PaperIV FlavioFernandez, DIgSILENT

Simulation of Power Systems with
Large Share of Renewable Generation
International Seminar on
„Impact of Generation from Renewable Sources
on Conventional Power Generation and Grid“
31st January 2014, New Delhi - India
Flavio Fernández
DIgSILENT GmbH, Germany
1
Overview
• Background
– What makes renewable generation different?
– Integration challenges from the power system analysis perspective
• Power system analysis functions

– Overview typical analysis needs
– Simulation tools for the steady-state and time domain
• Modelling aspects
– Considerations about dynamic models of renewable generation
• Conclusions
Impact of Generation from Renewable Sources on Conventional Power Generation and Grid, New Delhi -31.01.2014 2
Integration challenges
• Wind and solar energy is non-dispatchable and less predictable

– Limited prediction accuracy using weather forecasting
– Influenced by geographical distribution
– Impact on capacity building
– Impact on grid expansion engineering and strategic planning
• Power output inherently variable

– Increased variability (of the net load) in the power system
– Stepper ramp rates with increasing penetration levels
– Needs for higher load-following flexibility (interconnection between
balancing areas, demand response, smart grids)
– Impact on the system operation procedures
• Renewable energy is mainly based on inverted fed generation

systems that gradually replace traditional synchronous generation
– Technology allows a higher control flexibility
– New dynamic characteristics, influencing short-circuit currents and
network stability
• Wide range of manufacture-specific technical solutions in order to

comply with grid code requirements worldwide
– Testing and validation of simulation models has become necessary to
guarantee high model accuracy
– Benchmarking
Variable generation Smarter transmission network Active distribution networks
MW
1,600 Smart grids &
1,400
1,200
1,000
meters,
800
600
energy
storage
400
200
0
01-Jan
10-Jan
20-Jan
30-Jan
05-Jan
25-Jan
15-Jan
Large generation Active demand

• System operation
- Optimized operation close to 60
Peak Commuting Time
Peak Commuting Time

real time
55
Electricity Demand (GW)

50
2020 Demand ~ 15
GWh (daily) - 1.5
- Efficiently and accurately

45 million vehicles
Optimal Charging
Period Typical winter daily
demand
40
12,000 miles p.a.
create the required system

35
30
00:00
01:00
02:00
03:00
04:00
05:00
06:00
07:00
08:00
09:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
23:00
Time of Day
configuration Time of use tariffs

- Increased data volume
Inflexible generation Distributed generation

• System planning
- Efficient use of existing
transmission capacity (getting
closer to technical limits)
- Investment in new transmission
assets
Increased need for detailed network analysis
Strategic planning Operation
Long Term
Planning
LDF, N-1, SHC, Scheduled outages
Protection, etc Annual, month,
week Grid Safety Analysis;
Dynamics, LDF, N-1 DACF, D2CF
transients,
Congestion Management
harmonics, etc
Intraday
operation
ATFA
Outage planning Operational planning Control Center
D-years D-365 D-7 D-1 D D+1
PowerFactory
EMS
System integration: architecture solution
Power System
Analysis Functions
Load flow analysis
• Stochastic load flow calculations

– Inherent variability of renewable
generation
– Time dependent characteristics,
based for instance on wind speed
measurements or solar radiation
patterns
• Quasi-dynamic simulation
– Medium to long term simulations
based on steady-state analysis
– Synthetic signals via flexible
scripting functions
– Statistical analysis (e.g. network
losses)
Contingency analysis
• Work out post-fault actions to secure

network operation under (n-x)
contingencies
– Optimization of phase shifters,
transformer taps, reactive power
compensation devices, etc.
• Simulation model shall account for:
– Dynamic circuit ratings (short term and
continous ratings)
– Wide area balancing mechanisms
(power/frequency controllers)
– Demand side response / smart grids
• … hence maximize usage of existing

transmission assets
Short-circuit calculation
• Dynamic voltage support (reactive power injections) not currently

supported by standard short-circuit calculation standards (IEC/ANSI)
Short-circuit calculation
• Static generator model in PowerFactory

