Title : Using artificial intelligence to predict the performance of Al2O3 supported Mixed-Metal catalysts for the oxidative dehydrogenation of n-Butane
The development of efficient and selective catalysts for the oxidative dehydrogenation of n-Butane to produce alkenes with higher value has been a subject of intense research in recent years. In this project, we present a novel approach for predicting the performance of mixed metal catalysts supported on Al2O3 for this reaction using artificial intelligence (AI). Specifically, ANN, NuSVR, XGBR, and GBR machine learning algorithms were trained with a dataset of consistent experimental data to build the model using reaction temperature, reaction pressure, feed ratio, and catalyst composition as input features to predict the yield of dehydrogenation products as a measure of the catalyst performance. The results show that the AI-based model can accurately predict the performance of mixed metal catalysts for oxidative dehydrogenation of n-Butane, with prediction accuracy of 83%, 87%, 89%, and 91% based on test data achieved by the ANN, NuSVR, XGBR, and GBR models, respectively. Feature importance Analysis also revealed that the amount of Fe in the catalyst has the greatest influence on the yield of dehydrogenation products. These findings demonstrate that accurate predictions of catalyst performance can be made even with simple and easily accessible features, thus paving the way for the development of more efficient catalyst discovery and design methods.
Audience Take Away:
- Accurate predictions of catalyst performance can be achieved with simple and easily accessible features.
- The AI-based approach opens avenues for more efficient catalyst discovery, design and tuning methods.
- The amount of Fe in the catalyst has the greatest influence on the yield of dehydrogenation products.
- Machine learning algorithm implementation (including ANN, NuSVR, XGBR, and GBR) were trained with experimental data and input features to build the predictive model.