with the collaboration of Iranian Society of Mechanical Engineers (ISME)

Evaluation of a Laboratory-Scale Anaerobic Digestion System with ANFIS-Based Adaptive Control and Hybrid Biogas Production Modelling

Document Type : Research Article

Authors

Department of Biosystems Engineering, Faculty of Agriculture, University of Tabriz, Tabriz, Iran

10.22067/jam.2026.100403.1517
Abstract
Introduction
Anaerobic digestion (AD) is a well-established biochemical process for organic waste valorisation and renewable energy generation. However, the inherent complexity, nonlinearity, and time-varying dynamics of AD pose significant challenges for process stability and optimisation. Conventional control strategies (ON/OFF, PID) often fail due to limited adaptability to complex microbial interactions and varying substrate characteristics. In Iran, despite abundant biomass resources, biogas technology remains underdeveloped due to low fossil fuel costs, insufficient technical expertise, and limited public awareness. This study aimed to design, construct, and evaluate a laboratory-scale AD system with an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based intelligent controller. The core innovation lies in integrating fuzzy logic reasoning with neural network learning in real time and self-adaptive regulation of process variables. Additionally, a novel hybrid ANFIS-Gompertz model was developed as a proof-of-concept for dynamic kinetic parameter estimation, and the feasibility of using a low-cost optical sensor for early warning monitoring, along with theoretical scale-up analysis, was investigated.
Materials and Methods
Substrate and Reactor: Fresh cattle manure (1:1 with water) was characterised for moisture (90.54%), TS (9.46%), VS (78.21%), COD (910 mg L-1 O₂), and C/N ratio (21:1). A 2 L borosilicate glass reactor (H/D = 1) with variable-speed agitation (50–150 rpm) and a hot water jacket heating system was fabricated.
ANFIS Control System: The ANFIS controller had three inputs (temperature, pH, and pressure) and three outputs (agitation speed, heating power, and gas release), with two Gaussian membership functions per input (8 fuzzy rules) to prevent overfitting. It was implemented on an ESP32 microcontroller with sensors (LM35, MPX5700DP, E201C9, MQ4, and TCS34725) for turbidity monitoring.
Experimental Design: Response Surface Methodology (RSM) with a Box-Behnken design (17 runs) optimised temperature (30–40°C), agitation speed (50–150 rpm), and HRT (15–35 days). Kinetic models (modified Gompertz, logistic, and first-order) and a hybrid ANFIS-Gompertz model were developed. Model evaluation used R², RMSE, AIC, and BIC, with 5-fold cross-validation and three independent reproducibility runs. Theoretical scale-up to 330 L and 1,000 L was analysed based on hydrodynamic similarities.
Results and Discussion
ANFIS Control Performance: The ANFIS controller achieved 98.5% prediction accuracy, steady-state temperature error of 0.12 ± 0.03°C, settling time of 4.2 ± 0.5 min, and an overshoot of 2.1 ± 0.3%. Energy consumption was reduced by 28.9% compared to PID control (768 ± 25 W vs. 1,080 ± 38 W). pH error was maintained at ±0.02.
Biogas Production: Total biogas production over 30 days was 33.6 L (19% increase over simple control), with a maximum methane content of 62.5%. Average daily production was 1.12 ± 0.25 L day-1 (CV = 1.19% across replicates). Volatile fatty acids (VFAs) peaked at 1,800 mg L-1 (days 10–15), with the VFAs/alkalinity ratio reaching 1.8, indicating transient acidification followed by recovery. Ammonia remained below the inhibitory threshold (1,050 mg L-1 < 3,000 mg L-1 ).
Kinetic Modelling: The hybrid ANFIS-Gompertz model provided the best fit (R² = 0.960, RMSE = 0.85, AIC = 35.2, and BIC = 39.2), outperforming modified Gompertz (R² = 0.994, RMSE = 0.95), logistic (R² = 0.986, RMSE = 1.40), and first-order (R² = 0.950, RMSE = 2.55) models. Estimated Bₘ = 33.6 L, Rₘ = 0.132 L day-1, λ = 4 days. Five-fold cross-validation confirmed good generalisability (mean R² = 0.955 ± 0.006, RMSE = 0.92 ± 0.06). However, due to limited training data, this model is presented as a proof-of-concept requiring further validation with larger datasets for industrial applications.
RSM Optimisation: Optimal conditions were 36.6°C, 110 rpm, and 24 days HRT, achieving 33.6 ± 0.3 L (error < 0.6% vs. predicted 33.8 L). ANOVA confirmed model significance (F = 38.5, p < 0.001, R² = 0.972). Reproducibility was excellent (CV = 1.19% for biogas, 1.44% for hybrid model RMSE).
Scale-Up Analysis: Theoretical scale-up to 330 L (pilot) and 1,000 L (industrial) while maintaining a constant Reynolds number (0.27×10⁴) was analysed. Recommended pilot parameters: agitation speed 100 rpm, heating power 1,200 W, and insulation thickness 5 cm. Daily biogas production projections: 185 L day-1 (pilot) and 560 L day-1 (industrial). Experimental validation at pilot scale and CFD simulations are required for confirmation.
Conclusion
An automated laboratory-scale AD system with ANFIS-based control was successfully developed and evaluated. The key achievements include: (1) ANFIS control with 28.9% energy savings and superior stability compared to PID; (2) RSM-optimised conditions (36.6°C, 110 rpm, and 24 days) achieving 33.6 L biogas with 62.5% methane; (3) a proof-of-concept hybrid ANFIS-Gompertz model (R² = 0.960, RMSE = 0.85) demonstrating the potential of integrating intelligent systems with kinetic modelling; (4) a low-cost optical sensor (TCS34725) as an effective early warning system for process instability; and (5) theoretical feasibility of scale-up to 330 L based on hydrodynamic similarity.
Acknowledgement
This research was financially supported by the University of Tabriz. The authors sincerely thank the Department of Biosystems Engineering and the Vice-Presidency for Research and Technology for their support.

