On Defining Information and Communication Technology Requirements and Associated Challenges for ‘Energy and Comfort Active’ Buildings

Abstract

The intention of this article is to highlight considerations and ensuing challenges encountered in attempts to define communication requirements for the proposed multi-agent based building energy management systems. This is within the framework of ‘TKI-Smart Grid BEMS’ project which aims at developing new generation intelligent Building Energy Management Systems having capacity to interact with the utility power systems distribution network. The article identifies the development of comfort and energy active buildings as key to deriving maximum benefits from electrical smart grids for the built environment. These buildings require well specified information and communication technology for operational success. The paper is based on critical literature review. This is followed by a discussion on the challenges associated with specifying ICT infrastructure for multi-agents systems-based energy and comfort active buildings.

Full text

P r o c e d i a C o m p u t e r S c i e n c e 3 2 ( 2 0 1 4 ) 9 7 9 – 9 8 4
Avai lab le on li ne at www.s cie nc edire ct .com
1877-0509 © 2014 Published by Elsevier B.V. Open access under CC BY-NC-ND license.
Selection and Peer-review under responsibility of the Program Chairs.
doi: 10.1016/j.procs.2014.05.521
ScienceDirect
The 4th International Symposium on Frontiers in Ambient and Mobile Systems (FAMS-2014)
On defining information and communication technology
requirements and associated challenges for ‘energy and comfort
active’ buildings
Kennedy O. Aduda
*
, Wim Zeiler, Gert Boxem, Timilehil Labeodan
†
Department of Built Environment
Eindhoven University of Technology
Netherlands
Abstract
The intention of this article is to highlight considerations and ensuing challenges encountered in attempts to define
communication requirements for the proposed multi-agent based building energy management systems. This is within the
framework of ‘TKI-Smart Grid BEMS’ project which aims at developing new generation intelligent Building Energy
Management Systems having capacity to interact with the utility power systems distribution network. The article identifies the
development of comfort and energy active buildings as key to deriving maximum benefits from electrical smart grids for the built
environment. These buildings require well specified information and communication technology for operational success. The
paper is based on critical literature review. This is followed by a discussion on the challenges associated with specifying ICT
infrastructure for multi-agents systems-based energy and comfort active buildings.
© 2014 The Authors. Published by Elsevier B.V.
Selection and peer-review under responsibility of Elhadi M. Shakshuki.
Keywords: ICT, s mart grids , buildings, multi-agents system
* Corresponding author. Tel.: +31 40 247 2039
E-mail address: k.o.aduda@tue.nl
© 2014 Published by Elsevier B.V. Open access under CC BY-NC-ND license.
Selection and Peer-review under responsibility of the Program Chairs.

980 Kennedy O. Aduda et al. / Procedia Computer Science 32 ( 2014 ) 979 – 984
1. Introduction
The now common Net Zero Energy Buildings (NZEB) policy has consequently introduced the need to integrate
building level energy generations to power grids [1]. This has in turn led to emergence of ‘smart electrical grid’
characterised by multi-directional flow of power at low voltage levels and requirement for an elaborate informational
exchange between stakeholders and components. Smart electrical grids is defined as ‘electricity network that can
cost-efficiently integrate the behavior and actions of all users connected to it in efforts towards economic efficiency,
energy sustainability and better service delivery to end users’ [2]. Implications of smart grid to buildings are: (i)
flexible operations for energy advantage, and (ii) dynamic exchange of information by component subsystems and
actors. Informational exchange includes: energy generation and consumption profiles, occupants’ comfort profiles
and preferences, building behavior, market behavior and activity flows for different environmental scenarios. This
leads to a key term used in this paper: ‘energy and comfort active buildings’; this term is used to describe buildings
that proactively use indoor comfort requirements to define limits of dynamic interactions with electrical grid.
General guidelines on indoor comfort requirements for buildings are often described in terms of thermal comfort,
indoor air quality and visual comfort [3-5]. Operations in electrical smart grids involve interactive coordination
amongst multiple actors, processes and devices, this requires agile and robust controls. Some scholars propose multi-
agent systems (MAS) for interactions between buildings and electrical smart grids [6, 7]; this is also important in
enabling robust user interactions and real time decision making in buildings [16]. However, these have implications
in terms of communication requirement such as need for greater robustness, proactivity and real time dynamics [8].
Little attention has been given towards addressing these new developments. This paper outlines the requirements and
challenges associated with defining ICT for multi-agent based energy and comfort active buildings. Discussions are
further divided into the following subsections: ICT requirements; the challenges, illustrative cases and conclusion.
2. ICT requirements
Multi-agent based energy and comfort active buildings are essentially knowledge based systems whose operations
are characterised by informational flows illustrated in Figure 1. Subsequently their ICT requirements can be
categorised as unique to the 4 stages of informational flows (see Figure 1).
Fig. 1. Figure 1: Informational flows for energy & comfort active buildings in a smart grid.
Further details on the ICT requirements in relation to Figure 1are as discussed below.
Use level communication (stage 1)
This is stage connects the real world with the Building Management System. Information communicated at this
stage is modal, singular in objective and is delivered as a signal. There are two categories of communications at this
IAN/NAN/FAN
concentrator
/ level 1
t
WAN /Level 2
aggregator
Grid Control Centre
BUILDING domain illustrations are with black outline, those for UTILITY side domain are in red outlines.
ICT
PARAMETERS
Data rates<50 kbps
Latency<1 s
Range >10000m
BUILDING SIDE
COMMUNICATIONS
COMMUNICATIONS
BUILDING SIDE
`
UTILITY GRID SIDE
COMMUNICATION

