Computational cost reduction of the simulation-based optimization process using neural networks applied to an energy consumption problem with artificial conditioning systems
DOI:
https://doi.org/10.46421/entac.v19i1.2122Keywords:
Neural networks, Genetic Algorithms, Computational simulation, Building performance assessment, GrasshopperAbstract
Simulation-based optimization (SBO) processes can enhance building performance. The combination of OBS and machine learning methods appears as an alternative, capable of reducing the computational cost of the process without losing the robustness of the solutions. This study used artificial neural networks associated to a single objective SBO process to minimize the energy consumption with cooling, heating, and lighting systems in an office building through modifications on the building envelope. The results showed a significant reduction in the computational cost, in situations that the reduction of up to 50% of simulations.
References
OXMAN, R. Thinking difference: Theories and models of parametric design thinking. Design Studies, v. 52, p. 4–39, 2017. Disponível em: <http://dx.doi.org/10.1016/j.destud.2017.06.001>.
FAROUK, A.; ELDALY, H.; DEWIDAR, K. Parametric Design As a Tool for Performative Architecture. Journal of Al-Azhar University Engineering Sector, v. 14, n. 50, p. 148–157, 2019.
KHEIRI, F. A review on optimization methods applied in energy-efficient building geometry and envelope design. Renewable and Sustainable Energy Reviews, v. 92, n. May 2017, p. 897–920, 2018. Disponível em: <https://doi.org/10.1016/j.rser.2018.04.080>.
SEYEDZADEH, S.; POUR RAHIMIAN, F.; RASTOGI, P.; GLESK, I. Tuning machine learning models for prediction of building energy loads. Sustainable Cities and Society, v. 47, n. March, p. 101484, 2019. Disponível em: <https://doi.org/10.1016/j.scs.2019.101484>.
MILLER, C.; ARJUNAN, P.; KATHIRGAMANATHAN, A.; FU, C.; ROTH, J.; PARK, J. Y.; et al. The ASHRAE Great Energy Predictor III competition: Overview and results. Science and Technology for the Built Environment, v. 26, n. 10, p. 1427–1447, 2020. Disponível em: <https://doi.org/10.1080/23744731.2020.1795514>.
MELO, A. P.; CÓSTOLA, D.; LAMBERTS, R.; HENSEN, J. L. M. Development of surrogate models using artificial neural network for building shell energy labelling. Energy Policy, v. 69, p. 457–466, 2014. Disponível em: <http://dx.doi.org/10.1016/j.enpol.2014.02.001>.
CARLO, J.; LAMBERTS, R. Development of envelope efficiency labels for commercial buildings: Effect of different variables on electricity consumption. Energy and Buildings, v. 40, n. 11, p. 2002–2008, 2008. Disponível em: <https://www.sciencedirect.com/science/article/pii/S0378778808001114>.
WETTER, M.; WRIGHT, J. A comparison of deterministic and probabilistic optimization algorithms for nonsmooth simulation-based optimization. Building and Environment, v. 39, n. 8, p. 989–999, 2004. Disponível em: <https://www.sciencedirect.com/science/article/pii/S0360132304000332>.
WAIBEL, C.; WORTMANN, T.; EVINS, R.; CARMELIET, J. Building energy optimization: An extensive benchmark of global search algorithms. Energy and Buildings, v. 187, p. 218–240, 2019. Disponível em: <https://doi.org/10.1016/j.enbuild.2019.01.048>.
RUTTEN, D. Galapagos: On the logic and limitations of generic solvers. Architectural Design, v. 83, n. 2, p. 132–135, 2013.
