EMS do not usually take into account local conditions and specificities - such as
individual preferences and behaviours, occupancy rates of buildings, rooms and of-
fices, dimensions of the control area, differences between the sensors, etc. As a con-
sequence, the local conditions in certain areas of a certain building may be out of the
comfort boundaries that the EMS is designed to control and optimize. Therefore, it is
important to add learning features to the EMS that would allow the building managers
to receive feedback from the users regarding the control actions that are being taken.
In this way, the building managers can adjust their control actions to answer better to
the users' needs, within the budget and regulation constraints.
Notwithstanding, building users are usually not fully aware of the impact of their
actions in terms of energy consumption; neither are they fully enabled to provide a
technical description about their comfort levels. In this context, the EMS shall not
only provide more tailor-made solutions to the users (according to their preferences),
but also help raising awareness about the consequences of their actions and guiding to
more resource-efficient and energy-saving attitudes.
The Bi-directional Learning Process thus, on one hand, enables EMS to more effi-
ciently respond to specific local conditions and particular user's behaviour and prefer-
ences. It thus intends to go beyond typical and standard parameters that the EMS is
designed to control and optimize.
On the other hand, this bi-directional nature intends to foster user behaviour trans-
formation. It means that the feedback provided by the EMS and the control actions
adopted - both accordingly to the user behaviour - are not exclusively oriented for
information purposes. They are also focused on raising the user's awareness about the
impact of their choices in terms of energy efficiency and costs, as well as on regula-
tions compliance, fostering change behaviour towards more energy efficient comfort
levels.
In order to trigger this bi-directional learning process, one fundamental milestone
to be achieved is the adequate provision of Real Time Information (RTI). The
SMART CAMPUS project will measure real time energy consumption in the pilot
campus areas and provide this RTI to the users. The information will be presented
and visualized in an easy-to-understand format in all the pilots, thus triggering the Bi-
directional Learning Process and fostering User Behaviour Transformation.
RTI is understood as a dataflow of some measured variable, shown to the user
relatively soon after the measurement. The term "real time" could have a comprehen-
sive scope - either meaning immediately, presently or a couple of days after the actual
consumption. Within SMART CAMPUS scope, "real time" will refer to the time
lapse of the immediately previous measuring unit. Thus if the energy consumption is
measured every hour, real time will mean the energy consumption undertaken in the
last hour; if the energy consumption is measured every minute, then real time will
refer to the energy consumption undertaken in the last minute. RTI will be provided
in a user-friendly manner, in order to most effectively inform the users, enabling a
fully-conscious decision making as well as triggering User Behaviour Transfor-
mation.
The use of Bi-direction Learning Processes in Energy Management Systems leads
to the creation of Intelligent Energy Management Systems (IEMS). Within the
SMART CAMPUS project, an IEMS is being developed for each of the pilots to be
run in each of the campuses involved - Helsinki, Lisbon, Luleå and Milano. These
19