Abstract Most current batch process monitoring methods have been implemented under a single operation mode, thus the batch uncertainty is assumed to be caused by batch-to-batch variations. However, due to the change o...
Abstract Most current batch process monitoring methods have been implemented under a single operation mode, thus the batch uncertainty is assumed to be caused by batch-to-batch variations. However, due to the change of market requirements, the batch operation mode should also be changed frequently, especially in the semiconductor manufacturing process. This paper proposes an efficient method for monitoring those batch processes. First, the data is partitioned into multiple clusters, which correspond to different operation modes. Second, a sub-statistical model is built for each operation mode. Then the Bayesian inference strategy is introduced for result combination in different operation modes. The monitoring performance of the proposed method is evaluated by a real semiconductor process application case study.
Wireless sensor networks consist of a large number of sensor nodes that have low power and limited transmission range and can be used in various scenario. The nodes can be deployed in the long and narrow region, such ...
详细信息
Hysteresis nonlinearity exists in many physical actuators and actuator failures seem inevitable in practice. However, there is still no result available to compensate for failures of hysteric actuators in the design o...
详细信息
This article researches the problem of finite frequency (FF) H ∞ filtering for linear discrete-time state-delayed systems. The disturbance is assumed to reside in low/middle/high frequency ranges. To reduce the con...
详细信息
This article researches the problem of finite frequency (FF) H ∞ filtering for linear discrete-time state-delayed systems. The disturbance is assumed to reside in low/middle/high frequency ranges. To reduce the conservatism of the results, delay-partitioning idea is used to derive a new FF bounded real lemma (BRL). By applying the generalized Kalman-Yakubovich-Popov lemma, two equivalent approaches to the proof of the proposed FF BRL are given, respectively, starting from transfer function and Lyapunov-Krasovskii functional. A new FF H ∞ filter design method is proposed in terms of solving a set of linear matrix inequalities. Finally, a numerical example clearly demonstrates the merits and effectiveness of the proposed method.
Fed-batch processes are inherently difficult to model owing to non-steady-state operation, small-sample condition, instinct time-variation and batch-to-batch variation caused by drifting. Furthermore, when the process...
详细信息
Fed-batch processes are inherently difficult to model owing to non-steady-state operation, small-sample condition, instinct time-variation and batch-to-batch variation caused by drifting. Furthermore, when the process switches to different operation phrases, global learning modeling methods would suffer poor performance due to the negative impact of overdue training samples. In this paper, a k nearest neighbor relevance vector machine (kNN-RVM) based lazy learning method is proposed to model the fed-batch processes to soft-sense the corresponding production indices. A recursive algorithm is developed to effectively obtain the kernel matrices used by previous kNN step and following modeling process. Simulative soft-sensors of penicillin production process and rubber mixing process are implemented to valid the proposed method. Comparative results indict that proposed method has better precision and much lower computational complexity than relevance vector machine (RVM) on soft-sensing modeling of fed-batch processes.
Based on deep investigation to the static and dynamic characteristics of Hammerstein and Wiener model structures, this paper proposes an included angle based nonlinearity measure for SISO Hammerstein-like systems. The...
详细信息
Based on deep investigation to the static and dynamic characteristics of Hammerstein and Wiener model structures, this paper proposes an included angle based nonlinearity measure for SISO Hammerstein-like systems. The proposed measure is instrumental for both analysis and synthesis of nonlinear systems, especially when a multilinear model controller is designed. A CSTR process is studied to illustrate the effectiveness of the proposed nonlinearity measure.
Results of the analysis of published measurements of the air bubbles content in sea ice are presented. The study of interrelation of porosity (gas content) and salinity of ice was carried out and it was revealed their...
详细信息
This article utilizes the independent component regression (ICR) algorithm which is capable of extracting non-Gaussian components from both input and output variables for monitoring of complex industrialsystems. The ...
详细信息
This article researches the problem of finite frequency (FF) {H} \infty filtering for linear discrete-time state-delayed systems. The disturbance is assumed to reside in low/middle/high frequency ranges. To reduce the...
详细信息
暂无评论