基于SVR-OCSVM模型的多联机系统用能评估与诊断

直接根据多联机系统能耗数据的变化来判断导致能耗大幅波动的因素是很困难的。本文提出一种有效的可用于多联机系统的能耗评估与诊断方法:将支持向量回归(SVR)算法与单类支持向量机(OCSVM)算法相结合,首先通过提取系统能耗数据集特征,去除非稳态数据,根据提取的特征变量与系统能耗建立SVR模型,预测多联机系统能耗;然后将实际能耗值与预测能耗值之差和之比分别标准化,作为输入变量,建立单类支持向量机(OCSVM)模型进行样本判别,确定是否为导致系统能耗异常的原因,以此评估诊断多联机系统能耗情况。本文基于多联机能耗正常的数据集构建了能耗评估与诊断模型,并用多联机系统能耗异常数据集验证了模型的可靠性。结果表明:基于SVR-OCSVM模型的能耗评估与诊断模型具有较高的准确度,基本能达到70%以上。     

Assessment of human reliability based on evaluation of plant experience: requirements and implementation

A major problem in assessment of human failures in probabilistic safety assessment is the lack of empirical data needed for human reliability analysis (HRA). This problem is aggravated by the fact that different HRA methods use different parameters for the assessment and that HRA is currently enforced to provide data and methods for assessment of human reliability in new technical environments such as computerized control rooms, in accident management situations, or in low-power and shut down situations. Plant experience is one source to deal with this problem. In this paper, a method is presented that describes how plant experience about human failures and human performance may be used to support the process of analyzing and assessing human reliability. Based on considerations of requirements of HRA, a method is presented first which is able to describe and analyze human interactions that were observed within events. Implementation of the approach as a database application is outlined. Second, the main results of the application of the method to 165 boiling water reactor events are presented. Observed influencing factors on human performance are discussed; estimates for probabilities are calculated and compared with the data tables of the THERP handbook. An outline is given for using the presented method for the analysis of cognitive errors or organizational aspects.     

基于驾驶员身体特征和操作姿态的汽车内饰按键操作便捷性建模

作为汽车驾驶过程中的次级任务,内饰部件操作是否便捷严重影响行车安全。针对汽车内饰按键,提出一种基于驾驶员身体特征和肢体操作姿态的操作时间建模方法。首先,利用汽车模拟驾驶器和人体运动捕捉系统,采集9名驾驶员在驾驶状态下的肢体操作姿态和操作时间;然后,统计分析影响操作时间的主要因素;接着,通过主成分回归方法建立操作时间预测模型;最后,利用测试按键和5名新驾驶员的操作数据验证模型。实验结果表明：上躯干侧倾和前倾、肩关节前后摆、肘关节屈伸以及手臂的内外旋转等肢体运动和驾驶员的身体特征均是影响操作时间的主要因素;测试按键的平均预测绝对误差为0.212 s,新驾驶员为0.226 s,验证了模型的有效性。本研究对于汽车内饰部件的便捷性评估提供一种新方法。     

Fundamental Theories and Key Technologies for Smart and Optimal Manufacturing in the Process Industry

Given the significant requirements for transforming and promoting the process industry, we present the major limitations of current petrochemical enterprises, including limitations in decision-making, production operation, efficiency and security, information integration, and so forth. To promote a vision of the process industry with efficient, green, and smart production, modern information technology should be utilized throughout the entire optimization process for production, management, and marketing. To focus on smart equipment in manufacturing processes, as well as on the adaptive intelligent optimization of the manufacturing process, operating mode, and supply chain management, we put forward several key scientific problems in engineering in a demand-driven and application-oriented manner, namely: ① intelligent sensing and integration of all process information, including production and management information; ② collaborative decision-making in the supply chain, industry chain, and value chain, driven by knowledge; ③ cooperative control and optimization of plant-wide production processes via human-cyber-physical interaction; and ④ life-cycle assessments for safety and environmental footprint monitoring, in addition to tracing analysis and risk control. In order to solve these limitations and core scientific problems, we further present fundamental theories and key technologies for smart and optimal manufacturing in the process industry. Although this paper discusses the process industry in China, the conclusions in this paper can be extended to the process industry around the world.     

