Machine-learning-aided life cycle assessment and techno-economic analysis of hydrothermal liquefaction of sewage sludge for bio-oil production

Hydrothermal liquefaction (HTL) is a promising technology for converting high-moisture sewage sludges into biofuels. To evaluate the energy, climate change, and economic performance of sludge HTL, this study integrated machine learning (ML) with life cycle assessment (LCA) and techno-economic analysis (TEA). First, ML models were employed to predict the product distribution and properties. These predictions were then used to support LCA and TEA, calculating global warming potential (GWP), energy return on investment (EROI), and minimum fuel selling price (MFSP). The ML model for bio-oil demonstrated high accuracy, with an average test R² value of 0.89. LCA results indicated that using hydrochar as fuel was more advantageous than using it for carbon sequestration. TEA results revealed that the MFSP of bio-oil was lower between 320 °C and 360 °C. Furthermore, the discount rate was identified as the most significant factor influencing MFSP. The EROI, GWP, and MFSP values ranged from 0.29 to 3.59, −361.89 to 418.22 CO₂ eq/t, and 693.35 to 2880.44 $/t, respectively. This integrated framework can help to identify the optimal processing parameters for energy production, carbon emissions, and economic viability. Future work could further integrate process simulation to refine energy and material consumption data for more accurate LCA and TEA.