Application of Neural Networks in the Optimization of Francis Turbine Draft Tubes

• Eduardo Fabian de Oliveira UTFPR
• Raquel da Cunha Ribeiro da Silva UTFPR
• Dhaianny Guizoni de Oliveira HACKER Industrial

DOI: https://doi.org/10.5753/latinoware.2025.16457

Resumo

This study proposes the application of artificial neural networks (ANNs) to optimize the performance of hydraulic turbines, with a specific focus on the draft tube. The research is motivated by the need to enhance energy efficiency in hydroelectric systems, in alignment with the Sustainable Development Goals. The methodology combines computational fluid dynamics (CFD) simulations and machine learning techniques, utilizing operational data from a Francis turbine. Results demonstrate the potential of ANNs to predict and improve key performance parameters, offering a novel approach to hydraulic design optimization. The integration of computational modeling and artificial intelligence emerges as a promising tool for the renewable energy sector.

Palavras-chave: hydraulic turbines, artificial neural networks, energy efficiency, draft tube, CFD

Referências

Chakrabarty, S., Sarkar, B.K., and Maity, S., 2016. CFD Analysis of the Hydraulic Turbine Draft Tube to Improve System Efficiency. MATEC Web of Conferences, 40, p. 02003.

Hacker Industrial, Design of the Draft tube. Technical Note, 2025.

International Hydropower Association (IHA), 2024 World Hydropower Outlook: Opportunities to advance net zero. London, UK: IHA, 2024. [Online]. p. 96 Available: [link]

Lalonde, E.R., Vischschraper, B., Bitsamlak, G., and Dai, K., 2021. Evaluation of a Neural Network-Based Surrogate Aerodynamic Wind Turbine Blade Model. Journal of Wind Engineering, 45(2), pp. 1-12.

Poudel, R., Chitrakar, S., Qian, Z., and Thapa, B., 2024. Experimental Investigations of Sediment Erosion in Francis Turbine Using Non-Recirculating Sediment Test Rig. Energy Science & Engineering, 13, pp. 700-713.

REN21, Renewables 2023: Analysis and forecast to 2028. Paris: International Energy Agency (IEA), 2023. p. 336.

Ristić, B., and Božić, I., 2021. Digital Technologies Emergence in the Contemporary Hydropower Plants Operation. International Conference on Sustainable Development of Energy, Water and Environment Systems( SDEWES).

Sheikh, H.M., et al., 2022. Optimization of the Shape of a Hydrokinetic Turbine’s Draft Tube and Hub Assembly Using Design-by-Morphing with Bayesian Optimization. Computer Methods in Applied Mechanics and Engineering, 401, p. 115654.

Tao, R., Lu, J., and Su, C., 2024. Influence of Runner Blade Number on Hydraulic Performance and Flow Control in Draft Tube of Francis Turbine. Advances in Mechanical Engineering, 16(3), pp. 1-20.

Tian, X., Pan, H., Hong, S., and Zheng, Y., 2015. Improvement of Hydro-turbine Draft Tube Efficiency Using Vortex Generator. Advances in Mechanical Engineering, 7(7), pp. 1-8.

Tiwari, G., Kumar, J., Prasad, V., and Patel, V.K., 2020. Utility of CFD in the Design and Performance Analysis of Hydraulic Turbines – A Review. Energy Reports, 6, pp. 2410-2429.

Trivedi, C., Gandhi, B., and Cervantes, J.M., 2013. Effect of Transients on Francis Turbine Runner Life: A Review. Journal of Hydraulic Research, 51(1), pp. 1-12.

Zhou, X., Huang, Q., Hu, X., and Dai, X., 2024. Study of the Effects of Modified Draft Tube with Inclined Conical Diffuser on Draft Tube and Upstream Region. Journal of Physics: Conference Series, 2707, p. 012064.