Several major research initiatives have examined hybrid ventilation over the past two decades. The International Energy Agency’s Annex 35 (Heiselberg, 2025) investigated hybrid ventilation in office and educational buildings across various climates, but did not specifically address the challenges of cold climates. Similarly, the EU-funded RESHYVENT project (RESHYVENT Consortium, 2005) explored demand-controlled hybrid systems in residential buildings, with limited focus on cold-climate performance. Annex 68 (Rode, 2025) placed more emphasis on IAQ in residential settings using natural and mechanical ventilation yet also lacked specific insight into hybrid strategies for colder regions.

Simulation dominates hybrid ventilation research, particularly using tools like IDA ICE and EnergyPlus. While simulations offer controlled comparisons of scenarios and control strategies, real-world data remains critical. Measurement studies like those by Buvik (2003), Rabani and Petersen (2023), and Deng et al. (2020) validate assumptions but reveal occupant behavior and environmental variability often deviate from models. This divergence highlights the need for more long-term, monitored case studies in cold climates.

Window operation plays a central role in hybrid ventilation. Studies like those by Naili et al. (2023), Algrøy (2014), and Haktorson (2018) show automated control based on CO₂, temperature, and weather inputs yields better performance than manual use. However, even with sensors, issues like over-ventilation, occupant discomfort, or high CO₂ levels during cold weather persist. Few studies directly assess window usability or occupant satisfaction, especially in educational settings.

Some studies simulate hybrid ventilation in the context of climate change. Bamdad et al. (2022) in Australia and Sánchez-García et al. (2019) in Spain found mixed-mode strategies could offset future cooling loads, yielding 38%–47% energy savings. In cold climates, Gilani and O’Brien (2021) and Wang et al. (2017) noted reduced energy savings under future scenarios and did not address IAQ. These findings suggest that hybrid ventilation may need reevaluation or redesign to remain effective in colder regions under warming trends. Table 4 summarizes the studies on hybrid ventilation with various climate scenarios, highlighting the methodologies employed and their key findings.

Despite growing interest and research on hybrid ventilation, significant gaps remain, especially regarding its effectiveness in future cold climates. Most existing studies focus on warmer regions where rising temperatures and extreme heatwaves are expected to increase cooling demands, thereby reducing the potential energy savings achievable through natural ventilation alone.

In contrast, buildings in cold climates, such as Norway, rely heavily on mechanical ventilation with heat recovery during long heating seasons. However, with the anticipated shortening of heating periods due to climate change, hybrid ventilation strategies may involve more frequent use of window ventilation. This shift prompts critical questions about the applicability and efficiency of hybrid ventilation in airtight, low-energy buildings, such as nearly Zero Emission Building (nZEBs), compared to conventional constructions.

Key research gaps include:

• Limited studies specifically addressing hybrid ventilation performance in cold climates under future climate scenarios.

Addressing these gaps is essential for optimizing hybrid ventilation strategies and supporting sustainable building design tailored to the evolving climate conditions in cold regions.

For modelling the building, IDA-ICE 5.0 has been employed. The building model was first imported via Naviate Simple BIM and then exported to IDA-ICE 5.0 using SketchUp as shown in Figure 4. Two zones are considered in this model called Amphitheater and Exhibition.

The building envelope and technical properties adhere to the stipulations outlined in the Norwegian Building Code, TEK 17 (TEK17, 2017) and the specific requirements of the building. Factors contributing to internal gains—namely occupancy, lighting, and equipment—along with their respective utilization profiles, were chosen based on the real usage in the building (Table 5).

The usage profile for the occupants, lighting, and equipment are considered based on the information collected from users. The opening hours of the Climate House are every day from 9:00 to 17:00 (Figure 5).

The current hybrid ventilation control strategy in the Climate House, shown in Figure 3b, relies only on CO₂ and indoor air temperature and schedule opening. User feedback indicates that the current hybrid ventilation system in the Climate House does not adequately maintain thermal comfort and indoor air quality. To address this, an optimized hybrid control strategy is proposed, prioritizing mechanical ventilation while utilizing natural ventilation as a supplementary approach. The window opening mechanism is regulated based on air change per hour (ACH), indoor temperature, CO₂ concentration, outdoor temperature, and a predefined operation schedule.

