24th International Conference on Electricity Distribution Glasgow, 12-15 June 2017

Paper 0061

A HOLISTIC NETWORK PLANNING APPROACH: ENHANCEMENT OF THE GRID

EXPANSION USING THE FLEXIBILITY OF NETWORK PARTICIPANTS

Lars JENDERNALIK Christoph ENGELS Johannes HIRY

Dominique GIAVARRA Univ. of Appl. Sciences Dortmund Chris KITTL
Westnetz GmbH [email protected] Christian REHTANZ
[email protected] TU Dortmund University
[email protected]

ABSTRACT
APPROACH
DSO’s are facing significant challenges in network
This contribution is integrating the previous approaches
planning due to the integration of DER, Smart Grid across all voltage levels in following steps:
technologies, E-Mobility, regulation and volatile market
conditions. In previous work it was shown (A) how DSO’s 1. Input: The provision of global market conditions is
are able to optimize planning of network assets in the complemented by regulatory rules. The expected
presence of high uncertainty [1] and (B) how to describe energy mixture in terms of installed capacity is
the interdependencies between all market participants in projected into the future. Weather and supply tasks
an agent-based fashion to derive forecasts of prices, describe a set of actual load and feeder situation in a
generation & demand [3]. Here the integration of these specific hour which have to be applied to the network
system.
solutions across all voltage levels by means of long-term
network planning is presented. Thereby smart grid 2. Analysis: The usage of the actual network in terms of
components extend the traditional grid planning process. capacity, condition (age) and behaviour of market
This extension merges time-series based network planning participants is analysed by means of agent-based-,
approaches with models of active and reactive innovative asset- and power-flow simulation across all voltage
network participants. This results in robust and future- levels using the input defined previously. The analysis
proofed network planning. leads to potential violations of the network capacities
or asset conditions in a so called hotspot analysis.
INTRODUCTION 3. Fitness Restoration: An optimal network adaptation
While there have been several publications in the past is proposed by optimization techniques to balance the
considering the optimal combination of asset strategy and resulting future grid between the options of
network planning for upper voltage levels under uncertain curtailment or structural change (e.g. the addition of
conditions [1,2], as well as different ways of generating (innovative) assets).
time series for grid planning, the presented approach
creates added value by merging these two fields to improve 4. Systems Results: As an outcome the network planner
the electricity grid planning process. is able to prioritize actions due to the probabilities of
scenarios. Furthermore, he is able to compare the real
Previous work proposed an optimised planning procedure network behaviour with the forecasted situations.
under consideration of assets' condition [1]. It provides
functional valid grid structures in terms of asset condition RESULTS
and asset capacity (induced by the load and feeder) by
means of replacement, enforcement and expansion of The Integration of the approaches has led to first results:
physical network elements. In contrast the dynamics of The resulting planning & forecasting system across the
market participants also allow the utilization of voltage levels is shown in Figure 1.
flexibilities in terms of demand side management,
adaptation of regulation frameworks, storage technologies A risk matrix (right-upper box) serves as a decision base
and market mechanisms. Earlier work lead to an agent for the prioritization of actions regarding the transformer
based simulation framework [3] capable of providing time between the high voltage and medium voltage level (black
series based grid utilization analysis. The combination of arrows): Actions in the lower-left quadrant must be
both approaches in the scope of “Agent.GridPlan” project executed, actions in the upper-right quadrant are unlikely
allows for an integrated assessment of structural and to be applied. The necessary grid condition information is
operational measures and thus will result in more realistic delivered by the underlying agent-based, distributed power
and optimal solutions for holistic target grid planning. flow calculation which provides technical information,
e.g. nodal voltage (right-lower box), as well as information
about the smart market and smart grid interactions.

