Iot For Smart Buildings - Long Awaited Revolution or Lean Evolution
Iot For Smart Buildings - Long Awaited Revolution or Lean Evolution
Abstract—It is expected that popularization of smart building a) Traditional building b) Intelligent building c) IoT based building
technology will redefine the way we work and live in the future.
Energy used in buildings represents significant part of global Lights Garden Lights Garden Lights Garden
energy consumption and humans spend most of the time indoors. HVAC Multimedia HVAC Multimedia HVAC Multimedia
Visualization
Even now, using proven and commercially available technology, Air quality Roller blinds Air quality Roller blinds Air quality
Predicitve maitenance
Machine Learning
Roller blinds
it is possible to achieve significant reduction in building mainte- Energy Surveillance Energy Surveillance Energy
Artificial Intelligence
Big Data Analytics
Storage
Surveillance
management management
nance costs and energy consumption providing more comfortable Renovables Internet Renovables Internet
management
Renovables Internet
living environment at the same time. Nevertheless, the promise Security Fire alarm Security Fire alarm Security Fire alarm
of intelligent buildings extends far beyond energy efficiency or
housing comfort. Together with concept of Internet of Things it Management consoles for Multifunction management
Multifunction management
may change the world more than the Internet did. Will it be so? individual subsystems consoles
consoles
How this process will look like? What are potential threats? This
paper will briefly cover current trends in building automation Fig. 1. Regular vs. Intelligent building (base on [4])
systems and try to answer all those questions. It will focus both
on currently implemented solutions, mainly from ABB portfolio,
as well as on potential disruptive technologies that will shape the •Reduction of building operation and maintenance costs,
future of intelligent buildings.∗
•Improvement of health, satisfaction and well-being of
I. I NTRODUCTION building users,
• Minimization of building environmental impact (not only
Currently, over half of the world’s population lives in urban by the reduction of energy/water consumption or waste
areas, it is expected by 2050, this coefficient will grown to 66 production, but also by sustainable architecture),
per cent [1]. The coming decades will bring changes in the way
Of course, there are many different types of buildings, each
space and buildings are developed and used. Land shortage
type has its own specific characteristics, but to some extend,
leads to a tendency to construct taller buildings and dig for
most of the considerations discussed in this text can be applied
space underground. Such buildings exploit the land, but have
to all of them. The core part of each smart building is Building
the negative effects on the environment and create new prob-
Automation System (also known as Building Management
lems including increasing population congestion, pollution,
System). It controls and monitors the building’s mechanical
reduce citizen access to fresh air and sunligh [2]. Those factors
and electrical equipment such as ventilation, lighting, power
will be important drivers for the evolution of smart buildings.
systems, fire systems, and security systems [3]. In traditional
Urbanization and shortage of natural resources requires build-
building, those subsystems operates separately and commu-
ing owners and the government to take responsibility not only
nication between them is limited (Figure 1a). If data remains
for building maintenance and operations, but also to care for
solely contained within the boundary of the original subsystem
such things as sustainability, energy management, pollution
it can not be fully utilized. Those so called information silos
or livability. To address new challenges construction industry
are typically operated by a unique, vendor specific software
must apply latest advances in technology (new materials, IT
system which additionally makes them difficult to manage and
solutions) as well as improvements for health and well-being
maintain. Moreover, it is common that, due to the lack of
of people.
integration, various subsystems duplicates the same sensors.
II. S MART BUILDING DEFINITION Smart buildings use information technology to connect
variety of subsystems in the way those subsystems can share
Unfortunately, there is no common definition of smart the information to optimize total building performance. The
building. Existing definitions focus on three major features communication between particular subsystems is more often
of smart buildings: standardized. Data integration in smart buildings goes beyond
∗ Publication is based on keynote speech lead by Marcin Bajer on 23rd
the equipment installed within building four walls. Smart
International Workshop of the European Group for Intelligent Computing in buildings interact with IT systems of utilities providers or
Engineering June 29th – July 1st, 2016, Kraków, Poland emergency services as well as with dedicated software which
Building Tenant
… Automation Management
Systems Software Fig. 3. Building automation protocols
Gas Police
Computer
Aided
Design
Geographic
Information new technology adoption and may foreclose achieving, by the
System
Water Software
Security new buildings, targets in energy, economic and environmental
performance [6]. It is much easier to find common solutions on
higher layers of communication. The are two protocols usually
Fig. 2. Smart building data integration layers
used for backbone of smart buildings - KNX and BACnet,
each of them provides a set of different gateways to allow
supports facility management (Figure 2). interoperability and integrate other communication protocols.
