About Duke Energy M&D Center Scope of Work There are separate - - PowerPoint PPT Presentation

M&D Center

About Duke Energy

M&D Center Scope of Work

• There are separate M&D Centers for the regulated and the commercial

sides of the business.

• The M&D Center for regulated generation is divided into three functional

areas:

• Thermal Performance and Testing
• Real time performance monitoring and analysis
• Field testing and support
• Dispatch heat rate curve automation
• Reliability and CBM
• CBM process development and ownership (EPRI Maintenance and Reliability Model)
• CBM analysis and support for the fleet
• Advanced Pattern Recognition Monitoring Program
• SmartM&D (new sensors monitored by APR group)
• Scope and project development
• Installation and implementation

M&D Center Thermal Performance Team Mission

• We help the plants maintain a competitive position in the market by

providing real time heat rate monitoring, on-site testing and in-depth analysis of plant performance.

• We use first principles modeling software to monitor plant and system thermal
• performance. The software compares actual plant efficiency to their desired targets.
• We provide turnkey expertise and resources to test equipment and system

performance on site. Our service ranges from individual components to total systems.

• We partner with the stations to help them understand what factors are hurting

performance and what they can do to restore it to the desired level.

• We perform studies for the plants to determine the efficiency and operating

performance impact of proposed changes in the plant configuration.

• We ensure that the unit heat rate curves used for dispatching accurately reflect unit

performance and seasonal conditions.

Real Time Cycle Performance Summary

Cycle Performance:

• Heat balance
• Component Drill

Down

Standard CT Screens

Performance Parameters:

• Gross MW
• Heat Rate
• Compressor Efficiency
• Pressure Ratio

Real Time Controllable Loss Summary

Controllable Loses:

• MW Impact
• $/hour

Compressor Performance Curve

Baseload Efficiency:

• Red – Target
• Yellow – Clean
• Blue - Actual

BOP Equipment - Pump Performance Curve

• Performance of

equipment can be monitored

• Design head and

design efficiency are compared to current

• peration

Dispatch Curve Automation Project

• Moves from an annually

updated dispatch curve to being performance driven

• Uses thermodynamic

modeling software to estimate energy input to cycle

• Reflects current equipment

performance and ambient conditions.

Trend of Input-Output Curves VirtualPlant vs. EtaPRO Calculation Yellow – Dispatch Curve Cyan – Current Operation

M&D Center Thermal Performance Deliverables

• Real-time controllable loss and cycle performance reports will be readily

available to any plant user.

• Monthly controllable loss summaries will be provided at the unit level and

the region level.

• Dispatch curve review and adjustment is triggered when the notified and

the real time heat rate curves deviate by 1% or more.

• The center is developing more rigorous cooling tower and heat

exchanger models.

• The virtual plant model is capable of “what-if” parametric studies.

M&D Center APR Team Mission

• Our mission is to help the plants improve long term reliability and

cost by predicting equipment problems that will evolve from fault to failure 24 hours or more into the future.

• We use Advanced Pattern Recognition software to monitor plant equipment
• peration. The software detects subtle deviations from normal operation that

can be used as early indicators of future problems.

• We partner with the stations and fleet technical support to capture their

knowledge of the equipment. We build this knowledge into our models to free the plants from repetitive monitoring. This allows them to focus on emerging issues instead.

• We are part of the fleet-wide Maintenance and Reliability process that

manages risk and reliability at the lowest possible cost.

Evolution of the APR Program

• Legacy Progress
• Started with the CT fleet in 2004
• Expanded to the fossil fleet in 2007
• Process Data – Prism APR tool
• Legacy Progress – 27 Fossil Units (now

15), 10 CC’s, 140+ CT’s

• >3500 APR Models
• >35,000 points monitored every 5

minutes

• Dedicated resources assigned to APR

and Thermal Performance monitoring

13

Evolution of the APR Program

• Expansion After the Duke merger
• Legacy Duke – 29 Fossil Units, 4 CC’s,

63 CT’s, 8 PS, 9 Hydro

• >3000 APR Models
• >30,000 Points monitored every 5

minutes

• Regional Locations, Charlotte, Raleigh,

Indianapolis

• 5 team members dedicated to APR. Mix
• f new employees, field support and plant

experience.

• Roughly 35 unique model types

14

Evolution of the APR Program

15

APR Philosophy

• Maintain credibility by reviewing and screening all APR alarms before

sharing with the stations

• Focus on reliability finds
• Avoid any commitment to operational finds that would evolve in less than 24 hours
• We are not an extension of the operators. We are an extension of CBM.
• Integrate with the plant Maintenance and Reliability process

16

• Vibration
• Thrust Position
• Bearing Temps
• Oil Temps
• Disc Cavity Temps
• Cooling Air Valve

Positions

• Rotor Cooling Air

Temp

• MW Output
• Ambient Temp
• Humidity
• Lube Oil Parameters
• Generator Data

• Exhaust Temps
• Exhaust Pressure
• Examples of data available
• Compressor Pres
• Compressor Temps
• Gas Temps
• Gas Pres

• Pull raw data from Pi
• Typically data sets will be 5 min samples for 1 year

APR Modeling Process – Cleaned Data Set

• Cleaned data set represents operation during all ambient / MW loading conditions

Summer Winter Spring Fall

• Feed historical data to algorithm
• Feed Real-time values to algorithm every 5 minutes
• Algorithm compares Real-time with historical and outputs

predicted values

Predicted values

• Fan OB Bearing Temp
• Fan IB Bearing Temp
• Motor IB Bearing Temp
• Motor OB Bearing Temp
• Motor Winding Temp
• Amps
• Discharge Press

APR Modeling Process – Model Algorithm

First Pattern Recognition alarm, residual exceeded 10 C

• Subtract actual value from the predicted value to calculate the residual

Traditional Alarm @ 140 C

M&D Center Issue Notification Process

• In the long term, the M&D Center will initiate Tech Exams for all finds as

part of the EPRI Maintenance and Reliability Process. Owners at the stations will utilize these Tech Exams to determine what needs to be done to the equipment.

