WhatsNew 2020 en
WhatsNew 2020 en
WhatsNew 2020 en
What’s New
F2020
January 2020
r6222
Welcome to PowerFactory 2020!
With increasing demand for power system calculations and simulations, the requirements from tools
such as PowerFactory are growing. Robustness, flexibility and performance are key requirements
for maximising the benefits of these analyses. DIgSILENT addresses these needs in its constant
development of new functionality, whilst enhancing the existing functions and tools in order to provide
our customers with the best possible solutions. We are pleased to introduce the latest annual version of
PowerFactory, which as well as extra features and functions this year comes with a fresh modern look.
A new application framework has been introduced, which ensures that PowerFactory is future-proof
and enables new opportunities for continued development in the coming years. The new interfaces offer
users greater flexibility in customising the windows layout, by rearranging windows and toolbars and
using split-screen working in the graphics window. Coupled with this, we have harmonised the icons
thoughout the application and improved the overall clarity of dialogs and the Network Model Manager.
We hope that you enjoy working with all these new features, described in Section 1.
Besides the new framework, PowerFactory 2020 of course includes new developments in analysis
functions. All the main developments are listed in Section 2, but some of particular interest are:
• The Dynamic Network Reduction, which allows the user to reduce the size of a network to
a minimum, while keeping the results of dynamic simulations as accurate as possible. This
new tool can be used to anonymise the main part of their network, as well as to increase the
performance of individual dynamic simulations. In order to mimic the dynamic behaviour of the
reduced area as well as possible, the tool can be combined with the completely overhauled
Parameter Identification function.
• Our protection suite has been updated in order to fully support the new IEEE 1584-2018 standard
for Arc Flash Hazard Analysis in order to help our customers in complying with the latest stan-
dard. Furthermore, the Protection Coordination Assistant offers even more flexibility, allowing
users to create their own rules for zone-reach determination.
• To support our users with increasing numbers of connection requests, the latest standards VDE-
AR-N 4105:2018 and VDE-AR-N 4110:2018 have been implemented in PowerFactory, allowing a
fast and reliable connection request assessment.
• The Unit Commitment and Dispatch Optimisation module now also supports optimisation of
hydro units, batteries and general storage models.
Graphical Interface: For users creating and developing networks via graphics, the Drawing Toolbox
has been redesigned to make it more user-friendly and customisable. There are new plots available for
various functions, and an Area Interchange Diagram has been introduced. And if graphics are to be
printed or exported, users will find that these features have also been updated, to improve usability and
provide new options. See Section 3 to find out more.
Data handling has been improved with the introduction of additional options for using data from external
databases, which will particularly benefit customers who want to use such external data sources to
supply load and generator characteristics, or store calculation results. For details of this and other data
management developments, see Section 4.
As always, many of the Power Equipment Models have been further developed, and new models
introduced. Among the enhancements described in Section 5 are: the introduction of a frequency-
dependent phase domain line model for EMT, switch event options for time domain simulations, allowing
opening by current threshold and closing by voltage threshold and, for single phase network modelling,
we have introduced a single phase synchronous machine model and a single phase three-winding
transformer model.
We wish you success with PowerFactory 2020 and we look forward to your comments and feedback.
Contents
2 Analysis Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.7.2 Frequency Sweep: New result variables for impedance neglecting ca-
pacitances in the network . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.7.3 Harmonic Load Flow and Frequency Sweep: New option for tempera-
ture dependency of lines and cables . . . . . . . . . . . . . . . . . . . . 12
5.3 Transformers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5.6 Series Capacitors: Introduction of Loss Factor and Zero Sequence Input . . . . . 37
8 Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
A new application framework now supports the graphical user interface, in order to substantially en-
hance the resilience of the software and its potential for future development. The change also has
the benefit of an improved interaction between the application and the Window DPI scaling and layout
settings.
In conjunction with this update, the user will also observe a new, more modern look to the GUI, which
now incorporates dockable tool windows and a tabbed-document concept for the Graphics Board.
Some of the features for users to explore are presented in the following sections.
Windows such as the Project Overview, the Output Window and the Drawing Tools can be undocked
and freely moved around by the user; a blue background is used to indicate where windows can be
re-docked, as shown here:
Similarly, individual parts of the main toolbar (separated by ) can be moved around to suit the user
(see Figure 1.2 below). By default, there are two parts to the main toolbar, but if customised user profiles
are used to create several sets of tools, each set can be moved individually.
It is also useful to note that under the Window menu, an option Reset window layout allows the user to
restore windows and toolbars to their default positions.
A great deal more flexibility has been introduced into the Graphics Board, offering the user many options
for customising the view. It is now easy to split the view horizontally or vertically, as shown below:
The development offers many features to make working with the graphical pages convenient and
intuitive, including:
• Browser-like navigation options using left- and right-arrows on the main toolbar to return to previously-
selected pages
• Using the mouse scroll-wheel to move though the pages
• A drop-down menu on the main toolbar to allow easy access to graphical pages and for opening
recently-closed pages
The icons used in the Data Manager and Network Model Manager have now been updated to align with
those in the main toolbar. This new look can be seen, for example, in Figure 1.4 below, where useability
is also improved through the use of alternating colour shading.
