Online process monitoring for complex systems with dynamic weighted principal component analysis

Conventional multivariate statistical methods for process monitoring may not be suitable for dynamic processes since they usually rely on assumptions such as time invariance or uncorrelation. We are therefore motivated to propose a new monitoring method by compensating the principal component analysis with a weight approach.The proposed monitor consists of two tiers. The first tier uses the principal component analysis method to extract cross-correlation structure among process data, expressed by independent components. The second tier estimates auto-correlation structure among the extracted components as auto-regressive models. It is therefore named a dynamic weighted principal component analysis with hybrid correlation structure. The essential of the proposed method is to incorporate a weight approach into principal component analysis to construct two new subspaces, namely the important component subspace and the residual subspace, and two new statistics are defined to monitor them respectively. Through computing the weight values upon a new observation, the proposed method increases the weights along directions of components that have large estimation errors while reduces the influences of other directions. The rationale behind comes from the observations that the fault information is associated with online estimation errors of auto-regressive models. The proposed monitoring method is exemplified by the Tennessee Eastman process. The monitoring results show that the proposed method outperforms conventional principal component analysis, dynamic principal component analysis and dynamic latent variable.

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.

INFLUENCE OF DOPING COMPONENT ON THE SOL-GEL PROCESS

Si-29 and Li-7 NMR have been used to study the sol-gel process. It has been found that the doping components (such as TlNO3, LiNO2, and Cd(NO3)(2)) could promote the hydrolysis and condensation reactions and their ratio, thus they could influence the gelation time. The lithium species has been demonstrated to remain as free hydration. ions during the whole sot-gel process and not to participate in forming the inorganic network.

Key Technology and Complete-set Equipment for Intelligent Construction of HSR Simply-supported Girder Reinforcement Framework

In view of the problems such as backward production mode,poor quality stability,high safety risk and incomplete control system during erection of the reinforcement framework of simply-supported box girders for high-speed railway(HSR),and in combination with the key points and main challenges in the reinforcement framework construction of Guangzhou-Zhanjiang HSR,the overall technical route for the intelligent manufacturing of reinforcement framework of simply-supported box girders is put forward.The component design of reinforcement framework of simply supported box girder is carried out based on BIM,and the feasibility of the scheme is verified through segment assembly test.The assembly techniques are studied in combination with the mesh design scheme to achieve rapid forming of the reinforcement framework.R&D of automatic processing equipment for components,material transshipment equipment,automatic hoisting equipment and technological equipment for assembly clamping fixture are carried out to realize the overall design of equipment production line.An intelligent control system is developed for the whole-process intelligent construction of reinforcement framework to realize the full life-cycle applications for the workshop production and visual management including intelligent layout and quality traceability.The research results systematically optimize and innovate the assembly and forming technologies of reinforcement framework in the prefabrication beam yard of high-speed railway,realize the component processing,automatic assembly and information technology management,improve the construction quality,efficiency and information technology level of intelligent manufacturing of reinforcement framework of railway prefabricated beam as a whole,and reduce the construction cost of the project.The research has realized a major breakthrough in the construction technology of railway prefabricated box girders,which has the extensive technical and market promotion values.

Improving the process forming limit considering forming defects in the transitional region in local loading forming of Ti-alloy rib-web components

The isothermal local loading forming technology provides a feasible way to form Ti-alloy large-scale rib-web components in aerospace and aviation fields.However,the local loading process forming limit is restricted by forming defects in the transitional region.In this work,the feasibility of controlling forming defects and improving the process forming limit by adjusting die parameters is explored through finite element（FE） simulation.It is found that the common cavum and folding defects in the transitional region are significantly influenced by the fillet radii of left rib and middle rib,respectively.The cavum and folding defects can be effectively controlled by increasing the fillet radii of left rib and middle rib,respectively.The process forming limits considering forming defects in the transitional region are determined by the stepwise searching method under various die parameters.Moreover,the relationship between the process forming limit and die parameters is developed through the response surface methodology（RSM）.The developed RSM models suggest that increasing the fillet radii of left and middle ribs is effective to improve the process forming limit during local loading forming of rib-web components.The results will provide technical basis for the design of die parameters and the reduction amount,which is of great importance to control forming defects and improve the process forming limit in local loading forming of Ti-alloy large-scale rib-web components.

Logistics Network Design for Manufacturing Enterprises with Component Processing Workshops

Logistics network design influences the efficiency and cost of Logistics directly.Some manufacturing enterprises not only have warehouse hubs,but also build component processing workshops which are usually located in those places where the costs of materials and workforce are lower.This paper establishes a logistics network design model for the manufacturing enterprises with component processing workshops based on 0-1 mixture integer programming.The model optimizes the logistics network in an integrated view,by which the selection of the nodes,the manufacturing plan,and transportation plan can be obtained.An example is given to verify its feasibility.The approach is helpful for designing of the logistics network in manufacturing enterprises.