Keywords

Subjects

Authors retain the copyright. This is an open access article distributed under Creative Commons Attribution 4.0 International License (CC BY 4.0)

1.       Amran, N. A. M., Mohamed, H., Rafaai, Z. F. M., Yacob, N. S., Junoh, H., & Shamsuddin, A. H. (2024). A review of machine learning models in predicting biogas production. AIP Conference Proceedings, 2934(1), 060006. https://doi.org/10.1063/5.0181016
2.       Angelidaki, I., Treu, L., Tsapekos, P., Luo, G., Campanaro, S., Wenzel, H., & Kougias, P. G. (2018). Biogas upgrading and utilization: Current status and perspectives. Biotechnology Advances, 36(2), 452–466. https://doi.org/10.1016/j.biotechadv.2018.01.011
3.       Bastos, J. A., Silva, F. L. H., Medeiros, L. L., & Santos, J. S. (2026). Moderate thermal, alkaline and thermal-alkaline pretreatment of cattle manure: Effects on methane production and economic feasibility. Brazilian Journal of Chemical Engineering, 43(4), 737–748. https://doi.org/10.1007/s43153-025-00592-x
4.       Conesa, J. A. (2019). Scale-up of anaerobic digestion: A review of hydrodynamic and mass transfer limitations. Renewable and Sustainable Energy Reviews, 112, 123-135. https://doi.org/10.1016/j.rser.2019.06.012
5.       Esrat, W., & Purushottam Kouleh, K. (2025). Anaerobic co-digestion of wastewater treatment plant sludge and bread waste to enhance biogas yield. Journal of Agricultural Machinery, (in Press), e47251. https://doi.org/10.22067/jam.2025.92821.1363
6.       Fröner-Lacerda, L. R., Sganzerla, W. G., Lacerda, V. F., Sillero, L., Solera, R., Pérez, M., & Forster-Carneiro, T. (2024). Design, fabrication, automation, and scaleup of anaerobic reactors for waste management and bioenergy recovery. Biofuels, Bioproducts and Biorefining, 18(5), 1245–1267. https://doi.org/10.1002/bbb.2674
7.       Gadhi, T. A., Channa, N., Mahar, R. B., Bhatti, S., Ali, B., Raheem, A., Demichelis, F., Tommasi, T., Freyria, F. S., & Bonelli, B. (2026). Integrating technical and sustainability perspectives: Anaerobic co-digestion from lab to pilot scale using textile wastewater sludge and cattle manure. Biomass Conversion and Biorefinery, 16(6), 1–17. https://doi.org/10.1007/s13399-026-07092-z
8.       Jang, J. S. R. (1993). ANFIS: Adaptive-network-based fuzzy inference system. IEEE Transactions on Systems, Man, and Cybernetics, 23(3), 665–685. https://doi.org/10.1109/21.256541
9.       Jiménez-Páez, E., Ding, F., Fermoso, F. G., & García-Martín, J. F. (2024). Monitoring of volatile fatty acids during anaerobic digestion of olive pomace by means of a hand held near infrared spectrometer. Science of The Total Environment, 957, 177603. https://doi.org/10.1016/j.scitotenv.2024.176979
10.   Kariyama, I. D., Zhai, X., & Wu, B. (2018). Influence of mixing on anaerobic digestion efficiency in stirred tank digesters: A review. Water Research, 143, 503–517. https://doi.org/10.1016/j.watres.2018.06.065
11.   Karimi, K., & Mohammadi, S. (2015). Principles of biogas production technologies. Isfahan University of Technology Press. [ISBN: 978-964-6664-91-5] [In Persian].
12.   Lima, G., Maria, I., Joachim, M., & Andreas, L. (2026). Scale-Up of a Two-Stage Anaerobic Digestion System: From Laboratory Reactor to Pilot Plant. Fermentation, 12(6), 255. https://doi.org/10.3390/fermentation12060255
13.   Mao, C., Feng, Y., Wang, X., & Ren, G. (2015). Review on research achievements of biogas from anaerobic digestion. Renewable and Sustainable Energy Reviews, 45, 540–555. https://doi.org/10.1016/j.rser.2015.02.032
