Field testing and demonstration of a smart grid ready charging park

Full text

CIRED workshop on E-mobility and power distribution systems Porto, 2-3 June 2022
Paper n° 1218
CIRED 2022 Workshop 1/5
FIELD TESTING AND DEMONSTRATION
OF A SMART GRID READY CHARGING PARK
Christoph KONDZIALKA Chen SHUO Leonie SCHMIDT
Ulm University of Applied Sciences Ulm University of Applied Sciences Ulm University of Applied Sciences
Germany Germany Germany
christoph.kondzialka@thu.de chen.shuo@thu.de leonie.schmidt@thu.de
Basem IDLBI Rouven TAUBMANN Heiko LORENZ
Ulm University of Applied Sciences Ulm University of Applied Sciences Ulm University of Applied Sciences
Germany Germany Germany
basem.idlbi@thu.de taubro01@thu.de heiko.lorenz@thu.de
Gerd HEILSCHER
Ulm University of Applied Sciences
Germany
gerd.heilscher@thu.de
ABSTRACT
This paper presents the results of the research project
"controlled charging cells", which was conducted in the
period from 10/2019 to 0672021 and funded by the
Ministry of the Environment of the State of Baden-
Württemberg. During the project, a public charging park
was set up in a municipal parking garage consisting of 16
charging points. This demonstration test facility was
integrated into the operating concept for charging
infrastructure of the municipal utility. The goal was to
create a blueprint for the intelligent grid connection of
parking garages, (company) parking lots and
underground garages (PPT) and thus to develop the basis
for a safe as well as cost-efficient grid integration of the e-
charging infrastructure into the distribution grid and also
to demonstrate it in a field test. Through the collaboration
of the municipal utility, the local distribution grid operator
and the Ulm University of Applied Science (THU), a safe
intervention option for specifying the maximum possible
charging power was implemented and demonstrated. A
smart meter infrastructure-based solution was used, as it
allows secure communication of a control command and
is also well suited for further rollout thanks to a high
degree of automated and standardized processes. For all
components, developed or used, attention was paid to the
use of e-mobility and energy network standards (OCPP,
IEC 61850) so that the solutions developed can be used
independently of individual manufacturers and are
transferable nationally and internationally. While ready-
to-use solutions are already available for all individual
problems, integration into an overall system that is usable
and tested for municipal services that operates PPT and
distribution grid operators is currently still missing.
INTRODUCTION
Problematically high energy consumption peaks due to the
charging infrastructure for e-vehicles occur in particular
when many charging columns are installed in a confined
space [1]. This is especially the case in parking garages on
a large scale. How these charging stations are frequented
has not yet been conclusively clarified and also depends to
a large extent on the surrounding area. Parking garages in
city centers, for example, are very busy during the opening
hours of the surrounding stores, but can also be relatively
underutilized for longer periods outside business hours. On
the other hand, company parking lots are very busy during
the day on weekdays, but are often virtually unused on
weekends and overnight. These examples illustrate the
range of factors that need to be taken into account when
planning and implementing a charging park. In addition,
the grid connection of the e-charging park and also
upstream grid resources can be overloaded if the
simultaneity factor is high. This can be countered by
expanding the network capacity, but high network
expansion costs are to be expected. The replacement of
network equipment (transformers, cables) is seen as a
classic network expansion measure. Alternatively,
intelligent control of the charging stations can ensure that
the power consumption does not exceed the maximum
capacity of the infrastructure. In the best case, this
succeeds without noticeable restrictions in the use of the e-
charging infrastructure. The state of the art here is local
charging management, which distributes the maximum
available power to the active charging points. If demand
exceeds the set limit, the power at individual (or all)
charging points is reduced according to a defined
distribution scheme. In this process, additionally recorded
user information is also particularly helpful, covering, for
example, the planned duration of use of the charging point

CIRED workshop on E-mobility and power distribution systems Porto, 2-3 June 2022
Paper n° 1218
CIRED 2022 Workshop 2/5
or the amount of energy to be charged. Communication
between the charging point and the user/vehicle is still
inadequate in this context, although Plug & Charge (ISO
15118) is a possible solution [2]. The cumulative effect of
several charging parks, which are operated in the same
segment of an electricity distribution network, is hardly
taken into account in current planning. Although statistical
methods can often prove that the probability of an overload
is extremely low, this worst-case scenario cannot be
completely ruled out. Therefore, the establishment of a
possibility for external power reduction by the responsible
network operator is currently a major topic in the
discussion about concrete measures for the expansion of
the e-charging infrastructure.
