method and the famous sequential forward floating selection method. Posing monitoring as a classification problem, the evaluation is carried out by the wrapper approach with four different algorithms serving as the classifier. Empirical test results were shown to illustrate the effectiveness of feature extraction and feature selection methods.     

选择性融合多尺度筒体振动频谱的磨机负荷参数建模

针对目前采用经验模态分解(empirical model decomposition,EMD)得到的系列子信号构建的磨机负荷参数软测量模型泛化性能差、难以进行清晰物理解释,以及EMD算法存在的模态混叠等问题,本文提出了基于选择性融合多尺度筒体振动频谱的建模方法.首先采用EMD、集合EMD(ensemble EMD,EEMD)、希尔伯特振动分解(Hilbert vibration decomposition,HVD)共3种多组分信号自适应分解算法获得磨机筒体振动多尺度子信号的集合,接着通过相关性分析剔除虚假无关部分,然后再将与原始信号相关性强的那部分多尺度子信号变换至频域,进而更有利于构建这些多尺度频谱与磨机负荷参数间的映射模型,最后通过改进分支定界选择性集成(improved branch and bound based selective ensemble,IBBSEN)算法建立软测量模型,实现对多源多尺度筒体振动频谱的最优选择性信息融合.基于实验球磨机运行数据的仿真实验表明所提方法在模型可解释性和泛化性能上均优于之前研究所提出方法.     

基于功率谱分析的球磨机负荷模型

建立了基于功率谱分析的球磨机负荷模型,分析了球磨机负荷、磨音声强和磨音频谱分布三者之间的关系:磨音检测点处的声压与球磨机筒体振动速度的平方成正比,声音频率与球磨机筒体振动频率成正比,随着球磨机负荷的增加,磨音信号逐渐减弱;磨音频谱主要分布在1 500Hz以下,当球磨机正常运行时,磨音频谱分布较为均匀,当球磨机欠磨运行时,磨音频谱主要分布在0～1 000Hz。Matlab仿真结果验证了所建立的球磨机负荷模型的可行性和有效性。     

基于即时学习策略的火电厂球磨机负荷软测量

针对电厂球磨机负荷难以进行有效预测的问题,从提高预测模型在线自适应能力的角度出发,提出一种基于即时学习策略的改进SVM建模方法。利用灰色关联分析方法对过程参数进行优化筛选,获得辅助变量;在即时学习策略建模框架下,采用多种群混合优化算法进行SVM预测模型参数的优化选取;基于电厂实际运行数据进行了仿真研究。仿真实验表明,与标准BP神经网络和SVM建模方法的比较,该算法具有更好的预测性能,虽然计算开销有所增加,但能够满足制粉系统球磨机负荷检测的实时性要求。     

基于SVD和TKEO的轴承振动信号特征提取

为了解决滚动轴承振动信号中微弱故障信息难以提取的问题,提出了一种基于奇异值分解（Singular Value Decomposition,SVD）和Teager-Kaiser能量算子（Teager-Kaiser Energy Operator,TKEO）的轴承振动信号特征提取方法。采用SVD将突变信息从背景噪声和光滑信号中分离,提取信号的突变信息;利用TKEO计算突变信息的瞬时能量,对该能量信号进行频谱分析,从而提取出轴承振动信号的能量频谱特征,用于故障检测。将该方法应用于轴承外圈、内圈局部故障状态下的振动信号特征提取,利用特征信息能够准确检测并识别出故障类型,表明了该方法的可行性和有效性。     

基于PSO的LS-SVM特征选择与参数优化算法

针对最小二乘支持向量机特征选择及参数优化问题,提出了一种基于PSO的LS-SVM特征选择与参数同步优化算法。首先产生若干种群（特征子集）,然后用PSO算法对特征及参数进行优化。在UCI标准数据集上进行的仿真实验表明,该算法可有效地找出合适的特征子集及LS-SVM参数,且与基于遗传算法的最小二乘支持向量机算法（GALS-SVM）和传统的LS-SVM算法相比具有较好的分类效果。     

基于振动信号和云推理的球磨机负荷软测量

采用振动信号对球磨机料位进行测量时,特征值具有散度大、随机性强的特点.对此,基于具有将随机性、模糊性与稳定倾向性相结合能力的云模型,提出一种利用云模型对球磨机料位进行概念表示和推理测量的方法.首先,利用逆向云发生器对振动信号的功率谱特征值进行概念提取以获得前件云;然后,由料位值信息建立相对应的后件云;最后,利用云模型的不确定推理实现球磨机料位的软测量.对比实验结果表明了所提出方法的有效性和可行性.     

基于奇异值分解的掘进机振动信号特征量提取

针对掘进机动载荷识别难度大的问题,提出了基于奇异值分解的掘进机振动信号特征量提取方法。对采集的振动信号进行小波包分解,重构底层各频带节点系数,进而构造时频矩阵;对该矩阵进行奇异值分解,并基于Fisher判据,利用基于散度矩阵的类可分性准则,选择对不同截割岩壁硬度较为敏感的奇异值作为振动信号的特征量,并利用散度矩阵准则值来解决无法定量衡量各阶奇异值对截割硬度敏感程度的问题。与小波包频带能量法提取的特征向量进行比较,结果表明,对于掘进机水平截割、垂直截割和纵向钻进3种工况下的振动信号,基于奇异值分解法提取的特征向量都具有更好的类可分性。     

Feature selection for fault detection systems: application to the Tennessee Eastman process

