A hybrid deep learning model for robust prediction of the dimensional accuracy in precision milling of thin-walled structural components

The use of artificial intelligence to process sensor data and predict the dimensional accuracy of machined parts is of great interest to the manufacturing community and can facilitate the intelligent production of many key engineering components. In this study, we develop a predictive model of the dimensional accuracy for precision milling of thin-walled structural components. The aim is to classify three typical features of a structural component—squares, slots, and holes—into various categories based on their dimensional errors (i.e., “high precision,” “pass,” and “unqualified”). Two different types of classification schemes have been considered in this study: those that perform feature extraction by using the convolutional neural networks and those based on an explicit feature extraction procedure. The classification accuracy of the popular machine learning methods has been evaluated in comparison with the proposed deep learning model. Based on the experimental data collected during the milling experiments, the proposed model proved to be capable of predicting dimensional accuracy using cutting parameters (i.e., “static features”) and cutting-force data (i.e., “dynamic features”). The average classification accuracy obtained using the proposed deep learning model was 9.55% higher than the best machine learning algorithm considered in this paper. Moreover, the robustness of the hybrid model has been studied by considering the white Gaussian and coherent noises. Hence, the proposed hybrid model provides an efficient way of fusing different sources of process data and can be adopted for prediction of the machining quality in noisy environments.     

A hybrid intelligent multi-fault detection method for rotating machinery based on RSGWPT,KPCA and Twin SVM

This paper proposes a hybrid intelligent method for multi-fault detection of rotating machinery, in which three methods, i.e. including the redundant second generation wavelet package transform (RSGWPT), the kernel principal component analysis (KPCA) and the twin support vector machine (TWSVM), are combined. Firstly, RSGWPT is used to extract feature vectors from representative statistical characteristics in the decomposition frequency band, and then the KPCA in the feature space is performed to reduce the dimension of features and to extract the dominant features for the following classification. Finally, a novel support vector machine, called twin support vector machine is used to construct a multi-class classifier. Inputting superior features to this classifier, the condition of the monitored machine component can be determined. Experimental results demonstrate that the proposed hybrid method is effective for multi-fault detection of rotating machinery. The TWSVM is also indicated that has better classification performance and faster convergence speed than the normal SVM.     

Fault diagnosis of rotating machine by isometric feature mapping

Principal component analysis (PCA) and linear discriminate analysis (LDA) are well-known linear dimensionality reductions for fault classification. However, since they are linear methods, they perform not well for high-dimensional data that has the nonlinear geometric structure. As kernel extension of PCA, Kernel PCA is used for nonlinear fault classification. However, the performance of Kernel PCA largely depends on its kernel function which can only be empirically selected from finite candidates. Thus, a novel rotating machine fault diagnosis approach based on geometrically motivated nonlinear dimensionality reduction named isometric feature mapping (Isomap) is proposed. The approach can effectively extract the intrinsic nonlinear manifold features embedded in high-dimensional fault data sets. Experimental results with rotor and rolling bearing data show that the proposed approach overcomes the flaw of conventional fault pattern recognition approaches and obviously improves the fault classification performance.     

Feature extraction for identification of drug and explosive concealed by body packing based on positive matrix factorization

In the analysis towards the energy dispersive X-ray diffraction (EDXRD) spectra of drug and explosive concealed by body packing, positive matrix factorization (PMF) was introduced to extract features from EDXRD spectra of samples in a set of drugs and explosive concealed in the anthropomorphic phantom, because PMF prevents the negative factors from occurring, avoids contradicting physical reality, and makes factors more easily interpretable. In order to compare with the features extracted by PMF, Principal Component Analysis (PCA) and robust PCA were investigated. Then, K-nearest neighbour (KNN) and support vector machine (SVM) were introduced to classify the samples according to the features extracted by PMF, PCA and robust PCA. It is shown that the recognition rates obtained by PMF are highest (above 99.5%) and insensitive to classifiers. This work demonstrates that PMF is effective in feature extraction for identification of drug and explosive concealed by body packing.     

