Using Digital Twins to Optimise Renovations and Drive Zero-Carbon
Why Renovate?
To reach ambitious zero-carbon targets that are set for 2050, we must start by improving the energy efficiency of existing buildings. Why? Because there are so many buildings already built that are wasting energy.
It seems simple on paper. But why then are renovation rates only around 1% in Europe? The answer comes down to cost. Renovation is a very uncertain investment. It requires high upfront costs, and the energy and cost savings benefits are usually only rough ballpark estimates, so there is a high level of uncertainty on what the Return on Investment (ROI) will be. Therefore, it poses a very high-risk investment that not many are willing to take.
How can we change this?
When you first look at renovating an existing building, the reality is that you’ll know very little about the building. The likelihood is that it’s been built at a time when there’s been poor documentation, or the documentation has been lost, so those involved in the renovation design have to make a lot of assumptions.
But, what if you didn’t have to make so many assumptions? What if you had an accurate model of your building design that would behave exactly like its real-world counterpart and you could use it to test different design options and make better decisions? What if you could rely on these results to see exactly what you would get back from your investment? This would make for a much better, lower-risk investment.
Image 1: Digital Twin illustration of San Francisco skyline
This is what Digital Twin technology allows you to do. It takes real-life data from the actual building, filling gaps with AI and Machine Learning, and uses it to feed an accurate physics-based model which can be used as an operational model, i.e. an accurate representation of your building at any point in time. This model can then be used to design the renovation solution, making much more accurate predictions than what is actually made today. Right now, designers are forced to make very broad assumptions, having to rely only on compliance reference values. This affects how realistic the design is, and leads to what is known as the ‘performance gap’, where the design estimates for energy consumption are often much lower than the actual consumption of the building.
Add to this the issue of the ‘rebound effect’, i.e. the building actually performs better than expected before renovation, and performs worse than expected after renovation. When this happens, the resulting actual savings are far less than anticipated. To avoid this, we must undertake a Measurement and Verification process. This allows us to first understand the baseline of how the building is performing before renovation, and then measure and verify the savings after renovation. Through the creation of the Digital Twin, which accurately represents the actual building energy consumption, a precise baseline can be created before renovation. This then ensures that the correct decisions are taken, and results can be adequately validated after renovation through approved M&V processes, such as the IPMVP or ASHRAE 14.0.
Image 2: IESVE model image
Using Digital Twins we can close this performance gap and do much better in designing interventions for renovation. The European Commission estimates that in order to meet Government Zero-Carbon target by 2050, renovation rates need to rise to 3%. Therefore, we need to find a way to encourage more renovations of existing buildings.
This is what the Horizon 2020 (H2020) StepUP project aims to do. Its objective is to develop a new process for deep renovation for decarbonisation, to minimise the performance gap, reduce investment risk and maximise value.
IES ICL Digital Twin
As project coordinator, IES is providing the technology to help achieve this goal. The Intelligent Communities Lifecycle (ICL) digital twin technology, is the key to ‘how’ we decarbonise the built environment.
National governments and international organisations, such as UNEP, have already provided excellent information on ‘how’ to drastically cut emissions across the sector, including:
- Establish a baseline for building energy consumption and propose a number of % savings for improving the building envelope; or improving energy equipment efficiency;
- Integrating buildings using community-wide solutions, such as district heating or cooling solutions;
- Deploying renewable energy solutions;
- Improving occupant use of the building e.g. not leaving lights on, etc.
This is all well and good, until you consider the potential adverse effects and barriers to making these concepts work. They do not define how to improve energy efficiency of the building, how to develop community-based district heating or cooling solutions, or how to enable uptake of renewable energy solutions, for example.
We also know that no single solution can solve the issue in isolation. A holistic community approach, which also engages the citizens and technology providers key to any change, must be taken. So how can we decide on the right combination of options to achieve the best outcomes, engage those key stakeholders and avoid making decisions which will result in secondary problems or be too cost prohibitive?
This is where IES, and our ICL Digital Twin technology, come in. Outdated methods such as spreadsheets and simplified mathematical approaches will never be able to answer this question for you. It is extremely difficult to make a coherent and effective decarbonisation plan, understanding the actual impact of the different solutions, how they act holistically, and, most importantly, to track and monitor the progress of the plan.
Image 3: Illustration of outputs from IES’ Intelligent Communities Lifecycle (ICL) Digital Twin Technology
Using our unique integration of digitised physics, artificial intelligence (AI) and machine learning, we can help create a digital twin of any building to analyse all the various decarbonisation options, identify and weigh up the risks and potential savings – prior to implementation – and monitor progress towards net-zero targets over time.
The IES ICL technology also utilises the concept of Augmented Intelligence – the all important ‘know how’ which enhances human intelligence, rather than replacing it – to de-risk the decarbonisation process.
The problem with data
One of the issues many have come across in the past, is data, or rather a lack of it. This posed a problem because the technology to accurately fill these gaps via virtual sensors was not available. With digital twin technology, this is no longer a problem. You can start with the physics based model and augment this model with data from the building to improve the accuracy of the physics model. In turn, you can also use this enhanced physics model to create virtual sensors in the building and reduce the number of actual sensors that need to be placed in the building. This exercise is carried out in an iterative process, utilising sensitivity analysis and results in a Digital Twin created at the lowest cost to the end user. Using Machine Learning and AI, data gaps can be filled and data from different sensor sources (e.g. IoT and BMS) and at different time steps can all be integrated within the one model. Furthermore, if only partial data is available, e.g. 3 months BMS or utility data, regression algorithms can be applied to create a 12 month profile of the energy demand, which can be applied to the physics based model. In this way, the end user does not need to wait for 12 months data to be available, before they can create an accurate baseline of their building.
Image 4: Data outputs from IES iSCAN technology
A physics-based digital twin model like those created by the ICL, takes away the requirement to collect massive amounts of data. In fact it minimises the amount of data we need to gather, which is not just cheaper, but also great for GDPR – we never want to collect more data than we need. With digital twin technology we don’t have to fill buildings with sensors; we can focus on getting the high priority data, fill in the gaps and then make smaller collections as required.
We all understand the importance of data privacy, and in a data-rich digitised world this is increasingly becoming a problem.
In our next blog post for the StepUP project we will be addressing this topic and the possible solutions in more detail.
For more IES content please visit www.iesve.com/discoverIES
Authors
Giulia Barbano – Project Manager, IES Ltd, MEng (Building Engineering), StepUP Project Coordinator
Ruth Kerrigan – Director, IES R&D, B.A., B.A.I., PhD, CEng
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 847053.
This website reflects only the author’s views. The European Climate, Infrastructure and Environment Executive Agency is not responsible for any use that may be made of the information it contains.