Related Experiment Videos
Artificial neural networks for computer-aided modelling and optimisation in micellar electrokinetic chromatography
J Havel1, M Breadmore, M Macka
1Department of Analytical Chemistry, Faculty of Science, Masaryk University, Brno, Czech Republic.
Journal of Chromatography. A
|August 24, 1999
Summary
Artificial neural networks (ANNs) effectively model capillary micellar electrochromatography (MEKC) separations. Combining ANNs with experimental design significantly reduces experiments needed for optimal separation conditions in MEKC.
Area of Science:
- Analytical Chemistry
- Chromatography
- Chemometrics
Background:
- Capillary micellar electrochromatography (MEKC) is a powerful separation technique.
- Optimizing MEKC methods often requires extensive experimental work.
- Non-linear relationships in MEKC separations are challenging to model.
Purpose of the Study:
- To model the separation process in MEKC using artificial neural networks (ANNs).
- To facilitate MEKC method optimization by integrating ANNs with experimental design.
- To propose a generalizable computer-aided optimization approach for separation techniques.
Main Methods:
- Utilizing artificial neural networks (ANNs) for non-linear modeling of MEKC response surfaces.
- Combining ANN modeling with experimental design principles.
- Developing a computer-aided optimization strategy.
Main Results:
- ANNs demonstrated effective non-linear modeling capabilities for MEKC.
- The combined approach significantly reduced the number of required experiments for optimization.
- A general approach for computer-aided optimization was successfully proposed.
Conclusions:
- ANNs offer a robust tool for modeling complex MEKC separation processes.
- Integrating ANNs with experimental design enhances efficiency in MEKC method development.
- The proposed optimization strategy is broadly applicable across various separation techniques.