Fractal-wavelet based classification of tribological surfaces

Classification of the topography of freshly machined, worn and damaged surfaces (e.g. damaged by adhesion, scoring, abrasion, pitting) is still a problem in machine failure analysis. Tribological surfaces often exhibit both a multiscale nature (i.e. different length scales of surface features) and a non-stationary nature (i.e. features which are superimposed on each other and located at different positions on a surface). The most widely used approaches to surface classification are based on the Fourier transform or statistical functions and parameters. Often these approaches are inadequate and provide incorrect classification of the tribological surfaces. The main reason is that these techniques fail to simultaneously capture the multiscale nature and the non-stationary nature of the surface data. A new method, called a hybrid fractal-wavelet method, has recently been developed for the characterization of tribological surfaces in a multiscale and non-stationary manner. In contrast to other methods, this method combines both the wavelets’ inherent ability to characterize surfaces at each individual scale and the fractals’ inherent ability to characterize surfaces in a scale-invariant manner. The application of this method to the classification of artificially generated fractal and tribological surfaces (e.g. worn surfaces) is presented in this paper. The newly developed method has been further modified to better suit tribological surface data, including a new measure of differences between initial and decoded images. The accuracy of this method in the classification of surfaces was assessed.     

Discharge prediction for rectangular sharp-crested weirs by machine learning techniques

The stage-discharge relationship of a weir is essential for posteriori calculations of flow discharges. Conventionally, it is determined by regression methods, which is time-consuming and may subject to limited prediction accuracy. To provide a better estimate, the machine learning models, artificial neural network (ANN), support vector machine (SVM) and extreme learning machine (ELM), are assessed for the prediction of discharges of rectangular sharp-crested weirs. A large number of experimental data sets are adopted to develop and calibrate these models. Different input scenarios and data management strategies are employed to optimize the models, for which performance is evaluated in the light of statistical criteria. The results show that all three models are capable of predicting the discharge coefficient with high accuracy, but the SVM exhibits somewhat better performance. Its maximum and mean relative error are respectively 5.44 and 0.99%, and 99% of the predicted data show an error below 5%. The coefficient of determination and root mean square error are 0.95 and 0.01, respectively. The model sensitivity is examined, indicative of the dominant roles of weir Reynolds number and contraction ratio in discharge estimation. The existing empirical formulas are assessed and compared against the machine learning models. It is found that the relationship proposed by Vatankhah exhibits the highest accuracy. However, it is still less accurate than the machine learning approaches. The study is intended to provide reference for discharge determination of overflow structures including spillways.     

Modeling subcritical multi-phase flow through surface chokes with new production parameters

The produced hydrocarbons from underground reservoirs must eventually pass through surface chokes installed to control the surface flow rate at an optimum value, which should regularly be checked against the recommendations of the production engineers to prevent problems such as water coning. Accurate prediction of the surface flow rate is, therefore, crucial as it will lead to fulfilling the development plan goals of the reservoir and production optimization. In this regard, many correlations have been developed to predict the flow rate through surface choke and most of them being developed from only one dataset gathered from a single reservoir, hence with limited prediction capability and high error. Furthermore, these correlations predict the oil flow rate only as a function of wellhead pressure, gas-oil ratio, and choke size. In this study, two machine learning techniques are used to develop models for better prediction of the multi-phase flow rate for the oil wells using two new parameters of basic sediment and water (BS&W) and fluid temperature which were overlooked previously. A total of 182 production tests were utilized in developing these models which are covering a wide range of data. Graphical and statistical approaches are utilized to compare the forecasted values against the field data. Furthermore, absolute error is used as a statistical approach to assess the developed models based on machine learning in comparison to conventional correlations available in the published literature. The findings illustrate that an acceptable relation exists between the field data and predicted values with coefficients of determination equal to 0.9840 and 0.9706 for artificial neural network (ANN) and least squares support vector machine coupled simulated annealing (LSSVM-CSA), respectively, based on total datapoints. The results from this study will greatly assist petroleum engineers to have particular estimations of liquid flow rates from wellhead chokes.     

