Buildings as Energy Banks: New Model Unlocks Hidden Storage Potential
In the global race toward carbon neutrality, the humble building is emerging as a powerful new player in the energy landscape. No longer just consumers of electricity, homes, offices, and commercial structures are being reimagined as dynamic energy hubs, capable of storing vast amounts of power and helping to stabilize the grid. A groundbreaking study from Tsinghua University has unveiled a novel framework that quantifies this hidden potential, positioning buildings as central to the future of renewable energy integration.
The challenge is clear. As wind and solar power become the dominant sources of electricity, their inherent variability—sunlight fading at dusk, wind dying down—creates a critical need for energy storage. While large-scale solutions like pumped hydro and grid-scale battery farms are essential, a new frontier lies closer to home: the demand side. The building sector, responsible for a massive share of global electricity consumption, holds an untapped reservoir of “virtual energy storage” (VES). This isn’t about installing more lithium-ion batteries in every basement; it’s about harnessing the energy flexibility already embedded in the devices and systems we use every day.
Researchers at Tsinghua University have developed a powerful new tool to make sense of this complex potential. Led by Liu Xiaochen, a doctoral researcher, and Professor Liu Xiaohua from the Department of Building Science, the team has introduced an “equivalent battery model.” This model, published in the prestigious Proceedings of the CSEE, translates the diverse and often chaotic world of building energy use into a standardized, engineer-friendly language. By defining parameters like equivalent charging power, equivalent discharging power, and equivalent storage capacity, they have created a universal metric to compare and integrate different forms of virtual storage, from electric vehicles parked in garages to the thermal mass of a building’s concrete structure.
This approach is a significant departure from traditional energy planning. Historically, buildings have been viewed as static loads on the grid, with demand peaking predictably in the morning and evening. The concept of “energy flexibility” flips this script. It recognizes that many building systems do not need to operate at peak power at the exact moment energy is consumed. An air conditioner can cool a room slightly more before a heat wave hits, storing “coolth” in the walls and furniture. An electric vehicle can delay its charging until solar generation is at its peak. A smartphone can be charged at a slower rate overnight. Each of these actions represents a small act of energy storage, a deferral of demand that can be coordinated at scale.
The “equivalent battery model” is the key to unlocking this potential. It provides a systematic way to calculate how much flexibility a building truly has. For engineers and urban planners, this is invaluable. It moves the conversation from theoretical potential to concrete design. Instead of asking, “How many batteries do we need?” the question becomes, “How much of our storage need can be met by the building itself before we install a single battery?”
The implications for electric vehicles (EVs) are particularly profound. The paper identifies the “EV + smart charger” pairing as one of the three primary sources of virtual storage in buildings. The average private car spends over 90% of its time parked, often in a building’s garage or parking lot. This parked vehicle, with its large lithium-ion battery, is a mobile energy storage unit waiting to be utilized. The study quantifies this potential, showing that the equivalent storage capacity of a fleet of parked EVs can rival or even exceed that of dedicated battery systems.
The model’s power lies in its ability to account for real-world constraints. Not every EV will be available for grid services at any given moment. The model incorporates factors like parking duration, vehicle availability, and battery management systems that protect the battery from deep discharge. It differentiates between one-way chargers, which can only draw power from the grid, and bidirectional chargers, which can also feed power back. A building with a fleet of EVs connected to bidirectional chargers becomes a powerful, distributed energy resource, capable of providing both power and capacity to the grid during peak hours.
The second major source of virtual storage is the Heating, Ventilation, and Air-Conditioning (HVAC) system coupled with the building’s thermal mass. HVAC systems are often the largest energy consumer in a building, making them a prime target for flexibility. The study details how the thermal inertia of walls, floors, and furniture can be leveraged. By pre-cooling or pre-heating a building during off-peak hours when renewable energy is abundant, the system can then reduce or even shut off during peak demand periods, effectively “discharging” the stored thermal energy. This is not a new concept, but the “equivalent battery model” provides a rigorous way to quantify it, turning a qualitative idea into a precise engineering parameter.
The researchers go further, breaking down HVAC systems into two distinct virtual storage components. The first is the “cold/heat source + distribution” system, which includes dedicated thermal storage tanks (like water or ice tanks) and the associated pumps and chillers. The second is the “terminal devices + building thermal mass” system, which uses the building structure itself as the storage medium. The model calculates the equivalent efficiency of these systems, which is often lower than that of electrochemical batteries due to the thermodynamic losses inherent in heating and cooling processes. Nevertheless, the sheer scale of this potential is immense, especially in large commercial buildings and data centers.
The third, and perhaps most ubiquitous, source of virtual storage is the myriad of “electrical appliances with energy storage.” This category includes everything from laptops and smartphones to cordless power tools and e-bikes. While the storage capacity of a single device is small, the aggregate potential of millions of devices is significant. The study highlights that a large portion of the energy consumed by these devices is wasted in “vampire” or “phantom” loads—power drawn when the device is fully charged or simply plugged in but not in use. By intelligently managing the charging cycles of these devices, a substantial amount of flexible demand can be created.
