Different optimization tools have been developed to find the best trade-off between competitive goals. The optimization problem is typical of the design process, where different design solutions have to be compared to achieve one or more objectives, often in contrast with each other. A quite novel application of optimization is building energy model calibration. The use of well-calibrated energy simulation models is key for successful buildings' retrofit or operation management and the optimization techniques can improve the reliability of the results. The typical optimization method consists in the analysis of all the alternatives' performances, developing a full factorial plan and simulating all the possible options (brute-force approach). However, this process could take unsustainable long time. That is why some optimization tools, based on evolutionary algorithms have been developed to speed up the process. This study compares results obtained through the brute-force approach and the evolutionary optimization methods applied on the calibration of a large educational building model located in the province of Treviso, north of Italy. The total design space consists of about 72 000 EnergyPlus building models. Two optimization-based calibrations have been repeated using a genetic algorithm by means of jEPlus+EA on a local computer and through parametric simulations implemented by jEPlus on a cloud service. The quality of results from the evolutionary optimization tools as compared to a full parametric study applied on calibration have been discussed. Scenarios of applicability are drafted. On a practical level, the research is a contribution for the selection of methods and tools for the preparation of models that can lead to optimized retrofit interventions and rationalization of building management and operation.

Optimization Tools for Building Energy Model Calibration

CAPPELLETTI, FRANCESCA;ROMAGNONI, PIERCARLO
2017-01-01

Abstract

Different optimization tools have been developed to find the best trade-off between competitive goals. The optimization problem is typical of the design process, where different design solutions have to be compared to achieve one or more objectives, often in contrast with each other. A quite novel application of optimization is building energy model calibration. The use of well-calibrated energy simulation models is key for successful buildings' retrofit or operation management and the optimization techniques can improve the reliability of the results. The typical optimization method consists in the analysis of all the alternatives' performances, developing a full factorial plan and simulating all the possible options (brute-force approach). However, this process could take unsustainable long time. That is why some optimization tools, based on evolutionary algorithms have been developed to speed up the process. This study compares results obtained through the brute-force approach and the evolutionary optimization methods applied on the calibration of a large educational building model located in the province of Treviso, north of Italy. The total design space consists of about 72 000 EnergyPlus building models. Two optimization-based calibrations have been repeated using a genetic algorithm by means of jEPlus+EA on a local computer and through parametric simulations implemented by jEPlus on a cloud service. The quality of results from the evolutionary optimization tools as compared to a full parametric study applied on calibration have been discussed. Scenarios of applicability are drafted. On a practical level, the research is a contribution for the selection of methods and tools for the preparation of models that can lead to optimized retrofit interventions and rationalization of building management and operation.
File in questo prodotto:
File Dimensione Formato  
ARAMBULAetal_SEB16.pdf

accesso aperto

Tipologia: Versione Editoriale
Licenza: Accesso ristretto
Dimensione 1.12 MB
Formato Adobe PDF
1.12 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11578/266949
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 35
  • ???jsp.display-item.citation.isi??? 32
social impact