SMART GRIDS AND SMART METER
INFRASTRUCTURE
The expansion of renewable energies and their high level
of acceptance among the population, local authorities and
many companies is already posing challenges for many
power grids. The starting point for this development in
Germany is the smart meter infrastructure based on smart
meter gateways and controllable local systems (CLS)
introduced by the Act on the Digitization of the Energy
Transition [3]. The smart meter infrastructure is composed
of a digital electricity meter and a so-called smart meter
gateway, a communication unit. Smart meters enable
consumers and companies to better and more conveniently
manage their electricity consumption or the feed-in of their
electricity, for example from solar cells on the roof or e-
charging columns, and to benefit from new tariffs. The
security aspect, both in the system (energy supply) and for
the required ICT (information and communication
technology) of the smart grid components, plays a decisive
role in the success of the energy transition.
Fig. 1 System Architecture using the German Smart Meter
Infrastructure - control of the charging park using the Smart
Meter Infrastructure CLS channel (purple)
DISTRIBUTION NETWORK OPERATION
Decentralized energy systems have so far been integrated
into an energy system that is organized top-down. Under
the Renewable Energy Sources Act (EEG), special roles
were created for local energy systems and the direct use of
this energy, up to and including the solar tax on the local
use of energy. An important component for the flexible use
of the energy to be distributed locally is, among other
things, the charging infrastructure for electric vehicles. A
future distributed energy system embedded in a smart grid
offers completely new possibilities - both for operational
grid operation and for energy trading and billing models.
However, in order to operate its grid safely and efficiently
in the event of a high penetration of e-charging
infrastructure, the grid operator needs comprehensive
knowledge of the state of the power grid upstream of the
charging park. In addition, he needs the ability to make a
decision on power reduction at short notice, for which the
network status calculation in low temporal resolution (less
than 15 min) is a prerequisite. The current state of the art
here is a design calculation of the power grid for new
construction or conversion work, which takes into account
statistical specifications for the expected power and energy
flows. The possibility of controlling a charging park can
be implemented in Germany via the Energy Industry Act
§14a, but is difficult to apply in practice due to the lack of
grid status determination at the distribution grid level [4].
SOLUTION CONCEPT
As a framework condition, the number of 16 charging
points with a charging power of 22 kW each was specified.
For the entire charging park, 100 kW maximum total
power is reserved at the connection point of the
underground car park at the associated transformer. A
local load management system is used for power
distribution, which dynamically distributes the available
total power to the individual charging processes according
to a pre-selected algorithm and without further external
intervention. In practice, this means that from the 5th
charging process, which charges at full capacity (5 x 22
kW = 110 kW), the load management intervenes and
reduces the power according to the selected distribution
algorithm at the individual charging points.
The objective was to ensure the smoothest possible use of
the charging park by visitors to the parking garage, while
keeping the costs per charging point as low as possible. In
addition, an optimized use of installation space and
materials was already aimed for during the planning phase.
The result of the planning was a solution combining 4
wallboxes with a capacity of 22 kW and 6 with 2x 11 kW
(max 22 kW). These double wallboxes distribute the
available power to the two built-in charging points, with
single occupancy 22 kW are also available there. The
purpose here was to create an opportunity to investigate
the extent to which these different wallbox types can cover
the different user profiles.
