Your browser does not support JavaScript!
http://iet.metastore.ingenta.com
1887

access icon openaccess Smart building uncertainty analysis via adaptive Lasso

Uncertainty analysis plays a pivotal role in identifying the important parameters affecting building energy consumption and estimate their effects at the early design stages. In this work, we consider the adaptive Lasso for uncertainty analysis in building performance simulation. This procedure has several appealing features: (1) We can introduce a large number of possible physical and environmental parameters at the initial stage to obtain a more complete picture of the building energy consumption. (2) The procedure could automatically select parameters and estimate influences simultaneously and no prior knowledge is required. (3) Due to computational efficiency of the procedure, non-linear relationship between the building performance and the input parameters could be accommodated. (4) The proposed adaptive Lasso can use a small number of samples to achieve high modeling accuracy and further reduce the huge computational cost of running building energy simulation programs. Furthermore, we propose a stable algorithm to rank input parameters to better identify important input parameters that affect energy consumption. A case study shows the superior performance of the procedure compared with LS and OMP in terms of modeling accuracy and computational cost.

References

    1. 1)
    2. 2)
      • 12. Chen, X., Li, X., Tan, S.X.-D.: ‘From robust chip to smart building: CAD algorithms and methodologies for uncertainty analysis of building performance’, IEEE/ACM International Conference Computer-Aided Design (ICCAD), 2–6 Nov2015, pp. 457464.
    3. 3)
    4. 4)
      • 15. McKay, M.D.: ‘Latin hypercube sampling as a tool in uncertainty analysis of computer models’. Proc. of Winter Simulation Conf., 1992, pp. 557564.
    5. 5)
    6. 6)
    7. 7)
    8. 8)
      • 20. Huang, J., Ma, S., Zhang, C.-H.: ‘Adaptive Lasso for sparse high-dimensional regression models’, Statistica Sinica, 2008, 18, pp. 16031618.
    9. 9)
    10. 10)
    11. 11)
      • 9. Zhao, H., Quach, D., Wang, S., et al: ‘Learning based compact thermal modeling for energy-efficient smart building management’, IEEE/ACM International Conference Computer-Aided Design (ICCAD), 2–6 Nov2015, pp. 450456.
    12. 12)
      • 16. Crawley, D.B., Lawrie, L.K., Winkelmann, F.C., et al: ‘EnergyPlus: creating a new-generalization building energy simulation program’, 2001, 33, pp. 319331.
    13. 13)
    14. 14)
    15. 15)
      • 18. Hastie, T., Tibshirani, R., Wainwright, M.: ‘Statistical learning with sparsity: the Lasso and generalization’ (CRC press, 2015).
    16. 16)
      • 10. Liu, W., Wang, H., Zhao, H., et al: ‘Thermal modeling for energy-efficient smart building with advanced overfitting mitigation technique’, 2016 21st Asia and South Pacific Design Automation Conference (ASP-DAC), 25–28 Jan2016, pp. 417422.
    17. 17)
    18. 18)
      • 19. Tibshirani, R.: ‘Regression shrinkage and selection via the Lasso’, Statistical Methodology, 1996, 58, pp. 267288.
    19. 19)
    20. 20)
    21. 21)
      • 1. ‘Buildings Energy Data Book’, Available at http://buildingsdatabook.eren.doe.gov.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cps.2017.0011
Loading

Related content

content/journals/10.1049/iet-cps.2017.0011
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading
This is a required field
Please enter a valid email address