CIRED 2017 1/5

 24th International Conference on Electricity Distribution Glasgow, 12-15 June 2017

Paper 0061

Figure 1: Scope of the integrated planning solution

SIMONA is facing similar challenges considering the

Findings on system’s state of the art performance and scalability as presented in [6]. This
The analysis stage (step 2) is based on the current state of challenge has to be addressed by innovative solutions to
the agent-based simulation tool for an optimal grid compute a large system in an acceptable time and with an
extension planning (SIMONA) developed among others acceptable computational effort.
by Technical University Dortmund. As described by [3] One possibility to reduce the computational effort and in
the current state of SIMONA focusses on the medium and consequence accelerate the simulation is identified in the
low voltage levels. way of modelling DG plants. A simplified extract of
SIMONA’s overall system concept is shown in Figure 2.
The agent based programming paradigm is designed to The DG agent communicates with a time agent, a market
divide a large and complex problem (like a detailed voltage agent and a weather agent (all not depicted in Figure 2).
level integrated time series based distribution grid The time agent announces the beginning of a time step and
simulation) into smaller problems, which can be modelled waits for the message of every grid agent, to ensure the
very easy. In particular it aims to represent individual goals load flow calculation of the given time step is finished. At
of distinct individual entities in an environment, where the the beginning of every time step the market agent informs
entities are able to alter their own goals and behaviours all other agents about the current wholesale market price
during a negotiation process with other individuals [4]. (spot market), whereas the weather agents broadcast
One example could be the low level modelling of every information about the most relevant weather measures
single distributed generation (DG) plant. As the simulation (solar irradiance, wind velocity and direction as well as
framework’s ability evolves to a voltage level integrated temperature). All the described communication steps are
time series based simulation from high to low voltage performed by sending or receiving JADE messages. For
level, the number of participants to be modelled and further information on the overall system concept please
therefore the computational effort raises. Already in this refer to [1], [7]-[8].
early stage of the “Agent.GridPlan” project it can be noted,
that, due to the use of the java agent developing framework The available level of detail of environmental data, among
JADE [5], there is a certain competition between others, justifies that every single DG plant inherits the geo
modelling simplicity and computational efficiency. position from its point of common coupling. In

CIRED 2017 2/5

 24th International Conference on Electricity Distribution Glasgow, 12-15 June 2017

Paper 0061

consequence every DG plant (e.g. wind, photovoltaic environment models every network node as a distinct
panels, etc.) connected to the same node has the same geo agent, whose main task is the determination of residual
location and therefore receives the same weather nodal powers for a distributed load flow calculation by
information. If every DG would be modelled as a distinct negotiations with the connected load and DG agents. It
agent, the weather agent would send several messages receives messages of the DG and load agents containing
containing the same information. In other words: no information about the current power infeed or
information surplus with more communicational effort. consumption and provides the latest nodal voltage
The increased message traffic would have a large impact magnitude, which could be used e.g. to adjust the reactive
on the run-time, if the SIMONA tool should be executed power infeed in dependency of the given nodal voltage
in parallel on distinct machines The interexchange over an (cf. [9]). With raising voltage levels the nominal apparent
added physical network would slow the overall power of DGs does also grow. Additionally special loads,
performance. Even if the tool is running on a single like industrial consumers, often have specific requirements
machine, the same effects as presented in [6] are to voltage quality and to reliability. However, for grids at
observable. MV level or higher it might happen, that such significant
consumers have a distinct point of common coupling, like
grid agent a WEC farm being connected to the grid via its own
secondary substation. From SIMONA’s modelling point
of view, this means that these node agents only need to
communicate to a single load or DG agent. In that case
there is no advantage in splitting up the different agent’s
functionalities. Here again a new aggregation stage is
introduced.
node agent
The SIMONA user is given the possibility to choose up to
which number of loads and DGs connected to a node their
functionalities should be taken over by the node agent (cf.
Figure 3). Those systems are further on called “offline
systems”, because they are no longer modelled as a distinct
agent. As of now the agents’ negotiations take place as an
DG agent load agent iterative process inside the node agent. By this the
computational effort is further reduced, which helps in
simulating a bigger amount of agents and thus enhances
SIMONA’s capabilities, which is now able to perform
Figure 2: Extract of SIMONA's orverall system concept time series based network state analysis for bigger grids
across several voltage levels.
To improve performance it is proposed to create only one
DG agent per node that aggregates the functionality of
several DG agents at the specific node. Hence, the new DG grid agent
agent receives time, weather and market information.
Following, it provides the sum of generated power to the
environment. A new feature of this agent concept is a
combination of intrinsic mathematical models,
representing different generation technologies (e.g. PV
and CHP plants), in a single agent implementation. With node agent
this aggregation of different plants and plant technologies
their potential individual market driven behaviour is not
aggregated. Only the interface to their environment is
bundled. This quantity reduction of agents, and therefore Figure 3: Concept of node agent with offline systems
the communicational effort, enables a distributed
execution of the multi agent simulation (MAS) on distinct First results on system integration
machines. Moreover it reduces the memory and CPU With the improved version of SIMONA, the analysis
usage on every client of the computation cluster and simulations to investigate the grid performance are
therefore accelerate the simulation time even if a executed. To do so, the input parameters of stage one are
simulation is running on a single machine. bundled to several future scenarios that should be
investigated. These scenarios and their parameters are
A second potential for performance enhancement is mainly based on different forecasts to ensure a range of
identified in the coupling between network nodes and likely future developments (e.g. [10], [11]). As a result of
network participants. The SIMONA simulation the simulation, different time-series of the grid utilization

