Forecasting systems reliability based on support vector regression with genetic algorithms

This study applies a novel neural-network technique, support vector regression (SVR), to forecast reliability in engine systems. The aim of this study is to examine the feasibility of SVR in systems reliability prediction by comparing it with the existing neural-network approaches and the autoregressive integrated moving average (ARIMA) model. To build an effective SVR model, SVR's parameters must be set carefully. This study proposes a novel approach, known as GA-SVR, which searches for SVR's optimal parameters using real-value genetic algorithms, and then adopts the optimal parameters to construct the SVR models. A real reliability data for 40 suits of turbochargers were employed as the data set. The experimental results demonstrate that SVR outperforms the existing neural-network approaches and the traditional ARIMA models based on the normalized root mean square error and mean absolute percentage error.     

Hybrid approach for remaining useful life prediction of ball bearings

Remaining useful life (RUL) prediction plays an important role in predictive maintenance systems to support decision‐makers for arranging maintenance tasks and related resources. We propose a hybrid approach that is combined an exponential weighted moving average (EWMA) control chart for anomaly detection and machine learning models such as support vector regression (SVR) and random forest regression (RFR) with differential evolution (DE) algorithm to predict the RULs of ball bearings. Here, DE algorithm is used to find the optimal hyperparameters of SVR model. The datasets of ball bearings from the Prognostics Data Repository of NASA are used to compare the prediction performance of different methods. The degradation behavior of training data from the anomaly time to the end of life is used to transfer learning for the testing data in the SVR and RFR models. The results indicate that the proposed methods outperform the other four existing methods in terms of score. Therefore, the proposed hybrid approach is a reliable tool for the RUL prediction of ball bearings.     

Joint optimisation of production,maintenance and quality for batch production system subject to varying operational conditions

This paper considers the integration problem of production, maintenance and quality for a capacitated lot-sizing production system subject to deterioration. The effects of varying operational conditions from batch to batch on system reliability and product quality are modelled by proportional hazards models, resulting in non-monotonic failure rate and defect rate. After each batch production, imperfect preventive maintenance (PM) is determined to mitigate the system deterioration, and inspection is taken to sort nonconforming items in the finished goods. Once the cumulative number of nonconforming items exceeds a predetermined threshold, an overhaul is performed to renew the system. An integrated model for optimising production plan, PM plan and overhaul strategy is developed to minimise the total cost while satisfying all product demands. A genetic algorithm is proposed to solve the integrated model efficiently. Numerical results validate the rationality of the model with varying operational conditions consideration and its applicability in economic benefits.     

基于相空间重构和遗传优化SVR的机械设备状态趋势预测

针对机械设备振动信号序列的非线性、非平稳性特点,提出了一种基于相空间重构与遗传优化支持向量回归机的设备状态趋势预测方法。首先,采用相空间重构技术将一维振动信号时间序列转化成矩阵形式,自适应地选取特征,以相点作为输入特征训练SVR预测器;然后应用自适应遗传算法对惩罚因子、不敏感系数以及高斯核宽度进行同步优化,自动获取最佳的建模参数;最后构建SVR预测模型,并将其应用于某机组振动信号预测。实验结果表明,无论是单步还是24步预测,本文所提遗传优化SVR模型的预测精度都要比标准SVR模型的预测精度高,说明该方法对机械设备的运行状态趋势具有较好的预测能力。     

Saturation Magnetic Induction Prediction for Amorphous Magnetic Alloys by Using Support Vector Regression

This paper illustrates an application of support vector regression (SVR) approach in forecasting the saturation magnetic induction (B _s ) of amorphous magnetic alloys. SVR was trained and tested with an experimental data set comprised of five input variables, comprising the average number of valence electrons of amorphous magnetic alloys, mixed entropy, ratio of radii, difference of electron density, and difference of work function. The prediction performance of SVR was compared with that of artificial neural networks’ (ANN) model. The results demonstrate that the prediction ability of SVR is superior to that of ANN. This investigation indicates that SVR-based modeling is a practically useful tool in prediction of the saturation magnetic induction of amorphous alloys. This study provides a novel methodology to foresee the saturation magnetic induction in sintering/development of novel amorphous magnetic alloys possessing high saturation magnetic induction.     