基于神经网络训练的战略分析和协商机制的绩效评价

论述了绩效管理模型的发展历程和发展趋势，介绍了BP神经网络评价模型形成的原理，综合运用多种方法建立了财务指标和非财务指标相结合的企业投资决策战略绩效评价的一般指标体系，采用了主成分分析和专家判断相结合的方法对企业投资决策绩效评价指标进行筛选。建立了基于SWOT战略分析和BP神经网络训练的企业投资决策绩效评价模型，并运用该模型对武钢焦化企业投资项目进行了实证分析。     

Assessment of the role of micro defect nucleation in probabilistic modeling of cleavage fracture

The present study is concerned with an experimental and numerical investigation of the local conditions for initiation of cleavage failure in ferritic steels. In the experimental analysis, a variety of SE(B), C(T) and CC(T) specimens has been tested. The cleavage triggering sites and the fracture mechanisms were determined in a subsequent fractographic investigation. In a finite element analysis, the local mechanical field quantities at the cleavage initiation spots at the instant of fracture were investigated. Based on the results, an enhanced local concept for cleavage assessment is proposed, accounting for both, the nucleation of critical micro defects and their possible instability due to local overloading. Together with a comparison to the prediction of previous probabilistic models for cleavage failure assessment, the results reveal the importance of considering the nucleation of possibly critical micro defects due to plastic straining as a second necessary criterion for cleavage fracture.     

Performance analysis of classifiers for colon cancer detection from dimensionality reduced microarray gene data

Cancer disease is accountable for many deaths that are over 9.6 million in 2018 and roughly one out of six deaths occur because of cancer worldwide. The colon cancer is the second prominent source of death of around 1.8 million cases. This research is inclined to detect the colon cancer from microarray dataset. It will aids the experts to distinguish the cancer cells from normal cells for appropriate determination and treatment of cancer at earlier stages that leads to increase the survival rate of the patients. The high dimensionality in microarray dataset with less samples and more attributes creates lag in the detection capability of the classifier. Hence there is a need for dimensionality reduction techniques to preserve the significant genes that are prominent in the disease classification. In this article, at first ANOVA method used to select the best genes and then principal component analysis (PCA) and fuzzy C-means clustering (FCM) techniques are further employed to choose relevant genes. The PCA and FCM features are classified using model, discriminant, regression, hybrid, and heuristic-based classifiers. The attained results show that the heuristic classifier with PCA features is encapsulated an average classification accuracy of 97.92% for classifying both the colon cancer and normal samples. Also, for FCM features, the Heuristic classifier is maintained at an average classification accuracy of 99.48% and 97.92% for classifying the colon cancer and normal samples, respectively. The Heuristic classifier outperforms with high accuracy than all other classifiers in the classification of colon cancer.     

Mixture probabilistic PCR model for soft sensing of multimode processes

Principal component regression (PCR) has been widely used for soft sensor modeling and quality prediction in last several decades, which is still very popular for both academy researches and industry applications. However, most PCR models are determined by the projection method, which may lack probabilistic interpretation for the process data. In fact, due to the inevitable process noise, most process data are inherently random variables. Several probabilistic PCA methods have already been proposed in the past years. Compared to the deterministic modeling method, the probabilistic model is more appropriate to characterize the behavior of the random variables in the process. This paper first presents a probabilistic derivation of the PCR model (PPCR) and then extends it to the mixture form (MPPCR). For quality prediction of processes with multiple operation modes, a mixture probabilistic soft sensor is developed based on the MPPCR model. Simultaneously, the information of the operation mode can also be located by the proposed soft sensor. To evaluate the performance of the MPPCR model, a numerical example and a benchmark simulation case study of the Tennessee Eastman process are provided. Different methods have been compared with the proposed model, including the global, local, and multi-local PCR models. As a result, the proposed MPPCR model performs the best among these methods.     