2 Analysis Functions
The data model of transformers is extended by operational tap limits in the transformer element itself.
These are in addition to the tap limits of the transformer type and can be enabled selectively for all
transformers. The operational tap limits might be used to model restricted tap ranges used in operational
planning processes, for example.
The modelling and consideration of temperature dependent impedances in the load flow calculation
has been extended. It is now possible to give each line and cable an individual operating temperature
besides the maximum operating temperature in the type. Also the temperature to be used can be
globally defined in the load flow command. The available options are:
• 20 ∘ C
• Maximum operating temperature
• Operating temperature (new)
• User-specified temperature (new)
New features have been developed for Contingency Analysis, and the reports dialog has been re-
designed to accommodate the selection of additional tabular reports.
The concept of dynamic contingencies is introduced. Complementing the use of standard fault cases,
it is now possible to create contingencies “on the fly” during the Contingency Analysis calculation,
according to user-defined criteria. These criteria are applied on an element basis, to determine whether
a contingency case should be created for that element.
A typical application could be to create contingencies based on the pre-fault loading of a circuit, as
shown in Figure 2.1 below. But the user has a great deal of freedom in setting the criteria, because this
is done via filters, rather than a limited list of options.
The results from the dynamic contingencies are included in the result file, so that they are reported
alongside the results for any other “static” contingencies already included in the ComSimoutage com-
mand.
Dynamic contingency definition is particularly useful when executing a contingency time sweep, as it
enables the user to restrict analysis to different critical circuit faults for different study times and thereby
reduce execution time.
A new option is available in Contingency Analysis, to record additional summary result variables. This
option allows the user to obtain an overview of the contingency analysis by recording the numbers of
cases which have the following outcomes:
• Processed
• Not solved
• Inactive
• Causing grounding
• Causing islanding
• Causing blackouts
The dispatch event, which can be used for example as a post-fault action in contingency analysis fault
cases or as a remedial action in a Remedial Action Scheme (RAS), has been extended. This event
was previously only available as a relative change, for example to increase or decrease the output of a
generator by a specified amount; now there is another option, to change the output to an absolute level.
In a separate development, the dispatch event can now also be applied to Ward equivalent objects
(ElmVac).
In addition to the existing load states, generation states can now also be considered for reliability
analysis.
The generation states cluster the networks feed-in power from all available generators into certain states
together with a probability of occurrence for each cluster. These generation states are considered for
resupply strategy of the distribution network and therefore leads to more precise results for the reliability
indices.
The Techno-Economical calculation is a well established and commonly used functionality in PowerFac-
tory, which is used for carrying out an economic assessment and comparison of network expansions
(projects). Existing analysis results can now be accessed by reports at a later stage without the need
to repeat the analysis. Using this new option, complete or partial reports can be created on demand.
The Parameter Identification command enables the user to identify parameters of elements subject to
some given reference measurements. Supported calculations are:
• Load Flow
• Dynamic Simulation (RMS/EMT)
This tool has now been updated and redesigned in order to enhance usability and functionality:
A large set of new optimisation algorithms have been provided. This allows the user to select the most
suitable method for the specific identification task. Among others, the user can select from:
In addition, user-defined stopping criteria can be used to avoid unnecessary iterations in the identifica-
tion procedure.
The definition of the reference curve and values has been improved:
• The comparison with some given reference is now possible in a direct and intuitive way by means
of a comparison object (ElmDiff ) as shown in Figure 2.4.
• Any signal may be selected directly. Custom weighting for each signal pair is supported.
• Concrete reference values may be given for load flow based identification.
• Result files may be chosen as reference curves (in ElmFile) for dynamic simulation based identi-
fication.
Figure 2.4: Definition of the signal comparison, with time dependent weighting signal
A highly flexible definition of controls will guarantee optimum results (Figure 2.5):
• Flexible definition of lower and upper bounds of controls.
• Direct selection of available model parameters for common models.
Recording of the error over the iteration cycles enables the user to monitor the progress of the identifica-
tion process and support the selection of the best optimisation algorithm. The recorded values can be
visualised as shown in Figure 2.6. The y-axis shows the error (deviation between reference signal and
simulated signal) and the x-axis shows the number of iterations. The swarm algorithm uses a certain
number of particles per iteration; the result of each particle is shown as a dot.
Figure 2.6: Visualisation of the error minimisation progress (swarm intelligence algorithm)
A new DSL procedure has been introduced within DSL level 6 which allows variables of a DSL model to
be changed at specific points in time in the simulation. As such, any DSL model variable can be reset
upon satisfying a certain user-defined “trigger” condition. This procedure ensures that the variable
names remain consistent between a DSL macro and the main model, allowing users to include the
reset functionality within DSL macros. The reset procedure (unlike the “event()” procedure) is regarded
as a DSL model specific action, and so only emits Output Window messages if the flag Display internal
DSL events is checked in the Run Simulation command dialog. An example of a limited integrator with
reset is shown in Figure 2.7.