基于粒子群支持向量回归的原位自生工艺参数优化

综合应用激光熔覆和原位反应增强金属基复合材料,是当前金属基复合材料研究领域的一个热点,本文采用该工艺制备铁基表面复合材料,重点考虑该工艺参数的确定问题.根据在不同工艺参数下合成的铁基表面的WC体积分数实测数据集,提出建立不同工艺参数下WC体积分数的支持向量回归预测模型,并与基于人工神经网络模型(ANN)的预测结果进行比较.结果显示:对于相同的训练样本和检验样本,SVR预测模型比ANN预测模型具有更强的泛化能力.最后根据建立的预测模型,应用粒子群算法寻优得到最优工艺参数,该工艺参数在实际实验过程中的应用,验证了该方法的有效性.     

基于OVMD与SVR的水电机组振动趋势预测

为更好地预测水电机组振动趋势,研究提出了一种基于最优变分模态分解（OVMD）与支持向量回归（SVR）的水电机组振动趋势预测模型。首先基于中心频率观察法与残差指标最小化准则确定OVMD的分解参数,采用OVMD将非平稳振动序列分解为一系列模态函数,并对各模态函数分别进行相空间重构,构建状态矩阵,进而得到SVR回归预测模型的输入、输出,再采用交叉验证的网格搜索策略优化各SVR模型的参数,并分别进行回归预测,最后对所有SVR预测结果进行求和,得到原始振动趋势的预测值。研究对某大型混流式水电机组的振动监测数据进行预测试验,并进行对比分析,结果表明该模型可有效预测水电机组振动趋势。     

改进MDSMOTE与PSO-SVM在汽车组合仪表分类预测中的应用

汽车组合仪表生产过程中质检项目多且检测时间长,这在一定程度上制约了其生产效率的进一步提升。为此,提出一种基于改进最远点合成少数类过采样技术（max distance synthetic minority over-sampling technique,MDSMOTE）的支持向量机（support vector machine, SVM）分类预测方法。首先,结合专家经验对汽车组合仪表的原始生产数据进行特征筛选,并在MDSMOTE中引入类不平衡率I_R,以对所筛选的特征数据进行扩充;然后,利用粒子群优化（particle swarm optimization, PSO）算法对SVM的误差惩罚因子C和核函数参数γ进行优化;最后,建立优化的SVM分类预测模型,并对汽车组合仪表进行分类。通过与其他分类预测模型在不同数据集上的预测结果进行对比可知,基于改进MDSMOTE的SVM分类预测模型的准确率、F值和几何平均值等评价指标均优于其他模型。所提出方法在汽车仪表产品分类上表现出较强的泛化能力和稳定性,可为仪表制造企业生产效率的提升提供有效参考。     

基于EEMD和SVR的单自由度结构状态趋势预测

为了解决结构早期损伤难以正确识别的问题,本文结合聚类经验模式分解（EEMD）解决随机不确定性问题和支持向量机（SVM）解决预测问题这两者的优势,提出了一种基于EEMD特征提取的支持向量机回归（SVR）结构状态趋势预测方法。先对单自由度结构渐进损伤的加速度振动信号进行EEMD,再进行希尔伯特变换（HT）,计算瞬时频率,然后用回归支持向量机对反映结构健康状态的瞬时频率进行趋势预测。研究表明：对于渐变损伤该方法可以准确地、高精度地预测结构状态趋势。     

Matrix techniques for modeling ultrasonic waves in multilayeredmedia

Research into ultrasonic NDE techniques for the inspection of multilayered structures relies strongly on the use of modeling tools which calculate dispersion curves and reflection and transmission spectra. These predictions are essential to enable the best inspection strategies to be identified and their sensitivities to be evaluated. General purpose multilayer modeling tools may be developed from a number of matrix formulations which have evolved in the latter half of this century and there is now a formidable number of publications on the subject. This paper presents a review of the main developments of the matrix techniques, and their use in response and modal models, with emphasis on ultrasonics applications     