The real innovation of the Tsinghua team’s work is not just in identifying these resources, but in providing a practical design methodology. The paper outlines a step-by-step process for engineers. First, they define the building’s typical electricity demand and the target “grid-friendly” supply curve (e.g., matching a local solar generation profile). The difference between these two curves defines the total “generalized energy storage” (GES) capacity needed. Then, using the “equivalent battery model,” they calculate how much of this total need can be met by the building’s virtual storage resources—the EVs, the HVAC system, and the appliances. The remaining gap is the amount of traditional, stationary battery storage that must be installed.
This methodology was demonstrated with two compelling case studies: a small office building and a residential apartment complex. In both cases, the results were striking. By fully utilizing the available virtual storage, the required capacity of traditional lithium-ion batteries was reduced by more than 60%. For the office building, the virtual storage from its EV parking fleet alone provided the majority of the needed flexibility. For the residential building, the combination of EVs and smart appliance management was sufficient to eliminate the need for any additional stationary batteries to achieve a zero-carbon operating potential with a 1:1 wind-solar power supply.
These findings have profound implications for the future of urban energy systems. They suggest a path to a more sustainable and cost-effective energy transition. By reducing the reliance on expensive, resource-intensive stationary batteries, cities can deploy renewable energy faster and at a lower cost. It also enhances grid resilience. A network of flexible buildings, each acting as a small energy bank, can respond rapidly to fluctuations in supply and demand, smoothing out the peaks and valleys that challenge grid operators.
The research also addresses a critical bottleneck: the challenge of coordinating millions of distributed devices. How do you control the charging of thousands of EVs and smartphones across a city? The paper points to emerging technologies like advanced IoT networks, 5G, and power-line communication as viable solutions. These technologies can broadcast signals to smart devices, enabling them to autonomously adjust their charging behavior in response to grid conditions, without requiring constant, direct control from a central authority.
This vision aligns perfectly with the “Photovoltaic, Storage, Direct Current, Flexible” (PSDF) concept championed by Professor Jiang Yi, another co-author of the study and a leading figure in China’s building energy research. The PSDF system is a holistic approach to building design, where on-site solar generation, integrated storage, and direct current power distribution are combined with deep energy flexibility. The “equivalent battery model” provides the crucial missing piece—the quantitative foundation—for designing and optimizing such systems.
The study also makes a strong economic case. The paper includes a detailed analysis of investment costs, comparing the cost of traditional battery storage with the incremental cost of enabling virtual storage. For EVs, the cost is primarily the smart charger, not the vehicle itself. For HVAC systems, it’s the cost of the control system, not the building structure. For appliances, it’s often zero additional cost. When these costs are amortized over the equivalent storage capacity they provide, the result is a fraction of the cost of a lithium-ion battery. This makes virtual storage an incredibly cost-effective way to add grid-scale flexibility.
The environmental benefits extend beyond carbon reduction. Manufacturing lithium-ion batteries requires significant mining for lithium, cobalt, and nickel, which has its own environmental and social costs. By minimizing the need for new battery production, the widespread adoption of virtual storage can reduce the overall environmental footprint of the energy transition.
The work from Tsinghua University is a paradigm shift. It moves the focus from adding hardware to optimizing software and control. It recognizes that the intelligence and flexibility of our built environment are as valuable as the physical infrastructure. It transforms buildings from passive energy sinks into active participants in the energy market.
For the automotive industry, this research is a wake-up call. The value of an electric vehicle is no longer just in its ability to replace a gasoline car. It is now a key component of a larger energy ecosystem. Automakers, charging infrastructure providers, and utility companies have a powerful incentive to collaborate on developing and deploying bidirectional charging technology. A car that can power a home (V2H) or feed the grid (V2G) is not just a technological marvel; it is a critical piece of national energy infrastructure.
For architects and building engineers, the message is clear: energy flexibility must be designed in from the beginning. The choice of HVAC systems, the provision of EV charging infrastructure, and the integration of smart building management systems are no longer just about comfort and convenience. They are fundamental decisions that determine the building’s energy storage potential and its contribution to a sustainable future.
The “equivalent battery model” is more than just an academic exercise. It is a practical, scalable, and economically sound framework for a smarter, more resilient, and truly sustainable energy future. It demonstrates that the solution to the world’s energy challenges may not lie in some distant, futuristic technology, but in the intelligent use of the systems and devices we already have, right where we live and work.
Liu Xiaochen, Liu Xiaohua, Zhang Tao, Li Hao, Jiang Yi, Department of Building Science, Tsinghua University, Proceedings of the CSEE, DOI: 10.13334/j.0258-8013.pcsee.222949