PI : Information on hand for Operational Analytics Presented by: - PowerPoint PPT Presentation

PI : Information on hand for Operational Analytics Presented by: Ranjith Ramachandran Chandra Prakash Dewangan 1

AGENDA • Information Enhancement • ABT System. • Auto Reports. • Enhancement of data. • Performance Monitoring, Analysis & Improvement • Plant Performance Monitoring. • Plant Performance Improvement. 3

PI System Integration Jojobera Wind Trombay Haldia Hydro Maithon Mundra T&D Central PI Enterprise Systems

PI System - Network KHP BHVP BHIRA U4 U5 U6 U7 U8 U2 U1 U2 U3 U4 PH#6 EMS LD U1 U3 U5 HYDRO TROMBAY HALDIA JOJOBERA SERVER SERVER SERVER SERVER TATA POWER WAN WIND MUNDRA T&D SERVER MAITHON SERVER SERVER (DHARAVI) SERVER CENTRAL PI SERVER REGE ENERCON SUZLON SCADA DAS AMR U2 U1 U1 U2 U3 U4 U5 N MULSHI RDBMS PI-PI ABT BELGAUM LODHIVLI INTERFACE 3MW INTERFACE Legend MIS / HYPERION - Existing - Proposed

PI System Monitoring

PI System Monitoring

PI System Monitoring

ABT SYSTEM 2

ABT System ∗ Acquisition of data from ABT System deployed for Mumbai Operations. ∗ Accurate, reliable and wide acceptance. ∗ My SQL database in the backend. ∗ Web client with limited features available for monitoring. No tools for analysis purposes. ∗ Not possible to import data into excel. ∗ Report creation facility is not available.

Integration of ABT System with PI ∗ RDBMS Interface connectivity with PI server ∗ Immediate and time triggered data collection. ∗ Easier to import into excel sheet for analysis. ∗ Displays and trend making with ProcessBook. ∗ Report creation became error free and less time consuming ∗ Reports from PI being utilized by Load Control Centre for planning and forecasting.

AUTOMATED REPORTS 2

Automated Reports Daily operational and maintenance report on E-mail  Created Visual Basic application which utilizes PI DataLink to generate reports.  Reports are in MS Excel format. The values are available as static. Hence, no need of PI DataLink on the recipients computer.  Reports saved on a network drive.  E-mail the report using PI notifications to a list of recipients by setting time based triggers.

Automated Reports

ENHANCEMENT OF DATA QUALITY 2

Solution to Energy Meter commn. error Before  Energy meter communication problems during shutdown which results in error of the source data affecting calculations . 8.902E+07 TR U7 SATE 8.9014E+07 KWH 8.9E+07 8.898E+07 8.896E+07 8.894E+07 8.892E+07 8.89E+07 8.888E+07 8.886E+07 8.884E+07 8.882E+07 15/12/2010 12:00:00 AM 5.37 hours 15/12/2010 5:21:55 AM

Solution to Energy Meter commn. error After  Created a performance equation for storing the last reading of the energy meter and the same tag used for energy calculations . 8.9029E+07 TR U7 SATE 8.9014E+07 KWH 8.9E+07 8.895E+07 8.89E+07 8.885E+07 8.88E+07 15/12/2010 12:00:00 AM 5.37 hours 15/12/2010 5:21:55 AM

Energy Meter reset Before  Energy meter resets to zero after reaching the maximum of cumulative reading which affects energy calculations. Result : Negative

Energy Meter reset After  Added a condition in performance equations to check for a meter reset for the entire day and provided two methods of calculation using an IF clause. Modified PE IF (TAGVAL('TR U7 SATEC STG KWH NO ERROR','T')-TAGVAL('TR U7 SATEC STG KWH NO ERROR','Y'))<0 THEN (TAGMAX('TR U7 SATEC STG KWH NO ERROR','Y','T')-TAGVAL('TR U7 SATEC STG KWH NO ERROR','Y')+TAGVAL('TR U7 SATEC STG KWH NO ERROR','T'))/1000000 ELSE (TAGVAL('TR U7 SATEC STG KWH NO ERROR','T')-TAGVAL('TR U7 SATEC STG KWH NO ERROR','Y'))/1000000 Result : Accurate

Multiple interface status Before  Single tag for OPC interface status based on a particular watchdog. No direct monitoring over the data received from various functional areas. PI Interface OPC Server PI Server PI ICU OPC tag Watchdog tag Interface status

Multiple interface status After  Introduced the concept of multiple interface status tags using multiple watchdog from different functional areas.  Better monitoring over the plant.  Improved the availability of plant data.  Helps the plant engineers for immediate identification of the affected area. PI Interface OPC Server PI Server PI ICU OPC tag Watchdog tag Interface status OPC tag Watchdog tag OR OPC tag Watchdog tag OPC tag Watchdog tag

Trigger based PE Before After  Performance  Created equations Triggers for the was clock scheduled and it desired time.  Used was consuming more triggers for event computing power and triggering of the calculated memory. tags.  Calculations are performed  It was not possible to only onetime in a day which output results at a pre greatly reduced the system defined time. load and made the results accurate.

THERMAL POWER PLANT PERFORMANCE MONITORING 2

Performance Monitoring  PI Data used for Monitoring Plant Performance  “Before & After” Performance Analysis  Boiler & Turbine Efficiency Monitoring  Monitoring of Design vs. Actual Performance  Performance Analysis at different operating conditions  Dashboard / Report having Last Hour, Today, Last Day & Monthly data.

Performance Monitoring - “Before & After” Analysis

Performance Monitoring - Boiler & Turbine Efficiency

Performance Monitoring - Design vs. Actual

Performance Monitoring – Diff. operating conditions

Performance Monitoring - Dashboard

PERFORMANCE IMPROVEMENT INITIATIVE - SANKALP 2

SANKALP • Structured, Time bound, Team based program with top management support & bottom up approach to impact the company’s bottom line with minimal investment in shortest possible time. • Uses the creativity and energy of the people of Tata Power and all its stakeholders • Utilizes Online Data from PI System to take informed decisions. 3

Online Data for 500 MW Coal Fired Unit

History Data for analysis

Trend format for analysis MS Temp 555.0 12.0 550.0 10.0 545.0 8.0 540.0 Temp 6.0 535.0 4.0 530.0 2.0 525.0 520.0 0.0 4:25 PM 6:25 PM 8:25 PM 10:25 PM 12:25 AM 2:25 AM 4:24 AM 6:24 AM 8:24 AM 10:24 AM 12:24 PM 2:23 PM 4:23 PM 6:23 PM 8:23 PM 10:23 PM 12:22 AM 2:22 AM 4:22 AM 6:22 AM 8:22 AM 10:21 AM 12:21 PM 2:21 PM Date & time SH Temp SH Temp SH Spr

Trend format for analysis Last 100 Data only captured in Graph Unit 8 250.0 -400.0 230.0 -350.0 210.0 -300.0 190.0 -250.0 170.0 -200.0 150.0 -150.0 130.0 -100.0 110.0 -50.0 90.0 70.0 0.0 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 93 95 Load Cur Avg Suc Pr Avg

Predictive Analysis

Questions