Validated transient CFD and thermal response evaluation for representative height bands and adaptive sampling in cold-climate radiant-heated residential buildings

Accurate indoor temperature sensing is critical to energy efficiency and thermal comfort in smart district-heating systems. In cold-climate radiant-heating dwellings, strong solar–radiative coupling and large envelope thermal inertia generate non-stationary disturbances that undermine the spatial representativeness and temporal responsiveness of conventional sensor layouts. This study aims to develop and validate a transient CFD–measurement fusion method to optimize temperature-sensing configuration, including representative sensor height and adaptive sampling interval, in a radiator-heated apartment in a severe-cold region. High-frequency field measurements under typical winter operating conditions are used to construct boundary conditions and to dynamically validate a transient CFD model; the mean absolute error across six reference points is below 0.4 °C. Height-resolved temperature time series from the validated model and measurements are analyzed using deviation maps, vertical profiles, and root-mean-square error metrics to identify a representative observational band. A maximum allowable sampling interval criterion is applied to spatiotemporal temperature histories to derive piecewise sampling recommendations. Results show that the mid-height band around 1.2–1.4 m experiences weak disturbances and provides the most representative room temperature. Sampling can be relaxed to 30–60 min during quiescent periods such as nighttime or in north-facing rooms, but should be tightened to 1–3 min under strong insolation to capture rapid transients. The proposed framework clarifies the spatiotemporal structure of temperature disturbances in cold-region residential heating and offers a reproducible pathway for designing sensor placement and sampling strategies in smart district-heating applications.