The hybrid ventilation control system is designed to prioritize mechanical ventilation over natural ventilation via window openings (Figure 6). However, windows are automatically activated when specific indoor environmental conditions are not met in designated control areas (2, 3, and 5). Specifically, window operation is triggered if the indoor air temperature exceeds 24°C and is at least 1°C higher than the outdoor temperature (control area 4), or if CO₂ concentration surpasses 950 ppm—provided this occurs between 7:00 a.m. and 6:00 p.m. (control area 7). Outside of occupancy hours (6:00 p.m. to 7:00 a.m.), windows will open if the indoor temperature exceeds 21°C, the outdoor temperature is above 12°C, and there is a minimum 1°C temperature difference favoring indoor warmth. Throughout the day, the window opening is further regulated by a maximum allowable air change rate (ACH), ensuring that natural ventilation does not exceed predefined limits (control area 1). The degree of window opening is modulated automatically to maintain optimal conditions.

Additionally, the heating system is activated when the indoor temperature drops below 21°C. However, during night-time ventilation, the heating setpoint is lowered to 20°C (control area 6) to prevent the simultaneous operation of radiators and window openings, thus improving energy efficiency (Rabani and Petersen, 2023). Minor adjustments have been made to reflect the occupants’ usage patterns. The hybrid ventilation approach is evaluated in comparison to a fully mechanical ventilation system, which includes heating and cooling coils as well as heat recovery, to assess its performance in reducing energy consumption while maintaining indoor thermal comfort. Table 6 presents the ventilation airflow rates for both scenarios. In the full mechanical ventilation case, night ventilation is additionally implemented from May 1st to September 1st.

To access the indoor thermal climate and thermal comfort requirements, the Fanger model (Fanger, 1970) has been used for scenarios with full mechanical ventilation, and the adaptive thermal comfort model for hybrid ventilation scenarios. This distinction is based on guidance provided in the RIF industry guidelines (Association of Consulting Engineers in Norway) and Norway’s technical building regulations (TEK17), which acknowledge the applicability of adaptive models when natural ventilation is present (Karlsen et al., 2021). For the MV, according to TEK17, specifically § 13-4 on thermal indoor climate, it is recommended that the indoor operative temperature should not exceed 26°C for more than 50 h per year under normal conditions.

The adaptive thermal comfort was applied based on the model specified in NS-EN 16798-1 (Norsk Standard, 2019). This standard provides a method suitable for buildings without mechanical cooling systems, where occupants can adapt to the indoor climate through behavioral, physiological, or psychological means.

The analysis focused on Category II comfort conditions, which represent a normal level of expectation and are recommended for existing buildings and typical applications. According to NS-EN 16798-1, Category II specifies acceptable indoor operative temperature ranges based on the running mean outdoor temperature, with ±3°C limits for naturally ventilated buildings.

The adaptive comfort temperature was calculated using the following Equation 1: T comf = 0.33 T rm + 18.8 (1) Where:

• T_comf: is the neutral (comfortable) operative temperature [°C]

For Category II, the acceptable temperature range is Equation 2: T comf − 3 ≤ T operative ≤ T comf + 3 (2)

In this study this criterion is used to assess the number of hours during which indoor temperatures exceeded the comfort limits in HV scenario.

The future climate files for this study were generated using Meteonorm (Meteotest, 2020), a comprehensive software tool that provides access to historical climate data from various weather stations. Meteonorm enables interpolation between multiple stations to enhance location-specific accuracy. It is fully compatible with IDA-ICE 5.0, allowing climate files to be exported in PRN format for seamless integration into simulations. Additionally, historical climate data can be utilized to model future climate scenarios. For this study, we used climate data from the Oslo Blindern weather station, referred to as “Historic.” This is the most recent TMYx 2007–2021 dataset available for this location. TMYx represents typical weather conditions and is similar to traditional TMY files but is based on a longer and more recent period than older TMY2 or TMY3 datasets (Crawley et al., 2020).

