Neural network aided design of Pt-Co-Ce/Al2O3 catalyst for selective CO oxidation in hydrogen-rich streams

Developing catalytic materials for the oxidative coupling of methane through statistical analysis of literature data

“…Consequently, quadratic or some special cubic models known as response surface have been traditionally used. 23 Owing to the increasing power of modern computers, it became possible to use more sophisticated and complex nonlinear models, such as artificial neural networks, 24 kernel regression, 25 or Gaussian processes 26 for catalyst research [27][28][29] and for mathematical modelling in catalysis. [30][31][32] Such empirical models can be trained on the basis of available catalytic data and hereafter used to predict performance of other catalysts that had not been used in the training.…”

Multi-component Catalyst Design for Oxidative Coupling of Methane Using ANFIS and ANN

“…Artificial neural network modeling, which was inspired from the functioning of biological nervous systems, is proved to be a powerful technique for complex and nonlinear problems with a strong ability to learn and predict (Gunay and Yıldırım, 2008). After "raw" neural network has been created (Figure 1), its weights and biases are changed from random initialized value to combination that provide the least amount of sum of square errors, these set of actions is called "training," and the classical method of training is the back propagation algorithm.…”

Modeling and optimization of synthesis condition of CuCr2O4 Spinel for CO oxidation

“…These metals are environmentally friendly [16][17][18][19]. Catalyst design is a complex process involving many steps and variables as well as complex interactions among these variables, which make the experimental studies quite expensive and time-consuming [20,21]. With the advancements in the scope of information technology, catalysts design by computer is very important [22,23].…”