There is a common misconception that smart buildings
are only limited to the Building Automation System (BAS) IV. I MPACT OF I OT ON SMART BUILDINGS
part. In fact, smart buildings are much more than just set The IoT uses one common Internet Protocol (IP) to connect
of automation systems. In the literature one may find also devices, every single device can communicate directly with
term intelligent building which is older and initially had more each other and/or with the cloud. To cover all specific topics
limited scope - to some extend similar to modern Building related to implementation of this concept for buildings special
Automation Systems. Currently, term intelligent building is term - Building Internet of Things (BIoT) was established.
being used interchangeably with smart building [5], but author BIoT can be defined as, the overlaying of an IP network,
of this publication would like to distinguish those two terms. connecting all the building services monitoring, analyzing
Smart buildings seems to be new, enhanced version of intelli- and controlling without the intervention of humans [7]. The
gent buildings, which provide wider integration with utilities potential high-level benefits deriving from the implementation
and city infrastructure (i.e. to realize smart grid/smart city of IoT for buildings seems to be demonstrable. BIoT is
idea) and may use sophisticated algorithms such as machine seen as an important tool that will ultimately improve the
learning or AI for advanced control and diagnostics. Smart efficiency of building automation systems and may solve one
buildings utilize also the Internet of Things (IoT) concept and of the major problem of intelligent buildings implementation
involves wider range of equipment including household goods - interoperability.
or multimedia and mobile devices. Currently, everybody recognize the potential of IoT. Signifi-
III. C OMMUNICATION IN SMART BUILDINGS - CURRENT cant funding is spend on IoT related R&D activities involving
STATUS
new players (i.e. IoT startups), but the manufacturers are very
often struggling to make a concrete business cases for im-
Like in Tower of Babel there are many different com- plementation. Deep understanding of user’s needs is required
munication protocols in field of smart buildings (Figure 3). to succeed. In addition, problems of interoperability, privacy
Over the years, proprietary solutions has been replaced closed and security seems to be not fully solved. Almost half of
standards, nowadays we can observe trend toward open pro- the future consumers have concerns about the use of smart
tocols. Unfortunately, there is no one major standard. Due to technologies, because of inadequate data protection and cyber
broad range of applications, it is difficult to establish common intrusion [8]. Consequently, the focus of the industry has been
communication standard for low level communication. Smart on manufacturing and producing the right types of hardware
building equipment need to deal with various constraints such to enable those solutions in the future. More and more devices
as energy usage, high level of security, limited computation installed in the building has capabilities to become edge
power and many others. In addition, devices installed in a devices for connection to the cloud. At the moment, those
building often comes from different industries each promoting devices are used to collect data locally and provide access to
own standard (lighting - DALI, metering - mBus, HVAC aggregated data. Very often this data is not only accessible for
- Opentherm/BACnet, industrial equipment - Modbus/OPC). BMS via standardized protocols (BACnet, Modbus, OPC), but
Separate category are the wireless protocols - again, there is also directly for the user via embedded web page. Due to the
no dominant standard in this field, but to indicate a few major, emerge of on client-side web site rendering techniques such
one can mention ZigBee, Bluetooth LE, Z-Wave or EnOcean as AngularJS or React, it is possible to provide sophisticated
which is dedicated for batteryless solutions. A lack of collabo- data visualization with relatively not powerful server device.
ration between the subsectors of the building industry prolongs
150
CLOUD
Long term maintenance
System operation
ABB Ability
Advanced analytics
User/Operator Control logic System
and control
Azure Predix
a) FOG b) CLOUD c) INTERCLOUD
High level control
Set points Control signals Actuators
and diagnostics
Fig. 4. IoT for smart buildings development stages
Such pre-cloud solution to some extend is covered by the term Fig. 5. Control, Operation and maintenance loops
"fog computing" (Figure 4a). Data is stored and analysed at
the edge of the network. This reduces problem of the privacy
and security. At the same time, it allows in the future to fast multiple open standardization bodies, but on the other hand,
implement communication to the cloud and offer new cloud they are interested in securing their market share. For that
based services to the user (Figure 4b). reason, it is not sure that fully open standards will become a
Current approach, in many cases, is to use the cloud mainly norm in IoT. In case of smart phone revolution, the market
for advanced control and diagnostics (Figure 5). As it was of OS for mobile phones has been dominated by Android
in the traditional building automation system, the core logic and iOS. Although, Android is currently an open standard,
is distributed locally. Building data is fed back to influence middleware and application layers (i.e. Google applications)
actions within minimal latency. Sensors, controllers and ac- use proprietary specifications - similar situation can occur in
tuators are in close proximity, so that any adverse shift in IoT [11].