981
Kennedy O. Aduda et al. / Procedia Computer Science 32 ( 2014 ) 979 – 984
stage. The first denoted as ‘1a’ in the Figure 1 deals with communication between occupants, appliances or building
equipment to the sensors or actuators. The second category (denoted as 1b in the Figure 1) transmits collected
information from real word to the Building Management System and vice versa. The performance parameters
desired at this level are: latency of less than a minute, data transfer rate of less than 100 Mbps and a coverage of
100 m. Technologies used could be wired or wireless ones; wireless technologies are preferred at this level due to
ability for re-use, cost effectiveness, high flexibility and elimination of cable trunking which is at times a nuisance
[9-11]. The content of information communicated at this stage include: environmental conditions at room and other
zone levels, user requirements, user behaviour, user preference and aggregated energy requirements at all levels of
operations.
Building Management Level Communication (stage 2)
The BMS exists to improve indoor comfort whilst also reducing costs through prudent energy management
strategies during building operations [12, 13]. In order to achieve this a web based BMS undertakes load profiling,
forecast energy demand, undertake preliminary energy consumption analysis, undertake automated control and
benchmark with other systems.This requires a database system, intelligent system and connectivity infrastructure.
Connectivity is enabled by:
x application based communication protocols such as LonWorks and BACnet which poll sensor based
data from the real world to the database and relays action oriented instructions to the real world.
x standard internet based protocols.
The BMS is equipped with a communication middleware to mediate in interactions between different protocols.
The ICT performance parameters for this level of operation remain similar to those at stage 1 except for latency
which should be to the level desired for onwards transmission to the smart meter. Also, information content
communicated at this stage are similar to those at building level with the addition to grid power systems information
and energy statuses and local energy sources.
Agents/Agent Platform-Utility Grid Level Communication (stage 3)
Services at this stage are data and intelligence based. Communication at this stage is based on standard internet
protocol. Two agent communications protocols may be used for building operations: ‘Foundation of Intelligent
Physical Agents-Agent Communication Language’(FIPA-ACL) and ‘Knowledge Query and Manipulation
Language’ (KQML) [14]. FIPA-ACL is based on the FIPA protocol and defines the environment and protocol in
which the agents acts and communicate [15]. In addition it specifies the agents management system with regards to
creation, deletion of agents and, access rights and privileges. KQML is designed en-suite with protocol to support
high level of communication amongst intelligent agents and having capacity to allow for interactions between
multiple intelligent systems [14, 15]. KQML defines the format of messages in such a way that agents are
recognized, data channels are created and informational exchange occurs; this is done in 3 communication layers (
that is communication, message and content layers). The ICT requirements for this stage are similar to those on
stage 2. However, the multi-agents based Building Energy Management System (SG-BEMS) is equipped with a
middleware to decipher communication from the Smart Meter; this is often in radio frequency or GSM based
protocol. The content of information exchanged at this stage is similar to that at building management level of
communication.
Utility grid side communication, stage 4
These occur between the utility side and the smart meter. The smart meter interconnects the utility side (grid) to
the Customer and market layers are interconnected by the smart meter. The smart meter transforms metering
concept from a post consumption billing gadget to a comprehensive and dynamic information collection and
processing infrastructure collectively known as ‘automated metering infrastructure’ (AMI). On request or pre-
defined schedule, the AMI measures, saves and analyses energy consumption data received from an elaborate
communication system and metering devices [16]. The content of information exchanged at this stage is similar to
that at building management level of communication with addition of asset and utility cost management information.
Smart meter communicates information in protocols that are either GSM or radio frequency based to the agent; this
is then interpreted to standard internet based protocol. The information is then received and analysed by the agents
for building control.
3. The challenges Indications from existing cases

982 Kennedy O. Aduda et al. / Procedia Computer Science 32 ( 2014 ) 979 – 984
Functionally, performance parameters such as latency, data range, data transfer rate, flexibility, bandwidth,
availability and priority be of significance for ICT in buildings connected to electrical smart grids [9, 17], see
previous section. Closely related to this is the choice for a protocol. Table 2, indicates that technologically Dash 7,
ZigBee, and WLAN would be adequate for the ICT tasks demanded.
Table 1. Wireless ICT technology evaluation for at building level operations
Technology &
Performance
Benefits
Disadvantages
References
ZigBee
r: 250 kbps
R: 100 m
L: <minutes
Appropriate for enabling wireless networking between
devices connected to the grid; Robust nature make
them ideal for hostile environments prone to node
failures; Can be easily used with actuators and sensors
& are hence ideal for load and conditions monitoring;
Low powered & energy efficient.
Current buildings need retrofitting to
accommodate usage thus leading to high
initial cost; Limited by memory size and
range for large buildings.
[18-20]
WLAN
r: 2-600 kbps
R: up to100 m
L: <minutes
Ideal for shared spectrum and noisy; Supports all IP
based protocols; Achieves required security & data
authentication requirements during communication;
Effectively communicate in HAN as it accommodates
several devices to operate at the same time; Devices
are cost effective plug & play type.
Operates in an unlicensed ISM band
whose spectrum is crowded; Limited
bandwidth availability; Industrial grade
Wi-Fi largely unavailable; Highly
susceptible to interference.
[18-21]
Z-Wave
r: 40 kbps
R: up to 30 m
L: <minutes
1. Simple yet reliable; Appropriate for control lights
and appliances in a house setting
Data range limits it to residential houses
or very small office buildings..
[22]
DASH7
r: up to 200
kbps
R: 250 to
5000 m
L: <minutes
Appropriate for enabling wireless networking between
devices connected to the grid; Robust nature make
them ideal for hostile environments prone to node
failures; Can be easily used with actuators and sensors
& are hence ideal for load and conditions monitoring;
Low powered & energy efficient.
Current buildings need retrofitting to
accommodate usage thus leading to high
initial cost; Limited by memory size and
range for large buildings.
[18-21]
Acronyms used
L: latency; r:data transfer rate R: range or coverage
Dash 7 seems most preferable as it surpasses others in terms of the number of devices it can connect, data range and
data rate [21]. It is however noted that Dash 7 is yet to penetrate the market and ZigBee remains popular for building
level operations.
Broadly speaking, the challenges ICT for comfort and energy active buildings emanate either as a result of
relatively new age technological practice in the area or associated complex nature of operations; the former is
evidenced by use of numerous standards and protocols which are continually shifting [17, 18]. The complex
manifestations in the ICT system for comfort and energy active buildings occurs as a result of the use of numerous
devices across equally numerous operational systems and protocols. These imply differential requirements and
hence heterogeneity in technology, protocols and standards must be embraced [18-20]. Added to this is the fact that
there has been increasing move to realise real time data processing and real time optimisation [8]. This further
underlies additional difficulties in terms of requirement for rapid processing of enormous amount of data[18]. In our
discussion in section 2, it is noted that the information flow between the stages rely heavily the use of middleware
software; this may be a cause of concern taken due to associated proprietary nature and consequential high initial
costs. However, standardisation of operations at different stages of information flow remains problematic: for
example at the interface between the smart meter and the agent platform no guideline exists for
interpretation/conversion of grid data to agent usable data and vice versa. Existing guidelines such as 1) IEC/TR
62051: Electricity metering - Data exchange for meter reading, tariff and load control and 2) IEC/TR 62059:
Electricity metering equipment are non-committal on specifics and offer very little to a bold proposal. This is not to
mention that as at the end of 2013, no comprehensive standards for smart metering for buildings existed [23]. Also
requiring attention are issues of security and reliability requirements for building operations which may become an
issue with increased traffic and increased cyber security scares. This may render present technologies very costly.
4. Comments from illustrative cases

