The Issue
According to the latest Intergovernmental Panel on Climate Change (IPCC) report, the Sixth Assessment Report by Working Group II, various indicators for climate change have now been observed with medium to high confidence. The latest report and fact sheet for Asia regarding Energy highlights: “Asian countries are experiencing a hotter summer climate, increasing energy demand at a rapid rate, together with the population growth (high confidence).[1]
A warmer climate will undoubtedly increase the energy demand as the need for cooling and air conditioning increases. This leaves us with the question, how will a warmer climate affect the peak load (RT) and the annual consumption (RTh) of a building?
In this case study, we will investigate how a warmer climate will affect the chiller load for two different projects, one located in Kuala Lumpur and one located in Singapore.
Our Approach
For each of the two projects we first collected the current weather files, then generated the future weather files, and lastly simulated each of the projects with the current and future weather files:
1. The closest weather files to each of the project sites were selected and/or generated with Meteonorm Software.
Table 1. Current weather files
2. Future weather files were generated with Meteonorm Software and CCWorldWeatherGen. Meteonorm Software gives you access to weather data for typical years and historical data and calculation tools for estimating missing data. CCWorldWeatherGen is a Climate change world weather file generator developed by the University of Southampton. The tool enables you to generate future weather files in Microsoft Excel. Each of the tools can generate future weather files based on the IPCC scenarios, in this case, the Special Report on Emissions Scenarios (SRES) used in the IPCC’s third and fourth assessment reports. The SRES scenarios are similar to the Shared Socioeconomic Pathways (SSPs) used in the latest IPCC sixth assessment report, refer figure 1.
Figure 1. IPCC SSP scenarios comparison
SRES scenario A2 was selected for the studies, the scenario is similar to the current SSP3 scenario: Regional Rivalry and a Rocky Road (High challenges and adaptation), describing “A low international priority for addressing environmental concerns leads to strong environmental degradation in some regions.” The scenario estimates that a temperature rise of between 1.2 to 4.6°C will be very likely by the end of the century, refer table 2.[2]
Table. 2. Comparison between scenario SERS A2 and SPP3
Table 3 summarizes the weather files used in the studies.
Table 3. Current and future weather files used in the study
3. The weather files were then changed and an annual hourly energy simulation was carried out in IES Virtual Environment (VE) for each of the cases.
Our Findings
The interactive dashboard below shows how the different weather files compare. Generally, it is seen that the weather will become hotter and more humid in line with the IPCC SERS A2 and the current IPCC SPP3 scenarios.
Figure. 2. Interactive graphs with current and future weather data. (Move the sliders to filter the data)
Case 1.: Retail Mall in Kuala Lumpur
The first case looks at a retail mall in Kuala Lumpur, Malaysia. Designed to and exceeding the requirements in the Malaysia Standard (MS) 1525. The results in figure 2 show the annual hourly maximum chiller loads for each case. The results show that the midday peak chiller load is expected to increase by 17.6% to 22.7% by Mid-century (2050) and 24.2% to 26.6% by the end of the century. We are here focusing on the midday peak, as the morning start-up peak load can be reduced in various ways, for example by gradually scheduling the cooling for different areas of the building in the morning.
Figure. 2 Chiller load curve, hourly maximum
Similarly, the following figure (3) shows how the overall annual hourly chiller load distribution will increase with a warmer climate.
Figure 3. Annual chiller load distribution
Finally, Table 4 summarises the results, showing an annual increased chiller load of 12.7% by mid-century.
Table 4. Case 1. Summary cooling loads
Case 2.: Campus project in Singapore
The second case looks at a school campus project in Singapore, consisting of existing and new buildings. The annual peak loads were once again observed during the morning due to the start-up load.
Figure 4. Chiller load curve, hourly maximum
By comparing the load for each hour of the year with the respective occupancy we can see that the maximum peak loads are mainly when the school is unoccupied, meaning in the morning, during start-up, when more heat needs to be removed from the buildings.
Figure 5. Annual chiller load and occupancy distribution
By zooming in on the top 300 loads, it becomes clear that the peak loads are during start-up without any occupancy.
Figure 6. Zoom to top 300 hours, chiller load and occupancy
By excluding the start-up loads and zooming in on the loads with occupancy we can investigate the midday peak loads.
Figure 7. Zoom to top 300 hours excluding morning peak, chiller load and occupancy
Table 5 summarises the results and the estimated increased midday peak load, which is expected to increase by 4.4% and the annual load by 12.1% by mid-century.
Table 5. Case 2. Summary cooling loads
Conclusion
The simulation results for both of the cases, a retail mall in Kuala Lumpur, Malaysia and a school campus project in Singapore, can be summarised as follows:
The peak cooling load is generally seen during start-up in the morning. However, as this cooling load can be reduced by incorporating various strategies, the midday peak cooling load should be seen as the more critical cooling load to design for.
The midday peak cooling load is expected to increase by 4% (School) to 22% (Retail Mall) by mid-century (2050) depending on the building type. And similarly, the annual cooling consumption (RTh) is expected to increase by approximately 12% by mid-century for both of the cases.
By end of the century (year 2080 and 2100) the midday peak cooling load is expected to increase by more than 26% and the annual cooling consumption (RTh) is predicted to increase by more to 24%.
As the current life expectancy for chillers is between 20 – 25 years and the building's overall lifespan is expected to be between 50 to 80 years the mid-century scenarios should be seen as the most crucial scenario to design for. But with the current practice of designing and modelling the cooling load as worst-case, including various safety factors, there is a high chance that the chiller plant will already be able to handle the future loads by mid-century (compare midday peak versus start-up peak load). So what can we do to futureproof our buildings:
Make sure the building envelope is optimized, shading and lower U-values for walls, roofs and glazing can help reduce the cooling demand.
Incorporate passive strategies and hybrid ventilation, where occupant controlled fans will allow for higher air conditioning set points, which will help to lower the cooling demand.
Take advantage of "free cooling", such as AHU condensate water or sky cooling from the cool sky radiation.
Ensure that the chiller plant part-load efficiencies are optimized for various load scenarios.
Carry out commissioning and retro-commissioning to make sure that the buildings and chiller plants are actually operating as designed.
Incorporate strategies and control logics in the Building Management System (BMS), which can help to limit the peak electricity demand during warmer periods when more cooling is needed.
While these studies suggest indicative values for how the cooling consumption and chiller loads will increase with future climate scenarios, each project should be modelled and studied individually. This will ensure that the models reflect the exact project location (including updated weather data), design and usage. Additional factors such as actual occupancy, occupant behaviour and operational schedules would further impact the results.
Sources: