Predicting nickel catalyst deactivation in biogas steam and dry reforming for hydrogen production using machine learning

Abstract

This study employs Random Forests (RF) and Artificial Neural Networks (ANN) to model the transient behavior of Ni catalyst deactivation during steam and dry reforming of model biogas containing H₂S, with a focus on hydrogen production. Deactivation, induced by carbon deposition and sulfur poisoning, is a complex and transient phenomenon demanding precise kinetic mechanisms for accurately predicting Ni catalyst behavior in biogas reforming. Black-box machine learning (ML) models are developed, incorporating catalyst properties, biogas composition, and operating conditions. Encompassing both dry and steam reforming, the ML models aim to predict catalyst behavior, expressed in terms of packed bed reactor exit mole fractions (H₂, CO, CH₄, and CO₂) and conversions (CH₄ and CO₂). The ML models are trained and tested across a temperature range of 700–900 C with 0–145 ppm of H₂S in model biogas (CH₄/CO₂ ratio varying from 1.0 to 2.0). RF outperforms the ANN across all performance metrics, including overall R² and root mean squared error (RMSE). The RF achieves a mean overall R² of 0.979, with training and testing RMSE equal to $6.7\times {10}^{-3}$ and $1.47\times {10}^{-2}$ respectively. In contrast, the ANN achieves a mean overall R² of 0.939, with training and testing RMSE equal to $2.6\times {10}^{-2}$ and $2.55\times {10}^{-2}$ respectively. Moreover, pre-trained RF models are validated with unseen data of dry reforming of biogas containing 30 ppm of H₂S (25 data points). It is suggested that 35 % of this unseen experimental data is required to train the RF model for it to predict catalyst deactivation, achieving a validation R sufficiently²> 0.9. The mean overall R² values attained by the RF fine-tuned on 35 % of the unseen experiment data for both CH₄ and CO₂ conversions, as well as for all mole fraction predictions, are 0.952 and 0.948, respectively.