TNO and the energy company Eneco are cooperating on the development of smart and “self-learning office buildings”. An office building that provides model-based insights and recommendations about energy consumption with regard to climate control in workplaces. This model addresses both energy conservation and a better indoor climate. The model helps to show whether specific energy conservation strategies can be used, endeavouring to generate the desired temperature settings at the right time with the lowest possible energy consumption.
For over twenty years, TNO has been working on building models that provide insight into energy usage, based on pre-set parameters relating to the building and to the number of occupants. Such parameters generally take little or no account of the varying number of people in the building, their location at any given time, and the characteristics of the building itself. The novel aspect of this self-learning model is that it automatically collects data (from readily available sources of energy data on gas, electricity, heat and from a limited number of sensors), from which it learns what kind of building it is and how the flexible energy demand can best be met. To this end, a minimum number of building characteristics (number of rooms, spaces, floor area and volume) are used to make the model as universally applicable as possible. The model provides insight into current and expected energy demand and into the options for managing it.
“We are not only focusing on energy conservation, we also want to provide a range of other services, such as systems to create healthier working climates in offices”
Healthier working climate
Over the past six months, the model has been tested in a number of office buildings, including Eneco’s headquarters. The model was primarily designed with the commercial built environment in mind. Peter van Buijtene, Director of Smart Buildings at Eneco Zakelijk, explains that the self-learning model is perfectly in line with Eneco’s aspirations for the future. In addition to supplying energy, the company wants to deliver various energy-related services that will benefit business customers and the energy network as a whole. “We are not only focusing on energy conservation, we also want to provide a range of other services. Such as systems to create healthier working climates in offices. That would improve both the welfare and labour productivity of the staff involved. To this end, we feel that a detailed and up-to-date insight into an office’s energy demands (for heating, cooling and ventilation) is extremely valuable. We hope and expect that this model will provide useful information that, in a few years, will enable us to better coordinate the energy demands of business customers who occupy the same building. As an added benefit, this will also result in lower energy consumption.”
“In addition to harnessing our knowledge and expertise to achieve energy neutral buildings, we are making every effort to create energy-positive buildings: buildings that actually produce energy”
At the same time, the use of this model is perfectly in line with TNO’s energy strategy, explains Huub Keizers, manager of TNO’s energy in the built environment programme. “In addition to harnessing our knowledge and expertise to achieve energy neutral buildings, we are making every effort to create energy-positive buildings: buildings that actually produce energy. About forty percent of all energy consumption (and the associated CO2 emissions) takes place within the built environment. Thus, with sustainability in mind, there is much to be gained by tackling the energy consumption of buildings.”
The use of the self-learning model in the various pilot projects is being supported by Urban Energy, a Top Consortium for Knowledge and Innovation within the Energy Top Sector. The model consists of a mathematical description of a building’s thermodynamics. Each day, it collects data on the energy performance of the building’s systems and on the prevailing weather conditions. Each day, the system learns more about the patterns in these parameters until it is sufficiently “intelligent” that it can be described as a genuine Smart Building. While the model is theoretically suitable for any conceivable type of building, it is most likely to have the greatest added value in office buildings.
Making energy demand more flexible
Mr Keizers points out that “Sometimes, for example, we see that different parts of the same building are being heated and cooled at the same time. The model will then chart the characteristic parameters of a building and will quickly identify any simultaneous heating and cooling. This model enables you to understand the energy demand at given times, and to manage it more astutely. In this way, it is easier to anticipate changes at building level. We know that the model works. The goal of the pilot project is to learn lessons from all kinds of practical situations. We aim to optimize the energy requirement of an office building, while making allowance for the number of occupants and for their physical comfort.”
“This model provides insights into the energy demand at given times, enabling you to manage it more astutely”
Mr van Buijtene adds that “We are delighted to have the opportunity of testing TNO’s model in an exploratory study. We believe that the results will help us to accurately identify the current energy demand, in terms of ventilation, heating and cooling. We are keen to help make the commercial built environment more sustainable, and to steer the energy transition. I believe that a thirty percent energy conservation improvement in buildings is well within our reach. For us, however, the key question is: how can we increase the flexibility of energy demand above the level of individual buildings? That would enable us to avoid peak demand at the start of the working day. It would also make the energy network more manageable. Decentralized energy generation is a major factor here. Another factor involves responding to differing energy demands within buildings (heating, cooling) at local level, by exchanging heating and cooling for example. In the future, the self-learning model may be used at a level above that of individual buildings.”