To evaluate the resilience of the hybrid ventilation control strategy, the Climate House will be simulated using future climate data. All setpoint values, system properties, and parameters will remain unchanged, with no adjustments or modifications. Simulations will be conducted for two future time periods: 2050 and 2100. For each period, three climate models will be used RCP 2.6, RCP 4.5, and RCP 8.5.

Figure 7 presents the average monthly temperatures for the years 2050 and 2100.

The purple line represents the baseline climate. While the year 2050 shows minimal variation among the different climate scenarios, significant differences emerge by 2100. In particular, under the RCP 8.5 scenario, the average temperature in July reaches 23.6°C, representing a 7°C increase compared to the current climate.

Due to the shared energy metering system between the Climate House and the adjacent Natural History Museum, isolating the Climate House’s specific energy consumption is not feasible. This limitation restricts the possibility of directly validating the simulated energy use with measured data. Consequently, the validation strategy was redirected toward indoor environmental parameters. The simulation model was evaluated using detailed monitoring data, with a focus on indoor air temperature and CO₂ concentration levels measured on 7 January 2023. The measurement equipment and sensor placement for both temperature and CO₂ are illustrated in Figure 8. Details of the equipment and sensors used to measure air temperature and CO₂ concentrations are provided in Supplementary Appendix 1.

A macro for window control was developed in IDA-ICE to validate measured data from the existing ventilation system of the Climate House, which combines natural ventilation via window openings with mechanical support (Figure 9). This macro operates using PI control, and when either the room temperature exceeds 21°C or the CO₂ level surpasses 650 ppm, the windows gradually open to maintain the setpoints. The “Max” control ensures that the higher of the two signals (0–1) takes priority and activates the window opening accordingly. The mechanical ventilation is set to activate at 700 ppm CO₂. In other words, the windows will open first, and if that is not sufficient, the mechanical ventilation will be activated. This is how the building currently operates.

Figure 10 presents a comparison between the measured and simulated air temperatures at the two measurement points illustrated in Figure 8. The temperature fluctuations fall within the expected sensor error range and generally align well with the measured variations in both the Amphitheater and Exhibition areas.

Similarly, Figure 11 shows a comparison between the measured and simulated air temperatures at the two measurement points illustrated in Figure 8. While the simulation results generally follow the measured trends, some deviations are observed. These discrepancies may be attributed to the fact that CO₂ levels in the measurements were recorded at specific points, whereas the simulation provided only an average value, which could account for the differences.

Figure 12 presents the variations in operative temperature and window opening signals on the coldest winter day in the Amphitheater, under the existing hybrid ventilation system in the Climate House. During winter operation in mixed-mode ventilation, the windows are required to open to control indoor CO₂ levels. However, this introduces significant drops in indoor temperature, making it difficult to maintain thermal comfort.

When the outdoor temperature reached −15°C, the operative temperature in the Amphitheater fell below 13°C. These temperature reductions occurred exclusively during occupancy hours when the Climate House primarily relies on natural ventilation through window openings. User feedback further confirms that the current hybrid ventilation control system is ineffective, likely due to the reliance on natural ventilation as the primary system, with mechanical ventilation serving only as secondary support.

The thermal comfort results for both new HV and MV are presented in Figure 13. The operative temperature predominantly remains near the lower threshold for extended periods. Temperature fluctuations are significantly more pronounced in the Amphitheater compared to the Exhibition across both ventilation strategies. Among the analyzed scenarios, the Amphitheater with HV presents the greatest challenge in maintaining a stable operative temperature, with a total of 41 h exceeding 26°C. According to the adaptive thermal comfort model, the operative temperature under HV remains within Category II for the majority of the time, with only 8 h of exceedance in the Amphitheater and 1 h in Exhibition (as shown in Table 7).

