基于自适应高效递推规范变量分析的多模过程软传感器建模

由于多模过程中各模式间的均值和协方差发生了改变,多变量单模高斯分布的基本假设不再成立.基于递推方法的多模过程软传感器建模存在两点问题:其一,递推建模方法不能及时的跟踪多模过程的改变;其二,递推建模方法的在线计算负荷非常高.为了解决上述问题,本文提出了一种基于自适应高效递推规范变量分析的多模过程软传感器建模方法.首先,采用指数权重滑动平均来更新过去观测矢量的协方差矩阵;然后,利用基于模型输出误差范数的可变遗忘因子来跟踪多模过程的动态变化;最后,通过引入一阶干扰理论(firstorder perturbation,FOP)来实现递推奇异值分解,与常规奇异值分解相比递推奇异值算法的计算负荷显著降低.将提出的方法用于田纳西伊斯曼(tennessee eastman,TE)化工过程进行仿真验证,其结果表明了该方法的可行性和精确性.

基于VMD-FCM-RLSSVM的多模过程故障诊断方法

为了保证工业过程的正常进行,需要及时地辨识出故障以实现故障诊断。然而,传统的故障诊断方法多是基于单模过程而不适应于多模态过程。为了解决这个问题,本文提出了一种变分模态分解(VMD)、模糊C均值(FCM)及递推最小二乘支持向量机(RLSSVM)相结合的集合型故障诊断方法。本文首先介绍了VMD方法对所采集的数据进行去噪处理,然后利用FCM进行模态区分,最后利用RLSSVM方法实现故障诊断。通过CSTH过程的仿真结果表明,该方法提高了诊断效率和性能,能够快速、有效地诊断出故障。

IGSA-KPCA邻域建模的多模过程故障检测方法

为提高多模过程故障检测的准确率,提出改进引力搜索算法-核主元分析邻域建模的故障检测方法。首先应用及时学习算法在参考数据集中找到待检数据的相关数据,再将相关数据和待检数据作为核主元分析检测模型的输入进行故障检测。核主元分析模型中的参数对故障检测性能有较大影响,提出改进引力搜索算法对模型中参数进行优化,提高检测性能。将所提方法应用于青霉素多模过程进行实验验证,仿真结果表明所提方法在多模过程故障检测中用时短、准确率高。

高斯型混合双多模噪声模型分析

本文以一类高斯型混合非高斯噪声双模噪声为背景噪声,详细分析了二进制数字调制系统的抗噪声性能。为研究更一般的情形,本文提出了窄带多模过程的数学模型,进行了较详细的研究,给出了混合噪声中信号检测的一般方法,是对原建立在高斯噪声基础上通信与信号处理理论的完善和补充,有一定的普遍意义。在理论分析的基础上,最后给出了仿真结果并进行了分析。

Finite element simulation of aluminum alloy cross valve forming by multi-way loading

Deformation behavior,temperature evolution and coupled effects have a significant influence on forming process and quality of component formed,which are very complex in forming process of aluminum alloy 7075 cross valve under multi-way loading due to the complexity of loading path and the multiplicity of associated processing parameters.A model of the process was developed under DFEORM-3D environment based on the coupled thermo-mechanical finite element method.The comparison between two process models,the conventional isothermal process model and the non-isothermal process model developed in this study,was carried out,and the results indicate that the thermal events play an important role in the aluminum alloy forming process under multi-way loading.The distributions and evolutions of the temperature field and strain filed are obtained by non-isothermal process simulation.The plastic zone and its extension in forming process of cross valve were analyzed.The results may provide guidelines for the determination of multi-way loading forming scheme and loading conditions of the forming cross valve components.

Multiple Model Soft Sensor Based on Affinity Propagation, Gaussian Process and Bayesian Committee Machine

Presented is a multiple model soft sensing method based on Affinity Propagation （AP）, Gaussian process （GP） and Bayesian committee machine （BCM）. AP clustering arithmetic is used to cluster training samples according to their operating points. Then, the sub-models are estimated by Gaussian Process Regression （GPR）. Finally, in order to get a global probabilistic prediction, Bayesian committee mactnne is used to combine the outputs of the sub-estimators. The proposed method has been applied to predict the light naphtha end point in hydrocracker fractionators. Practical applications indicate that it is useful for the online prediction of quality monitoring in chemical processes.