基于主成分层次分析的产品设计语义差分评价研究

目的 以外观设计评价为研究对象,提出一种语义差分评价方法。方法 运用主成分分析法,对初始多维语义进行主成分提取,得出综合语义指标和分值,再通过层次分析法构建综合语义指标的主观评价矩阵,最终得出综合语义分值。结论 针对46款热销女士洗发水瓶的外观设计样本,实现了语义差分评价。运用主成分—层次分析—语义差分评价的方法,实现了近似语义的归类。依据综合语义权值,得出了语义差分分值,并适用于较大样本的外观设计评价。     

基于偏最小二乘回归的军用飞机采购价格预测

考虑到军用飞机采购价格样本数据少、难于预测的特点和偏最小二乘回归方法在处理小样本多元数据方面的优势，提出一种基于偏最小二乘回归的军用飞机价格预测方法。偏最小二乘回归首先提取第一、第二主成分对采购价格样本的特异点进行剔除；然后进行变量投影重要度分析来筛选变量；最后，偏最小二乘回归对筛选的变量进行回归建立军用飞机价格预测模型，并对军用飞机价格进行预测，结果表明，在军用飞机价格预测方面，与未筛选变量的回归模型和逐步多元回归相比，经过变量筛选的偏最小二乘回归模型预测的精度更高，更能体现采购价格与飞机性能参数乏问的关系。     

基于刚度-阻尼器ATMD模型的性能评价

提出了基于刚度-阻尼器主动调谐质量阻尼器(SD-ATMD)新模型。利用导出的SD-ATMD-结构系统的传递函数，建立了SD-ATMD-结构系统的动力放大系数。于是SD-ATMD的最优参数与有效性的评价准则可定义为设置SD-ATMD结构最大动力放大系数中最小值的最小化。而SD-ATMD冲程的评价准则定义为SD-ATMD动力放大系数的最大值。基于上述两个评价准则，研究了刚度比(R)对SD-ATMD性能(最优参数、有效性和冲程)的影响。为比较，同时考虑了传统的ATMD(即R=0．0的SD-ATMD)的数值结果。。     

基于Volterra级数和KPCA的旋转机械故障诊断方法研究

提出了一种基于Volterra级数和核函数主元分析（KPCA）的故障诊断方法。在提出的方法中,首先利用量子粒子群优化（QPSO）算法辨识出正常、转子裂纹、转子碰摩、基座松动四种状态下的Volterra级数,然后将Volterra核函数作为特征向量输入到KPCA进行训练识别。实验结果表明,提出的方法是有效的,在只考虑一阶Volterra核不能进行很好地识别时,可以从二阶、三阶Volterra核上来区分。     

冶金生产过程危险识别及评价技术

生产过程危险识别及评价技术涉及人身安全和设备安全,是关系到国民经济正常运行和社会稳定的重要工作,其作用不是对发生的事故事后进行总结分析,而是在发生事故前就能识别,并评价其危害程度,可为管理决策层制定防治事故的对策提供依据,从而预先采取措施,防止事故的发生。     

On the Implementation of the Principal Component Analysis–Based Approach in Measuring Process Capability

The use of principal component analysis in measuring the capability of a multivariate process is an issue initially considered by Wang and Chen (1998). In this article, we extend their initial idea by proposing new indices that can be used in situations where the specification limits of the multivariate process are unilateral. Moreover, some new indices for multivariate processes are suggested. These indices have been developed so as to take into account the proportion of variance explained by each principal component, thus making the measurement of process capability more effective. Copyright © 2011 John Wiley & Sons, Ltd.     

Kernel principal component analysis and support vector machines for stock price prediction

Technical indicators are used with two heuristic models, kernel principal component analysis and factor analysis in order to identify the most influential inputs for a forecasting model. Multilayer perceptron (MLP) networks and support vector regression (SVR) are used with different inputs. We assume that the future value of a stock price/return depends on the financial indicators although there is no parametric model to explain this relationship, which comes from the technical analysis. Comparison studies show that SVR and MLP networks require different inputs. Furthermore, proposed heuristic models produce better results than the studied data mining methods. In addition to this, we can say that there is no difference between MLP networks and SVR techniques when we compare their mean square error values.     