output parameters can be immediately mitigated. Building One of key success factors which decided about global
users/operators, base on operational data, control system’s set- acceptance of iOS and Android was allowing users to de-
points to achieve the best comfort and overall efficiency. The velop competing software and monetize their ideas. Similarly,
cloud, is additional level of control loop, which looks at higher Microsoft with its Azure cloud platform provides vast set of
level patterns and correlations [9]. The output can be used by tools enabling enterprises to build sophisticated IoT solutions
building managers to make over-arching decisions relating to quickly and efficiently. The price one have to pay for this is
operations and maintenance or, providing that latency require- dependency on Microsoft API, limited customization options
ments allows that, for automatic control purpose. and cloud service costs.
The age of propritery solutions on a system level is largely Many traditional automation manufacturers, to some extend,
history. But what about the cloud level? Unfortunately, current use a cloud-based platform compatible with/on top of Mi-
approach of most IoT manufacturers is to build whole solution, crosoft Azure (ABB - Ability, Siemens - MindSphare, GE
including their own cloud, from the scratch. As a result, there - Predix). Those clouds offer a broad range of analytics and
is a lack of consistency and standards across the cloud services services. While Azure market share is growing, it is not the
used by the different IoT solutions [10]. While it is easy to only this type platform on the market - Google (Google Cloud
integrate the data within the scope of ecosystem provided by IoT) and Amazon (AWS IoT), IBM (Watson IoT) provides
the same vendor, it might be difficult to share data between similar solutions. In general, services such as machine learn-
clouds or collect data from multi-vendor devices and systems ing, cloud storage, voice recognition delivered by Microsoft,
into one cloud. If standardization will not proceed, the problem Google, Amazon or Intel are used as a new set of tools which
will only increase as we move toward greater IoT adoption. were not available before. The competitive edge lies in how
In the end, we may end-up with the same problem of data those tools are used [12]. It is important to indicate that all
silos which plagued in-house network, but this time at the the major cloud platforms, to some extend, support MQTT
cloud level. While, the multi-cloud environment seems to be as the communication protocol. For this reason, if one would
inevitable, it is necessary to develop standards to allow the like to find common standard for IoT to cloud communication,
inter-cloud communication (Figure 4c). MQTT would be the first answer.
151
Set-top boxes 1.16%
0.92% Imaging equipments
the current TRMS at the branch. With this information, system
External power supplies 1.94%
Air-conditioning 2.12% 1.24% Office appliances stand-by can calculate power for each branch and early detect any
Dishwashers 2.69% 2.84% Air-conditioning deviation in power consumption. The data can be accessed
3.42% Computers
Computers 2.75%
4.73% Street lighting
4
via standard protocols - Modbus RTU/TCP and SNMP or in
Ventilation 2.75%
Home appliances stand-by 5.37% 5.26% Cooking appliances aggregated, visual form, through the built-in web page. What is
Washing machines 6.37% 5.92% Pumps more, it is possible to integrate the system with ABB AbilityTM
6.26% Miscellaneous
6
Television 6.74%
6
6.84% Circulators (and other
cloud to store information and drive efficiency for the long
Electric ovens & hobs 7.49%
Others 7.57%
similar) term with advanced analytics.
8.66% Commercial refrigeration
8
Water heaters 8.59%
12.62% Ventilation
Electricity is the largest utility expense for most of the
Lighting 10.49%
19.73% Space and water heating buildings, but of course not the only one. To comprehensively
Cold appliances (refrigerators &
freezers) 15.24% 21.57% Office lighting solve the problem of realtime utility usage monitoring ABB
Heating/electric boilers 18.73% has developed EQmatic system [18]. This integrated solution
Residential - 800,72 TWh Tertiary sector - 760,43 TWh is capable of collecting data from all utility meters installed in
the building. For collecting the data from the metering devices
Fig. 6. Breakdown of residential and tertiary sector‡ electricity consumption
M-Bus protocol is used. M-Bus is a European standard and
in EU-27 in 2007 [16] widely accepted solution for Automatic Meter Reading (AMR)
therefore EQmatic allows to integrate multi-vendor devices.
