Machine Learning Reduced Order Model for Cost and Emission Assessment of a Pyrolysis System

Abstract

Biomass pyrolysis is a promising approach for producing economic and environmentally-friendly fuels and bioproducts. Biomass pyrolysis experiments show that feedstock properties have a significant impact on product yields and composition. Scientists are developing detailed chemical reaction mechanisms to capture the relationships between biomass composition and pyrolysis yields. These mechanisms can be computationally intensive. In this study, we investigate the use of a machine learning reduced order model (ROM) for assessing the costs and emissions of a pyrolysis biorefinery. We developed a Kriging-based ROM to predict pyrolysis yields of 314 feedstock samples based on the results of a detailed chemical kinetic pyrolysis mechanism. The ROM is integrated into a chemical process model for calculating mass and energy yields in a commercial-scale (2000 tonne/day) biorefinery. The ROM estimated biofuel yields of 65 to 130 gallons per ton of dry biomass. This results in biofuel minimum fuel-selling prices of $2.62 to $5.43 per gallon and emissions of -13.62 to 145 kg of CO2 per MJ. The ROM achieved an average mean square error of 1.8e-20 and a mean absolute error of 0.53%. These results suggest that ROMs can facilitate rapid feedstock screening for biorefinery systems.

A Review of Automation and Sensors: Parameter Control of Thermal Treatments for Electrical Power Generation

“…Reduced order models configured with machine learning as long and short-term memory networks (LSTM) are alternatives within an initial scope of obtaining autonomous plants because they are smaller programmed systems that can be added to chemical and environmental modeling from other software, forming a complex environment in which different programming languages are used. Among these, the Python language stands out for its results in bio-fuel minimum selling prices, evaluating mass and energy yields as necessary feed-stock volumes for commercial-scale power-plant planning [149]. As presented in [150] and in [151], the use of LSTM may be even enhanced by applying filters such as wavelet transform.…”

Advanced Thermochemical Conversion Approaches for Green Hydrogen Production from Crop Residues

“…The biological method involving microorganisms and sunlight to convert water and organic matter to green hydrogen is eco-friendly, nontoxic, and requires low energy consumption but suffers from a lack of scalability and industrial utilization. The electrolysis and direct solar water splitting techniques guarantee green hydrogen [76], Copyright ©2021, America Chemical Society production with minimal GHG emissions and moderate production conditions. However, the use of water for hydrogen production impacts water supply, endangers aquatic ecosystems, and poor product quality.…”

Comparative Techno-economic Analysis and Life Cycle Assessment of Producing High-Value Chemicals and Fuels from Waste Plastic via Conventional Pyrolysis and Thermal Oxo-degradation

“…23,24 The financial assumptions made in this study follow closely with assumptions made in other TEA studies where the nth plant assumptions are used. 17,25 The study assumes that 40% of the expenses of the facilities are covered by equity, while 60% will be financed at a 10% interest rate and a 10 year repayment period. In estimation of the NPV of the refineries, the study assumes a 10% internal rate of return (IRR) over a 20 year plant life.…”