AI-FORECASTED TECHNO-ECONOMIC AND ENVIRONMENTAL ASSESSMENT OF BIOGAS, METHANE (CH₄), HYDROGEN (H₂), AND ELECTRICAL POWER GENERATION AT A DAIRY FARM IN AL-DHLAIL, ZARQA, JORDAN

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Methodology combining techno-economic analysis with LSTM forecasting for biogas systems is novel and could be adapted for other contexts.

🏢実務担当者:The detailed economic metrics (payback, LCOE) provide a benchmark for similar dairy farm biogas projects, and AI forecasting aids in planning.

🏛政策担当者:The scalability analysis shows significant regional emission reduction potential, supporting renewable energy and waste management policies.

Jordan’s heavy reliance on imported fossil fuels and its rapidly expanding dairy sector create both an environmental burden and a largely untapped renewable-energy opportunity. While household-scale and laboratory-scale biogas studies dominate the Middle-Eastern literature, mid-scale fixed-dome plants designed for semi-arid Jordanian conditions have received very little attention. No published study has yet coupled the techno-economic and environmental assessment of such a plant with an artificial-intelligence (AI) forecasting framework that explicitly resolves biogas into its methane, hydrogen co-production, and electrical-power components. This work addresses that gap. A fixed-dome biogas plant is designed for a 200-cow dairy farm in the Al-Dhlail area of Zarqa Governorate, comprising a total plant volume of 262 m³ (170.3 m³ digester + 91.7 m³ gas holder) and a 32-day hydraulic retention time. Under base-case operation, the plant produces 38.64 m³/day of biogas (60% CH ₄ → 23.18 m ³ /day of methane, with a two-stage dark-fermentation H ₂ co-production envelope of about 8.89 m ³ /day), equivalent to 86,510 kWh/year of thermal energy and, after combined-heat-and-power conversion at 35% electrical efficiency, approximately 30,279 kWh/year of electricity ( ≈ 3.5 kW continuous). The capital cost of 50,065 JD is offset by annual revenues of 12,365 JD, giving a payback period of ~4 years, a levelized cost of electricity of ~0.066 JD/kWh (≈ 0.093 USD/kWh) and a negative carbon-abatement cost of approximately 235 JD per ton of CO ₂ avoided, while displacing 28.46 tons of CO ₂ per year (~6 passenger cars or ~470 mature trees). A Long Short-Term Memory (LSTM) recurrent neural network is then used to project plant performance over 2025 – 2035, forecasting biogas growth to 42.8 m ³ /day, CH ₄ to 25.7 m ³ /day, H ₂ potential to 9.8 m ³ /day, and electrical generation to 33,538 kWh/year by 2035. Scaled to all 250 dairy farms in the Al-Dhlail region, the design would deliver ~21.6 GWh/year (~2.5 MW continuous) more than four times Jordan’s existing UNDP-supported landfill biogas plant and abate ~7,115 t CO ₂ /year, supporting the National Energy Strategy and demonstrating that AI-augmented, multi-product farm-scale biogas is a strategically valuable building block for Jordan ’ s renewable-energy transition.

• Zenodo https://zenodo.org/records/20703631first seen 2026-06-16 04:15:02