983
Kennedy O. Aduda et al. / Procedia Computer Science 32 ( 2014 ) 979 – 984
It is important to use some illustrative cases to highlight further challenges for energy and comfort active buildings.
We discuss two illustrative cases that highlight some practical challenges associated with defining multi-agents
based comfort and energy active buildings.
Case 1: ‘HOMEBOTS’ system field test [24]
HOMEBOTS is an agent-based energy management services for buildings using power line
communication (PLC). In relation to this a series of tests was were performed in energy distribution area in
the South-East of Sweden to evaluate real-time requirements for agent communication. The system
connected various components a total of 28 electrical radiators able to communicate with each other over
power line communication using LonTalk. The purpose of communication was to ensure market based
direct load management. Communication test results indicates the following: (1) average latency of 1.29
seconds, (2) message lengths ranged from 10-160 bits, (3) response times for messages were from 0.54-
0.99 seconds and (4) 13201 messages were successfully exchanged. PLC proved adequate for the task.
Case 2: ABB-Zurich [25]
The ABB experiment was meant to offer a proof of concept for design of a user centric Building
Automation System based on multi-agents systems. The idea was to improve interactions with users and
actualise dynamic control reconfigurations for building systems. The system used utilised KNX operation
protocol for the building management system and ZigBee for the sensor and actuator network. Inter-
process communication was XML based. Key results indicated that using a common communication bus
by multiple devices led to congestion of messages and hence message delay. Also, multi-agents system
need to operate on servers external to the BMS infrastructure to ensure continued basic functionality during
server downtimes.
Two main lessons from above cases are: (1) simple architectural frameworks with target specific devices connection
such as the case with HOMEBOTS [24] has higher chances of success than generalised systems that are associated
with technological capabilities such as in ABB-Zurich [25], (2) use of multiple communication buses and servers
could decrease communication latency and by default increase response times. Few practical instances are reported
for multiagent system in comfort and energy management in office buildings [26, 27, 28]; few practical learning
opportunities therefore exist. However studies from other fields suggest that agent to agent communication in
comfort and energy active buildings may be complicated by the following [14, 29]:
x existence multiple intelligent systems, for example a grid based control system, comfort system and
building energy management systems. This requires communication specification in a manner that
allows interoperability amongst agents,
x elaborate protocols for dialogue, negotiation and competition amongst agents to increase reliability,
x administration of communication and intelligent processes such as modalities of handling
inconsistencies and mismatches ensuing from different world views and ontological framework is
necessary to guarantee timely and universal interpretation of the exchanged information.
5. Conclusions
We have defined the requirements for ICT in comfort and energy active buildings within the context of smart
grid operations and use of multi-agent systems. Little practical evidence exist for similar scenarios thus emphasising
the need for practical studies to evaluate the performance of the ICT systems in energy and comfort active buildings.
We have also highlighted some key challenges that are encountered in ICT systems for comfort and energy active
systems. Central to these challenge is the heterogeneity in devices, technologies and systems for comfort and energy
active buildings. It is acknowledged that heterogeneity in protocols, standards and technologies is essential for
realisation of desired performances [18, 20]. Consequently open platforms, standards and protocols remain a key
concern. This need to be considerate of both the present and future needs of the building and electrical grids.
References
1. Li DH, Yang WL, Lam JC. Zero energy buildings and sustainable development implications-A review, Energy 2013; 54:1-10.