Simulation of Continuous Esterification Process of Polyester Polyols

Based on the kinetic and thermodynamic equations, a comprehensive mathematical model for the con- tinuous esterification process of polyester polyols was developed, which was carried out in an innovational bub- bling reactive distillation tower （BRDT） at atmospheric pressure. In this new type of reactor, direct esterification between ethylene glycol and adipic acid was accomplished efficiently and rapidly. A bench BRDT with the height of 2 m was applied for the esteriflcation process of l^oly （ethylene adlpate） （P＇EA）. In the continuous operation, Hn- ear oligomers were discharged from the bottom of the column, while water passed a few column trays and a pack- ing section as a condensation byproduct. The influence of major operating conditions on reactor performance was also simulated. Simulation results were in good agreement with experimental data, providing a strategy for devel- oping and optimizing this process.

Modeling and monitoring of nonlinear multi-mode processes based on similarity measure-KPCA

A new modeling and monitoring approach for multi-mode processes is proposed.The method of similarity measure(SM) and kernel principal component analysis(KPCA) are integrated to construct SM-KPCA monitoring scheme,where SM method serves as the separation of common subspace and specific subspace.Compared with the traditional methods,the main contributions of this work are:1) SM consisted of two measures of distance and angle to accommodate process characters.The different monitoring effect involves putting on the different weight,which would simplify the monitoring model structure and enhance its reliability and robustness.2) The proposed method can be used to find faults by the common space and judge which mode the fault belongs to by the specific subspace.Results of algorithm analysis and fault detection experiments indicate the validity and practicability of the presented method.

Thermodynamic simulation of complex Pb-Bi concentrate oxidative bath smelting process

The element partitioning in a Pb-Bi concentrate oxygen-rich bath smelting process was studied using thermodynamic equilibrium simulation method. Effects of oxygen to feed ratio(OFR) and sulfur dioxide partial pressure(pSO2) on the partitionings of Bi, Pb, As, Sb, Cu and Ag were analyzed and compared with industrial data. The results suggested that the optimal OFR was between 6.3 and 6.8 kmol/t to maximize Bi, Pb, Cu and Ag partitioning in the metal phase. Further increase of OFR led to the drop of metal partitioning and increase of slag liquidus temperature. High pSO2 led to high deportment of Bi and Pb in the gas phase mainly in the form of sulfides, suggesting that a low pSO2 was conducive for reducing the dust ratio.

An aligned mixture probabilistic principal component analysis for fault detection of multimode chemical processes

A novel approach named aligned mixture probabilistic principal component analysis(AMPPCA) is proposed in this study for fault detection of multimode chemical processes. In order to exploit within-mode correlations,the AMPPCA algorithm first estimates a statistical description for each operating mode by applying mixture probabilistic principal component analysis(MPPCA). As a comparison, the combined MPPCA is employed where monitoring results are softly integrated according to posterior probabilities of the test sample in each local model. For exploiting the cross-mode correlations, which may be useful but are inadvertently neglected due to separately held monitoring approaches, a global monitoring model is constructed by aligning all local models together. In this way, both within-mode and cross-mode correlations are preserved in this integrated space. Finally, the utility and feasibility of AMPPCA are demonstrated through a non-isothermal continuous stirred tank reactor and the TE benchmark process.

A novel multimode process monitoring method integrating LCGMM with modified LFDA

Complex processes often work with multiple operation regions, it is critical to develop effective monitoring approaches to ensure the safety of chemical processes. In this work, a discriminant local consistency Gaussian mixture model(DLCGMM) for multimode process monitoring is proposed for multimode process monitoring by integrating LCGMM with modified local Fisher discriminant analysis(MLFDA). Different from Fisher discriminant analysis(FDA) that aims to discover the global optimal discriminant directions, MLFDA is capable of uncovering multimodality and local structure of the data by exploiting the posterior probabilities of observations within clusters calculated from the results of LCGMM. This may enable MLFDA to capture more meaningful discriminant information hidden in the high-dimensional multimode observations comparing to FDA. Contrary to most existing multimode process monitoring approaches, DLCGMM performs LCGMM and MFLDA iteratively, and the optimal subspaces with multi-Gaussianity and the optimal discriminant projection vectors are simultaneously achieved in the framework of supervised and unsupervised learning. Furthermore, monitoring statistics are established on each cluster that represents a specific operation condition and two global Bayesian inference-based fault monitoring indexes are established by combining with all the monitoring results of all clusters. The efficiency and effectiveness of the proposed method are evaluated through UCI datasets, a simulated multimode model and the Tennessee Eastman benchmark process.