The connection of such an energy system to the network
control center of a distribution network operator is critical
to safety. The main challenge here is the provision and
configuration of the various system instances involved
(smart meter gateway admin, CLS management backend,
experimental distribution network control center), so that
the connection of e-charging infrastructure using the

CIRED workshop on E-mobility and power distribution systems Porto, 2-3 June 2022
Paper n° 1218
CIRED 2022 Workshop 3/5
encrypted CLS channel of the intelligent metering system
(iMSys) could be investigated and tested [5].
In addition to the secure communication link via the smart
meter infrastructure, the connection of the systems within
the property must be ensured. In the project, a Restful API
of the central master unit of the charging park was used as
an interface to the local charging management.
Furthermore, a real-time simulation environment [source]
developed at THU is used. This was developed with the
goal of evaluating the feasibility of real-time state
estimation in existing low-voltage feeders of the test areas,
as grid cells, based on iMSys and CLS infrastructure. The
simulated grid cells and the predefined metering points
were modelled in the distribution grid simulation
environment within the distribution grid automation agent.
Based on the calculated grid state, the distribution grid
automation agent sends a grid congestion measure to the
grid control room when needed. The devices in the field
receive the request as control commands from the network
control room of the THU when the occurring congestion
situations need to be eliminated. The network control room
communicates these control commands as well as the
measurement data to the CLS gateways in the field via the
IEC 61850 protocol.
IMPLEMENTATION
An essential part of the project was the procurement and
installation of the charging infrastructure. Extensive
planning and preparatory work had to be carried out by the
municipal utility.
For the substructure, a curtain-type supporting structure
including an installation road was produced. A T-shaped
facing with an air space behind it was used. In addition,
there is a load-bearing hollow profile along the entire
length of the charging park, which can accommodate the
power cables as well as the communication cables.
For the electrical installation, existing cable routes could
be used for the most part, but suitable solutions had to be
found at neuralgic points such as cable routes that were too
narrow or fire compartment passages.
The distribution network operator installed a measuring
device on the associated local network transformer, which
can record both the power at the grid connection point of
the parking garage and the overall utilization of the
transformer.
For integration into the existing parking guidance
system/charging guidance system of the municipality,
LoRaWan parking sensors, user information displays and
the connection to the higher-level systems for
authentication and billing of charging processes were
implemented.
Fig. 2 Overview of the 16 charging points installed in the
demonstrator, implemented using 6 twin wall boxes and 4 single
wall boxes (picture ©Ulmer Parkbetriebs-GmbH)
Furthermore, a floor coating in green with a black border
was implemented, which at the same time delimits the
parking space as such, but also refers to the e-charging
infrastructure. A photorealistic pattern in leaf optics was
chosen as the graphic design in the form of a foil for the
curtained installation level, which is intended to visualize
the contribution of e-mobility to the reduction of CO2
emissions. This design is also to be used for the charging
stations to be built in the future in order to create a
recognition effect among users.
Parameterization of local controllers and setup of
interfaces
When it comes to power distribution within the charging
park, there is a wide range of options for implementation.
Simultaneous charging of many electric vehicles in a
parking garage or parking lot leads to various challenges.
This is mainly due to the limited power availability, which
is limited by the limited capacity of the existing lines and
transformers. In addition, each vehicle has different
charging requirements due to the different parking time,
battery charge level, and driving distance after parking. To
this end, THU has developed a simulation model of the
charging infrastructure for electric mobility. The focus was
on the simulation of a charging park controller (LPR),
which controls the interaction of several charging columns
and the interaction with the network. In the process, the
interfaces to the outside of the grid as well as to the inside
of the individual charging stations were also addressed.
One step in the implementation of a LPR, is the definition
of the interface between the charging station and the
charging station controller and between the charging
station controller and the control unit. In order to find the
right communication model for the interface between the
LPR and the CLS gateway, criteria have been developed
that should be met. The interface should be as standardized
as possible, take into account the specifications for
integration in the THU Smart Grid Lab, use open
communication interfaces for networked mobility services
and offer the possibility of real-time communication [6].

