After proving the value of our SentientSystem solution for real-time asset optimisation in the power generation industry, we decided to adapt our digital twins services to the built environment.
The built environment, involving:
- A high dependency on natural resources
- Complex and growing infrastructures
- Intensive energy consumption assets
appeared to be a perfect candidate for our hybrid digital twin. Likewise, it perfectly aligned with Synengco’s philosophy to reduce human impact on the environment.
Naturally, the complex operations of a building and its infrastructure can result in functionality inefficiencies. This leads to unnecessary expenditure and avoidable emissions production. However, with an increasing amount of data and knowledge available, intelligent data-driven analysis tools are able to support decision-makers with identifying and eliminating these inefficiencies.
By combining the knowledge acquired by our team in complex systems modelling, data science and software development with R&D projects (realised for the building industry), we started developing our own building-focused solution. The goal being to provide buildings owners and property managers with tools to decrease operational costs while enhancing their environmental credentials.
Our solution: quantify energy saving opportunities through real-time digital twins
The optimisation process within the built environment is designed to develop an adequate understanding of an asset and its environment before executing improvements. It is important to analyse a whole portfolio of assets to gain insights about operating and functional inefficiencies within a building. These insights inform decisions for building improvements. Our digital twins are then implemented into the improvement process to monitor behaviour and provide predictive analytics. In utilising digital twins in the built environment, they create ongoing improvements for your assets.
To create additional value in the built environment, there are three important stages to undergo.
Stage 1: Analysing a whole portfolio of assets
Before starting an advanced asset modelling process, decision-makers must be provided with insights as to where development opportunities lie in their building portfolio. To do this, it is important to understand how the whole building portfolio is performing comparatively to other buildings. There also must be consideration as to how assets compare to each other within the portfolio. These considerations highlight information about a building’s costs, emission ceilings and energy consumption. By iterating these steps, decision-makers receive a range of quantified improvements that they can choose to implement depending on their goals and constraints. It also allows decision-makers an insight into the buildings with the biggest improvement opportunities. This is particularly useful for property managers who want to improve their portfolio performance within a limited budget and timeframe.
There are two methods of analysing opportunities…
Through data science
Analysing the existing buildings data is the first step in identifying issues and opportunities within a portfolio.
One can easily compare energy use intensity (EUI) between buildings and see which ones perform well. However, with many variables impacting EUI such as climate, occupancy or building shape, this simple metric is not enough to accurately evaluate a building’s improvement opportunities.
Luckily, data science gives us a wide range of tools for advanced data processing, analysis and the development of actionable insights. Data is key in establishing accurate correlations between variables such as a building’s functionality and energy consumption. For example, heat and location maps are effective visuals in analysing the efficiency of a building.
Once these multi-criteria correlations are established within a portfolio, or through existing buildings databases, we can evaluate how each building compares to the baselines. This results in a hierarchy of buildings with the highest potential for improvement.
Through high-level twin modelling
A second approach to portfolio analysis is to model all assets with minimal details. This form of modelling enables rapid development of digital buildings with associated theoretical consumption values. The digital replicas are then tuned with data from the actual asset behaviour.
In utilising these digital models, inefficiencies are identified by comparing asset performances over a specific period with specific external conditions.
Which tool to pick?
Both tools are effective and viable in the development of digital twins. They utilise different methodologies to cater for the user’s requirements. If not requiring a digital model, the data science tool is a fast method of identifying inefficiencies, useful for those with a small timeframe. On the other hand, high-level twin modelling utilises a digital model to provide thorough analysis. This tool is effective for those with a larger budget and a larger timeframe.
Stage 2: Focusing on higher opportunity buildings
Once the underperforming buildings are identified, they can start being modelled at finer-grained levels. Documents such as blueprints, diagrams of mechanical services, system specifications and historical data are used to establish an information model. This creates the basis of the building’s digital twin. The digital model evaluates how the buildings performance and expenditure impacts on its economic and natural capital.
A digital twin for retro-commissioning
A built digital twin enables retro-commissioning, which is the process of improving the efficiency of a building’s equipment and systems. It does so by evaluating all systems and operations through operational and historical data.
These data pieces include measurements of multiple systems’ variables. Comparing these variables with the theoretical values calculated by the digital twin, performance inefficiencies and defects are identified. These issues reflect improvement opportunities within the building.
The digital twin goes one step further by accurately predicting costs and levels of performance improvement as a result of these uncovered opportunities. Its scenario analysis capabilities (also called a what if analysis) simulate internal and external changes to identify their impact on the whole building system. This allows a building owner to minimise risks on a decision-making basis.
Stage 3: Enabling ongoing commissioning with real-time monitoring and prediction
The value of the digital twin is fully uncovered through its integration of real-time data. This data, collected through onsite monitoring devices, informs early warning systems. An early-warning system alerts a user when the system’s actual behaviour deviates from its predicted behaviour. As a result, the user can undertake relevant actions in accordance with the fault diagnosis. Analytic fault systems suggest maintenance activities or equipment upgrades, avoiding additional costs and impracticalities that result from unexpected disruptions.
Consequently, real-time data produced from the digital twin system supports ongoing commissioning (the process of continuously improving a buildings performance). The capabilities for continuous improvement allows for an asset that functions more effectively, is more cost-effective and has a larger lifespan.
To see how our digital twin technology is applied in other industries, check out our SentientSystem solution case-studies.