CIRED 2017 3/5

 24th International Conference on Electricity Distribution Glasgow, 12-15 June 2017

Paper 0061

for each scenario for several years are generated. In the might occur in the investigated grid in the next years as
next step, these time series are taken to detect grid well as a possible solution to prevent them. Furthermore,
elements that will face critical situations in the future (e.g. information about the estimated costs, depending on the
violation of the (n-1)-criterion or an overload of a considered scenario and the different operational actions
transformer). Not only the usage of time series identifies and structural changes. These results will be accessible via
the critical situations, it also allows to determine the a user interface which allows analytical visualization of
number of hours a violation of operational requirements simulated data even in big grids with large volumes of
and their exact occurrence. The result of this so called measures (e.g. load factors) in a multidimensional fashion.
hotspot analysis are then used to optimize the existing grid. Technologically this is achieved by use of GeoJSON,
Openstreetmap, Java and a Model View View Model
As a third step of the proposed approach, a fitness (MVVM) concept in a four tier architecture. To ensure
restoration of the investigated grid is performed. To do analytical efficiency the database layer provides NewSQL
so, the determined hotspots (critical grid elements) are features like In-memory computing and columnar
investigated in detail and a problem analysis of the critical organization. A first mockup of a geographical
situation is performed. Then, a predefined catalogue of representation is shown in Figure 4.
measures is used to remove all critical situations. These
predefined actions are a user specific catalogue of
operational (e.g. curtailment of the feed of RES) as well as
structural (e.g. addition of (innovative) assets) possibilities
to improve fitness of the grid. They have to be defined at
the beginning of the optimization process.

The described optimization process not only restores the

fitness of the grid but also estimates the expected costs
resulting from the different operational and structural
actions. Hence, an optimization problem with different
restrictions (e.g. maximal costs allowed) is formulated and Figure 4: Blueprints of geographical HV-grid views
has to be solved.
The values of resulting measures are represented directly
As an outcome of the steps two and three, the network using appropriate colour schemes or are accessible in
planner gets a full overview over the different scenarios, dedicated views (e.g. load duration curves for individual
their impact on the grid utilization, possible issues that assets). A schematic overview on the described four stage
approach is presented in Figure 5.

