Status Update: Using NMMS for Building SSWG Cases

1
Status Update Using NMMS for Building SSWG Cases

• Wes Woitt
• CenterPoint Energy
• 2010 SSWG Chair

2
Current Process

• TSP Data
• Some have internal databases to compile data
• Others take previous case and add projects or
  corrections as necessary
• Incumbent upon each TSP to coordinate and track
  what is contained in data sent to ERCOT

3
NMMS Process
MOD
TSP Data PMCR
TSP Data PSSE RAW format
Yes
Posted to ERCOT Website
No
ERCOT Using PSSE
Compile Data and Solve
Output PSSE Cases
Case need tuning?
Ready for use in planning studies
PSSE
ERCOT Determines Dispatch Using UPLAN
Topology from NMMS Topology Processor RAW file to
MOD
TSPs Provide Loads and Device Control
Profiles RAW file to MOD
TSPs Provide Future Projects PMCR
4
Topology Processor Functionality

• Topology Processor (TP)
• Converts Operations formatto Planning format
  (RAW) files
• MOD produces (.PRJ Project)
• Converts Bus/breaker to Bus/branch model
• Case Builder (CB)
• Builds time-based models

4
5
Whats been done so far?

• IMM work
• ERCOT prepopulated much of the planning related
  data in 2009
• TSPs cleaned up over 1000 incorrect Connectivity
  Node Groups (CNGs)
• Moved incorrect Connectivity Nodes into the
  correct CNGs
• Worked with ERCOT to correct CNGs and nodes where
  TSPs did not have ownership
• TSPs cleaned up planning related data (Area s,
  Zone s, ckt IDs, etc.)
• Submitted SCR 759
• Worked with ERCOT to modify TP
• MOD training in July
• Currently testing new build process using
  Topology Processor (TP) and MOD
• Identified issues with TP/IMM

6
(No Transcript)
7
IMM/TP output PSSE v30TSP Combined Issues List
50

• TP is already dated
• SSWG moved to PSSE v31 in May 2009
• New data types and parameters
• PSSE generally puts out a new version annually
• For TP to be able to include new data types or
  new parameters, IMM will need to be changed not
  just TP

8
No Distribution Cap Banks in TP CaseTSP Combined
Issues List 38

• TSPs model distribution cap banks at transmission
  buses in PSSE
• Allows cap banks to be switched during planning
  studies
• IMM does not model equipment below transmission
  voltage
• Hundreds of these cap banks in ERCOT system
• Cannot solve through MOD Device Control Profile

9
Distribution Cap banks

• Fall Planning Case

• Topology Processor Case

10
Zero Impedance Lines Created by TP Have 9999 MVA
RatingsTSP Combined Issues List 35

• TSPs model about 250 zero impedance lines which
  have real ratings in SSWG cases
• See examples

11
Zero Impedance Line Ratings

• Planning Case

• TP Case

12
Zero Impedance Line Ratings
Zero impedance branch 46110 46111 95 Rate A
360 MVA Rate B 440 MVA Rate C 440 MVA
North Belt 138 kV
Non-zero impedance branch 46111 46002
95 Rate A 455 MVA Rate B 478 MVA Rate C 580
MVA
13
Associating Loads and Cap Banks with the Correct
CNGTSP Combined Issues List 33 and 34

• It was somewhat of an art to assign nodes to CNGs
  to create model that we wanted to see
• See example

14
Figure 1 Modeling of Cap Banks and Load Taps in
IMMScenario 1
Bus 40002
Ckt. 01
A
D
To Bus 70000
Bus 50000
B
E
G
C
F
Bus 60000
Ckt. 02
Bus 40001
To Bus 80000
Bus 40000

• LEGEND
• Node Switch
  AC Line Segment
  Series Compensator Breaker
• With the current method of modeling, the cap
  bank and distribution load are associated with
  Bus 40000. To match the topology of our cases,
  the cap bank and distribution load should be
  associated with Buses 40001 40002,
  respectively. With the cap bank and load modeled
  as part of multi-section lines off of the main
  bus (40000), we are able to switch the lines,
  de-energize the cap bank, and roll the load to
  another load bus. This is impossible with the
  current method, because when the lines are
  de-energized, the cap bank and load remain on the
  main bus (40000).
• See Figure 2 for the resulting PSSE diagram when
  the above (current) method of modeling is run
  through the Topology Processor.
• See Figure 3 for the desired PSSE topology of
  the above substation.

15
Figure 2 Resulting PSSE Topology Using Scenario
1 Modeling Method
16
Figure 4 Modeling of Cap Banks and Load Taps in
IMMScenario 2
Bus 40002
Ckt. 01
A
D
To Bus 70000
Bus 50000
B
E
G
C
F
Bus 60000
Ckt. 02
Bus 40001
To Bus 80000
Bus 40000

• LEGEND
• Node Switch
  AC Line Segment
  Series Compensator Breaker
• With the current method of modeling, the cap
  bank and distribution load are associated with
  Bus 40000. To match the topology of our cases,
  the cap bank and distribution load should be
  associated with Buses 40001 40002,
  respectively. With the cap bank and load modeled
  as part of multi-section lines off of the main
  bus (40000), we are able to switch the lines,
  de-energize the cap bank, and roll the load to
  another load bus. This is impossible with the
  current method, because when the lines are
  de-energized, the cap bank and load remain on the
  main bus (40000).
• See Figure 2 for the resulting PSSE diagram when
  the above (current) method of modeling is run
  through the Topology Processor.
• See Figure 3 for the desired PSSE topology of
  the above substation.

