Chemie und Prozesstechnik
Filtern
Erscheinungsjahr
- 2020 (2) (entfernen)
Dokumenttyp
- Zeitschriftenartikel (2) (entfernen)
Sprache
- Englisch (2)
Referierte Publikation
- ja (2)
Schlagworte
- Artificial neural networks (1)
- Automation (1)
- Exploratory analysis (1)
- FTIR (1)
- Microplastics (1)
- Online NMR spectroscopy (1)
- Process industry (1)
- Real-time process monitoring (1)
Organisationseinheit der BAM
- 1 Analytische Chemie; Referenzmaterialien (2) (entfernen)
Industry 4.0 is all about interconnectivity, sensor-enhanced process control, and data-driven systems. Process analytical technology (PAT) such as online nuclear magnetic resonance (NMR) spectroscopy is gaining in importance, as it increasingly contributes to automation and digitalization in production. In many cases up to now, however, a classical evaluation of process data and their transformation into knowledge is not possible or not economical due to the insufficiently large datasets available. When developing an automated method applicable in process control, sometimes only the basic data of a limited number of batch tests from typical product and process development campaigns are available. However, these datasets are not large enough for training machine-supported procedures. In this work, to overcome this limitation, a new procedure was developed, which allows physically motivated multiplication of the available reference data in order to obtain a sufficiently large dataset for training machine learning algorithms. The underlying example chemical synthesis was measured and analyzed with both application-relevant low-field NMR and high-field NMR spectroscopy as reference method. Artificial neural networks (ANNs) have the potential to infer valuable process information already from relatively limited input data. However, in order to predict the concentration at complex conditions (many reactants and wide concentration ranges), larger ANNs and, therefore, a larger Training dataset are required. We demonstrate that a moderately complex problem with four reactants can be addressed using ANNs in combination with the presented PAT method (low-field NMR) and with the proposed approach to generate meaningful training data.
Exploratory analysis of hyperspectral FTIR data obtained from environmental microplastics samples
(2020)
Hyperspectral imaging of environmental samples with infrared microscopes is one of the preferred methods to find and characterize microplastics. Particles can be quantified in terms of number, size and size distribution. Their shape can be studied and the substances can be identified. Interpretation of the collected spectra is a typical problem encountered during the analysis. The image datasets are large and contain spectra of countless particles of natural and synthetic origin. To supplement existing Analysis pipelines, exploratory multivariate data analysis was tested on two independent datasets. Dimensionality reduction with principal component analysis (PCA) and uniform manifold approximation and projection (UMAP) was used as a core concept. It allowed for improved visual accessibility of the data and created a chemical two-dimensional image of the sample. Spectra belonging to particles could be separated from blank spectra, reducing the amount of data significantly. Selected spectra were further studied, also applying PCA and UMAP. Groups of similar spectra were identified by cluster analysis using k-means, density based, and interactive manual clustering. Most clusters could be assigned to chemical species based on reference spectra. While the results support findings obtained with a ‘targeted analysis’ based on automated library search, exploratory analysis points the attention towards the group of unidientified spectra that remained and are otherwise easily overlooked.