First release of lignet

Comprehensive models of biomass pyrolysis are needed to develop renewable fuels and chemicals from biomass. Unfortunately, the detailed kinetic schemes required to optimize industrial biomass pyrolysis processes are too computationally expensive to include in models that account for both kinetics and transport within reacting particles. lignet is a project to use machine learning to train neural nets and decision trees that can reproduce the results of my detailed ODE-based kinetic model for lignin pyrolysis, ligpy. The trained neural networks generalize very well, predicting the outputs of the detailed kinetic model with over 99.9% accuracy on new data, and reduce the computational cost by four orders of magnitude.

First release of lignet

Comprehensive models of biomass pyrolysis are needed to develop renewable fuels and chemicals from biomass. Unfortunately, the detailed kinetic schemes required to optimize industrial biomass pyrolysis processes are too computationally expensive to include in models that account for both kinetics and transport within reacting particles. lignet is a project to use machine learning to train neural nets and decision trees that can reproduce the results of my detailed ODE-based kinetic model for lignin pyrolysis, ligpy. The trained neural networks generalize very well, predicting the outputs of the detailed kinetic model with over 99.9% accuracy on new data, and reduce the computational cost by four orders of magnitude.