To allow a more flexible dynamic model initialisation, it is now possible to automatically execute a user-
defined DPL script during the initialisation procedure of a DSL model. The DPL script is able to read any
information of the power system elements, including load flow results, process these quantities and set
various parameters of the DSL model to which it is linked. The configuration script changes the values
of DSL model parameters only during a dynamic simulation (via the initialisation process), whereas the
values of the DSL model parameters (as shown in the DSL Model edit dialogs) are not influenced, i.e.
they remain persistent. The DPL script can also generate output window messages in order to provide
additional information at the initialisation stage, as shown in Figure 2.8. The configuration scripts for
initialisation can be encrypted and password-protected.
In previous versions of PowerFactory, the DSL special function picontrol_const was introduced to
support the IEEE 421.5 compliant PI regulators. This function requires that the upper and lower limits
are constant quantities (e.g. parameters). With PowerFactory 2020, a new built-in DSL special function
(picontrol) is available for use in DSL level 6 upwards, which allows users to program variable limits for
this PI controller type, as shown in Figure 2.9.
Figure 2.9: Example demonstrating the use of the DSL function “picontrol”.
The main functions of the Power Quality and Harmonic Analysis toolbox, Frequency Sweep and Har-
monic Load Flow, are enhanced by additional options and result variables, increasing the flexibility and
performance of the calculations.
The Frequency Sweep command is enhanced by an option to run the impedance evaluation of the
network over the selected frequency range without initialising the network components with a load flow
calculation. This setting leads to a high performance increase, when executing the Frequency Sweep
in large networks. The performance benefit will even be higher for the case when the impedance
evaluation also considers contingencies.
When deselecting the option Load Flow Initialisation, the power frequency impedances of all network
components are considered as base for the subsequent sweep over the selected frequency range,
evaluating the frequency dependencies of all components. For equipment models where the impedance
is affected by a load flow calculation, such as tap dependent impedances for transformers with automatic
tap adjustment, the actual tap position of each transformer will be considered for determination of the
power frequency impedance. Other models like the Static Var System are enhanced by additional
initialisation parameters to assign them an effective impedance without the need of load flow results.
2.7.2 Frequency Sweep: New result variables for impedance neglecting capacitances in the
network
The Frequency Sweep calculation now allows the evaluation of the network impedance for the case of
completely neglected capacitances. If the additional result variables Rnc, Xnc, Znc and phiznc (with
the suffix “nc” for “no capacitance”’) are selected for the nodes of interest, this will trigger a separate
evaluation of the network impedance for those nodes while neglecting any capacitive component in
all network elements. The impedance without capacitances can be used as a base to evaluate and
visualise the impact of capacitive components on the impedance considering capacitances (see Figure
2.10).
40,00
[Ohm]
30,00
20,00
10,00
0,00
-10,00
0,000 200,0 400,0 600,0 800,0 [Hz] 1000,
Terminal: Network Impedance, Magnitude
Terminal: Network Impedance, Magnitude (without C)
Figure 2.10: Magnitude of the network impedance with and without the consideration of capacitances
2.7.3 Harmonic Load Flow and Frequency Sweep: New option for temperature dependency of
lines and cables
The Frequency Sweep and the Harmonic Load Flow will take into account the temperature setting of the
referenced Load Flow calculation (see also Section 2.1.2) and evaluate the impedance over the selected
frequency range for that temperature. For the Frequency Sweep without the Load Flow Initialisation, the
temperature is selectable directly in the command dialog. The corresponding selection offers the same
options as are available for the Load Flow calculation.
The evaluation of the impedance at a certain temperature allows a further increase in accuracy of the
results by considering more realistic conditions in the network model.
The Harmonic Load Flow for balanced network representation calculates the voltage and current dis-
tortion for all frequencies, where one or more harmonic sources is connected to the network. For
injections at the harmonic orders 5, 11, 17, etc., the harmonic currents and voltages naturally appear
in the negative sequence; their calculation and impedance consideration are done therefore for the
negative sequence as well and the results are stored in separate negative sequence variables.
With the newly introduced option “Only positive sequence”, the calculation of these negative sequence-
orders will be forced into the positive sequence. For the balanced Harmonic Load Flow calculation, all
harmonic orders will then be covered by positive sequence result variables.
In PowerFactory 2020, the Protection Coordination Assistant has been completely updated, in order
to provide the user with full flexibility for the definition of setting rules for the zone reach calculation of
distance relays.
Users may now specify their own rules for the zone-reach determination, using equations with an
intuitive syntax to control a topological search within the network by means of predefined keywords.
Previously available common settings rules continue to be available as pre-defined options but now
may also be adapted according to user needs.
Special consideration can also now be given to the rules applied when the primarily protected lines in
zone 1 are parallel lines in contrast to when they are single circuit lines. The new version will allow the
user to easily specify their own rules for parallel lines in particular or omit them if required.
Figure 2.11: Definition of custom setting rules for Zone 3 using simple syntax based on keyword
variables.
