Peter Hintenaus

Evolving chemometric models for predicting dynamic process parameters in viscose production

In viscose production, it is important to monitor three process parameters in order to assure a high quality of the final product: the concentrations of H(2)SO(4), Na(2)SO(4) and Z(n)SO(4). During on-line production these process parameters usually show a quite high dynamics depending on the fiber type that is produced. Thus, conventional chemometric models, which are trained based on collected calibration spectra from Fourier transform near infrared (FT-NIR) measurements and kept fixed during the whole life-time of the on-line process, show a quite imprecise and unreliable behavior when predicting the concentrations of new on-line data. In this paper, we are demonstrating evolving chemometric models which are able to adapt automatically to varying process dynamics by updating their inner structures and parameters in a single-pass incremental manner. These models exploit the Takagi-Sugeno fuzzy model architecture, being able to model flexibly different degrees of non-linearities implicitly contained in the mapping between near infrared spectra (NIR) and reference values. Updating the inner structures is achieved by moving the position of already existing local regions and by evolving (increasing non-linearity) or merging (decreasing non-linearity) new local linear predictors on demand, which are guided by distance-based and similarity criteria. Gradual forgetting mechanisms may be integrated in order to out-date older learned relations and to account for more flexibility of the models. The results show that our approach is able to overcome the huge prediction errors produced by various state-of-the-art chemometric models. It achieves a high correlation between observed and predicted target values in the range of [0.95,0.98] over a 3 months period while keeping the relative error below the reference error value of 3%. In contrast, the off-line techniques achieved correlations below 0.5, ten times higher error rates and the more deteriorate, the more time passes by.

An Infrared Spectrometer for Process Monitoring I, Spectroscopy

In this paper we discuss the design of a Fourier transform infrared spectrometer for the monitoring of chemical processes. We contrast our approach to computing interferograms, which relies heavily on numerical procedures computed on digital signal processors, with the more traditional of deriving triggers from the movement of the mirror in hardware. A presentation of our implementation attempts rounds out this paper.

Enhanced genetic operators design for waveband selection in multivariate calibration based on NIR spectroscopy

Nowadays, the techniques employed in data acquisition provide huge amounts of data. Some parts of the information are related to the others, making dimensionality reduction desirable, and losing less information as much as possible, in order to decrease computational times and complexity when applying any ensuing data mining technique. Genetic algorithms offer the possibility of selecting which variables contain the most relevant information to represent all the original ones. The traditional genetic operators seem to be too general, leading to results that could be improved by means of designed genetic operators that employ some available problem‐specific information. Especially, when dealing with calibration by means of near‐infrared spectral data, which use to contain thousands of variables, it is known that not isolated wavelengths but wavebands allow a more robust model design. This aspect should be taken into account when crossing individuals. We propose three crossover operators specifically designed for calibration with near‐infrared spectral data, based on a pseudo‐random two‐point crossover, where the first point is chosen randomly, and the selection of the second point is guided by problem‐specific information. We compare their performance with that of state‐of‐the‐art operators. We combine these new genetic algorithm‐based variable selection designs with partial least squares regression and fuzzy systems based calibration.

An infrared spectrometer for process monitoring II, chemometry and automatization

We describe a hard and software architecture for process control applications in the chemical, pharamceutical and food industries based on spectroscopic measurements. We argue for the tight integration of the spectrometer itself, the data analysis software and the measurement automatization to achieve situation awareness and predictible real-time behavior and to be able to handle complicated sampling situations, while keeping the programming effort for an individual installation low.

A fast and accurate near infrared spectroscopy method for the determination of cellulose content of alkali cellulose applicable for process control

A novel near infrared (NIR) spectroscopy method for the determination of the cellulose content of alkali cellulose, an intermediate product in viscose fibre production, is presented. This method is especially suitable for the purpose of process control. The method is realised in diffuse reflectance mode. A fast sample preparation step consisting of the compression of the coarse bulky alkali cellulose is introduced, in order to homogenise the sample material. The measurement takes 10 min, including the preparation step. For calibration of the method, samples taken directly from the production line have been used in combination with samples prepared in the laboratory. Validation conducted in a production environment yields a root mean square error of prediction of 0.36% w/w for cellulose content, which is sufficient for the detection of deviation from standard production parameters. It is found that the source material for the alkali cellulose influences the NIR spectra. Analysis of the source material indicates that the hemicellulose composition, i.e. a poly- or oligosaccharide dissolvable in caustic lye, has an impact on the NIR spectra. Different types of source materials can be discerned from their NIR spectra.

Ensembled Self-Adaptive Fuzzy Calibration Models for On-line Cloud Point Prediction

In this paper we investigate the usage of non-linear chemometric models, which are calibrated based on near infrared (FTNIR) spectra, in order to increase efficiency and to improve quantification quality in melamine resin production. They rely on fuzzy systems model architecture and are able to incrementally adapt themselves during the on-line process, resolving dynamic process changes, which may cause severe error drifts of static models. The most informative wavebands in NIR spectra are extracted by a new variant of forward selection, termed as forward selection with bands (FSB) and used as inputs for the fuzzy models. A specific ensemble strategy is developed which is able to properly compensate noise in repeated spectra measurements. Results on high-dimensional data from four independent types of melamine resin show that 1.) our fuzzy modelingmethodologycan outperform state-of-the-art chemometric modeling methods in terms of validation error, 2.) the ensemble strategy is able to improve the performance of models without ensembling and 3.) incremental model updates are necessary in order to prevent drifting residuals.

Dynamic Quantification of Process Parameters in Viscose Production with Evolving Fuzzy Systems

In viscose production, it is important to monitor three process parameters as part of the spin-bath in order to assure a high quality of the final product: the concentrations of H 2 SO 4, Na 2 SO 4 and ZnSO 4. During on-line production these process parameters usually show a quite high dynamics depending on the fibre type that is produced. Thus, conventional chemometric models, kept fixed during the whole life-time of the on-line process, show a quite imprecise and unreliable behavior when predicting the concentrations of new on-line data. In this paper, we are demonstrating evolving chemometric models based on TS fuzzy systems architecture, which are able to adapt automatically to varying process dynamics by updating their inner structures and parameters in a single-pass incremental manner. Gradual forgetting mechanisms are necessary in order to out-date older learned relations and to account for more flexibility and spontaneity of the models. The results show that our dynamic approach is able to overcome the huge prediction errors produced by various state-of-the-art static chemometric models, which could be verified on data recorded on-line over a three months period.

2.5.4 Reliable Online-Prediction of Characteristic Process Parameters by FTNIR-Spectroscopic Analysis

Within the industrial research project “Process Analytical Chemistry” (PAC) we are working on FTNIRspectroscopic measurement systems predicting characteristic parameters of industrial production processes. Those parameters are usually monitored offline or at-line with time consuming and expensive laboratory methods. In this contribution, we present a spectroscopic measurement configuration together with the required chemometric analysis, acting as an online-monitoring system. In order to demonstrate the potential of such a system we use the example of melamine resin production in an industrial process. At company partner Dynea the predicted value of the turbidity point is used as an indicator for the end of the batch reaction (turning off heating). Furthermore, we illustrate a way to verify the chemometric prediction by calculating a confidence interval for each predicted value.