14.   Mahmoodi-Eshkaftaki, M., Lotfalian Dehkordi, A., Khafajeh, H., & Bahram Shabahrami, S. H. (2026). Improving anaerobic fermentation of agricultural waste using ultrasonic pretreatment to increase biohydrogen production: A study of the metabolic pathway of reactions. Journal of Agricultural Machinery, (in Press), e48525. https://doi.org/10.22067/jam.2026.98121.1477
15.   Maluta, P., Paglianti, A., & Montante, G. (2024). Scale-up criteria for stirred tanks with non-Newtonian fluids: A CFD-based approach. Chemical Engineering Science, 287, 119132. https://doi.org/10.1016/j.ces.2023.119132
16.   Mata-Alvarez, J., Mace, S., & Llabres, P. (2000). Anaerobic digestion of organic solid wastes: An overview of research achievements and perspectives. Bioresource Technology, 74(1), 3–16. https://doi.org/10.1016/S0960-8524(00)00023-7
17.   Mohammadzadehchali, F., Ostovari Deylamani, A., Shirazi Karameh, A., Ebrahimi, M., & Karimi, K. (2026). Biogas emissions and renewable energy potential of municipal solid waste landfills: A case study in Kashan, Iran. Pollution, 12(2), 685–701. https://doi.org/10.22059/poll.2026.405717.3188
18.   Negahban, V., Ebrahimi-Nik, M., & Rohani, A. (2025). Higher anaerobic digester performance by the strategical increase in the feeding rate of cow manure in laboratory continuous stirred tank reactor. PLOS ONE, 20(10), e0332972. https://doi.org/10.1371/journal.pone.0332972
19.   Rego, A. S. C., Leite, S. A. F., Leite, B. S., & Santos, B. F. (2025). Study of neural models for optimization of biogas production. Biomass and Bioenergy, 185, 107365. https://doi.org/10.1016/j.biombioe.2025.107365
20.   Sami, M., Akram, A., & Sharifi, M. (2021). Potential assessment and modeling of biogas production from apple juice pomace. Journal of Agricultural Machinery, 11(2), 305–316. https://doi.org/10.22067/jam.v11i2.80322.
21.   Sánchez-Aguirre, M., Espinosa-Solares, T., Toledo, J. U., & Ramírez-Arpide, F. R. (2026). Enhancing renewable energy performance of biogas systems through Sargassum co-digestion and methane recovery: An integrated life cycle assessment. Energy Conversion and Management, 325, 120656. https://doi.org/10.1016/j.enconman.2026.120656
22.   Taghinejad, J., Abdi, R., & Adl, M. (2018). Modeling of biogas production process from cattle manure in a completely mixed digester with semi-continuous feeding. Journal of Agricultural Machinery, 8(1), 159–169. https://doi.org/10.22067/jam.v8i1.57758.
23.   Zhu, Y., Tsui, T.-H., Zhang, L., Smoliński, A., Xiang, G., Lyu, Q., & Liu, R. (2026). Metagenomic insights into microwave-enhanced anaerobic co-digestion for maximizing biomethane and securing pathogen-free biofertilizer from agricultural waste. Bioresource Technology, 420, 132107. https://doi.org/10.1016/j.biortech.2026.132107
24.   Zughaibi, A., Al-Momani, F., & Al-Hajri, R. (2025). Hydrodynamic similarity and scale-up of anaerobic digesters: A comprehensive review. Biomass and Bioenergy, 185, 107120. https://doi.org/10.1016/j.biombioe.2025.107120
Send comment about this article
Enter Name.
Enter a valid email address.
Enter a vaid affiliation.
Enter comments (At leaset 10 words)
CAPTCHA Image
Enter Security Code Correctly.

Articles in Press, Accepted Manuscript
Available Online from 16 November 2026

  • Receive Date 18 August 2026
  • Revise Date 25 September 2026
  • Accept Date 30 September 2026
  • First Publish Date 16 November 2026