In order to calculate resistive reach for phase and ground distance protection, a maximum arc re-
sistance is considered (the so-called arc reserve). The specification of this quantity in the tool has
been enhanced so that multiple arc reserve values are now specifiable and subsequently automatically
selected according to the applicable nominal voltage level.
For circular characteristics the characteristic angle can now be calculated in one of two ways. Either
the Zone 1 line impedance angle is used for each zone or an angle based on the summation of the line
impedances involved in the calculation of the reach of each individual zone is used.
In addition to the existing user-defined zone time delays, users are now able to specify a grading time.
This grading time can apply an additional constraint to the setting calculation for the case where a
distance zone reaches over a position containing other protection devices, such as overcurrent relays
at tee-off feeders. In such a case the zone delay can optionally be coordinated with the minimum
or maximum trip time of the overcurrent relay so as to achieve the specified grading margin. Where
directional and starting elements provide backup protection, time delays for these backups can now
also be specified in the tool and will be coordinated with other zones calculated for this relay.
An improved reporting layout has been provided, facilitating easy and flexible transfer of the coordination
results to the relay models and facilitating interactive marking of elements traversed by the topological
search in the single line diagram for each calculated relay setting.
The new version of the IEEE Guide for Performing Arc-Flash Hazard Calculations (IEEE 1584-2018) is
now implemented in PowerFactory.
Once the IEEE Std. 1584 calculation method is selected, the user can choose between the Standard
versions published in 2002 and 2018. The input window of affected elements (such as Terminal,
Substations, Bays) has been updated so that additional information such as “Electrode configuration”
and “Dimensions of enclosure” can be entered.
In addition, an advanced and more detailed tabular report has been introduced. For each calculation
object of interest an intermediate results table can be generated, containing details of the full arc and
reduced arc current. In this way the PPE-Category selection and calculation of Arcing Energy can be
easily reproduced.
All load and generation models in PowerFactory are now supported by the Voltage Profile Optimisation
calculation. Some of the different options are shown in Figure 2.15. This gives the user more modelling
flexibility and enables the optimisation of distribution transformers’ tap positions in all networks.
Figure 2.15: Use of different transformer representations for the Voltage Profile Optimisation
The Hosting Capacity analysis has been enhanced and now allows parallel computing using multiple
cores, resulting in a significant reduction in the calculation time.
The representation of the results for the Hosting Capacity has also been enhanced, as shown below
in Figure 2.16. In geographic diagrams circles can be displayed around substations, sites, secondary
substations and nodes whose sizes represent the power that can be connected without violating the
user-defined constraints.
The network reduction has been enhanced to create dynamic network equivalents to support balanced
RMS calculations. The reduction is based on pre-defined boundaries as the static network reduction.
The implemented method focusses on the synchronous generators of the system and aggregates them
to equivalent machines. Thus, dynamic equivalents can only be obtained if a system has synchronous
generation.
Figure 2.17 below shows the dialog page and options for this new feature.
The dynamic equivalents are created based on the aggregated machines of coherent clusters of syn-
chronous generators. In order to find the coherent clusters the network is excited, either by noise
injections or user-defined events, and the machines are grouped based on their response. Then the
coherent machines are aggregated and the passive network is reduced by a REI Reduction. In the
next step generic or user selectable controllers with default parameter settings can be added to the
equivalent machines and an optional parameter identification can be carried out to adjust the controller
parameters to match the dynamic response of the reduced network to the pre-reduction behaviour.
The verification of the reduction results is extended by a dynamic verification. The before and after
reduction simulation results are compared in verification plots, as illustrated in Figure 2.18 below.
Please note that the calculation of a dynamic equivalent will require a licence for the “Stability Analysis
Functions (RMS)” module.
Several newly introduced result variables for the Quasi-Dynamic Simulation give the user a better view
of the current network status.
• The Capacity Factor is the unitless ratio of the actual to the maximum possible electrical energy
output over a given period of time. It is available for all generation models in PowerFactory and
helps to analyse the utilisation of a generation unit.
• The Load Factor is defined as the average load divided by the peak load in a specified time
period. It is available for all load models in PowerFactory and additionally for objects such as
Grids, Areas, Zones and Feeders.
• The PV Penetration Factor is the ratio of the total peak PV power to the peak load apparent
power of a feeder. This variable gives the user a good sense of the extent of currently installed
photovoltaic plants within a defined feeder.
• The Maximum Reverse Power quantifies the maximum power export of a feeder and helps the
user to analyse the impact of generation units on the load flow of a feeder.
For a better allocation of different generation types as various renewable and fossil energies, new plant
categories have been introduced for each generation model. In the wake of this, several new plots have
also been added to the Quasi-Dynamic Simulation to visualise the different generation types, as shown
here:
With the addition of annual duration curves to the range of plots in PowerFactory, the user is now able to
analyse the utilisation of specific elements such as transformers or different power plants over a given
time period. Further details are given in Section 3.4.2.
