Existing studies on the thermal effect of building-integrated photovoltaic (BIPV) curtain walls are mostly numerical and mainly use absolute PV temperature; few have compared the relative roles of ambient temperature and solar radiation across seasons, or their temporal expression, in cold regions. This study conducted continuous field measurements on a south-facing ventilated CIGS PV curtain wall of a free-running demonstration building in Jinan, China, over 62 summer and 60 winter days at 10-min intervals. Single and two-factor regressions, standardized coefficients and partial correlations were used to compare the two drivers, including an incidence-angle-corrected radiation input; diurnal profiles, cross-correlation and positive/negative temperature-excess integrals characterized the temporal response and day-night asymmetry. Results show that: (1) the daytime mean PV surface temperature was higher in summer (35.2 vs. 13.8 °C), whereas the daytime temperature rise above ambient (ΔT) was 2.4 times higher in winter (+10.7 vs. +4.5 °C); (2) within the two-factor model, the dominant driver of ΔT shifted from ambient temperature in summer (75.5% relative statistical contribution) to solar radiation in winter (62.4%), a ranking unchanged with the corrected radiation input; (3) in winter, the warming rate was 2.26 times that in summer and the peak occurred 1.3 h earlier, while normalized diurnal profiles were similar in shape; (4) the PV surface was a daily net heat source relative to ambient air, with positive-to-negative integral ratios of 6.16 in summer and 2.48 in winter. Scientifically, the seasonal variation reflects a shift in the relative importance of the two boundary conditions rather than a change in heat-transfer mechanism. Theoretically, absolute temperature and temperature rise should be chosen according to the analysis purpose. In practice, summer operation should limit daytime heat gain, winter operation should limit nocturnal heat loss, and heating-season analyses should allow for the apparent response lag.

This study investigates the soil thermal imbalance of ground-source heat pump (GSHP) systems in residential buildings in cold regions and evaluates their economic and low-carbon performance. A case study is presented of a two-star green-certified residential building in Qingdao. The building exhibits a high heating load in winter, a low cooling load in summer, a long heating season, and large load fluctuations. To tackle these characteristics, a composite energy system combining a ground-source heat pump, a peak-shaving chiller, and a peak-shaving boiler is proposed. Three scenarios are designed, in which the ground-source heat pump covers 45%, 50%, and 52.6% of the winter peak heating load, respectively. These are compared with a conventional municipal heating scheme. Load simulation, techno-economic analysis, and carbon emission assessment are performed. The results show that the scheme in which the ground-source heat pump handles 50% of the peak heating load achieves the best overall performance. It reduces the soil thermal imbalance rate from 34.47% to 7.1% and obtains the lowest 10-year life-cycle cost. The annual carbon emission reaches 32.58 kgCO₂/(m²·a), representing a 33% reduction compared with municipal heating. Seasonal and diurnal optimized operation strategies are further proposed based on the optimal solution. The results provide theoretical and engineering guidance for the design and operation of low-carbon energy systems in green residential buildings in cold regions.