– Dynamic voltage support acc. to k-factor settings of the converter, for
instance acc. to ENTSO-E
– Active for transient current calculation: Iks
– Based on a current iteration method
i
imax
k
1
10%
Du
Capacity value and reserve margins
• The Capacity value is the measure of the “firm” capacity of a

renewable source that can be counted on as a reliable contribution to
the sum of all grid capacity
– „Firm“ capacity is the generation capacity (MW) available at all times
– Capacity Credit is usually expressed in percentage of the rated power.
• The capacity credit of a variable power source is system dependent

parameter determined by:
– Penetration level of variable power sources
– Availability of conventional generation
– Load
– Transmission constraints
• Purpose of the calculation:

– General assessment of the value of variable generation.
– Capacity Planning.
– Generation Reliability assessment.
– Remuneration strategies definition
• Determination of adequate reserve margins (or reserve capacity)

– Available capacity above the normal capacity needed to meet peak
demand
– Regulatory bodies usually require producers to maintain a constant
reserve margin of 10-20% of normal capacity
• Calculation based on reliability-based model of the power system

– Probabilities for each generator to drop out
– Variable power source model based (non-dispatchable generation)
– System Load curves
• Loss of load probability (LOLP, typically <10%, on seasonal peak):

– Probability that yearly/seasonal peak load cannot be covered by available
generation capacity.
• Loss of load expectancy (LOLE, typically < 6 hs/year))

– Expected number of hours per year (time probability) during which the
available capacity cannot cover the load.
Harmonics analysis and flicker emission
• Harmonic calculation, as traditionally used for

distorting loads, may lead to excessively
conservative results
• Therefore, alternative methods have been

proposed (as eg. IEC 61000-3-6):
– Summation laws with weighting factors to
better assess the contribution of large wind
parks or distributed generation
N N
 U m, h  I m, h
 
Uh   Ih  
m1 m1
• Voltage Flicker Assessment

– Short- and long term flicker emission factors
– Relative voltage changes
Resonances
• Risk of parallel and/or series resonances

– In particular offshore wind farms have vast underground cable systems
and reactive power compensation equipments
– Resonances may therefore arise due to the interaction with the power
system (inductive)
– Frequency scans can be used to identify resonance frequencies
• Special attention shall be driven to:

– The impedance of the network at the PCC can vary significantly (large
number of system operation scenarios, outages, etc.)
– Impedance locus of the network at PCCs shall be used instead of
classical max/min short-circuit power equivalent
– Automatic frequency scan for all possible configurations
Stability analysis
• Fault ride-through (FRT) capability

– With increasing penetration levels, the system cannot afford a large loss of
renewable generation
– Simulations shall verify that the generator stay connected during a grid
fault causing a voltage dip at the PCC (requirement acc. to grid codes)
• Frequency response
– Response of the renewable generation to a change in the grid frequency
– Contribution to frequency control (primary control, synthetic inertia)
– Minimum frequency range of operation (50.2 Hz problem)
• Small Signal Stability

– Assist in damping of power system oscillations (enough synchronizing
torque in the system?)
Stability analysis: fault ride-through
• Assessment of grid code requirements, typically:

– Do not disconnect above the red line
– Max active power recovery rate
– Temporary disconnection and re-synchronisation
• Analysis is done by means

of time domain simulations
– Accurate dynamic models
of the renewable
generation required
Stability analysis: simulation
• RMS/EMT time domain simulations

– Multi-phase AC/DC networks
– Fast, adaptive step-size algorithms
– A-stable numerical integration algorithms
– Simulation events of any kind of event
• Small signal analysis

– Selective eigenvalue calculation
Dynamic models of renewable generation
• Development of dynamic model not an easy task

– Wide range of control strategies to comply with grid code requirements
worldwide
– Technology is continuously evolving, so that models have to be
continuously updated
– The model shall be adequate to the simulation purpose (trade-off between
accuracy and complexity)
• Several efforts worldwide started