Figure 5: Schematic overview of the four stage approach

CIRED 2017 4/5

 24th International Conference on Electricity Distribution Glasgow, 12-15 June 2017

Paper 0061

 consideration of different scenarios over multiple

CONCLUSION years,
 automatic optimization of the critical situations in
This paper presents an approach to enhance the
the grid via a pre-defined measure catalogue with
distribution grid planning process while taking into
operational and structural actions,
account multiple voltage levels. With the widely discussed
 estimated costs of the different actions depending
emerging volatility of grid loading and increasing (market
on the scenario and its occurrence which allows a
oriented) interaction of distribution grid participants, the
financial evaluation as well as
need for sophisticated, cross-voltage grid state analysis
arises.  a state-of-the-art and user-friendly interface.
Such a detailed analysis can be carried out by a time series
based multi agent simulation, which will produce a huge OUTLOOK
amount of data from which information have to be With the systems’ integration and functionality
extracted. To achieve an optimal distribution grid planning enhancement some major challenges arise. With emerging
process those resulting information have to be processed voltage levels the number of network participants has to be
further, to derive adequate measures. Those steps can taken into account. Thus the bidirectional interaction
hardly be performed in a manual way. between market-oriented load and feed in systems with a
wholesale market needs to be represented. The higher
The project “Agent.GridPlan” develops an approach for amount of individuals that have to be simulated also
assisting distribution grid planners by integrating the challenges computational and data efficiency. Here,
agent-based grid analysis and simulation tool SIMONA adequate solutions have to be found in order to keep the
and an optimized decision making process into an simulation time and computational requirements in an
interactive system. The tool is either be capable of deriving acceptable size. As already mentioned, the planning
adequate measures from sophisticated grid state analysis, engineer is challenged in analysing the gathered
as well as of giving a powerful visualisation module. information. Here a suitable solution for human machine
Hence planning engineers would be able to understand interaction has to be found.
given complex interdependencies and their resulting
consequences. ACKNOWLEDGEMENT
The developed approach enables network planners to This project is supported by the European Regional
improve their network planning process due to Development Fund (ERDF).
 multiple voltage level analysis in one step, See also http://www.agent-gridplan.net

REFERENCES

[1] L. Jendernalik, D. Giavarra, C. Engels, A. Maier; [7] J. Kays, C. Rehtanz, 2016, “Planning process for
“Resilient and comprehensive network planning is key to distribution grids based on flexibly generated time series
a successful integration of renewable energy resources”, considering RES, DSM and storages”, IET Generation,
2016, CIRED, Helsinki. Transmission & Distribution, vol. 10, no. 14, 3405-3412
[2] C. Engels, L. Jendernalik, J. von Haebler, S. [8] J. Kays, A. Seack, C. Rehtanz, “Consideration of
Schimmeyer, A. Maier, D. Giavarra, "Smart Planning – smart-meter measurements in a multi-agent simulation
Optimal Balance between Risk & Costs”, 23th CIRED Intl. environment for improving distribution grid planning”,
Conference, 2015, Lyon. 2016, Proceedings IEEE Power & Energy Society
[3] J. Kays, A. Seack, U. Häger, “The potential of using Innovative Smart Grid Technologies Conference (ISGT),
generated time series in the distribution grid planning 1-5
process”, 2015, CIRED, Lyon [9] A. Seack, J. Kays, C. Rehtanz, “ Time series based
[4] S. D. J. McArthur et al., "Multi-Agent Systems for distribution grid planning approach with decentralised
Power Engineering Applications—Part I: Concepts, voltage regulation“, 2014, Proceedings Power Systems
Approaches, and Technical Challenges", 2007, IEEE Computation Conference, Wroclaw
Transactions on Power Systems, vol. 22, no. 4, pp. 1743- [10] German Federal Ministry for Economic Affairs
1752 and Energy, “Entwicklung der Energiemärkte –
[5] “JADE v4.4.0.” [Online]. Available: Energiereferenzprognose”, Berlin, June 2014
http://jade.tilab.com/. [Accessed: 12-Dec-2016] [11] 50Hertz Transmission, Amprion, TenneT TSO,
[6] E. Cortese, F. Quarta, and G. Vitaglione, “Scalability TransnetBW, “Szenariorahmen für die
and Performance of JADE Message Transport System,” Netzentwicklungspläne Strom 2030 – Entwurf der
AAMAS Work. AgentCities, Bol., vol. 16, pp. 52–65, Übertragungsnetzbetreiber”, January 2016.
2002.

CIRED 2017 5/5