The former “Load Flow Sensitivities” function, now called “Sensitivities / Distribution Factors”, has been
greatly extended in order to provide a range of sensitivity calculations commonly used in electricity
analysis and markets.
In addition to the existing sensitivities, the following standard quantities may be calculated:
• PTDF (Power Transfer Distribution Factor): The change in power flow on a circuit for a given
power injection at one or more nodes.
• LODF (Line Outage Distribution Factor): The change in power flow on a circuit as a result of
the (fault) outage of another circuit.
• OTDF (Outage Transfer Distribution Factor): The change in power flow on a circuit for a given
pre-fault power injection at one or more nodes, under fault outage conditions. This quantity is
obtained by running a PTDF calculation for each contingency case.
• PSDF (Phase Shift Distribution Factor) or TCDF (Tap Change Distribution Factor): The
change in power flow on a circuit as a result of tapping a transformer.
The functionality allows for single sensitivity calculations (for example to power injections at a single
busbar or multiple busbars concurrently) and multiple sensitivities (for example to power injections
at many busbars in turn). In addition to the base load flow sensitivities, there are also options to
consider contingencies, which lends another dimension to the results. The various options are shown
in Figure 2.20 below.
Note that the calculation of sensitivities for contingency cases requires a licence for the corresponding
module “Contingency Analysis”.
If a single sensitivity calculation is being executed, the results can be viewed directly in a Network
Model Manager. For multiple sensitivities, a result file is used to store the results of the calculation.
These results can then be accessed using inbuilt reports. To assist with the handling of potentially large
quantities of data, the reports offer not only the usual tabular reporting filters but also filtering by element
and by class.
The new functionality also includes the possibility to only record results for selected elements, using a
filter, which offers considerable benefit in terms of speed of execution.
The existing PTDF analysis, provided as part of the Transmission Network Tools module, is still sup-
ported. In contrast to the “Sensitivity / Distribution Factors”, the PTDF Analysis of the Transmission
Network Tools function is capable of identifying PTDFs across flowgates whilst scaling generation and
load demand in predefined regions.
The Transfer Capacity Analysis tool has been enhanced with addtional options. Previously, the scaling
was always carried out in selected importing and exporting regions of the network, for example grids or
zones. The boundary between the two regions was automatically determined and the transfer across
this boundary reported. Some new features have now been added:
• If scaling according to region, the user has an additional option to select the boundary for which
the transfer should be calculated and reported.
• As an alternative to specifying importing and exporting regions, the user can instead define precise
sets of generators and/or loads to be scaled, together with the boundary for which results should
be reported.
The Unit Commitment and Dispatch Optimisation function was released in PowerFactory 2019 and the
wide range of functionality is further extended in PowerFactory 2020. The major new enhancements
are:
The Unit Commitment and Dispatch Optimisation offers the possibility for a cost optimised operation of
storage units. Generating units can be linked via efficiency curves to newly-introduced storage model
objects (see Section 5.9) .
Each storage model can be associated either with a single generating unit or with a group of generating
units via a virtual power plant. The storage model is generic and can represent any kind of storage. The
considered figures are the capacity, limits, initial energy, self discharge and natural inflow if used as a
hydro unit. The values can be given as energies and powers or as equivalent hydro figures.
The influence of storage elements in a unit commitment can be analysed via pre-configured table
reports:
Figure 2.21: New storage report of the Unit Commitment and Dispatch Optimisation
A negative active power dispatch is possible when the consumption is enabled in the linked generating
units. This can be visualised in a energy plot (see also section 3.4.1):
Figure 2.22: Energy plot of active power dispatch sorted by plant categories with storage
In a Unit Commitment and Dispatch Optimisation the determined controls and resulting load flow
results are stored in a result file. Writing back the resulting controls as characteristics to the network
model is now possible. This means that the user can execute further calculations based on the Unit
Commitment results directly. Thus, this offers the possibility to execute a market simulation without
network constraints in a first Unit Commitment and then execute a second Unit Commitment with
enabled network constraints to do a redispatch calculation based on the results of the market simulation.
New energy plots such as the aggregated infeed per plant category are also available for the Unit
Commitment calculation (see 3.4.1). These are the available plots:
Due to the increasing penetration of renewable energies in distribution networks and the growing
challenge of integrating new plants, PowerFactory 2020 offers the newest guidelines for the connection
request assessment. For low-voltage networks an updated version of the VDE-AR-N 4105 is now
available and for medium-voltage networks the new VDE-AR-N 4110 is introduced. A detailed report as
shown in Figure 2.24 gives the user the ability to analyse the assessment.
The updated version of the VDE-AR-N 4105 is a guideline for the connection of new generation plants
within low-voltage networks. Storage units can also be assessed according to the new standard. The
various assessment criteria are shown in Figure 2.25.
For the assessment of plants in medium-voltage networks, the VDE-AR-N 4110 standard has been
introduced. This guideline not only considers multiple generation units, but also storage units as well
as load types in a single connection request in PowerFactory. This guideline can therefore also be
applied to evaluate new charging stations, which makes this tool very valuable for planners and network
operators of medium-voltage networks.