Phase Analysis and Identification Method for Multiphase Batch Processes with Partitioning Multi-way Principal Component Analysis (MPCA) Model

Multi-way principal component analysis (MPCA) is the most widely utilized multivariate statistical process control method for batch processes. Previous research on MPCA has commonly agreed that it is not a suitable method for multiphase batch process analysis. In this paper, abundant phase information is revealed by way of partitioning MPCA model, and a new phase identification method based on global dynamic information is proposed. The application to injection molding shows that it is a feasible and effective method for multiphase batch process knowledge understanding, phase division and process monitoring.

Continuous FEM simulation of multi-pass plate hot rolling suitable for plate shape analysis

In order to continuously simulate multi-pass plate rolling process,a 3-D elastic hollow-roll model was proposed and an auto mesh-refining module with data passing was developed and integrated with FE software,Marc.The hollow-roll model has equivalent stiffness of bending resistance and deformation to the real solid and much less meshes,so the computational time is greatly reduced.Based on these,the factors influencing plate profile,such as the roll-bending force,initial crown,thermal crown and heat transfer during rolling and inter-pass cooling can be taken into account in the simulation.The auto mesh-refining module with data passing can automatically refine and re-number elements and transfer the nodal and elemental results to the new meshes.Furthermore,the 3-D modeling routine is parametrically developed and can be run independently of Marc pre-processing program.A seven-pass industrial hot rolling process was continuously simulated to validate the accuracy of model.By comparison of the calculated results with the industrial measured data,the rolling force,temperature and plate profile are in good accordance with the measured ones.

Conventional and Predictive Control Algorithms for Controlling Nonlinear Processes Using Multiple-Model Approach

The objective of this work is to formulate and demonstrate the methodology of multi-models for improving the performance of existing advanced control strategies. Multiple models are used to capture the nonlinear process dynamics relating to gain and time constant variations. The multi-model strategy was implemented on several controllers such as Smith-Predictor using PI （Proportional-lntegral） and GPC （Generalized Predictive Control）. Computer simulations and experiments were conducted on several nonlinear systems and compared to the original form of these controllers. The enhanced approach was tested on controlling the screw speed of an injection molding machine and temperature of a steel cylinder.

Adaptive predictive functional control based on Takagi-Sugeno model and its application to pH process

In order to obtain accurate prediction model and compensate for the influence of model mismatch on the control performance of the system and avoid solving nonlinear programming problem,an adaptive fuzzy predictive functional control(AFPFC) scheme for multivariable nonlinear systems was proposed.Firstly,multivariable nonlinear systems were described based on Takagi-Sugeno(T-S) fuzzy models;assuming that the antecedent parameters of T-S models were kept,the consequent parameters were identified on-line by using the weighted recursive least square(WRLS) method.Secondly,the identified T-S models were linearized to be time-varying state space model at each sampling instant.Finally,by using linear predictive control technique the analysis solution of the optimal control law of AFPFC was established.The application results for pH neutralization process show that the absolute error between the identified T-S model output and the process output is smaller than 0.015;the tracking ability of the proposed AFPFC is superior to that of non-AFPFC(NAFPFC) for pH process without disturbances,the overshoot of the effluent pH value of AFPFC with disturbances is decreased by 50% compared with that of NAFPFC;when the process parameters of AFPFC vary with time the integrated absolute error(IAE) performance index still retains to be less than 200 compared with that of NAFPFC.

The correlation between nitrogen species in coke and NO_x formation during regeneration