17
Figure 5 Resulting PSSE Topology Using Scenario
2 Modeling Method
18
Figure 3 Desired PSSE Topology
19
Autotransformer Modeling IssuesTSP Combined
Issues List 41, 42, 48, and 52

• 48 and 52
• CNP reported that the TP was incorrectly
  calculating off-nominal 3-winding transformer
  impedances.
• Subsequent discussion with ERCOT/Siemens leads us
  to believe that we are inputting incorrect Ohmic
  impedances into IMM Still testing
• Siemens has used this method in other regions and
  had the same arguments
• No documentation about what we are expected to
  provide in IMM or what calculation TP is using
• 41
• CNP Operations standardizes all autos SCADA
  does adjustment automatically
• PSSE does not have that same intelligence
  therefore, IMM auto models must be adjusted to
  reflect real equipment characteristics
• Some parameters can be changed in Device Control
  Profile, but not all
• Not talking about ratings and impedances
  differences
• 42
• Allow a phase shift to be applied to transformers
  and output in the case. Useful when creating
  cases for fault duty studies to have this already
  in case.
• This has nothing to do with phase-shifting
  transformers

20
Autotransformer Ops Model vs. Planning Model

• The device control profile is not robust enough
  to handle the majority of differences between the
  Topology Processor output case and the
  Transmission Planning case. Instead, at least 50
  3-winding transformers in CenterPoint Energys
  area will require PRJ files to alter the Topology
  Processor output case to a typical Transmission
  Planning case.
• Differences Resolved with PRJ
• Moving the LTC from the low winding to the high
  winding.
• Number of taps
• Specifying a different voltage control bus
• The tap range of the LTC Rmax, Rmin
• Differences Resolved with Device Control Profile
• Winding voltage for off-nominal NLTC settings.
• The auto-adjust code for LTC windings
• Voltage control bus regulation values Vmax, Vmin

21
FACTS Device ModelingTSP Combined Issues List 32

• PSSE allows modeling FACTS devices (STATCOMs,
  SVCs, TSCs, etc.) in a number of different ways
• Generator
• Switched shunt
• FACTS Device
• TP only outputs a FACTS device as a generator
  model
• Important to model correctly for dynamic studies

22
FACTS Devices

• Planning Case CNPs FACTS Device

• TP Case model CNPs FACTS Device

23
FACTS Devices

• Planning Case AENs FACTS Device

• TP case AENs FACTS Device, currently on wrong
  bus

24
PSSE IDs for Branches

• PSSE allows any two character CKT ID for any
  branch
• IMM has validation rules that series devices PSSE
  CKT IDs have to be Sx
• TP automatically creates PSSE CKT ID of BC or BO
  for any branch that represents and breaker or
  switch
• Would prefer to leave PSSE CKT ID open and left
  to the discretion of TSPs

Line IDs with BO/BC ID
25
RARF Data TSP Combined Issues List 30

• TSPs have seen lots of basic problems with RARF
  data
• MBASE
• Zsorce
• Other RARF data problems
• Reactive data does not correspond with biennial
  tests
• RARFs submitted with erroneous PSSE bus numbers
• ERCOT Network Modeling creates NOMCRs
• Creates problems with TP output as SSWG recently
  discovered
• Need process to allow TSP review
• Even more issues with wind resource RARFs
• NPRRs currently being drafted or in process

26
Equipment Associated with SPS/RAPs Modeled in
IMM TSP Combined Issues List 49

• SPS/RAP models are not real equipment
• TSPs see additional data in TP output

27
Remaining Issues TSP Combined Issues List 26,
27, 28, 29

• 26 Each of the items identified above can be
  fixed in SSWG base cases through Standard PMCRs
  
• Taken as a whole, this represents an enormous
  amount of corrections to make to the case
• SCR being drafted to propose changes to many of
  these
• 27 Changes described in 26 are not
  implemented by next spring
• 28 SCR 759 is never implemented
• Multi-section line creation is not functional
• 29 Case building process is still very unclear
• Due to these uncertainties
• Just begun finding issues with MOD

28
Remaining Issues TSP Combined Issues List 40,
48, 51, 52

• 40 Specific modeling added for contingency
  analysis, dynamic analysis, short circuit
  analysis, etc., that is lost by creating cases
  from IMM data
• This punishes TSPs who have made a concerted
  effort to align their base cases to serve the
  needs of planning, protection, and stability
  studies without having to maintain separate
  cases.
• 48 Topology Processor issues
• Software is still being changed
• Radial line reduction still not working properly
• 51 Not all TSPs have PSSE which is creating
  problems with their ability to create PRJ files
• Potential resolution is being discussed, but it
  is a workaround
• 52 Lack of documentation about TP