SOMRuler: a novel interpretable transmembrane helices predictor

Transmembrane helices (TMH) identification is one of the most important steps in membrane protein structure prediction. Existing TMH predictors tend to pursue accurate computational models without carefully considering the interpretability of these models and thus act as a black box. In this paper, a novel TMH predictor called SOMRuler with excellent interpretability while possessing high prediction accuracy is presented. The SOMRuler uses a self-organizing map (SOM) to learn helices distribution knowledge, which is encoded in the codebook vectors of the trained SOM, from the training samples. Human interpretable fuzzy rules are then extracted from the codebook vectors of the trained SOM. By extracting fuzzy rules from the learned knowledge rather than the original training samples, on the one hand, the computational burden of extracting fuzzy rules can be greatly reduced; on the other hand, the reliability of the extracted rules can also be enhanced since noise contained in the original samples can be smoothened by the learning procedure of SOM. The validity of the fuzzy rules extracted by SOMRuler is qualitatively and quantitatively analyzed. Experimental results on the benchmark dataset show that the SOMRuler outperforms most existing popular TMH predictors and is flexible to suite for a wide variety of problems in bioinformatics. The SOMRuler software is implemented by Java and Matlab and is available for academic use at: http://www.csbio.sjtu.edu.cn/bioinf/SOMRuler/.     

Customer demand prediction of service-oriented manufacturing using the least square support vector machine optimized by particle swarm optimization algorithm

Many nonlinear customer satisfaction-related factors significantly influence the future customer demand for service-oriented manufacturing (SOM). To address this issue and enhance the prediction accuracy, this article develops a novel customer demand prediction approach for SOM. The approach combines the phase space reconstruction (PSR) technique with the optimized least square support vector machine (LSSVM). First, the prediction sample space is reconstructed by the PSR to enrich the time-series dynamics of the limited data sample. Then, the generalization and learning ability of the LSSVM are improved by the hybrid polynomial and radial basis function kernel. Finally, the key parameters of the LSSVM are optimized by the particle swarm optimization algorithm. In a real case study, the customer demand prediction of an air conditioner compressor is implemented. Furthermore, the effectiveness and validity of the proposed approach are demonstrated by comparison with other classical predication approaches.     

Modeling of Non-uniformly Sampled Systems by Support Vector Regression with Applications

This paper presents a modeling method for a non-uniformly sampled system based on support vector regression (SVR).First,a lifted discrete-time state-space model for a non-uniformly sampled system is derived by using the lifting technique to reduce the modeling difficulty caused by multirate sampling.Then,the system is divided into several parallel subsystems and their input-output model is presented to satisfy the SVR model.Finally,an on-line SVR technique is utilized to establish the models of all subsyste...     

山东出口危险货物包装检验监管分析与建议

目的提升山东地区出口危险货物包装检验监管水平,帮扶企业规避贸易风险、提高产品质量、扩大出口。方法根据山东地区2016年出口危险货物包装业务基本情况(包装检验46 424批,8799万件),总结业务变化(同比分别增长28.9%,27.1%)和检验的不合格信息。结果 2016年危险货物包装性能检验批次、数量分别同比增长了17.4%,13.8%;包装使用鉴定批次、数量分别同比增长了32.2%,64.5%。危险货物包装性能检验与使用鉴定之间成正比关系,性能检验逐年增长,相应的危险货物包装使用鉴定也随之出现了井喷式增长态势。结论归纳了天津"8.12"事件后山东地区业务激增原因和性能检验与使用鉴定不合格情况,从信息化、技术性贸易措施、安全特性检验管理3个方面提出了有针对性的对策建议。     

NDTE using pulse thermography: Numerical modeling of composite subsurface defects

The ability to predict subsurface defect information in composite materials through a non-invasive, efficient inspection protocol is fast becoming a vital research area. In numerically modeling the thermographic process associated with an infrared (IR) technique we can afford inspectors the ability to predict subsurface defect information associated with a specific material configuration. The research involved in this study looks specifically at the finite element modeling (FEM) of delaminations in a composite flat plate setup. To date the modeling of delaminations has been restricted to only two dimensional (2D) numerical representations and associated primarily with rear faced detection. The results of this research, however, clearly show that the rear faced detection technique has limitations in defect depth prediction and the 2D approximation associated with this technique ignores a paramount effect in the form of lateral thermal diffusion. It is also made clear that the representation of experimental flat plate models with flat bottomed holes, under pulse phase thermographic inspection, in simulating delaminations is misguided.     