In case the particular meter does not support M-Bus, it is
KNX based building automation systems, but also one of the possible to use Modbus instead or install additional M-Bus
largest providers of electricity grid equipment, key player in impulse adapter. The entire user interface is fully web based
providing industrial automation and power conversion systems and accessible through the browser. Each of registered users
(EV chargers, PV inverters). Integration of devices installed can create own configurable dashboards to quickly access the
both inside the building and the external one (i.e. electrical required data. Besides the instantaneous values, the device
grid) requires cooperation of all stockholders, but it seems to provides several analysis functions such as historical data
be essential to fully unlock potential of smart buildings - both analysis, cost analysis or benchmark functions. All those
in terms of cost savings and reducing building environmental features make building energy flows and cost transparent.
impact. With a broad portfolio, it is easier to provide compre- They also can help building owners to implement energy
hensive solutions, which later very often determine standards management systems and put in place standards such as ISO
in the field. 50001 or ISO 14001. In addition, the device provides various
VII. E NERGY & COST EFFICIENCY export functions for further processing of the data.
Heating, ventilation and air conditioning (HVAC) is re-
Increasing price of oil, gas and electicy are the main driver
sponsible for a significant proportion of total building energy
conducive to reduce energy usage, but energy efficiency is not
consumption (see Figure 6) therefore running it properly and
only about cost cutting. Buildings are responsible for one third
at peak efficiency is the first step towards reduction of overall
of total global energy consumption and an equally important
building energy usage. Implementation of efficient HVAC
share of CO2 [13]. Global climate change is forcing us to
system requires few components: hardware which allows fine-
reduce energy consumption to save the planet.
grained control over the HVAC system, sensors for mon-
Key to energy savings is energy consumption awareness.
itoring various physical attributes (occupancy/door&window
Householders can find it difficult to reduce their domestic
contacts/environmental conditions/energy usage) and logic to
electricity consumption because the energy cost is billed in
maintain optimal thermal comfort and indoor air quality using
aggregate, at long intervals and retrospectively [14]. Moreover,
minimal energy. In intelligent buildings HVAC installation is
energy meters are installed at high level of the circuit, for that
tightly coupled with the rest of the system. The integration
reason it is not possible to verify electricity consumption of
can be done both on low level incorporating individual HVAC
particular electric appliances. Providing feedback of usage that
actuators to smart building backbone such as KNX and/or on
disaggregated by sub-appliances lead to more efficient energy
higher level - bridging the subsystems.
usage and overall reduction in consumption [15]. Of course, it
The IoT revolution is also influencing HVAC systems. IoT
is easy to find the general statistics such as Figure 6, but the
smart thermostats and environmental sensors can be easier
ability to see in-house consumption has a much greater impact
deployed and integrated into system. This makes system more
on one’s behaviour. In order to address this requirement,
flexible and allows to better adjusts space conditions based
ABB provides Circuit Monitoring Systems (CMS). The system
on occupant feedback. In addition, new sensing capabilities
consists of control unit and sensors (Figure 7). The control unit
triggers the implementation of more complex control logic and
CMS-700 allows to measure the currents, voltages and power
allows holistic system approach that accounts interaction of all
at the connection to the grid [17]. In addition, each circuit
subsystems in the building - i.e. with dynamic precooling or
breaker can be equipped with current sensor which measure
preheating system can reduce energy usage during consump-
‡ Tertiary sector refers to the public sector, education, healthcare, services tion peaks, adopt to occupants daily habits and optimize usage
and commerce. of renewable energy.
152
- base on movement detection. The key aspect in this area
is integration of light systems such as DALI, Philips Hue or
Osram Lightify with building automation system [22] [23] and
further on with the cloud.