In fault detection systems, a massive amount of data gathered from the life-cycle of equipment is often used to learn models or classifiers that aims at diagnosing different kinds of errors or failures. Among this huge quantity of information, some features (or sets of features) are more correlated with a kind of failure than another. The presence of irrelevant features might affect the performance of the classifier. To improve the performance of a detection system, feature selection is hence a key step. We propose in this paper an algorithm named STRASS, which aims at detecting relevant features for classification purposes. In certain cases, when there exists a strong correlation between some features and the associated class, conventional feature selection algorithms fail at selecting the most relevant features. In order to cope with this problem, STRASS algorithm uses k-way correlation between features and the class to select relevant features. To assess the performance of STRASS, we apply it on simulated data collected from the Tennessee Eastman chemical plant simulator. The Tennessee Eastman process (TEP) has been used in many fault detection studies and three specific faults are not well discriminated with conventional algorithms. The results obtained by STRASS are compared to those obtained with reference feature selection algorithms. We show that the features selected by STRASS always improve the performance of a classifier compared to the whole set of original features and that the obtained classification is better than with most of the other feature selection algorithms.     

一种基于LS-SVM的特征提取新方法及其在智能质量控制中的应用

提出一种基于最小二乘支持向量机(LS-SVM)的特征提取新方法,并将其成功应用到智能质量控制领域。首先,将线性特征提取公式表达成与LS-SVM回归算法中相同的形式;再遵循SVM方法将数据集由原输入空间映射到高维特征空间,进而使用该技巧通过线性形式实现非线性特征提取。然后,用常规控制图提取出一个含有6种模式、50维特征的仿真数据集用于测试,通过LS-SVM特征提取后,原数据集的特征被降到了3维并保留了原80%的分类信息。最后,用BP分类器对特征提取后的样本进行识别,其结果优于新型RSFM网络直接对原始样本进行识别的效果。仿真实验结果表明了LS-SVM特征提取方法的可行性和有效性。     

基于最大均值差异多源域迁移学习的 湿式球磨机负荷参数软测量

针对湿式球磨机工况改变时,实时数据与建模数据分布不一致,不满足数据同分布的假设,传统软测量模型难以适应数据分布变化,造成模型性能恶化的问题,有针对性地引入迁移学习策略,并通过多源域集成机制提高模型的鲁棒性,实现多工况下湿式球磨机负荷参数测量.首先,对多工况数据进行预处理并提取频谱特征,经过联合分布适配对多工况数据进行边缘、条件分布适配;然后,使用最大均值差异对适配后的数据进行分布度量并为源域构建的回归器加权;最后,对目标域数据进行负荷预测.通过对比实验与交叉实验表明了模型的实用性和有效性.     

一种基于核函数特征提取改进方法的应用

针对标准的核典型相关分析（KCCA）方法在对训练样本增大的情况下相应计算机复杂度剧增、内存占用量大的缺陷,在对标准的KCCA特征提取方法分析推导的基础上,提出了一种改进的核函数特征提取方法。该方法首先根据特征值的大小对训练样本重要程度进行判断,进而完成对应特征向量的提取;然后通过与SVDD分类器的结合,在对图像识别率影响不大的情况下,提升了对图像特征提取的效率,节省了系统的存储量;最后通过在Yale标准人脸库上进行仿真对比实验,验证了该方法的可行性,从而为提高图像模式识别效率提供了一种有效的途径。     

Effects of the friction coefficient of a barrel on the grinding performance of a planetary ball mill

Microsystem Technologies - We investigated the effect of the friction coefficients of a barrel on the grinding performance of multi-layer ceramic capacitors (MLCCs) in a planetary ball mill. We...     

基于加窗光谱积分的高光谱遥感图像特征提取

提出了一种基于加窗光谱积分的高光谱遥感图像特征提取算法.该算法借用子波变换多分辨率分析的思想,设计了一组波段相互重叠的窗函数来提取光谱曲线特征,然后进行有监督RBF神经网络分类实验,在实验过程中确定了相关参数的取值范围.实验结果表明,加窗光谱积分特征可以有效地描述光谱曲线,获得了比较好的正确分类率.     

基于EMD与奇异值差分谱的侧信道信号特征提取

为了从强烈的背景噪声中提取侧信道信号的特征信息,提出了一种基于经验模式分解（EMD）与奇异值差分谱相结合的信号特征提取方法。该方法首先对原始侧信道信号进行EMD分解,计算各个特征模态函数（IMF）与原始信号的相关系数,找到最大相似特征分量;再对该分量进行奇异值分解求出对应的奇异值差分谱;最后根据差分谱进行重构和消噪,进一步提取分量的特征信息。实验结果表明,该方法可以有效应用于侧信道信号的特征提取,成功提高信号的信噪比和攻击成功率。     

Combining rough set and case based reasoning for process conditions selection in camshaft grinding

Case Based Reasoning (CBR) is a novel paradigm that uses previous cases to solve new, unseen and different problems. However, redundant features may not only dramatically increase the case memory, but also make the case retrieval more time-consuming. Furthermore, camshaft grinding process is controlled by a number of process parameters, and it is more complex comparing with the ordinary cylindrical grinding. The process conditions are achieved by skilled and professional workers. Therefore, this research combines Rough set (RS) and CBR for process conditions selection in camshaft grinding, and Genetic Algorithm (GA) is developed to discretize condition features. Through the approach an optimal subset of process conditions can be selected quickly and effectively from a large database with a lot of cases, and complexity of computation of the similarity testing is significantly reduced. Moreover, the validity of the proposed solution is verified by the application of practical experiments for the process conditions selection in camshaft grinding.