The Drawing Toolbox has been redesigned to make it more user-friendly. By default, the toolbox is
seen on the left-hand side of the graphic window, where it shares the space with the Project Overview
window. The Drawing Toolbox is automatically shown when the graphical freeze mode is turned off.
For users who are less familiar with the icons, there are settings to show group headings and icon
names. Filters allow the user to choose which icons are displayed, according to name or by group.
These options are shown below in Figure 3.1.
For graphical pages, the options for exporting the graphic and for printing have been enhanced. The
preview page offers a wider range of options such as selecting part of the graphic, or creating a multi-
page PDF document from selected graphics, as shown here in Figure 3.2:
For graphics with titles or legends, these can be moved as required within the print-preview dialog,
Area interchange diagrams are a new feature which has been developed in the Diagram Layout Tool. It
enables users to create diagrams which are designed for visualising the power flows between different
parts of the network model, or to add this graphical representation to an existing diagram.
Interchanges between grids, areas or zones can be shown. In the example shown below, the inter-
change diagram is an easy way to see the power flows between the eight grids which make up the
network.
Figure 3.3: Interchange diagram used to visualise power flows between grids
The relative thickness of lines indicates the relative level of active (default) or reactive power flow. The
interchange boxes and arrows are contained in a dedicated graphical layer, where the user can select
the power quantity and configure the representation.
In PowerFactory 2020 several new plots are available for the Quasi-Dynamic Simulation and for the Unit
Commitment, to display the different types of generation units.
The plot Complete Generation illustrates the entire generation of the whole grid or a defined area.
With the help of user-defined generator groups it is possible to modify the plot and to display only the
relevant plant categories. In Figure 3.4 the complete generation of different areas within a network is
shown.
With the plot Plant Categories the user can visualise all plant categories that currently exist in the grid
or a specified area as seen in Figure 3.5. An additional feature can be seen when holding the cursor
over the plot. Then the user is not only able to see the actual power output of the selected category
compared to all categories but also the complete generated energy over the simulated time frame.
To analyse the penetration of renewable energies in a network the Renewable and Fossil plot can be
used, to illustrate how much renewable and fossil generated power there is in the network.
The representation of all plots mentioned can be stacked to show the actual generation of the different
types or in percent to see the percentage share of the various plant categories.
Another new plot is the Annual Duration Curve which is available for different elements in PowerFac-
tory. This visual enhancement makes the analysis of the utilisation of power plants or transformers, for
example much easier. In Figure 3.7 an example of the loading of a transformer is illustrated.
In earlier versions, any plots which the user removed were deleted and no longer available. Now, the
user has the option to either delete or just close the plot page. Plots which are closed but not deleted
are retained in the study case, and can be reopened at any time using the option “Open Plot Page...”
from the Window menu in the main toolbar, or as shown below from the Graphics Board toolbar.
The basic structure of a PowerFactory project has been changed to make it more intuitive. As can be
seen in the figure below, the Diagrams, Network Data, Operation Scenarios and Variations all now sit
at the same level within the Network Model folder, because these are the components which together
define the network model in conjunction with any data used from the project or global libraries. The
default behaviour of variations, to record just changes to the Network Data, is unchanged.
The former project hierarchy is still supported without any loss of functionality. Existing projects from
earlier versions which are migrated to PowerFactory 2020 will not automatically be migrated to the new
project structure; however, on the Migration page of the project, a button is provided to allow the user to
update the structure as and when required.
The Study Case Manager window is now accessed using a new icon , in the main toolbar.
When applying time characteristics to object parameters, users have the option to store the values either
locally or in an external location. Previously, this external location would be a text or CSV file, but now it
is also possible to use data stored in a database. Oracle, PostgreSQL and SQL Server databases are
all supported. Access details to the database are configured by the user and the time and data values
may be stored in one table or separate tables for each element, within the referenced database.
This possibility to store characteristics data in an external database will be especially useful for users
working with very large quantities of data, such as demand or generation profiles for a sizeable network,
particularly if long time periods and high resolution are also required.
Another option is to use data from result files (ElmRes) to provide the characteristic values. The user
selects the result file, the element and the relevant variable to provide the data for the characteristic.
A range of options have already been available for the export of calculation results. Now it is possible
to export results to an external database. Oracle, PostgreSQL and SQL Server databases are all
supported. Access details to the database are configured by the user, then the database configuration
is selected for the results export.
When users wish to have access to additional information about network elements, for example, it has
long been possible to create an IntDocument object, which contains a pointer to an external document
such as a .pdf file. Now it is possible to import the actual document so that it is stored within the
PowerFactory database. As all network elements have an Additional Data parameter on the Description
page, it is then convenient to use this to reference the imported document.
The Network Model Manager now distinguishes between the type of a common model (model definition)
and the type of a composite model (composite model frame) as shown in Figure 4.2. This simplifies the
management of dynamic models.
The automatic tap changer capabilities for shunt elements are extended in the new version. Shunt
elements can now be controlled continuously in addition to the already existing discrete control option.
