Superstructure Optimization with Embedded Neural Networks for Sustainable Aviation Fuel Production

2025-09-11

AI Summary

Sustainable aviation fuel (SAF) superstructure optimization faces challenges in jointly optimizing discrete process selections and continuous operational parameters under stringent carbon constraints. Method: This study proposes a data-driven mixed-integer nonlinear programming (MINLP) framework that embeds artificial neural network (ANN) surrogate models into the superstructure optimization architecture, integrating Fischer–Tropsch synthesis, biomass gasification, and direct air capture with carbon storage (DAC-CS). The framework supports variable-composition material streams and customizable product specifications. Contribution/Results: By overcoming the limitations of fixed-process configurations, it enables simultaneous optimization of process topology and operating conditions. Under zero-carbon constraints, the autothermal reforming (ATR)-biomass gasification hybrid pathway achieves the lowest production cost at $2.38/kg—78% lower than DAC-CS—and reduces total system cost by up to 20%.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Constraint Satisfaction and Optimization: Mixed Discrete/Continuous Optimization
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Agentic search

Abstract

This study presents a multi-objective optimization framework for sustainable aviation fuel (SAF) production, integrating artificial neural networks (ANNs) within a mixed-integer quadratically constrained programming (MIQCP) formulation. By embedding data-driven surrogate models into the mathematical optimization structure, the proposed methodology addresses key limitations of conventional superstructure-based approaches, enabling simultaneous optimization of discrete process choices and continuous operating parameters. The framework captures variable input and output stream compositions, facilitating the joint optimization of target product composition and system design. Application to Fischer-Tropsch (FT) kerosene production demonstrates that cost-minimizing configurations under unconstrained CO2 emissions are dominated by the fossil-based autothermal reforming (ATR) route. Imposing carbon emission constraints necessitates the integration of biomass gasification and direct air capture coupled with carbon sequestration (DAC-CS), resulting in substantially reduced net emissions but higher production costs. At the zero-emission limit, hybrid configurations combining ATR and biomass gasification achieve the lowest costs (~2.38 $/kg-kerosene), followed closely by biomass gasification-only (~2.43 $/kg), both of which outperform the ATR-only pathway with DAC-CS (~2.65 $/kg). In contrast, DAC-only systems relying exclusively on atmospheric CO2 and water electrolysis are prohibitively expensive (~10.8 $/kg). The results highlight the critical role of process adaptability: configurations exploiting flexible process parameters, facilitated by embedded ANNs, consistently outperform fixed setups, achieving up to 20% cost savings. Sensitivity analyses elucidate the influence of process conditions, such as FT reactor pressure and gasification temperature, on economic and environmental performance.

Problem

• Optimizing sustainable aviation fuel production with neural networks
• Integrating discrete process choices and continuous operating parameters
• Minimizing costs under carbon emission constraints for SAF

Innovation

• Embedding neural networks in optimization framework
• Integrating data-driven surrogate models mathematically
• Simultaneously optimizing discrete and continuous parameters