Validation of a Computer Code for the Energy Consumption of a Building, with Application to Optimal Electric Bill Pricing

Merlin Keller, Guillaume Damblin (), Alberto Pasanisi, Mathieu Schumann, Pierre Barbillon () and Fabrizio Ruggeri
Additional contact information
Merlin Keller: EDF - Electricité de France
Guillaume Damblin: Université Paris-Saclay
Alberto Pasanisi: Edison
Mathieu Schumann: EDF - Electricité de France
Pierre Barbillon: MIA Paris-Saclay - Mathématiques et Informatique Appliquées - AgroParisTech - Université Paris-Saclay - INRAE - Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement
Fabrizio Ruggeri: IMATI - Istituto di Matematica Applicata e Tecnologie Informatiche - CNR - National Research Council of Italy | Consiglio Nazionale delle Ricerche

Post-Print from HAL

Abstract: In this paper, we present a case study aimed at determining a billing plan that ensures customer loyalty and provides a profit for the energy company, whose point of view is taken in the paper. The energy provider promotes new contracts for residential buildings, in which customers pay a fixed rate chosen in advance, based on an overall energy consumption forecast. For such a purpose, we consider a practical Bayesian framework for the calibration and validation of a computer code used to forecast the energy consumption of a building. On the basis of power field measurements, collected from an experimental building cell in a given period of time, the code is calibrated, effectively reducing the epistemic uncertainty affecting the most relevant parameters of the code (albedo, thermal bridge factor, and convective coefficient). The validation is carried out by testing the goodness of fit of the code with respect to the field measurements, and then propagating the posterior parametric uncertainty through the code, obtaining probabilistic forecasts of the average electrical power delivered inside the cell in a given period of time. Finally, Bayesian decision-making methods are used to choose the optimal fixed rate (for the energy provider) of the contract, in order to balance short-term benefits with customer retention. We identify three significant contributions of the paper. First of all, the case study data were never analyzed from a Bayesian viewpoint, which is relevant here not only for estimating the parameters but also for properly assessing the uncertainty about the forecasts. Furthermore, the study of optimal policies for energy providers in this framework is new, to the best of our knowledge. Finally, we propose Bayesian posterior predictive p-value for validation.

Keywords: uncertainty quantification; Bayesian analysis; energy contracts; uncertainty quantification Bayesian analysis energy contracts (search for similar items in EconPapers)
Date: 2022-12
New Economics Papers: this item is included in nep-des, nep-ene and nep-reg
Note: View the original document on HAL open archive server: https://hal.inrae.fr/hal-04071903v1
References: View references in EconPapers View complete reference list from CitEc
Citations:

Published in Econometrics, 2022, 10 (4), pp.34. ⟨10.3390/econometrics10040034⟩

Downloads: (external link)
https://hal.inrae.fr/hal-04071903v1/document (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:journl:hal-04071903

DOI: 10.3390/econometrics10040034

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().