基于PCA与贝叶斯决策的人脸识别算法

研究了主元分析与贝叶斯决策相结合的人脸识别方法。利用主元分析提取人脸图像训练集的特征子空间,将训练图像和测试图像投影到该子空间,提取特征向量及计算统计特性,利用最小错误率贝叶斯决策规则对测试图像进行分类,从而实现人脸识别。大量实验表明:主元分析能将人脸图像的特征信息有效地映射在特征子空间,同时采用贝叶斯决策规则能够快速准确地对人脸图像进行分类。     

Fault diagnosis method of rolling bearing using principal component analysis and support vector machine

To effectively extract the fault feature information of rolling bearings and improve the performance of fault diagnosis, a fault diagnosis method based on principal component analysis and support vector machine was presented, and the rolling bearings signals with different fault states were collected. To address the limitation on effectively dealing with the raw vibration signals by the traditional signal processing technology based on Fourier transform, wavelet packet decomposition was employed to extract the features of bearing faults such as outer ring flaking, inner ring flaking, roller flaking and normal condition. Compared with the previous literature on fault diagnosis using principal component analysis (PCA) and support vector machine (SVM), one-to-one and one-to-many algorithms were taken into account. Additionally, the effect of four kernel functions, such as liner kernel function, polynomial kernel function, radial basis function and hyperbolic tangent kernel function, on the performance of SVM classifier was investigated, and the optimal hype-parameters of SVM classifier model were determined by genetic algorithm optimization. PCA was employed for dimension reduction, so as to reduce the computational complexity. The principal components that reached more than 95 % cumulative contribution rate were extracted by PCA and were input into SVM and BP neural network classifiers for identification. Results show that the fault feature dimensionality of the rolling bearing is reduced from 8-dimensions to 5-dimensions, which can still characterize the bearing status effectively, and the computational complexity is reduced as well. Compared with the raw feature set, PCA has a higher fault diagnosis accuracy (more than 97 %), and a shorter diagnosis time relatively. To better verify the superiority of the proposed method, SVM classification results were compared with the results of BP neural network. It is concluded that SVM classifier achieved a better performance than BP neural network classifier in terms of the classification accuracy and time-cost.     

A new sense-through-foliage target recognition method based on hybrid particle swarm optimization-based wavelet twin support vector machine

In order to improve the accuracy of sense-through-foliage target recognition, a new recognition method based on sparse representation-based adaptive feature extraction and hybrid particle swarm optimization (HPSO)-optimized wavelet twin support vector machine (WTSVM) is proposed in this paper. First, an adaptive feature extraction approach based on sparse representation is applied to extract the target features from the measured radar echo waveforms, the target feature set is constructed by sparse coefficients that contain most target information. Then, a new recognition method based optimized WTSVM is developed to perform target recognition. Twin SVM (TSVM) is a powerful tool in the field of machine learning, but the kernel and parameters selection problem still affects the performance of TSVM directly. A novel HPSO is developed in this study to determine the optimal parameters for WTSVM with the highest accuracy and generalization ability. As a hybridization strategy, local search is integrated in the PSO algorithm to further refine the performance of individuals and accelerate their convergence toward the global optimality. Finally, the performance of the proposed method is verified by experiments taken in the forest, and the results conform the improved accuracy of target recognition.     

基于支持向量机运动链同构识别的方法研究

介绍了由Vapnik等人提出的统计学习理论和由此发展的支持向量机,提出一种基于支持向量机的机构运动链同构识别的新方法。算例表明,在机构运动链同构识别中,采用支持向量机这一新方法,具有其他传统方法不可比拟的优势。     

[成形过程的数据挖掘与深度学习方法]基于机器学习的管材弯曲回弹有效预测与补偿

采用基于优化的误差反向传播(BP)神经网络的机器学习算法建模,提出了考虑材料参数、几何参数等多因素的弯管回弹精确预测和高效控制方法。该方法通过引入非线性惯性权重及遗传算法的杂交算子,改进了粒子群优化(PSO)算法,进而通过改进的PSO算法对BP神经网络进行优化,构建了基于改进的PSO-BP神经网络机器学习回弹预测和补偿模型。以多种规格的铝合金数控弯管构件为对象,将实际生产中不同规格、批次、成形参数下回弹数据作为训练样本,实现了所建机器学习预测模型的应用验证。所建模型获得的预测结果平均相对误差为6.3%,与未优化的BP神经网络等传统模型相比,预测精度最大提高了18.5%,计算时间可从1.5 h缩短至300 s,同时实现了回弹预测与补偿精度以及计算效率的显著提高。     

Bearing performance degradation assessment using locality preserving projections and Gaussian mixture models