As a result, the tap controller can precisely comply with the specified control condition. In this case, the
corresponding voltage or power range settings are disregarded.
The circuit breaker element ElmCoup now includes a series reactance 𝑋 that can be defined in the
element type TypSwitch. Thus, the user can now define both a series resistance and reactance for the
switching element, which is especially suitable for EMT time domain simulations.
New options for switching events in RMS and EMT time domain simulations are available. Figure 5.1
provides an overview of closing and opening event configurations. Enhancements in PowerFactory
2020 include:
Closing Event
Figure 5.1: Closing and opening events for time domain simulations in PowerFactory 2020
Switching operations in power systems are very common and must not jeopardise a system’s reliability
and safety. Both fault and load switching cause transient voltages across the circuit breaker contacts as
shown in Figure 5.2. These voltages are called "Transient Recovery Voltage"(TRV).
𝑢 𝑇𝑅𝑉
System 1 System 2
During the switching operation, the circuit breaker needs to withstand the dielectric stress caused by
the TRV. According to international standards such as the IEC 62271-100, the dielectric strength of high
voltage circuit breakers is modelled by 2-parameter or 4-parameter envelope curves as shown in Figure
5.3. The 2-parameter curve is specified by 𝑢1 and 𝑡1 and the 4-parameter curve by 𝑢1 , 𝑡1 , 𝑢2 and 𝑡2 .
The corresponding parameter are defined in the standards for all relevant switching duties.
u1
u2
u1
t1 t1 t2
Figure 5.3: Circuit breaker switching capability curve defined by 2-parameter and 4-parameter envelope
curve
The TRV envelope curve of the circuit breaker can now be defined in the circuit breaker type TypSwitch.
The envelope curve is then automatically generated for each phase of the breaker during the simulation.
With this enhancement, the user can directly validate the circuit breaker ratings with regard to the TRV
for switching transients.
An example for a 400 kV shunt reactor load switching transient is shown in Figure 5.4. The simulation
shows the TRV across the circuit breaker contacts and the envelope curves of the corresponding circuit
breaker during the transient.
5.3 Transformers
A single phase 3-winding transformer model has been introduced in PowerFactory 2020. The model
is required for many applications in railway systems but also for special transformers in three phase
systems.
The model supports any winding configuration such as D-D-D, Yn-Yn-Yn and Yn-D-D for example. The
user can define both the winding configuration and the type of each winding (with or without centre
tap). Autotransformer connection can be selected for D-D windings. The model supports saturation and
hysteresis.
Internal grounding and neutral connection are also supported. Tap controllers with the same features
as for the 3-phase three-winding transformer are available for load flow calculation and RMS simulation.
For harmonic load flow calculation and impedance frequency sweep analysis, the magnetising impedance
of transformers (ElmTr2, ElmTr3, ElmTr4) can now be modelled frequency dependent. The frequency-
dependency is defined in the element type, as shown below in Figure 5.5.
Figure 5.5: Example of frequency dependent core resistance gm(f) for a two-winding transformer
Simulations of electromagnetic transients such as transformer and shunt reactor energisation or fer-
roresonance require a detailed model of the non-linear behaviour of the iron core (saturation) in the
corresponding element.
In order to improve the modelling capabilities, the Frolich equation for saturation models has been
implemented. The Frolich equation is a convenient way of fitting a saturation curve to only a few
measurement points. The basic equation with 2 parameters and a modified version with 3 parameters
is available in PowerFactory 2020.
Using the Frolich equation for the saturation, the user can choose to define a final slope (saturated
reactance) for the current-voltage values of the saturation curve, which is very important for inrush
current studies.
Figure 5.6 shows the inrush current during the energisation of a 110/33 kV 50 MVA transformer. The
magnetising reactance of the iron core is modelled based on the Frolich equation. The parameters for
the Frolich equation are fitted to the open circuit measurements of the transformer (after transforming
the measurement of RMS open circuit current vs RMS open circuit voltage to peak current-flux values).
1,2
Magnetising flux in the core in p.u.
0,8
0,6
Frohlich Equation
Measurement
0,4
0,2
0
0 0,02 0,04 0,06 0,08 0,1
Figure 5.6: Energisation of a 110/33 kV 50 MVA transformer using the Frolich equation for the saturation
characteristic
Single-phase synchronous machines in the high-power range are mainly used as generators for the
railway power supply in some European countries.
PowerFactory 2020 introduces a new model for single phase synchronous machines that can be used
for steady state, frequency domain and RMS/EMT time domain analysis. The single phase synchronous
machine can be defined in the TypSym by defining the number of phases for the type.
The single phase synchronous machine rotor d-axis is modelled by two rotor loops representing the
excitation (field) winding and the 1d-damper winding. The q-axis is modelled using one 1q-damper
winding.
The Model 3.3 model already takes into account the negative sequence torque in the equation of motion
by default. It is now also possible to include the negative sequence torque for the Standard model.
This can be done by using the newly added option “Consider negative sequence torque” located in the
Advanced tab of the RMS-Simulation page of the synchronous machine type.