基于集成学习的FRP加固混凝土梁抗弯承载力预测研究

为解决当前纤维增强复合材料(FRP)加固钢筋混凝土梁抗弯承载力预测中模型不统一、计算繁琐、精度有限等问题,建立了统一化的抗弯承载力预测模型。根据既有文献收集外贴式、端锚式和嵌入式3种FRP典型加固方式加固钢筋混凝土梁试验数据,确定影响加固梁承载力的关键因素,通过XGBoost(极限梯度提升树)算法训练回归各影响因素与加固后梁抗弯承载力间的非线性映射关系,得到统一化的FRP加固钢筋混凝土梁抗弯承载力预测模型。随后在测试样本集上对该模型的预测精度进行了验证,与基于支持向量回归(SVR)和人工神经网络(ANN)两种代表性机器学习算法得到的预测模型进行了横向对比,并分析了不同加固方式下的预测精度。研究结果表明：该文得到的基于XGBoost的抗弯承载力预测模型拟合优度R²=0.9417,可见整体精度较高,有良好的性能;相比基于传统机器学习算法SVR和ANN建立的预测模型,基于集成学习算法XGBoost的拟合优度分别提升了8.00%及6.70%,均方根误差减少了33.94%和30.72%,平均绝对误差减少了32.38%和30.51%,表明基于XGBoost的模型精度更高,远优...     

Modular artificial neural network for prediction of petrophysicalproperties from well log data

An application of Kohonen's self-organizing map (SOM), learning-vector quantization (LVQ) algorithms, and commonly used backpropagation neural network (BPNN) to predict petrophysical properties obtained from well-log data are presented. A modular, artificial neural network (ANN) comprising a complex network made up from a number of subnetworks is introduced. In this approach, the SOM algorithm is applied first to classify the well-log data into a predefined number of classes, This gives an indication of the lithology in the well. The classes obtained from SOM are then appended back to the training input logs for the training of supervised LVQ. After training, LVQ can be used to classify any unknown input logs. A set of BPNN that corresponds to different classes is then trained. Once the network is trained, it is then used as the classification and prediction model for subsequent input data. Results obtained from example studies using the proposed method have shown to be fast and accurate as compared to a single BPNN network     

Optimal Antibody Purification Strategies Using Data-Driven Models

This work addresses the multiscale optimization of the purification processes of antibody fragments. Chromatography decisions in the manufacturing processes are optimized, including the number of chromatography columns and their sizes, the number of cycles per batch, and the operational flow velocities. Data-driven models of chromatography throughput are developed considering loaded mass, flow velocity, and column bed height as the inputs, using manufacturing-scale simulated datasets based on microscale experimental data. The piecewise linear regression modeling method is adapted due to its simplicity and better prediction accuracy in comparison with other methods. Two alternative mixed-integer nonlinear programming (MINLP) models are proposed to minimize the total cost of goods per gram of the antibody purification process, incorporating the data-driven models. These MINLP models are then reformulated as mixed-integer linear programming (MILP) models using linearization techniques and multiparametric disaggregation. Two industrially relevant cases with different chromatography column size alternatives are investigated to demonstrate the applicability of the proposed models.     

Development of planning level transportation safety tools using Geographically Weighted Poisson Regression

A common technique used for the calibration of collision prediction models is the Generalized Linear Modeling (GLM) procedure with the assumption of Negative Binomial or Poisson error distribution. In this technique, fixed coefficients that represent the average relationship between the dependent variable and each explanatory variable are estimated. However, the stationary relationship assumed may hide some important spatial factors of the number of collisions at a particular traffic analysis zone. Consequently, the accuracy of such models for explaining the relationship between the dependent variable and the explanatory variables may be suspected since collision frequency is likely influenced by many spatially defined factors such as land use, demographic characteristics, and traffic volume patterns. The primary objective of this study is to investigate the spatial variations in the relationship between the number of zonal collisions and potential transportation planning predictors, using the Geographically Weighted Poisson Regression modeling technique. The secondary objective is to build on knowledge comparing the accuracy of Geographically Weighted Poisson Regression models to that of Generalized Linear Models. The results show that the Geographically Weighted Poisson Regression models are useful for capturing spatially dependent relationships and generally perform better than the conventional Generalized Linear Models.