Nitrogen oxides （NOx） emission during the regeneration ofcoked fluid catalytic cracking （FCC） catalysts is an en- vironmental issue. In order to identify the correlations between nitrogen species in coke and different nitrogen- containing products in tail gas, three coked catalysts with multilayer structural coke molecules were prepared in a fixed bed with model compounds （o-xylene and quinoline） at first. A series of characterization methods were used to analyze coke, including elemental analysis, FT-IR, XPS, and TG-MS. XPS characterization indicates all coked catalysts present two types of nitrogen species and the type with a higher binding energy is related with the inner part nitrogen atoms interacting with acid sites. Due to the stronger adsorption ability on acid sites for basic nitrogen compounds, the multilayer structural coke has unbalanced distribution of carbon and ni- trogen atoms between the inner part and the outer edge, which strongly affects gas product formation. At the early stage of regeneration, oxidation starts from the outer edge and the product NO can be reduced to N2 in high CO concentration. At the later stage, the inner part rich in nitrogen begins to be exposed to 02. At this period, the formation of CO decreases due to lack of carbon atoms, which is not beneficial to the reduction of NO. There- fore, nitrogen species in the inner part of multilayer structural coke contributes more to NOx formation. Based on the multilayer structure model of coke molecule and its oxidation behavior, a possible strategy to control NOx emission was discussed merely from concept.

Two-Degrees-of-Freedom Decoupling Control for Stable Multivariable Processes

This paper proposes a decoupling control scheme with two-degrees-of-freedom (2DOF) control structure. In the proposed scheme, two multivariable controllers are designed based on Internal Model Control (IMC) theory for setpoint tracking and disturbance rejection independently. An analytical approximation method is utilized to reduce the order of the controllers. By adjusting the corresponding controller parameter, the setpoint tracking and disturbance rejection of each control loop can be tuned independently. In the presence of multiplicative input uncertainty, a calculation method is also proposed to derive the low bounds of the control parameters in order to guarantee the robust stability of the system. Simulations are illustrated to demonstrate the validity of the proposed control scheme.

The effects of vegetation on runoff and soil loss:Multidimensional structure analysis and scale characteristics

This review summarizes the effects of vegetation on runoff and soil loss in three dimensions: vertical vegetation structures(aboveground vegetation cover, surface litter layer and underground roots), plant diversity, vegetation patterns and their scale characteristics. Quantitative relationships between vegetation factors with runoff and soil loss are described. A framework for describing relationships involving vegetation, erosion and scale is proposed. The relative importance of each vegetation dimension for various erosion processes changes across scales. With the development of erosion features(i.e., splash, interrill, rill and gully), the main factor of vertical vegetation structures in controlling runoff and soil loss changes from aboveground biomass to roots. Plant diversity levels are correlated with vertical vegetation structures and play a key role at small scales, while vegetation patterns also maintain a critical function across scales(i.e., patch, slope, catchment and basin/region). Several topics for future study are proposed in this review, such as to determine efficient vegetation architectures for ecological restoration, to consider the dynamics of vegetation patterns, and to identify the interactions involving the three dimensions of vegetation.

A generalized model of island biogeography

MacArthur and Wilson's equilibrium theory is one of the most influential theories in ecology.Although evolution on islands is to be important to island biodiversity,speciation has not been well integrated into island biogeography models.By incorporating speciation and factors influencing it into the MacArthur-Wilson model,we propose a generalized model unifying ecological and evolutionary processes and island features.Intra-island speciation may play an important role in both island species richness and endemism,and the contribution of speciation to local species diversity may eventually be greater than that of immigration under certain conditions.Those conditions are related to the per species speciation rate,per species extinction rate,and island features,and they are independent of immigration rate.The model predicts that large islands will have a high,though not the highest,proportional endemism when other parameters are fixed.Based on the generalized model,changes in species richness and endemism on an oceanic island over time were predicted to be similar to empirical observations.Our model provides an ideal starting point for re-evaluating the role of speciation and re-analyzing available data on island species diversity,especially those biased by the MacArthur-Wilson model.

A novel multimode process monitoring method integrating LDRSKM with Bayesian inference

A local discriminant regularized soft k-means （LDRSKM） method with Bayesian inference is proposed for multimode process monitoring. LDRSKM extends the regularized soft k-means algorithm by exploiting the local and non-local geometric information of the data and generalized linear discriminant analysis to provide a better and more meaningful data partition. LDRSKM can perform clustering and subspace selection simultaneously, enhancing the separability of data residing in different clusters. With the data partition obtained, kernel support vector data description （KSVDD） is used to establish the monitoring statistics and control limits. Two Bayesian inference based global fault detection indicators are then developed using the local monitoring results associated with principal and residual subspaces. Based on clustering analysis, Bayesian inference and manifold learning methods, the within and cross-mode correlations, and local geometric information can be exploited to enhance monitoring performances for nonlinear and non-Gaussian processes. The effectiveness and efficiency of the proposed method are evaluated using the Tennessee Eastman benchmark process.