In contrast, MV demonstrates better thermal stability, with just 5 h outside Category II in the Amphitheater and none in the Exhibition. The larger window areas and a greater reliance on window openings contribute to the increased temperature variability in the Amphitheater, particularly in the HV scenario. The maximum operative temperature is relatively consistent across both zones and ventilation strategies, reaching approximately 27°C. While MV ensures a more stable indoor temperature, much of the temperature variation occurs outside operating hours, reducing its impact on overall thermal comfort assessment.

To assess the risk of draft (DR%), a detailed zone analysis was conducted using IDA-ICE 5.0. The grid resolution of the model is 0.2 m × 0.2 m with time interval 0.25 h. The analysis was performed for two different days, one in winter and one in summer. Figure 14 presents the DR% for the coldest winter day (23rd January), with an outdoor temperature of −15°C. The analysis was conducted at 15 O’clock, during which the window opening percentage was limited to 15%.

During this time period, the ACH reached three, achieved through window opening in response to CO₂ levels, as indicated by the control signals, temperature control was inactive, and window operation followed the CO₂ signal. A noticeable draft was observed near the door to the Exhibition area, while draft levels near the windows remained minimal. DR% across the zone was approximately 10%–13%. The concentrated draft in the center corresponds to the location of the single supply air vent in the Amphitheater, which leads to elevated airflow velocities and explains the draft sensation in that area. The results show a significant DR% under HV during extremely cold winter days when using the proposed new control strategy, an outcome not achievable with the current model.

Figure 15 illustrates the average surface temperature across horizontal cross-sections at different heights within space. The simulation was performed for a winter day with an outdoor temperature of approximately −9°C. According to NS-EN ISO 7730 (Standard Norge, 2005), the maximum permissible vertical temperature difference is defined as heights ranging from 0.1 to 1.1 m above the floor. The average surface temperature values shown are calculated by IDA-ICE, following the method described in Javed et al. (2021). The results indicate that while opening the windows leads to a decrease in air temperature, it remains within a comfortable range. Notably, the vertical temperature difference is lower when the windows are open, approaching ∼0°C. This reduction is likely attributed to enhanced air mixing, as the substantial temperature gradient between indoor and outdoor air drives cold air inflow, promoting effective ventilation.

The air quality in terms of CO₂ concentration (ppm) for the Climate House is presented in Figure 16 for HV and MV scenarios. The data indicates occasionally high CO₂ levels during occupancy but never exceeds the Norwegian Labour Inspection Authority’s guideline (Norwegian Labour Inspection Authority Arbeidstilsynet and in Norwegian, 2023), which sets a maximum CO₂ level of 1,000 ppm. For HV, the upper threshold is 950 ppm, at which point the CO₂ control system signals window openings. HV operates with an upper threshold of 950 ppm, triggering window openings, whereas MV is set at 900 ppm.

In HV scenario, elevated CO₂ levels persist longer in the exhibition area, likely due to fewer windows compared to the amphitheater. In contrast, for MV, higher CO₂ levels are more pronounced in the amphitheater, driven by a higher occupant density relative to room size. Notably, HV demonstrates superior CO₂ reduction efficiency during night ventilation, effectively lowering concentrations to 400 ppm over extended periods compared to MV.

Furthermore, air quality improvements are more pronounced in HV during the warmest summer periods, benefiting from enhanced ventilation rates that not only regulate temperature but also optimize air exchange. This suggests that HV offers a dynamic and responsive strategy by leveraging natural airflow to enhance indoor air quality while maintaining thermal comfort. This is supported by Table 8, which shows a lower number of occupancy hours with CO₂ concentrations exceeding 800 ppm in the HV scenario.

Figure 17 illustrates the energy consumption for heating, cooling, and fan operation in the Climate House equipped with HV and MV systems.

Figure 18 presents a diagram illustrating the percentage of occupancy time during which windows are open in the Amphitheater. The innermost circle serves as a reference, representing window opening under the current climate conditions. By 2050, there is already an increase of approximately 10%, with minimal variation between different climate models—window opening times range from 26% for RCP 2.6%–29% for RCP 8.5. However, by 2100, a more significant divergence emerges between future climate scenarios, with windows expected to be open for 27% of the time under RCP 2.6 and 39% under RCP 8.5. This trend is anticipated, as higher outdoor temperatures increase cooling demand.

