The Influence of Input and Output Measurement Noise on Batch-End Quality Prediction with Partial Least Squares

Vanlaer, Jef; Kerkhof, Pieter Van den; Gins, Geert; Impe, Jan Van

doi:10.1007/978-3-642-31488-9_11

Cited by 2 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This enables unrealistically tight control of the process around its temperature and pH set points, greatly facilitating SPM. Many authors already recognized this problem and manually added measurement noise to the simulation data (e.g., [2][3][4]6,14,32,35,37,50,79,87,93]). In addition, Pensim only simulates a limited set of process upsets.…”

Section: Practical Implementationmentioning

confidence: 99%

An extensive reference dataset for fault detection and identification in batch processes

Impe¹,

Gins²

2015

Chemometrics and Intelligent Laboratory Systems

Self Cite

View full text Add to dashboard Cite

Close process monitoring (i.e., detection and identification of disturbances) is important to achieve high process efficiency and safety. The Tennessee Eastman process is an extensive benchmark dataset for fault detection and identification, but it is only representative for continuous processes because it does not contain the inherent nonstationarity that complicates monitoring of batch processes. Nevertheless, batch processes also play an important role in many types of industry. This paper therefore presents an extensive reference dataset for benchmarking data-driven methodologies for fault detection and identification in batch processes. The original Pensim model [10] is expanded with sensor noise. By changing the properties of the initial conditions and/or model parameters, four subsets of different complexity are generated, each containing 400 batches with normal operation. To correctly assess the fault detection and identification in batch processes, 15 faults are simulated with various amplitudes and onset times for a total of 22,200 faulty batches for each subset, or 90,400 batches in total. Analysis of the data indicates that the presented types of process faults and their various amplitudes in each of the four subsets present a suitable benchmark for fault detection and identification in batch processes. The dataset is freely available at http://cit.kuleuven.be/biotec/batchbenchmark.

show abstract

Section: Practical Implementationmentioning

confidence: 99%

An extensive reference dataset for fault detection and identification in batch processes

Impe¹,

Gins²

2015

Chemometrics and Intelligent Laboratory Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…Measurement noise is simulated in the form of additive random numbers sampled from a normal distribution with zero mean and standard deviation (stdev) as indicated in Table 2. 51 Process variability is generated by randomly changing the values of some initial conditions and some operating variables, as detailed in Table 3. Further variability is generated by assuming that the threshold glucose concentration determining the switch between operating steps 1 and 2 randomly varies between 0.3 and 7 g/L.…”

Section: Case Study #2mentioning

confidence: 99%

“…Measurement noise is simulated in the form of additive random numbers sampled from a normal distribution with zero mean and standard deviation ( stdev ) as indicated in Table . Process variability is generated by randomly changing the values of some initial conditions and some operating variables, as detailed in Table .…”

Section: Case Studiesmentioning

confidence: 99%

Optimal Indicator-Variable Approach for Trajectory Synchronization in Uneven-Length Multiphase Batch Processes

Sartori,

Facco,

Zuecco

et al. 2023

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Partial least-squares regression models assessing the end-point product quality in batch processes require that all of the measured variable trajectories across the historical batches have the same length. Most of the conventional and advanced methodologies for batch synchronization need some prior knowledge about the process to carry out one or more of the following activities: partitioning of the batches into phases, selection of an appropriate indicator variable that is then used to synchronize the batches, or selection of a reference batch to which all other batches are matched. We present an optimal indicator-variable approach for phase partitioning and trajectory synchronization in uneven-length multiphase batch processes. The main advantages are that partitioning into phases and selection of the most appropriate indicator variable within each phase are performed automatically rather than manually and are carried out simultaneously rather than disjointly based on a surrogate optimization framework that maximizes the performance of the product quality assessment model under development. Therefore, differently from conventional and advanced synchronization methodologies currently available, the proposed method is completely process-agnostic, which enhances applicability to complex batch processes. Also, in terms of computational times, it scales favorably with the calibration data set size. An industrial fed-batch process for the manufacturing of a specialty chemical and a simulated fed-batch process for the manufacturing of penicillin are used as test beds and demonstrate that the new indicator-variable approach has a superior performance than models built using other synchronization strategies.

show abstract

The Influence of Input and Output Measurement Noise on Batch-End Quality Prediction with Partial Least Squares

Cited by 2 publications

References 13 publications

An extensive reference dataset for fault detection and identification in batch processes

An extensive reference dataset for fault detection and identification in batch processes

Optimal Indicator-Variable Approach for Trajectory Synchronization in Uneven-Length Multiphase Batch Processes

Contact Info

Product

Resources

About