The sensitivity of various features that are characteristics of machine performance may vary significantly under different working conditions. Thus it is critical to devise a systematic feature extraction (FE) approach that provides a useful and automatic guidance on using the most effective features for machine performance prediction without human intervention. This paper proposes a locality preserving projections (LPP)-based FE approach. Different from principal component analysis (PCA) that aims to discover the global structure of the Euclidean space, LPP is capable to discover local structure of the data manifold. This may enable LPP to find more meaningful low-dimensional information hidden in the high-dimensional observations compared with PCA. The effectiveness of the proposed approach for bearing defect and severity classification is evaluated experimentally on bearing test-beds. Furthermore, a novel health assessment indication, Gaussian mixture model (GMM)-based negative log likelihood probability (NLLP) is developed to provide a comprehensible indication for quantifying bearing performance degradation. The proposed approach has shown to provide better performance than using regular features (e.g., root mean square (RMS)). The experimental results indicate potential applications of LPP-based FE and GMM as effective tools for bearing performance degradation assessment.     

基于混合智能新模型的故障诊断

为了解决机械设备中早期故障和复合故障识别的难题,提高故障诊断的准确率,利用经验模式分解(Empirical mode decomposition,EMD)、改进的距离评估技术、自适应神经模糊推理系统(Adaptive neuro-fuzzy inference system, ANFIS)和遗传算法(Genetic algorithm, GA)等技术,提出一种综合运用多征兆域特征集和多个分类器的混合智能诊断模型。该模型在特征提取之前,利用滤波、EMD、解调等预处理技术挖掘潜藏在动态信号中的故障信息;从原始振动信号和预处理信号中,分别提取从不同侧面表征设备运行状态的时域和频域统计特征,得到6个特征集。采用提出的一种改进的距离评估技术选择特征,从6个原始特征集中相应地筛选出6个敏感特征集。将6个敏感特征集输入到基于GA组合的多个ANFIS分类器以得到最终的诊断结果。该模型在电力机车轮对轴承的故障诊断中实现了轴承不同故障类型、不同故障程度,以及复合故障的可靠识别,获得了满意的诊断结果。应用结果也验证了基于改进的距离评估技术的特征选择方法的有效性。     

Classification and authentication of unknown water samples using machine learning algorithms

This paper proposes the development of water sample classification and authentication, in real life which is based on machine learning algorithms. The proposed techniques used experimental measurements from a pulse voltametry method which is based on an electronic tongue (E-tongue) instrumentation system with silver and platinum electrodes. E-tongue include arrays of solid state ion sensors, transducers even of different types, data collectors and data analysis tools, all oriented to the classification of liquid samples and authentication of unknown liquid samples. The time series signal and the corresponding raw data represent the measurement from a multi-sensor system. The E-tongue system, implemented in a laboratory environment for 6 numbers of different ISI (Bureau of Indian standard) certified water samples (Aquafina, Bisleri, Kingfisher, Oasis, Dolphin, and McDowell) was the data source for developing two types of machine learning algorithms like classification and regression. A water data set consisting of 6 numbers of sample classes containing 4402 numbers of features were considered. A PCA (principal component analysis) based classification and authentication tool was developed in this study as the machine learning component of the E-tongue system. A proposed partial least squares (PLS) based classifier, which was dedicated as well; to authenticate a specific category of water sample evolved out as an integral part of the E-tongue instrumentation system. The developed PCA and PLS based E-tongue system emancipated an overall encouraging authentication percentage accuracy with their excellent performances for the aforesaid categories of water samples.     

A hybrid fault diagnosis approach based on mixed-domain state features for rotating machinery

