At the home office in the summer cottage. How are the customers doing? In just a few hours I check the situation at almost one hundred properties.
- Are there any problems anywhere with energy use? Is it due to ventilation, heating or a poorly functioning heat recovery system? Is the operating efficiency of the heat pumps sub-optimal?
- What about the conditions in the building?
These properties usually have 500 ventilation machines and a heat recovery system, one hundred heating networks and over 30 heat pumps and ten solar energy plants.
You probably think that sounds like a hell of a job. In the past this would have taken a week. Now I get all the data directly from the cloud on my screen, organised, and in real time. When it’s the technology finding and recognising the problems for me, I can use my time for solving them.
Tesla conquered the electric car market partly due to the fact that it was the first to truly connect the cars to the cloud. The cars send a huge amount of data to be analysed every day. Thanks to that data we can recognise possible faults before they cause problems and create new features for the cars.
You drive your car into the garage in the evening, and in the morning you notice that it has a new feature. The car feels new again.
Data is fuel
Data is the fuel for machine learning and AI. The more data, the better machine learning and AI can be used in developing new automated processes.
The same is true for buildings. The brains of a building have traditionally been the building services systems that can be likened to the engine management system in a car. They vary in both cars and buildings according to age, make, type, technology and type of control. In cars, however, the established user interfaces are very similar in terms of their core functions. It is difficult to imagine a car without a steering wheel, pedals or a speedometer. Once you have driven one car, after a little practice you can drive any car.
Buildings do not have this kind of established user interface, rather each building is individual. That’s why the person in charge of maintenance has to gain a deep understanding of the technology and processes of each building so that they can control the building as well as possible. It takes time, and often the learning is done the hard way. Even hardened professionals sometimes struggle to understand the effects different adjustments have on different functions.
It’s the same as if you would try to drive a car without controls or dials. Instead of using the speedometer and accelerator you would have to control the speed by changing the fuel mixture ratio or controlling the engine’s electrics. And what if the speed and fuel dials took a few days to respond, as is the case in many building energy-use reports? So when you drive to work today, you only find out a day later if you kept to the speed limits and used fuel efficiently.
This is not a particularly smart way of doing things. That’s why we have developed a platform where the technical processes of the properties are modelled with the help of what is known as a digital twin. We can intervene in events instantly rather than at the stage where the problem becomes visible to the user resulting in dissatisfaction or an over-inflated energy bill.
Underpinning all this is the same engine management logic as in a car. If a red light goes on, the maintenance person is immediately called to diagnose the problem and if necessary adjust the controls.
A self-driving property
The data harvesting possibilities are only one side of the coin. Harvested data can be enriched with external data. In this way the enriched data can be used to create new features for the existing technology used in the property. How would it sound if your property, for example, would learn to connect to the demand-side energy market or take the weather forecast into consideration all based on its own data?
We are not necessarily too far from the day when a property can do its own long-term plan. It monitors the operating efficiency of equipment and predicts when is the best time to perform maintenance or order new equipment. In this way the property saves energy and levels out the need for peak power. At the same time the comfort of the user is improved and the conditions remain stable.
The first steps have already been taken. Our digital twin already actively collects, reports and manages properties. As the collected data increases, we develop and create new smart models that open the way to functionalities that previously seemed impossible. We create added value throughout the entire lifetime of the property.
It’s not a question of if there will be a platform economy for properties or not. Rather it is a question of when will it happen and who will be the first to benefit from its potential. They will be the first to enjoy the benefits and competitive advantages that the cloud brings.
Author, Ari Taiponen, is in charge of digital services at LeaseGreen. Ari makes sure that the new smart solutions offered to our customers provide the best possible added value, support the achievement of lifecycle benefits and keep our customers satisfied.