Seeing Double

by Stacy Rinella | August 16, 2024 7:29 am

[1]

[2]

A love-hate relationship with technology is common. It comes not only with the promise of greater freedom and productivity, but also the frustration of not living up to expectations and the fear of a slippery slope toward a dystopian future. These are all valid feelings and should inform a sober adoption of appropriate technology, because regardless of how we feel, technology is here to stay and, with thoughtful application, can provide tremendous benefits. One such example is digital twins.

We’re drowning in data, most of it unused. Do you ever wish you could clone yourself to manage it all? While this is not possible, creating a virtual clone of the things we must manage offers hope. It can help us leverage data, simplify our work, and improve results. So, what exactly is a digital twin?

Simply put, it’s a virtual representation of a real-world entity or process. A digital twin uses historic data and real-time observations to optimize systems and inform decisions. Digital twinning involves a two-way digital communication channel that provides regular updates and intelligence to the virtual representation, which then processes and interprets the data and presents it in actionable form to users, thus guiding optimization of the real-world entity. This application of technology is promising enough that the digital twin market is expected to triple
by 2030.

In buildings, a digital twin can be applied to individual buildings or entire portfolios to optimize performance and reduce cost. This is important because, over the life of a building, operation and maintenance will cost at least eight times as much as initial construction.

The primary use cases for digital twins in buildings include:

• Guiding predictive maintenance to reduce downtime and maintenance costs. For example, optimizing the timing of air filter replacements and avoiding performance losses due to clogged filters, while avoiding the cost of premature filter replacement.
• Providing fault detection by flagging the difference between predicted and actual performance of systems, an indicator of a potential component or system failure.
• Enabling predictive control of systems based on variable conditions. Traditionally, building systems are run on a set schedule, like a daily morning HVAC start-up, which may not always be optimal. A digital twin can draw from occupant arrival patterns and variable weather conditions to optimize when, where, and how much to temper indoor environments.
• Supporting autonomous operation by responding to data on changing conditions automatically within established parameters.
• Assisting in disaster management by monitoring real-time conditions to inform timely decisions.

 

A digital twin is created by combining static and dynamic data in a model that represents the real-world entity to which it’s connected. A digital twin of a building is comprised of four components:

1. A geometric model, typically derived from the building information model (BIM) generated during design and construction. Existing buildings adopting digital twins may need to create a BIM if one does not already exist.
2. A simulation to model behavior, sometimes based on engineering modeling or analytics (e.g. energy and air flow, etc.). Artificial intelligence (AI) may be applied in the simulation.
3. A data lake of historic (e.g. past performance, weather, occupancy patterns, etc.) and real-time data (e.g. sensors) typically compiled and updated using the internet of things (IoT) and AI.
4. Defined loads and restraints such as weather conditions or code requirements, set by the user to establish required performance parameters or outcomes. APIs from building systems and IoT play a role here as well.

 

Four areas where digital twins offer a promising option to improve building performance include energy performance, indoor environmental quality, resilience, and circularity.

Designing a high-performance building to reduce energy use and greenhouse gas emissions is just the start. Just like an automobile, without careful management and regular maintenance, performance will decline over time, undermining the promised return on investment (ROI). Digital twins are a robust means to maintain and optimize performance.

People now have higher expectations for the indoor environments they occupy. Changing use patterns and dynamic outdoor conditions mean that static controls aren’t adequate to maintain healthy and comfortable indoor environments while also saving energy. Sensor technology has improved and helps meet this goal, but it’s not perfect. A digital twin can optimize performance based on data derived from multiple sensors and provide fault detection if any of the sensors fail or fall out of calibration.

Natural and anthropogenic hazards and climate change present increasing physical risks to buildings and their occupants. The systems in resilient buildings need to react quickly to these unpredictable hazards, and a digital twin provides an ideal assistant for emergency managers. From automatically adjusting HVAC system operation during severe wildfire smoke events to providing real-time condition assessments to a remote building manager during acute hazard events (e.g. hurricanes and floods), digital twins can support informed decision-making and timely actions that could preserve occupant safety and limit property damage, as well as speed recovery.

While most use cases for digital twins focus on building operations, they can also be applied in construction. A digital twin can optimize material use and reduce waste, thus reducing embodied carbon. It can also predict potential constructability conflicts that could lead to costly and wasteful delays. Applying a digital twin during construction sets the building up for deconstruction and material recovery during future renovations and at the end of its service life, as it can serve as a catalog of the types, sizes, quantities, and, most importantly, values of the materials that can be recovered and sold.

There are some barriers to adopting digital twins in buildings today. Currently, creating a digital twin requires an upfront cost with a potentially slow ROI through performance savings. The cost is not necessarily proportional to the building size, so digital twins are currently more applicable to larger buildings. While digital twins can be maintained entirely on-premises, linked digital systems can create cybersecurity and privacy risks. Digital twins require maintenance, as changes to building uses and physical alterations to the building will also require updates to the virtual model. Finally, human-nature can be an obstacle to the adoption of digital twins, as the users of the system may be concerned with its potential to expose their poor performance.

Digital twins corral and harness the copious data we have, both historic and real-time, and use a virtual model to test potential outcomes, identify optimal responses, and guide automated responses or inform human decision and action to maintain and improve building performance.

This optimization can significantly reduce greenhouse gas emissions, improve occupant comfort and wellness, and increase resilience. Ultimately, even greater benefits will be derived by creating ecosystems of digital twins to provide integration and optimization of the complex and interdependent public and private infrastructure of smart cities. It may be time for us to double down on digital twins.

Alan Scott, FAIA, LEED Fellow, LEED AP BD+C, O+M, WELL AP, CEM, is an architect and consultant with more than 35 years of experience in sustainable building design. He is director of sustainability with Intertek Building Science Solutions. To learn more, follow him on LinkedIn at linkedin.com/in/alanscottfaia/.

Source URL: https://www.metalarchitecture.com/articles/columns/seeing-double/

---