Determination of carbamazepine polymorphic contents in double-layered tablets using transmittance- and reflectance-near-infrared spectroscopy involving chemometrics

Background: Since polymorphs exhibit differences in chemical and physicochemical stability, characteristics, and dissolution rate of the bulk powder, they may significantly affect on the bioavailability of pharmaceutical compounds. Aim: The purpose of the present study is to establish a method for determining the carbamazepine (CBZ) polymorphic content of a double-layered tablet containing various ratios of forms I and III by using transmittance- and reflectance-near-infrared (TNIR and RNIR) spectroscopy involving chemometrics. Methods: Both TNIR and RNIR instruments were used to analyze both top (form I) and wire (form III) sides of the compacts, respectively. NIR spectra were analyzed to predict polymorphic content by a principal component regression analysis. NIR data of the tablets were divided into two wavelength ranges: between 860 and 1680 nm (FW), and 1245 and 1285 nm (NW). Results: The calibration models for polymorphic content based on TNIR had a linear relationship, but those based on RNIR did not. The accuracy of the calibration models suggested that the double-sided data set is more robust than the single-sided data set. Since the spectra of FW involved various information, the calibration models showed a linear correlation, but it is difficult to understand their model. In contrast, those of NW provided limited information on polymorphic forms making it very easy to understand the model. Conclusion: Limiting the wavelength of the spectra is useful to help understand the calibration-complicated model.     

基于广义回归神经网络的工程公司风险评估

针对工程公司日常管理中由于缺乏风险评估工具,经常造成公司资源的浪费,甚至造成大量损失的现状,结合其行业特点,构建风险评价体系模型。在模型的基础上,提出一种基于广义回归神经网络(GRNN)的工程公司风险评估方法,通过矩阵实验室中的神经网络工具对其进行仿真计算,以某安防系统工程公司的实例证明了其有效性。该研究为同类型公司的风险评估提供了一种有效的管理工具。     

重点行业烟气脱硫脱硝除尘装备运行效果评价指标体系研究

当前日益严峻的大气污染物控制形势和节能减排相关产业政策要求的日趋严格,对重点行业开展烟气脱硫脱硝除尘装备运行效果评价工作提出了迫切需求。本文从分析当前我国火电和钢铁行业大气污染物排放及控制装备运行现状入手,论述了研制重点行业烟气脱硫脱硝除尘装备运行效果评价技术要求系列国家标准的必要性和紧迫性,阐述了相关评价标准的编制思路和原则;重点介绍了燃煤烟气脱硫脱硝装备和钢铁烧结烟气除尘装备运行效果评价指标体系的建立,以及定性评价要求和定量评价指标要求,借此推动我国重点行业大气污染物控制装备运行效果评价机制的建设,以标准化手段促进我国重点行业大气污染物控制装备运行可靠性和有效性的提升。     

基于改进主元分析方法的空调系统传感器故障检测和诊断研究

传感器在空调系统中主要起着监测和控制的作用,影响空调系统的正常运行,从而带来能耗增加等不良影响。本文提出了结合小波变换的数据优化,以及基于神经网络的故障诊断优化的改进主元分析方法,用于空调系统传感器故障检测和诊断研究。通过对比数据优化前后主元分析的结果,发现同样0. 850 0累计贡献率原则上,采用小波变换去除噪声后,主元个数减少了两个,蒸发器进口温度传感器的固定偏差、漂移、精度下降等故障检测效果分别提升了0. 020 7、0. 020 8、0. 041 5,风量传感器固定偏差故障检测效果提升了0. 160 6。为了进一步找出故障源,在小波变换和主元分析的基础上,将求得的主元作为神经网络的输入,对5个传感器固定偏差故障进行测试,故障诊断结果分别为0. 766 7、0. 866 7、0. 900 0、1. 000 0、1. 000 0。