984 Kennedy O. Aduda et al. / Procedia Computer Science 32 ( 2014 ) 979 – 984
2. European Commission. Smart Grids: from Innovation to deployment, Communication from the Commission to the European Parliament. The
Council, the European Economic and Social Committee and the Committee of the Regions, COM(2011)202 final, 2011.
3. ANSI/ASHRAE Standard 55, Thermal Environmental Conditions for Human Occupancy, ASHRAE Inc., Atlanta, 2004.
4. ANSI/ASHRAE Standard 62-2001-Ventilation for Acceptable Indoor Air Quality, American Society of Heating, Refrigerating and Air-
Conditioning Engineers, 2001.
5. Mills E, Borg N. Trends in recommended illuminance levels: An international comparison. Journal of the Illuminating Engineering Society
1999; 8: 155-163
6. Pipattanasomporn, M., H. Feroze, and S. Rahman. Multi-agent systems in a distributed smart grid: Design and implementation. In Power
Systems Conference and Exposition, 2009 (PSCE'09), IEEE, 2009. p. 1-8.
7. Dounis AI, Christos C. Advanced control systems engineering for energy and comfort management in a building environment-A
review. Renewable and Sustainable Energy Reviews 2009; 13: 1246-1261.
8. Klein L, Kwak JY, Kavulya G, Jazizadeh F, Becerik-Gerbe r B, Varakantham P, Tambe M. Coordinating occupant behavior for building
energy and comfort management using multi-agent systems, Automation in Construction 2012; 22: 525-536.
9. Doh YM, Kim SJ, Hoe TW. A Trend Analysis of Smart Grid Technology: The Convergence of Electric Power Network and IT Technologies.
ETTRENDS 2009;24:74-86.
10. Kuzlu M, Manisa P. Assessment of communication technologies and network requirements for different smart grid applications. In 2013
IEEE PES Innovative Smart Grid Technologies (ISGT), IEEE, 2013. p. 1-6.
11. Al-Omar B, Al-Ali AR, Ahmed R, Landolsi T. Role of information and communication technologies in the smart grid, Journal of Emerging
Trends in Computing and Information Sciences 2012; 3: 707-716.
12. Figueiredo J, and José S. A SCADA system for energy management in intelligent buildings. Energy and Buildings 2011; 49: 85-98.
13. Han J, Jeong YK, Lee I. Efficient Building Energy Management System Based on Ontology, Inference Rules, and Simulation, International
Conference on Intelligent Building and Management, Proceedings of CSIST, IACSIT Press, Singapore, 2011; 5: 295-300
14. Berna-Koes M, Illah N, Katia S. Communication efficiency in multi-agent systems. Proceedings of ICRA'04. 2004 IEEE International
Conference on Robotics and Automation 2004;3:2129-2134.
15. Fang L, Panos JA. On communication requirements for multi-agent consensus seeking. Networked Embedded Sensing and Control. Springer
Berlin Heidelberg; 2006. p.53-67.
16. Siano, P. Demand response and smart grids-A survey. Renewable and Sustainable Energy Reviews 2014; 30: 461-478.
17. Jürgen Heiles, D1.3.1 Smart Grid Standardization Analysis Version 2.0 (February 2012), Nokia Siemens Networks. Helsinki, 2012.
18. Singh, A, Jyotsna B, and Debabrata D. Two tier communication architecture for smart meter. (COMSNETS), 2013 Fifth International
Conference on Communication Systems and Networks. IEEE, 2013.
19. W. Kastner, G. Neugschwandtner, S. Soucek, H. M. Newmann. Communication systems for building automation and control, Proceedings of
the IEEE 93 (2005): 1178-1203.
20. Khamphanchai W., M. Kuzlu, M. Pipattanasomporn, M. A smart distribution transformer management with multi-agent technologies, 2013
IEEE PES Innovative Smart Grid Technologies (ISGT), IEEE, 2013.
21. Ma JR, Chen HH, Huang YR, MengW. Smart grid communication: Its challenges and opportunities. IEEE Transactions on Smart Grids
2013; 1:1-11.
22. Haase J. Wireless Network Standards for Building Automation, Embedded Systems for Smart Appliances and Energy Management, Springer
New York; 2013. p. 53-65.
23. The Open Meter Consortium. Description of Current State of the Art Technologies and Protocols -General Overview of State of the Art
Technological Alternatives Task 2.1.0, 2009.
24. Ygge FJ, Akkermans M, Andersson A, Krejic M, Boertjes E. The HOMEBOTS system and field test: A multi-commodity market for
predictive power load management, Proceedings 4th Int. Conf. on the Practical Application of Intelligent Agents and Multi-Agent
Technology PAAM-99, 1999. p. 363-382.
25. Yu D, Ettore F, Hadeli H. An intelligent building that listens to your needs. In Proceedings of the 28th Annual ACM Symposium on Applied
Computing, ACM; 2013. p. 58-63
26. Kwak J, Pradeep V, Rajiv M, Yu-Han C, Milind T, Burcin B, and Wendy W. TESLA: An energy-saving agent that leverages schedule
flexibility. In Proceedings of the 2013 international conference on Autonomous agents and multi-agent systems. International Foundation for
Autonomous Agents and Multiagent Systems; 2013. p. 965-972
27. Clearwater SH, Bernardo A. H. Thermal markets for controlling building environments. Energy Engineering 1994; 91: 26-56.
28. Kok K, Warmer C, Kamphuis R. PowerMatcher: multiagent control in the electricity infrastructure. Proceedings AAMAS ’05, Int. conf. on
Autonomous Agents and Multiagent Systems, volume industry track, New York; 2005:75–82.
29. Finin, T, Don M, Rich F. An overview of KQML: A knowledge query and manipulation language. Technical report, Department of
Computer Science, University of Maryland Baltimore County, 1992.