The station controller now also supports single-phase network elements. With this new functionality it
is possible to simulate the behaviour of automatic control devices and/or operator action in load flow
calculations also for single-phase elements.
The Station Control object acts on sources of reactive power and, optionally, on transformers and shunts
with tap changers, to achieve a target voltage at a certain bus or a target reactive power flow through a
cubicle/boundary, or a target power factor at a cubicle/boundary.
5.6 Series Capacitors: Introduction of Loss Factor and Zero Sequence Input
The Series Capacitor model has been enhanced by a parallel path of conductance to allow the consid-
eration of the Loss Factor.
In addition new parameters for the ratio of zero to positive sequence susceptance and conductance are
introduced to account for a possible coupling between the phases.
For analyses in the frequency domain, frequency characteristics can be applied to capacitance and
conductance for both positive and zero sequence separately and considered in the Frequency Sweep
and the Harmonic Load Flow calculations.
The active power setpoint for load flow calculations can now be defined by the mechanical torque or
mechanical power of the asynchronous machine. This additional option enables the user to configure
the operating point of the machine by the mechanical load and a corresponding speed-torque curve.
The effects of dead-time in the converter valves of a two-level PWM converter can now be included in
EMT simulations. The dead-time can be represented either via a dedicated parameter in the built-in
model or by including the PWM modulation strategy in detail as a DSL model (see Section 5.8.2). The
effects of this improved modelling are mainly observed in the magnitude of the low-order harmonics
of the output current as illustrated in Figure 5.7. Further detailed control techniques for dead-time
compensation can be integrated into the converter control dynamic models.
Figure 5.7: EMT simulation of a two-level PWM converter including dead-time - phase currents and
harmonic contents
The detailed two-level PWM Converter model (EMT simulation) can now be controlled via six separate
input signals (g1, g2, ... g6). Each firing signal controls the switching state of one transistor within the
bridge. This converter control mode allows user-defined PWM modulation techniques to be used e.g.
inclusion of valve dead-time within the PWM modulation algorithm.
Storage models (ElmStorage) can be linked with generators such as pumped-storage units in order to
manage the generated and consumed energy. A storage model can be linked directly to one generating
unit or to several generating units via a virtual power plant (ElmBmu).
• Battery: This represents an storage unit without any inflow but with a self-discharge. The Capacity
can only be entered as energy in MWh.
• Hydropower: This represents an storage unit with the possibility to have a natural inflow. The
Energy figures can be entered directly in MWh or as Water volumes which are then converted
using a Head of water parameter.
Once assigned to a generator or virtual power plant, the storage model is relevant for these calculation
functions:
• Quasi Dynamic Simulation
• Unit Commitment and Dispatch Optimisation
Efficiency curves are used in conjection with the storage models to define the efficiency of the unit in its
generation and consumption modes. An example for a hydro turbine in shown in Figure 5.8 below.
The analysis of power systems transients such as switching and lightning requires highly accurate
distributed parameter line models that account for the frequency dependency of the line parameters
over the complete frequency range.
In addition to the modal domain frequency dependent model that has been available in PowerFactory
for several years, a new wideband transmission line model is introduced in PowerFactory 2020. The
model is based on the Universal Line Model proposed by Gustavsen which utilises vector fitting for the
rational function approximation of the transfer function and is solved in the phase domain.
The new phase domain model provides highly accurate results for simulating single-circuit and multi-
circuit coupled lines and avoids the shortcomings of the frequency dependent modal domain model
when representing asymmetrical tower configurations.
Figure 5.9 shows the transient voltages at the sending and receiving end of of a 400 kV overhead
line during energisation, using the new Frequency Dependent Phase Domain Line Model. The model
accurately accounts for the distributed nature of the travelling wave phenomena including attenuation
and distortion during the transient.
Figure 5.9: Overhead line energisation using the new EMT Frequency Dependent Phase Domain Model
The interface with PSS/E now supports PSS/E versions up to and including version 34 for import and
version 33 for export.
From PowerFactory 2020 SP3 it will be possible to export or import graphics in well-known text format
(WKT, defined by the Open Geospatial Consortium). The graphical representation of network elements
will be given as simple text, which describes the geometrical form (e.g. point or line) and the corre-
sponding coordinates. The WKT converter can be used as a stand-alone converter or in combination
with the DGS converter. Both converters support the same formats (ASCII file, Oracle Database server,
SQL server, etc.) that are structured in a similar way. So it is possible to export or import the electrical
network data via DGS first and the graphic network data via WKT subsequently.
8 Installation
PowerFactory 2020 extends the support of Windows Servers to Windows Server 2019.
27
SERVING MORE THAN 150 COUNTRIES
DIgSILENT customers
DIgSILENT offices
Partners and representatives
DIgSILENT GmbH
Heinrich-Hertz-Straße 9
DIgSILENT GmbH is certified
72810 Gomaringen (Germany)
to the ISO 9001:2015 standard.
T: +49 7072 9168-0 More information is available at
mail@digsilent.de www.tuv-sud.com/ms-cert.