Noise level Although the acoustical environment in the build-
ing is mainly result of design, smart buildings control systems,
can play some role in the providing acoustic comfort in
the building. The source of the noise in the building can
Fig. 7. ABB Circuit Monitor System be internal or external. In case of internal one, intelligent
control of HVAC, central vacuum, shatters or other noisy
building equipment can reduce noise nuisance. As for the
VIII. C OMFORT external sounds, windows and glazing, normally used to allow
Smart building implementation is not only about cost saving daylight to enter the space and reject heat, can be also used to
and environmental impact. It also involves using all the build- protect from external noise. Another option to address acoustic
ing automation to improve the overall user experience. The issues is to use sound masking systems. Sound masking using
system sense the internal and external conditions and responds engineered background sounds distributed via loudspeakers in
in the way users could feel safe and comfortable. The comfort the ceiling to reduce impact of distracting sounds [24].
metrics can be characterised in 3 main metrics [19]: Indoor air quality Indoor air quality management is rela-
Thermal comfort There are several factors which impact tively new topic. Studies shown that indoor air pollution levels
overall thermal well-being. Air temperature and velocity, hu- are often many times higher than outdoor [25]. Many of the
midity, mean radiating temperature effect in complicated way indoor pollutants came from the structure of the building itself
how we perceive temperature. Having in mind that the system (asbestos, radon, formaldehyde) so they can not be eliminated
has inertia and each user has its own preferences it is not easy by the building control system, therefore building construction
to determine the algorithm for automatic control. or retrofit phases are crucial to ensure safety of the occupants
The Nest Learning Thermostat is innovative, IoT solution, in the future. The most important task of building automation
for this problem. The control is based on user’s feedback. is to ensure that the air is circulated effectively in all parts
Temperature can be set manually via local HMI or through a of the building. Inadequate ventilation of a building impairs
mobile device and the cloud. The device learns the occupant’s the quality of the room air. Important factor which influence
habits and adjust the control accordingly. After a bit of time, it perception of indoor air quality is CO2 concentration. A
knows when to turn on heating and cooling or when nobody high CO2 level in the air influences badly the well-being
is at home and system can be put to energy saving mode. of occupants, therefore more and more building automation
Further more, the system can be integrated with the local systems integrates CO2 sensors [26] to include it into HVAC
energy provider in the way, local utility company can reduce control loop. To reduce energy consumption while maintain-
the load during consumption peak, offering discounts on the ing adequate air quality, demand controlled ventilation can
bill in return [20]. In addition, Nest provides comprehensive be implemented. Instead of fixed air replacement rate, CO2
API for further integration with other systems. sensors can be used to adjust ventilation throughput to actual
Visual comfort Providing visual comfort is also a complex emissions of building occupants.
issue with a lot of influencing factors. In general, it depends While discussing comfort in smart buildings, it is important to
on having appropriate amount of light for the performed task. include safety and security. Traditionally, methods of linking
Good visual comfort depends also on light source and light security systems with smart home installation were limited.
distribution. Direct lighting tends to create more shadows and Security installation was isolated from the other systems to
contrast and is more harsh than indirect lighting, therfore light meet tough reliability and security requirements. Despite the
colored walls and cailings and indirect light may improve difficulties, the integration of data from security system into
visual comfort [21]. Also the light color spectrum influences smart building seems to be required. Sensors in the security
our perception of the environment. In general, artificial light system offer entirely new possibilities for data collection.
spectrum should be as close as possible to that of sunlight, Not only does this increase the capabilities of the smart
which varies on time of the day and year and influences human building logic, but also allows savings related to installation
circadian system. All those factors are challenging not only for cost (less wiring, no sensors duplication). Typically, such an
interior designer, but also for smart building lighting system. integrated safety systems are working in 2 modes. During
First of all, it should give the users ultimate control over normal security systems support the logic of the intelligent
their own visual needs by providing the required light sources, building. For example, door and window sensors, together
second it should be equipped with required sensors and logic with the motion sensors integrated into security system can
to customize the internal lighting to the user’s needs. Current help in effective control of light or heating. In second mode,
solutions concentrate mainly on providing individual light in case of emergency situation, safety system can use all the
switch and dim as well as on calling up predefined dimming devices installed in the intelligent building to react properly
values and light scenes both via local HMI and automatically to the threat. For example, alarm of gas detector can result
153
Magnet
read Glass-break Gas Smoke Water Motion
Although, the idea IoT for smart buildings seems to be
contact sensor detector detector detector detector
Motion
promising, we are just on the beginning of the road towards
detector for
security bus
Setting
device
its broad implementation. The technology is already available,
Zone
bus
but security and privacy issues, which were neglected in this
module
Signalling
device paper, need to be solved first. Also IoT manufacturers need to
Security bus get more customer acceptance for new solutions.
KNX bus
154