To make further improvement in the diagnosis accuracy and efficiency, a mixed-domain state features data based hybrid fault diagnosis approach, which systematically blends both the statistical analysis approach and the artificial intelligence technology, is proposed in this work for rolling element bearings. For simplifying the fault diagnosis problems, the execution of the proposed method is divided into three steps, i.e., fault preliminary detection, fault type recognition and fault degree identification. In the first step, a preliminary judgment about the health status of the equipment can be evaluated by the statistical analysis method based on the permutation entropy theory. If fault exists, the following two processes based on the artificial intelligence approach are performed to further recognize the fault type and then identify the fault degree. For the two subsequent steps, mixed-domain state features containing time-domain, frequency-domain and multi-scale features are extracted to represent the fault peculiarity under different working conditions. As a powerful time-frequency analysis method, the fast EEMD method was employed to obtain multi-scale features. Furthermore, due to the information redundancy and the submergence of original feature space, a novel manifold learning method (modified LGPCA) is introduced to realize the low-dimensional representations for high-dimensional feature space. Finally, two cases with 12 working conditions respectively have been employed to evaluate the performance of the proposed method, where vibration signals were measured from an experimental bench of rolling element bearing. The analysis results showed the effectiveness and the superiority of the proposed method of which the diagnosis thought is more suitable for practical application.     

共空间模式和超限学习机的模拟电路故障诊断

主成分分析属于代数特征分析方法,是一种线性映射方法,降维后的表示是由线性映射生成的,更主要的信息保留在投影空间里,而剩余的信息则被过滤掉,但保留的信息是一体的,而不是每个特征向量分别表示一个主成分,在一定程度上影响了PCA方法的效果。提出了一种基于共空间模式对主成分方法改进的模拟电路故障诊断方法。此方法利用CSP算法对PCA得到的特征向量进行处理,然后将得到的主成分输入到超限学习机以进行网络训练或故障判断。通过Sallen-Key带通滤波器电路的实例,结果表明该研究方法的有效性。     

Machine-learning classification of environmental conditions inside a tank by analyzing radar curves in industrial level measurements

There are several solutions to measure the tank level in industrial applications. However, the environmental conditions inside this tank, such as turbulence and foam, can jeopardize measurement accuracy and precision. This article proposes a methodology to identify the presence of turbulence and foam in a fermentation tank. The proposal is based on the extraction, selection, and classification of statistical features by machine learning methods. The use of machine learning strategies and statistical features guarantees the necessary robustness and generality for industrial applications. Actual data obtained from a must fermentation tank of a sugar-alcohol industrial plant were used for training and verifying one Artificial Neural Network-based and three Support Vector Machine-based classifiers. These classifiers obtained accuracy over 98% for different environmental conditions proving the effectiveness of the proposed methodology.     

Numerical methodologies for investigation of moderate-velocity flow using a hybrid computational fluid dynamics — molecular dynamics simulation approach

Numerical approaches are presented to minimize the statistical errors inherently present due to finite sampling and the presence of thermal fluctuations in the molecular region of a hybrid computational fluid dynamics (CFD) — molecular dynamics (MD) flow solution. Near the fluid-solid interface the hybrid CFD-MD simulation approach provides a more accurate solution, especially in the presence of significant molecular-level phenomena, than the traditional continuum-based simulation techniques. It also involves less computational cost than the pure particle-based MD. Despite these advantages the hybrid CFD-MD methodology has been applied mostly in flow studies at high velocities, mainly because of the higher statistical errors associated with low velocities. As an alternative to the costly increase of the size of the MD region to decrease statistical errors, we investigate a few numerical approaches that reduce sampling noise of the solution at moderate-velocities. These methods are based on sampling of multiple simulation replicas and linear regression of multiple spatial/temporal samples. We discuss the advantages and disadvantages of each technique in the perspective of solution accuracy and computational cost.     

基于机器视觉的表面缺陷检测方法研究进展

在半导体、PCB、汽车装配、液晶屏、3C、光伏电池、纺织等行业中,产品外观与产品性能有着千丝万缕的联系。表面缺陷检测是阻止残次品流入市场的重要手段。利用机器视觉的技术进行检测效率高、成本低,是未来发展的主要方向。本文综述了近十年来基于机器视觉的表面缺陷检测方法的研究进展。首先给出了缺陷的定义、分类以及缺陷检测的一般步骤;然后重点阐述了使用传统图像处理方式、机器学习、深度学习进行缺陷检测的原理,并比较和分析了优缺点,其中传统图像处理方式分为分割与特征提取两个部分,机器学习包含无监督学习和有监督学习两大类,深度学习主要囊括了检测、分割及分类的大部分主流网络;随后介绍了30种工业缺陷数据集以及性能评价指标;最后指出缺陷检测方法目前存在的问题,对进一步的工作进行了展望。