tt dafea

1
Wind Energy Penetration in Islands Limitations
and Prospects N. Hatziargyriou S.
Papathanassiou, T. Chaviaropoulos, I.
Tsipouridis NTUA
CRES
PPC
2
Island Systems
3
RES in Greece (1)
4
RES in Greece (2)
5
Kythnos Island

• Small isolated Power system 2MW.
• 11 Annual RES penetration Wind PV
• Up to 100 instantaneous RES Penetration

6
Kythnos Power System-today
7
  Crete Island
Largest isolated power system in Greece 3 Power
Stations (Chania,Linoperamata,Atherinolakos)-26
thermal units, Peak 540 MW(2004), 87.1MW Wind
capacity-Wind power penetration 10
8
Crete System Single-Line Diagram
9
W/F Production 1993-2000 KWh
10
Penetration during the day of max. inst. output
( Dec.28.2000)
11
Rokas W/F output (13 May 1998)
12
Rokas W/F output ( 20 May 1998 )
13
Frequency and production due to Gas Turbine Outage
14
Summary of Experience from Wind Power Penetration
in Crete

• High efficiencies (Capacity Factor in
  200037,33)
• Stable output but with serious exceptions
• Normal Voltage fluctuation at connecting W/F bus
• Higher yields during summer months but lower,or
  zero, at evening peaks
• No problem for Flicker or Harmonics
• Economical considerations.

15
Large Scale Penetration of Increased Wind Power
in Island Systems

• Wind Power - Intermittent Power Source
• Wind Power partly non-controllable
• Effects
• Operation with increased uncertainty,
  requirements for reliable short-term forecasts of
  renewable source production.
• Effects on power system security both
  steady-state and dynamic, inability of
  conventional directly coupled Wind Turbines to
  provide ancillary services, requirements for
  on-line security assessment.
• Effects on economic operation
• Power Quality ???

16
Fault Ride-Through Capability

• Modern WG are required to remain connected during
  a fault
• Various standards for voltage dips to be sustained

Frequency Requirements?
17
CAREMORE CARE EC Projects Advanced Control
Advice for Power Systems with Large-Scale
Integration of RESJOULE JOR3-CT96-0119 ENERGIE
ERK5-CT1999-00019

• Minimise production costs through unit commitment
  and on-line economic dispatch of both
  conventional and RES also considering market and
  increased security.
• Provide short-term forecasts of load and wind for
  short-term (0-8 hr) and medium-term horizon (4-48
  hrs).
• Maximise power system security by on-line
  security assessment modules, that supervise
  scheduling of the power units and monitor the
  system security margin. Also provide corrective
  control actions.

18
The CARE System Architecture
19
Execution Cycles of MORE CARE For Crete
H48 hrs, h4 hrs, t11 hr, t220 min
20
Current System Operation and Economic Dispatch
21
Economic Appraisal of Wind Power Forecasting

• Off-line execution of Economic Scheduling
  functions assuming
• Spinning Reserve at least equal to wind power
  no wind power forecasting
• Spinning reserve at least equal to 50 of wind
  power 50 maximum forecasting error
• Spinning reserve at least equal to 20 of wind
  power to face very short wind fluctuations very
  advanced WF with 20 maximum forecasting error.
• Evaluation of economic impacts

22
Economic Appraisal of Wind Power Forecasting -
Results

• More accurate wind power forecasting reduces
  operating costs as follows
• Monthly cost savings with thermal spinning
  reserve equal to 50 of wind power
• 1.8-3.5 (163-480
  kEuro)
• Monthly cost savings with thermal spinning
  reserve equal to 20 of wind power
• 2.3-5.3 (217-758
  kEuro)

23
Economic Scheduling Evaluation

• Unit Commitment
• Average gain 3.1
• Economic Dispatch
• Average gain 1.7

24
Economic Scheduling Evaluation
Relative gain provided by GA Economic Scheduling
Functions
25
Dynamic Security Evaluation
Actual Operation
Proposed Operation
26
Short term flicker (Pst) at W/P bus of Sitia S/S
( 12 Aug. 1998 )
27
Long term flicker (Plt) at W/F bus of Sitia S/S
( 12 aug. 1998 )
28
PROJECT Integration of Wind Turbines into
Electricity Supply Networks with Limited Power
Transportation Capacity Wiring (JOR3-CT98-0245)
Electrical network of the Greek island Samos
VESTAS
WINDMATIC
NEG MICON
NEG MICON
29
Frequency behaviour of the island grid of Samos
during one week

30
Diesel generator production during 13 days
31

• The measurement campaign gave some unexpected
  results
• Most important is the fact that the voltages are
  highly distorted in the flicker producing
  frequency region, due to firing harmonics of the
  power station diesel engines. This background
  distortion is large enough to completely mask the
  effect of the flicker emitted by the WTs. The
  connection of the WTs leaves more or less
  unaffected the overall flicker levels, and in
  certain cases even reduces them.
• The effect of the Vestas wind turbines on the
  frequency spectrum of voltage and power at the
  diesel power station was investigated. The same
  spectrum of voltages occurs with and without the
  wind turbines.

32
Voltage PSD before cut-in of WTs
Voltage PSD after cut-in of WTs
Firing of diesel engines
33

• CONCLUSIONS
• Wind power installations in islands provide
  economical benefits compared to diesel
• Increased requirements for robust Wind Turbines
  equipped with Low Voltage Ride Through
  capabilities and able to withstand frequency
  varaitions
• Need for advanced control systems to deal with
  optimal operation and management of power systems
  with large integration from wind power.
• Application of the MORE CARE control software on
  Crete has shown clear economic gains and timely
  warning for more secure operation.
• Flicker levels do not seem to be affected by the
  operation of WTs.