CIRED workshop on E-mobility and power distribution systems Porto, 2-3 June 2022
Paper n° 1218
CIRED 2022 Workshop 4/5
Control of the e-charging park
To implement the desired functionality, an adaptation of
the software on the CLS component was carried out. The
software serves as a protocol converter between the local
instance in the field and the experimental distribution
network control center at THU. The distribution network
control center uses IEC61850 for the telecontrol
connection. In the meantime, the IEC61850 is also used for
the connection of decentralized components in the low
voltage [7]. Thus, the software was adapted to address the
Restful API of the master unit of the charging park in order
to communicate the regulation command. Furthermore, the
software on the CLS gateway was extended to include the
ability to read measured values from the separately
installed measuring device. The software combines this
information and makes it available to the control system
via IEC61850 communication.
The system design and development of the IT systems at
THU includes in particular an experimental distribution
network control center, the smart meter gateway admin
and the EMT platform as SaaS as well as the CLS backend.
In addition, proprietary tools are used to support
parameterization of the CLS components and device
management, tools for parameterization of the control
system (automated creation of configuration and data
model), test systems for function verification as part of a
pre-deployment test, and a cross-system tool for
monitoring and alerting. For the implementation of
business and application logics, the system landscape was
supplemented by a state estimator for distribution
networks and a logic module for bottleneck elimination,
including a mockup for development support [8].
CONDUCT OF THE PILOT TEST
The pilot test at the campus of the THU consists of two
logical charging parks, each with a wallbox and a charging
management system. In addition, a three-phase external
socket was installed in order to be able to flexibly test the
products of other manufacturers.
The integration of the charging stations as well as the
charging management system was carried out with the help
of an open-source OCPP server [9] [10]. This software is
published as open source and is operated on a locally
installed embedded computer. This component is used for
the initial configuration and authentication of the charging
station users. To test the load limitation, a special interface
of the charging management system was used, which
works according to the Restful principle.
In this case, the measurement and control is acquired by a
CLS gateway and connected to the experimental
distribution network control room of the university via
iMsys and secured channel. The software of the CLS
gateway was extended for this purpose in order to be able
to send the command used for load limitation. The
measuring device can also be read out by the software
running on the CLS gateway.
Fig. 3 Extension of the Smart Grid Laboratory by two separately
controllable wallboxes, each of which is configured as a
charging park with an associated master unit
FIELD TESTING
The solution previously designed and already tested in the
pilot test is used for the loading park demonstrator.
Adjustments were still necessary due to a different
firmware status of the devices used. In addition, the CLS
application was extended to support the Janitza-UMG801
measuring device used on site. In order to test the
connection of the measuring device in advance, a
corresponding test setup was set up in the smart grid
laboratory. The effort for on-site integration of the CLS
gateway in the demonstrator could thus be limited to two
dates. As already tested in the pilot test, the demonstrator
also uses an SMGW with a mobile radio connection for the
communication link. The connection proved to be stable in
this application, even in practical use. On a positive note,
the solution now in use automatically re-establishes the
data connection after brief interruptions.
Fig. 4 Visualization of the charging park control by THU network
control system (controlled limitation of the charging park power
in green; actually used charging power in orange)

CIRED workshop on E-mobility and power distribution systems Porto, 2-3 June 2022
Paper n° 1218
CIRED 2022 Workshop 5/5
RESULTS
The demonstrator charging park and its upstream
distribution network were incorporated as a permanent test
facility in cooperation with the network operator.
Solutions for network status recording were developed in
order to gain experience for the controlled charging of e-
mobiles in PPT. This enables, for example, the
implementation of dynamic load management in the
parking garage based on the network condition of the
upstream local network. The evaluation of parking, user
and charging behaviour as well as customer feedback
regarding the controlled charging of e-mobiles has also
started. This evaluation will be continued in order to be
prepared in particular for the increasing spread of e-
vehicles. Particular attention was paid to developing a
standardized list of requirements for local load
management and interfaces to upstream control levels.
