Integrating data‐based modeling and nonlinear control tools for batch process control

Aumi, Siam; Mhaskar, Prashant

doi:10.1002/aic.12720

Cited by 63 publications

(14 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Actually, RSW is a dynamic process involving changes of current and voltage in the welding machine together with changes of the nugget and HAZ sizes of the welding sample. The dynamic considerations should be involved to better model the realistic process, monitor the final quality and handle possible dynamics (Aumi and Mhaskar, 2012; Aumi et al, 2013). In this section, our scheme for welding quality estimation will be presented, where the HAZ dynamic increment and expulsion time detection are explored, respectively.…”

Section: Methods For Welding Quality Estimationmentioning

confidence: 99%

Quality monitoring of spot welding with advanced signal processing and data-driven techniques

Wang

Zhou

Yan

et al. 2017

Transactions of the Institute of Measurement and Control

View full text Add to dashboard Cite

As resistance spot welding is a crucial and widely used metal joining technique nowadays, a cheap and highly accurate online quality monitoring scheme is strongly demanded in industry. In this paper, a novel framework for welding quality examination is proposed by using advanced signal processing and artificial intelligence techniques. Our proposed framework consists of two objective schemes, a general welding quality classification scheme and a more detailed and advanced welding quality estimation scheme. To achieve a fast, convenient and cheap monitoring strategy, only the easily obtained electrical signals are monitored for data acquisition. For the welding quality classification, a self-organizing map under a windowed feature extraction is applied. Results of the spot welding experiment from a portable welding machine show that the classification accuracy can be 92.9%. For welding quality estimation, variation of the welding time and detailed aspects of welding quality are introduced. In particular, a modified recurrent neural network is utilized for the size estimation of the heat affected zone, while a novel self-organizing map type classifier is used to detect the expulsion condition along with its occurrence time. During the spot welding experiments, the average error percentage of the heat affected zone size estimation is around 6.7% and the accuracy on expulsion detection reaches 93.3%.

show abstract

Section: Methods For Welding Quality Estimationmentioning

confidence: 99%

Quality monitoring of spot welding with advanced signal processing and data-driven techniques

Wang

Zhou

Yan

et al. 2017

Transactions of the Institute of Measurement and Control

View full text Add to dashboard Cite

show abstract

“…The central concept of multimodel approaches is the division of the process data into several groups such that each represents a single operation mode/phase. For most papers, mode/phase division is based on clustering algorithms; for example, Aumi and Mhaskar developed a multi‐model approach that integrates the theories of ARX modelling, PCA/PLS methods, fuzzy c‐means clustering, and multiple linear models. Outside of clustering algorithms, Zhao and her co‐workers detected the phase‐division points by comparing the control limits of two adjacent time‐segment models.…”

Section: Issues In Process Monitoringmentioning

confidence: 99%

Survey on the theoretical research and engineering applications of multivariate statistics process monitoring algorithms: 2008–2017

et al. 2018

View full text Add to dashboard Cite

Multivariate statistical process monitoring (MSPM) methods are significant for improving production efficiency and enhancing safety. However, to the authors’ best knowledge, there is no survey paper providing statistics of published papers over the past decade. In this paper, several issues related to MSPM methods are reviewed and studied. First, the annual publication numbers of journal articles concerning MSPM are provided to show the active development of this important research field and to point out several promising directions in the future. Second, the annual numbers of patents are also shown to demonstrate the practicality of different MSPM methods. Particularly, this paper also lists and analyzes the number of MSPM‐related publications in China. The statistics indicate that Chinese researchers and engineers may have different viewpoints from those of other countries, which results in different development trends of MSPM in China.

show abstract

“…The batch-to-batch quality control is an off-line control strategy, and it adjusts the entire input trajectory of the following batch runs. After the input trajectory is derived, the proportional-integral-derivative (PID) algorithm may be employed to adjust the manipulated variable to track the predefined trajectory. , However, PID is not an ideal choice for the process with multidimensional coupling variables. Golshan applied the predictive control method based on the latent variable model to achieve the batch-to-batch trajectory tracking. , This method calculates the adjustment in the low-dimensional latent variable space, then derives the manipulated variables according to the latent variable model, which performs better in the trajectory tracking than PID.…”

Section: Introductionmentioning

confidence: 99%

Input Trajectory Adjustment within Batch Runs Based on Latent Variable Models

Zhao

Wang

et al. 2019

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

In a batch process, the input trajectory is usually predefined before the beginning of a batch run through the product design according to the desired output. However, within a batch run, the trajectory tracking up to the current sampling instant may be influenced by process drifts and disturbances, so even the perfect track of remaining trajectory may not achieve the desired product. The projection to latent structure (PLS)-based trajectory adjustment method guarantees the final desired output by keeping the latent variable constant through the online adjustment of the remaining trajectory within a batch run. However, the latent variable extracted by PLS may contain the outputrelated information as well as the output-irrelevant information, and it is not appropriate to ensure the output by keeping the output-irrelevant information constant. Three contributions are proposed in this work: the condition to obtain the final desired output is analyzed and derived; a novel trajectory adjustment method is proposed based on the total projection to latent structures (TPLS) model, which further decomposes the latent variable extracted by PLS into the output-related latent variable and the output-irrelevant latent variable; and a trajectory adjustment space is proposed within which any adjustment can achieve the desired output. It is concluded that the outputrelated latent variable should be kept constant within a whole batch run to achieve the desired final output, and the outputirrelevant input part provides the freedom of adjustment, while the implemented trajectory constitutes the constraints on the adjustment. Through the applications in a numerical case and a continuous stirred reactor process, it is verified that the proposed method can effectively achieve the desired final output.

show abstract

Integrating data‐based modeling and nonlinear control tools for batch process control

Cited by 63 publications

References 30 publications

Quality monitoring of spot welding with advanced signal processing and data-driven techniques

Quality monitoring of spot welding with advanced signal processing and data-driven techniques

Survey on the theoretical research and engineering applications of multivariate statistics process monitoring algorithms: 2008–2017

Input Trajectory Adjustment within Batch Runs Based on Latent Variable Models

Contact Info

Product

Resources

About