Figure 19 presents the thermal comfort analysis for the Climate House with both HV and MV. Dissatisfaction percentages (PPD) are calculated exclusively during occupancy hours, omitting non-operational periods. For MV, the percentage of dissatisfied occupants rarely exceeds 25%, aligning with the Category IV threshold in NS-EN 16798-1. Similarly, for HV, PPD values above 25% are infrequent. However, under the RCP 8.5 scenario for 2100, HV faces challenges in maintaining acceptable indoor temperatures during summer months, resulting in elevated PPD levels. Specifically, 6.7% of occupancy time exceeds the 25% PPD threshold, primarily due to increased indoor temperatures. Nevertheless, considering the recommended commercial building limit of 10%, HV still demonstrates a clear advantage over MV under future climate conditions.

Based on the adaptive thermal comfort model according to NS-EN 16798-1 (Norsk Standard, 2019), the thermal indoor climate is satisfactory for all climate scenarios with hybrid ventilation. For the year 2100 with RCP 8.5, the Climate House with HV is only 23 h outside Category II. These results are not depicted in a figure but are derived from simulation data based on the adaptive model.

Figure 20 presents a 3D carpet plot illustrating indoor air quality under HV and MV for the Historic climate. Corresponding plots for future climate scenarios are provided in Supplementary Appendix 2. It is essential to distinguish the upper CO₂ concentration limits for each strategy: 950 ppm for HV and 900 ppm for MV. In the plot, the width represents the hour of the day, the length corresponds to the day of the year, and the height indicates CO₂ concentration. Notably, the blue region appears to expand in future scenarios, likely due to an increased demand for night ventilation driven by rising outdoor temperatures.

The results show that both ventilation strategies demonstrate improved air quality during summer, with this effect becoming more pronounced as temperatures rise, consequently increasing ventilation demand. Climate scenarios RCP 4.5 and RCP 8.5 show lower CO₂ concentration (wider blue area) due to higher window opening percentage. Overall, the findings suggest that future climate change is unlikely to negatively impact indoor air quality in cold climate.

Figure 21 presents the simulated energy performance of the Climate House under mechanical ventilation (MV) and hybrid ventilation (HV) across various future climate scenarios. When simulated using historical climate data, HV achieves a 13.1% energy savings compared to MV. While both ventilation strategies contribute to energy reductions, HV demonstrates a more pronounced reduction.

In the near future (2050), the projected energy savings with HV increase to approximately 20% compared to MV, marking a 7% point improvement relative to current conditions. By 2100, the energy savings vary significantly across different climate models, with HV yielding energy reductions of 19%–31%, depending on the scenario.

Rising outdoor temperatures amplify cooling demands. For MV, this results in higher fan power and increased cooling energy requirements. In contrast, HV can often rely on natural ventilation through window openings, reducing the need for mechanical cooling. This explains why energy savings with HV become more substantial as outdoor temperatures rise.

Figure 22 presents the energy consumption patterns for fan operation, cooling, and heating in the climate house with HV across various climate scenarios. The results indicate a clear decline in heating demand over time, accompanied by an increase in cooling requirements. In the historic climate scenario, heating constituted 96% of the building’s total energy use; however, under the RCP 8.5 scenario in 2100, this share is projected to drop to 66%. Despite variations in climate conditions, energy savings across different scenarios remain relatively stable. By 2100, under the RCP 8.5 scenario, energy savings for fan operation and heating improve by 5% and 1.6%, respectively, relative to the current climate, while cooling energy use increases by 5.1%. These findings highlight the HV system’s resilience and efficiency in future climate scenarios. Furthermore, Figure 23 reinforces these results by drawing comparisons with the MV system.

Figure 23 demonstrates that HV continues to outperform MV in warmer future climates, with increasing energy savings in heating, cooling, and fan operation.