Artificial neural networks experimental design

Plumb AP, Rowe RC, York P, Doherty C. The effect of experimental design in the modelling of a tablet coating formulation using artificial neural networks. Eur J Pharm Sci 2002 16 281-8. 

Chapter 17 - Vapor-liquid equilibrium (VLE) data are important for designing and modeling of process equipments. Since it is not always possible to carry out experiments at all possible temperatures and pressures, generally thermodynamic models based on equations on state are used for estimation of VLE. In this paper, an alternate tool, i.e. the artificial neural network technique has been applied for estimation of VLE for the binary systems viz. tert-butanol+2-ethyl-l-hexanol and n-butanol+2-ethyl-l-hexanol. The temperature range in which these models are valid is 353.2-458.2K at atmospheric pressure. The average absolute deviation for the temperature output was in range 2-3.3% and for the activity coefficient was less than 0.009%. The results were then compared with experimental data. 

Optimization can be simplified by employing the predictive capabilities of an artificial neural network (ANN). This multivariate approach has been shown to require minimal number of experiments that allow construction of an accurate experimental response surface (5, 6). The apposite model created from an experimental design should effectively relate the experimental parameters to the output values, which can be used to create an ANN with a strong predictive capacity for any conditions defined within the experimental space (4). 

However, no book on experimental design of this scope can be considered exhaustive. In particular, discussion of mathematical and statistical analysis has been kept brief Designs for factor studies at more than two levels are not discussed. We do not describe robust regression methods, nor the analysis of correlations in responses (for example, principle components analysis), nor the use of partial least squares. Our discussion of variability and of the Taguchi approach will perhaps be considered insufficiently detailed in a few years. We have confined ourselves to linear (polynomial) models for the most part, but much interest is starting to be expressed in highly non-linear systems and their analysis by means of artificial neural networks. The importance of these topics for pharmaceutical development still remains to be fully assessed. 

Cadmium is sometimes determined together with other metals such as zinc, lead, and mercury. Generally, these applications are based on the formation of metal-ligand complexes, and some kinds of statistical or chemometric tools (full factorial designs, sequential experimental Doehlert designs, artificial neural network, partial least square) are used for distinguishing the concentrations of each analyte. In addition, some applications are based on the kinetic differentiation of the reactions related to metals and the same ligands. 

Neural Networits for Optimization of High-Performance Capillary Zone Electrophoresis Methods. A New Meth Using a Combination of Experimental Design and Artificial Neural Networks. 

Lee et al. [36] developed an artificial neural network model for use to design and analysis PEM fuel cell power systems. The artificial neural network model can simulate the experimental data for different operating conditions and hence can be used to investigate the influence of process variables. 

The generated surfaces are measured using the Talysurf instmment and further processed to get fractal dimension (D). The experimental resnlts are used for further analysis nsing artificial neural networks to model fractal dimension. For ANN, full factorial design of experiments is considered and the design matrix and the experimental resnlts of cylindrical grinding of mild steel work-pieces are presented in Table 5.13 (Barman and Sahoo, 2011). 

The main reason for the utility of artificial neural networks (ANN) for modeling purposes is related to the incomplete knowledge of the mechanisms of polymerization processes and to the fact that the available phenomenological models are being developed and solved with difficulty or do not provide accurate results. For these types of processes, designing neural networks represents modeling alternatives for providing predictions useful for experimental practice. 

In general, it should be borne in mind that artificial neural networks do not consider the physics of processes in any way they simply reflect the experimental information used for their training. However, if properly applied, they can significantly support the design of processes and formulations. 

Artificial Neural Network (ANN) is another assistant method for the catalyst design, which is used more often presently. Compared with the expert system, ANN is under the case of a rather obscure catal3ftic mechanism based on the experimental data, and is able to establish the reflecting relationship between the catalyst compositions, preparing conditions and catalytic properties (including selectivity and conversion) and then obtain such parameters as the proportions through with the optimizations. This kind of method is simple and universally applicable, and is especially suitable for the design of the multi-component catalysts. 

Artificial neural networks and response surface methodology in combination with experimental design were developed. The optimisation by ANN and RSM in combination with a central composite design (CCD) was compared. 

---