Furthermore, the focus was on establishing a scalable
charging solution with load management for both long-
term and short-term parkers in a parking garage of the
municipal parking garage operator together with the local
energy supplier in the parking garage. This includes in
particular the demonstration of a secure control via the
German smart meter infrastructure by distribution network
operators or market participants. For this purpose, it will
be shown how a maximum retrievable value for the
charging infrastructure can be specified and the ongoing
charging processes regulated by means of a secure
communication gateway.
OUTLOOK
Currently, the implementation effort for creating the
distribution grid simulation of a cell is still high due to the
complexity of the system design and the lack of metering
points in low-voltage grids. This complexity is mainly due
to the synchronization of multiple systems and data objects
on the grid control room and in the distribution grid
automation agent. However, these use cases will become
more and more important for future scenarios with a very
high share of distributed generators and will therefore be
further investigated and optimized. In the future, the
developed grid simulation in combination with the
determined communication approaches should lead to an
increase in the reliability of the grid state determination
with simultaneous optimization of the grid operation and
will be further evaluated.
The results of the work carried out will be used by
municipal services in the construction of the further
charging infrastructure. In a recently built parking garage,
30 charging points will be created with a reserve for
expansion to 100 charging points. Furthermore, the other
central public parking garages will then be successively
retrofitted with charging parks; here, a size of 20 charging
points will be implemented in the first expansion stage. In
addition, the results will also be available to the private
housing sector and companies. Here, there is still a need
for adaptation to the developed blueprint, as user
behaviour and the technical boundary conditions differ.
There is still a need for further research, particularly in the
development of operating strategies for a charging park as
part of a local energy management system, especially if
other energy technology components such as battery
storage or flexible consumption facilities such as industrial
plants are to be taken into account in the control system.
Acknowledgments
This research was co-funded by the federal state of Baden-
Württemberg (Germany), INPUT gLadeZellen grant
number BWINP 19007; Federal Ministry for Economic
Affairs and Climate Action (Germany), SINTEG C/sells
grant number BMWi 03SIN136; European Union's
Horizon 2020, SerendiPV grant number 953016
REFERENCES
[1] ‘Metastudie Forschungsüberblick Netzintegration
Elektromobilität’. FGH e.V., Dec. 2018.
[2] M. Mültin, ‘ISO 15118 as the Enabler of Vehicle-to-
Grid Applications’, in 2018 International Conference
of Electrical and Electronic Technologies for
Automotive, Jul. 2018, pp. 1–6. doi:
10.23919/EETA.2018.8493213.
[3] Bundestag (The Parliament of the Federal Republic of
Germany), Gesetz zur Digitalisierung der
Energiewende. 2016.
[4] Bundestag (The Parliament of the Federal Republic of
Germany), Energiewirtschaftsgesetz - EnWG. 2005.
[5] Bundestag (The Parliament of the Federal Republic of
Germany), Gesetz für den Ausbau erneuerbarer
Energien (Erneuerbare-Energien-Gesetz - EEG
2021). 2021. [Online]. Available:
https://www.gesetze-im-
internet.de/eeg_2014/inhalts_bersicht.html
[6] F. Ebe et al., ‘Comparison of Power Hardware-in-the-
Loop Approaches for the Testing of Smart Grid
Controls’, Energies, vol. 11, p. 3381, Dec. 2018, doi:
10.3390/en11123381.
[7] M. Wunsch, ‘Vorbereitung des Stromnetzes für die
Mobilitätswende’, MTZ - Motortechnische
Zeitschrift, vol. 81, no. 12, pp. 64–67, Dec. 2020, doi:
10.1007/s35146-020-0339-0.
[8] F. Ebe, J. Morris, S. Chen, B. Idlbi, D. Graeber, and
G. Heilscher, ‘Test and evaluate an automated low
voltage grid management system through utilization
of CLS-Gateways to control a decentralized energy
resource’, Jul. 2020.
[9] ‘OCPP 1.6’. Open Charge Alliance, 2015. [Online].
Available: https://www.openchargealliance.org
[10] ‘SteVe’. RWTH Aachen University, Dec. 19, 2021.
[Online]. Available: https://github.com/RWTH-i5-
IDSG/steve