• In the short term, the M&D Center uses Email and a notification matrix to

determine who to notify at each station.

• The center is developing a report tool that can customize reports to users

based on attributes such as region, unit, system, component or technology type.

• The Thermal Performance group issues monthly Controllable Loss

reports for major plant systems and provides real time thermal performance displays to the plants.

• Performance testing groups provide turnkey testing and analysis for a

wide range of performance issues.

CR 1&2 Fire & Major Damage – Jan 2010

• Major Catastrophic Failure
• $MM’s+ 4 months loss generation
• SmartM&D Technology could have

prevented this failure

23

$200 RF Bus Temp Device

• n SmartM&D Network

Existing Program Gaps

• Equipment Specialists
• Current specialists spread too thin – hard to replace
• Current manual predictive programs
• 80% data collection, 20% diagnostic
• 60,000 manual vibration data collections per month!!
• M&D Center Limitations
• Currently limited only to DCS available data
• Need more instrumentation to expand capabilities

Industry Gap – lack of cost effective automated

monitoring technologies on critical equipment.

• 2010 Progress Energy Executive Challenge: to better leverage technology to address

increasing reliability demands and to optimize our workforce

• 2012 Duke Energy Long Term Enterprise Strategy: Modernize generation and grid

infrastructure

• 2013 IT Strategic Goal: Enable Business Model Evolution
• Build an advanced M&D infrastructure capable of

a significant advancement in remote equipment monitoring, smart diagnostics, and prognostics

• Provide a “Force Multiplier” to leverage fewer

specialists to analyze fleet critical equipment

Shaping our Future – Technology Innovation & Workforce Optimization

SmartM&D Strategy

Evolution of the APR Program – SmartM&D

• Additional Expansion planned for

Phase 2 (SmartM&D) implementation

• >5000 APR Models for SmartM&D
• Advanced Sensor Types

– Vibration – Temperature – Electromagnetic Interference (EMI) – Motor Diagnostics – Thermography – Visual Cameras – Additional sensors for performance monitoring (press, flow, etc.) – National Instruments Compact Remote I/O Devices (CRIO)

• >65 new model types

26

Avoided Cost Chart Tracks Value

M&D Find – CT Combustion Hardware Failure

• Blade Path temperature spread

began to increase

• Spread increase was caused by 1
• f 16 thermocouples trending low

Filtered Data > 40 MW 4/22 – 5/16/2012 Blade Path Spread Blade TC #2

M&D Find – Boiler Feed Pump Motor Bearing

• Increasing temperature detected on NDE Motor bearing on BFP (alert set to +/-

15F)

• Site contacted with issue
• PdM tech determined the oil flow was restricted to the bearing
• The diverter stack was closed on the CT and the pump was shutdown

Site contacted and pump was shutdown

M&D Find - Boiler Feed Pump Motor Bearing

• After pump was shutdown, a ball of Teflon tape was found in the oil supply line
• rifice (possible construction waste)
• Quick action by the site saved the unit from a possible trip or bearing damage to

the motor

• The diverter damper was shut on the unit and the CT stayed online in simple

cycle mode while the issue was corrected

Teflon tape found in the supply line orifice

32

M&D Find - CT Generator ‘J strap’ failure

• APR detected a 2X shift in vibration on all the

generator proximity and seismic probes

• Occurred on the last run before the Fall outage
• DCS does not alarm until 8.66 mils on the Bently

proximity probes Typical Run High Vibration Run

X - Normal X - After Y - Normal Y - After Gen Brg1 0.95 2.60 0.56 1.10 Gen Brg2 0.70 1.81 0.81 1.40

M&D Find - CT Generator ‘J strap’ failure

• Site decided to remove manhole cover to inspect (not in
• riginal outage scope)
• Both J-straps were found damaged
• Site decided to remove field and repair J-straps
• Had issue not been addressed, high probability of failure
• f J-Strap during winter run season

2 Visible Leads 1 Visible Leads

M&D Find - CT Generator ‘J strap’ failure

• Found shorted #1 coil, shorted #6 coil

M&D Find – Gore Filter Performance Analysis

• Conventional filters at one CC site required frequent washing due to

contamination from the cooling tower

• Gore filters were tested as an alternative
• M&D Center compared filter performance over time to find that the Gore

filters were cost justified in this application

M&D Find – CC Ambient Air Correction Factor

• Third Party PPA supplier to Duke
• Performance test based on

loosely related ambient air sensors

• M&D review of test results

revealed the need for additional ambient air and RH temp sensors.

• Station installed the sensors

identified by the M&D Center

• The second test was satisfactory

and Duke accepted the results.

• The revised ambient temperature

will reduce our annual capacity costs from the plant by $250K.

36