– IEC TC88 WG27 working group on wind generation models
– Western Electricity Coordinating Council (WECC), Renewable Energy
Modeling Task Force on wind and PV generation models
Dynamic models of renewable generation
• Vendor-specific models
– Normally available from the manufactures
– Highly accurate models
– Not always publicly available but may require non-disclosure agreements
– In PowerFactory, models of all major manufactures are available
• Generic models
– Not vendor specific
– It can be parameterized to reasonably simulate the dynamic behavior of
the generator
– Standard generic models are needed for portability across different
software simulation packages and data exchange between TSOs.
Dynamic modelling environment in PowerFactory
• Models of renewable generation are implemented in PowerFactory

using the dynamic modeling environment DSL (DIgSILENT Simulation
Language)
– General-purpose dynamic modeling language allowing to describe any
system of linear or non-linear differential equations
– Modeling environment with a graphical editor, with which any model can
be defined by drawing a block diagram
• Especially in wind generation applications, a flexible modeling

environment is essential because of the large variety of generator
types and control concepts used by different turbine manufacturers
that have to be reflected in the corresponding models.
Dynamic Modelling Environment in PowerFactory
S
G/
IG WTG with fully rated converter:
0
Pin 0
PQ
StaPqmea* Qin
1 1
0
id_ref 0
PQ Control 0
u1r_in 0
u ElmDsl* iq_ref
0 2 1 1
Generator
ElmGen*,ElmVsc*
pred 3 1
u1i_in 1
Vac
StaVmea*
1
2
ActivePowerReduction
0
ir 2
Grid Side Converter Step Up Transformer Iac
StaImea* ii
1 3
ElmDsl
cosref Current Controller
0 4 ElmDsl*
PLL 1 sinref 5
ElmPhi*
2 Fmeas
Fmeas
Slow PLL
ElmPhi*
6
7
Wind Power Integration – Modelling in PowerFactory 24

Conclusions
• High integration level of renewable generation challenges classical

power system planning and operation
– In general, increased need for detailed system analysis and optimization
– Analysis automation (to allow system optimization closer to real time
operation)
– Higher level of integration of power system simulation tools with external
systems (EMS, DMS, weather forecast systems, etc.)
• Large variety of power system analysis functions available such as
steady state, quasi-dynamic, dynamic and transient network analysis
• Ongoing efforts towards the standardization of dynamic models of
renewable generation (wind/PV generators)
Thanks for your attention
Flavio Fernández
DIgSILENT GmbH
f.fernandez@digsilent.de

PaperIV FlavioFernandez, DIgSILENT

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PaperIV FlavioFernandez, DIgSILENT

Uploaded by

Copyright:

Available Formats

Simulation of Power Systems with

Large Share of Renewable Generation

31st January 2014, New Delhi - India

• Power system analysis functions

• Wind and solar energy is non-dispatchable and less predictable

• Power output inherently variable

• Renewable energy is mainly based on inverted fed generation

• Wide range of manufacture-specific technical solutions in order to

Variable generation Smarter transmission network Active distribution networks

Large generation Active demand

Peak Commuting Time

Peak Commuting Time

Electricity Demand (GW)

- Efficiently and accurately

12,000 miles p.a.

create the required system

configuration Time of use tariffs

Inflexible generation Distributed generation

Strategic planning Operation

D-years D-365 D-7 D-1 D D+1

• Stochastic load flow calculations

• Work out post-fault actions to secure

• … hence maximize usage of existing

• Dynamic voltage support (reactive power injections) not currently

• Static generator model in PowerFactory

• The Capacity value is the measure of the “firm” capacity of a

• The capacity credit of a variable power source is system dependent

• Purpose of the calculation:

• Determination of adequate reserve margins (or reserve capacity)

• Calculation based on reliability-based model of the power system

• Loss of load probability (LOLP, typically <10%, on seasonal peak):

• Loss of load expectancy (LOLE, typically < 6 hs/year))

• Harmonic calculation, as traditionally used for

• Therefore, alternative methods have been

• Voltage Flicker Assessment

• Risk of parallel and/or series resonances

• Special attention shall be driven to:

• Fault ride-through (FRT) capability

• Small Signal Stability

• Assessment of grid code requirements, typically:

• Analysis is done by means

• RMS/EMT time domain simulations

• Small signal analysis

• Development of dynamic model not an easy task

• Several efforts worldwide started

• Models of renewable generation are implemented in PowerFactory

• Especially in wind generation applications, a flexible modeling

Wind Power Integration – Modelling in PowerFactory 24

• High integration level of renewable generation challenges classical

You might also like