Paper Conference

Proceedings of Building Simulation 2021: 17th Conference of IBPSA


Uncertainty analysis of hardware in loop setup for testing building technologies: A case study

Prasaant Balasundaram 1,3, Benoit Delinchant 2, Stephane Ploix 3, Cristian Muresan 1
1 ENGIE, France
2 Grenoble INP - G2E Lab
3 Grenoble INP - GSCOP

Abstract: Hardware in the loop simulation (HIL) offers an intuitive real-time procedure for the validation and evaluation of building technology. The effect of uncertainty due to the components present in the test bench of the HIL is taken into account since it might affect the performance of the product that is measured. In the case of measuring the seasonal performance/annual performance, one should extrapolate the performance based on short-term testing which makes it important to consider the uncertainties. In this case, the probability distributions are used to model the uncertainties. The uncertainty propagation is based on the Monte Carlo Simulation. The model of the entire HIL test setup is developed in Modelica as well as propagation of uncertainties, which is not common. In the future, the modeled HIL setup compared with the real experimental setup for discrepancies(simulated efficiency vs real-time efficiency) and calibrated so. The aim is to confirm thanks to Monte Carlo Simulations that the test bench parameters don’t affect the efficiency of the product being tested. In the future, if one cannot specify input values for the parameter distribution, the shape of the probability distribution, or the uncertainty due to dependencies between the inputs then a second-order probability can be used.
Keywords: Uncertainty Propagation, Uncertainty, Hardware in Loop, Probability box
Pages: 2141 - 2146