11. August 2014 23:59
Our GADS Open Source software has been included in the New York State Reliability Council’s Policy No. 5-8 - Procedure for Establishing New York Control Area Installed Capacity Requirements (see attached or go to http://www.nysrc.org/policies.asp )
"The multi-state model for each unit is derived from the collection of forced and partial outages that occur over the most recent five-year period using benchmarked GADS Open Source (OS) software as described in Appendix E."
Search the document for “GADS”, but pay specific attention to Appendix E - Development of Generator Transition Rate Matrices for MARS that are Consistent with the EFORd Reliability Index
Beginning on page 33 and continuing on page 34:
“In order to generate the metrics needed to populate the transition rate matrices for the generating units modeled in MARS, the above methodology was coded into the GADS Open Source software package but modified as described above. The GADS Open Source (GADS OS) is the software utilized by the NYISO to analyze generator performance data. It is used to calculate the generator performance indexes used in determining a generator’s UCAP value and, now, for developing the transition rates that are used in the MARS model. GADS OS allows electric generating companies to collect and report validated GADS performance data and event data. GADS OS can be found at this link:
“GADS OS, written by industry veteran Ron Fluegge, consists of two open source applications— GADS OS Data Entry and GADS OS Analysis & Reporting to analyze the GADS data. Besides the use of this software for submittal of GADS data to NERC and the NYISO, the software is also used for submittal of data to the ISO New England, PJM, and the MISO.”
Final POLICY 5-8 approved July 11 2014 v.2.pdf (811.67 kb)
1. June 2012 16:35
Dr. Singh has concluded that the GADS OS has implemented Method 2 as intended and the software appears to be producing the expected transition rates.
Based on Dr. Singh’s findings in conjunction with the EFORd comparisons, the NYISO has concluded that MARS state transition rate matrices whose probabilities are consistent with concept of the unit being demanded as specified in Method 2 has been successfully implemented.
The objective of the EFORd calculation is to:
1. Find the time spent in various states during demand
2. Converting these times into conditional probabilities
3. Adjusting the times in derated states to equivalent times in the full forced outage state
This results in the LOLE calculation for MARS being based on EFORd and transition rate matrices that satisfy the conditional state probabilities of the units being demanded. The NYISO recommends that this approach be implemented for the upcoming IRM study
Transition Rate Validation White Paper.pdf (302.91 kb)