kinetic data evaluation
Multi-output machine learning models for kinetic data evaluation : A Fischer–Tropsch synthesis case study
Machine learning model like Lasso regression is not sufficient for multi-output Fischer Tropsch synthesis prediction. Artificial Neural Network regression able to capture complex non-linearity in Fischer Tropsch synthesis kinetic data. Shap interpretation technique finds process variable ranking in model prediction. Predicting the impact of input process variables on chemical processes is key to assess their performance of the latter. Models explaining this impact through a mechanistic approach are rarely readily available, complex in nature and/or require long development time.