Search

This study proposes a novel system designed to turn waste masks (WM) into hydrogen while aiming at carbon neutrality. The system utilizes plasma gasification to primarily convert WM into syngas, which is subsequently upgraded through water gas shift to enhance hydrogen fraction. To mitigate carbon emissions, carbon capture techniques including monoethanolamine-based carbon absorption and oxy-fuel combustion are incorporated into the system. A high-fidelity model of the designed process is developed in Aspen Plus. Machine learning-driven optimization is applied to identify the optimal operating conditions. Utilizing Gaussian process regression-based surrogate optimization, the system achieves around 0.5 kg CO 2 eq/kg H 2 of levelized emissions per unit hydrogen (LEH) in the reaction stage and the execution time required for optimization is only 4 s, outperforming detailed process models-based optimization. Energy assessment highlights plasma gasification and CO 2 absorption as the primary sources of energy loss. Furthermore, life cycle analysis indicates that CO 2 absorption is the primary contributor to carbon emissions, accounting for approximately 60% of the total emissions. With levelized emissions of 1.27 kg CO 2 eq/kg WM and 5.18 CO 2 eq/kg H 2 , the designed system demonstrates superiority over conventional medical waste treatment and hydrogen production methods. • A waste mask-to-H 2 system with carbon capture is designed and modeled. • Zero direct carbon emission is achieved. • GPR-based surrogate optimization demonstrates high efficiency. • The optimal system shows an emission of 1.27 kg CO 2 eq per kg waste mask. • The system is potential in carbon reduction by comparative analysis. [ABSTRACT FROM AUTHOR]

A novel co-valorization process integrating plasma gasification and Fischer-Tropsch synthesis is designed for converting medical waste (MW) and biomass waste (BMW) into mixed e-fuels where the additional required hydrogen is supplied by solar-based electrolysis. Optimization and comprehensive assessments have been conducted for three scenarios with different ratios of BMW. Operations optimization based on genetic algorithm (GA) has led to a remarkable enhancement in the quality of syngas, with a more than 10% increase in H 2 mole fraction. Techno-economic analysis reveals the net present values (NPVs) of three scenarios are −2.56 MM$, 3.38 MM$ and 40.1 MM$ with internal rates of return of 3.4 %, 9.5 % and 14.0 %. By improving the subside of MW treatment, the economic viabilities of all scenarios have been enhanced significantly with higher positive NPVs. Environmental assessment shows ∼2.5 kg eqCO 2 /kg waste and ∼0.1 g eqCO 2 per MJ fuel have been generated across the system boundary and the operation of the designed process has been identified to be the largest source of emissions. These findings underscore the potential of our integrated approach to efficiently convert MW and BMW into valuable e-fuels while maintaining a focus on environmental sustainability. [Display omitted] • Co-valorization process was designed for converting wastes into mixed e-fuel. • Plasma gasification and Fischer-Tropsch synthesis were integrated. • Scenarios with different blending ration were analyzed. • The H 2 fraction in syngas has been improved by over 10% via optimization. • Economic and environmental assessments were performed. [ABSTRACT FROM AUTHOR]

An innovative valorization process for biomass waste treatment has been developed in this study. The basic simulation process developed considering experimental studies, and it has been optimized by application of a pattern search algorithm to obtain the optimum output of energy efficiency, economic performance, and process safety index (PSI). Mainly, three different scenarios have been analyzed based on the sustainability index (SI). These SI has been determined through energy, economy, safety, and electric power potential. According to the results of the SI, the optimized processes are more sustainable (SI = 0.517 to 0.563) in all scenarios as compared to the basic process (SI = 0.503). Similarly, the economic analysis in terms of internal rate of return (%) of the basic process varies from 37 to 2% when the process efficiency dropped from 100 to 80%, and it varies from 44 to 6% for the optimized process without any subsidy. The optimized process has a potential of 2788–4044 kW of electric power generation from thermal energy of the process while it is only 2302 kW for basic scenario. Therefore, the optimized process has better energy, economic, electric power, and safety performances, thus it is more sustainable as compared with the basic process. [Display omitted] • Integration of pyrolysis and gasification for biomass valorization. • Pattern search algorithm application for valorization process optimization. • A novel sustainability index has been proposed. • Electricity potential from optimized process thermal energy is 2788–4044 kW. • Energy efficiency of the processes varies from 46% to 66%. [ABSTRACT FROM AUTHOR]

Operations optimization in an organic Rankine cycle (ORC) based combined system is important while computationally difficult by using mechanistic models due to complex nonlinearities and constraints. In this study, a hybrid framework integrating machine learning and mathematical programming has been proposed to optimize the operations of the system for the best exergy performance. The combined system is first decomposed into two single ORCs for reducing computational complexity. Classification models and regression models based on artificial neural network (ANN) and linear regression are developed using simulation data, where classifications can be employed for high-throughput screening feasible inputs which meet the mechanistic constraints in ORC. The results demonstrate high performances of machine learning with at least 99% accuracies for classifications and with mean relative errors of less than 1% for regressions. These data-driven models and the relation of two ORCs were then embedded with mathematical programming for optimization and maximum net exergy of 28.66 MW is obtained. By linear expansion of ReLU operators in ANN, mixed-integer linear programming (MILP) based on machine learning models achieve high efficiency with ∼0.1 s required for optimization compared to mixed-integer nonlinear programming (MINLP) (>1000 s) and heuristic optimization based on mechanistic models (>10 h). • Machine learning and math programming are integrated for ORC optimization. • Classification is used for high-throughput identifying the mechanistic constraints. • Machine learning achieve >99% accuracy and < 1% mean relative error. • ReLU activation function is expressed by linear equations. • The method's efficiency is validated by optimization of ORC-based combined system. [ABSTRACT FROM AUTHOR]

This study evaluated three blue hydrogen production processes - solar-grid powered, wind-powered, and thermal power grid (TPG) - considering the context of Hong Kong, SAR, China. A process sustainability analysis was performed based on energy, economic, and environmental (3E) factors. The energy efficiency analysis indicates that the TPG system is the most energy-efficient with 64% efficiency, followed by the wind power system at 63% and the hybrid solar-grid powered system at 60%. The economic analysis results indicate that the levelized cost of hydrogen (LCH) is 2.165 $/kg for the TPG system, 2.132 $/kg for the hybrid solar-grid powered system, and 2.060 $/kg for the wind power system. The environmental assessment suggests that wind powered are eco-friendly with a unit point total (µPt) of 1.11, compared to the solar-grid 1.50 µPt and TPG system's 1.70 µPt. Therefore, 3E analysis proposes wind powered process is more sustainable for blue H 2 production., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)