neuco's maintenanceopt great catches

Great Catches:How Equipment HealthDetection & DiagnosticSystems are Improving

Great Catches:How Equipment HealthDetection & DiagnosticSystems are ImprovingPlant AvailabilityPlant Availability

Coal Gen 2010Ray JohnsonNeuCo, Inc

Optimization Software & Availability

� Optimization helps improve availability/reliability by:� Using predictive analytics to identify potential

equipment health issues� Providing early warnings for impending failures� Minimizing unnecessary outages� Minimizing unnecessary outages� Sponsoring better use of planned down-time� Reducing tube leak outages

Proprietary and Confidential

MaintenanceOpt®

� What it does:Automatically alerts users to symptoms of potential equipment healthissues (efficiency, capacity or reliability based) and allows quick easydrill down to uncover actionable responses to diagnose and remedythe underlying problems.

� Using:� Neural Networks


� Expert Rules� Other Optimizer Alerts

� To Improve:� Unit up-time� Speed of issue resolution� Efficiency

Equipment Types

� Air heaters� Auxiliary equipment (ESP, FGD, SCR, crusher, etc.)� Boiler

MaintenanceOpt may be used to detect problems in a fossilgenerating power plant associated with the following types ofequipment:


� Condenser� Fans and drives� Feedwater heaters� General (flows, heat rate, etc.)� Pulverizers� Pumps� Turbine

Examples of Potential Problems

The types of problems that can be detected aredependent upon the type of equipment. For example, forair heaters, the following provides a list of potentialproblems that may be identified:

� Inaccurate leakage calculation due to stratification or instrument error


g� Basket damage� Ash or salt plugging� Bearing inadequate oil cooling� Shaft misalignment� Bearing failure

Indicators of Potential Problems

� High AH air or gas side pressure drop� High AH cold end average temperature� Low AH cold end average temperature

The following process values associated with an airheater are monitored as part of an early warning system:


Low AH cold end average temperature� Air Heater leakage increasing� Increasing APH coil differential pressure� Bearing temperature increasing

Problem Identification


� Model is used to predict equipment performance� Model output is compared to actual performance� If difference is larger than a threshold, an anomaly is

detected

Diagnosis Flow Chart

A flow chart for diagnosinga potential problem isshown


How Alerts areManaged & Maintained


Managed & Maintained

NeuCo Customer Support

� Answer any questions related to the products� Provide software updates� Update system configuration� Provide weekly reports� Monitor for high urgency alerts


g g y� Monitor performance of the optimizers� Provide additional training� and more…

Automated Alerts Management

TriggerService

EmailComposer

ProcessLink at Site

EmailServer

CustomerSupportInterface

Customer Support

EmailServer


Service

HistoricalData

Composer at Site

CompositeTrigger

Database

Interfaceat NeuCo

Automated Reporting HelpsExtract Maximum Value

� KPIs Reflect Your Asset Management Metrics� EFOR/EFOR-D� Heat Rate (rolling average, vs. targets, etc.)� Emissions (cumulative tons, rolling average, vs. targets, etc.)� Revenue (vs. targets, historical benchmark, etc.)� Profitability (gross & net margins, vs. target, etc.)

B ilt I B h ki� Built-In Benchmarking� What benefits have been achieved?� Additional benefits that could be achieved� Changes in performance over time and implications

� Demystification Sheds Light on Key Processes� What optimizers are doing and why� Where the most important trade-offs are occurring� How are constraints limiting benefits and what can be done?


MaintenanceOpt Weekly Report


Example Recent Catches

� High ID fan motor temperature

� Low oil on a fan bearing

� Condenser air evacuation problem

� Increasing turbine bearing vibrationg g


MaintenanceOpt: ID Fan Motor Temperature

� Unit: PRPA Rawhide Unit #1, 300 MWs.� Usage: The following problem was identified

during a routine review of MaintenanceOptalerts.

� Incident: The temperature of an ID MotorStator Temperature had increased over apseveral week period. This issue wasrecognized by Customer Support. Earlywarnings were sent to the client about thisissue. Based on the warnings, the clientcleaned the filter and performance returned tonormal.

ID Fan Problem

� ID Fan problem is shown on the MaintenanceOptHome page.


Issues to Screen

The ID Fan problem is also shown on the MaintenanceOptIssues to Screen page along with the context data.


Alert DataReviewing the alert data, the problem had started threeweeks earlier and had been getting progressively worse.The client had been notified about this issue.


Diagnosing the Problem� By clicking “show causes”, we are able to investigate the

potential causes of the problem. The rule for Dirty Filter isshown below. Variables associated with this rule are highlightedincluding data for the other ID stator motor.


Identifying the Problem� The temperature of both ID fan motor stators are analyzed. The

temperature on ID Fan 102 is increasing while the temperatureon ID Fan 101 is normal. This is consistent with the rule for adirty filter.


Customer Support Sends Warning


Verifying the Problem

� An operations crew inspected the filter. It was foundto be dirty and is thus subsequently replaced. Thetemperature returned to normal.


Good Catch


Low Oil Example

� Unit: NRG Limestone Unit 2, 913 MW� Users: Maintenance Supervisor and Performance

Engineer.� Usage: Both work 4/10 shift Monday to Thursday.

They check MaintenanceOpt each morning whenthey are at site.

� Incident: Low oil on a primary air fan bearingsystem was found on Monday during routine login toMaintenanceOpt.

� Fix: Oil was added to the bearing system severalhours after the problem was identified. Thetemperature returned to normal.

Problem Alerted by MaintenanceOpt

� In a routine login to MaintenanceOpt, the users find an alert for abearing on a primary air fan. The bearing is at 174 degrees whileit is expected to be at 150 degrees.

Checking Alert� Chart shows the actual temperatures for all bearings on this fan.

Also shows expected value for the bearing of interest. Only thetemperature of bearing associated with the alert looks abnormal.

Actual vs. Expected� Actual and expected (neural network) value of applicable bearing

temperature shown. Incident starts Friday afternoon at around 1pm with nomaintenance personnel on site. Value stayed under the points DCS alarmtrigger value. This view is from Monday morning during a routine check.

Diagnosing the Problem

� One likely cause is inadequate lubrication or cooling. Rule clauses arehighlighted below. Temperature is increasing but vibrations are normal.

Fixing the Problem

� Customer fixes the low oil problem within hours of finding it. Temperaturereturns to normal.

Closing the Alert

� The alert is closed and a comment is posted in MaintenanceOpt’s database.

UserComment

Turbine Bearing Vibration

� Unit: APS Four Corners Unit #4, 800 MW� Usage: Routine checks of MaintenanceOpt made

daily� Incident: During a routine check, it was observed

that a turbine vibration was increasing. Noimmediate action was necessary but problem neededt b it d d ilto be monitored daily.

� Fix: Problem was fixed at the next outage

Vibration Alert� Vibration on bearing #7 has triggered. Note that alert had been

triggering >50% of the time over the past week and month.


Increasing Vibration� Turbine bearing vibration increasing for past month (green). Neural model

adapting to the change but not as fast as vibration increase (blue).


Monitoring the Problem� User has flagged problem so that it stays on the Issues to Screen tab. The

user also entered a comment. Monitoring continued until next outage.

Flagged


ggitem

UserComment

Benefits of ProcessLink® Integration

� PerformanceOpt indicates equipment degradation through high-fidelitymodel of thermal processes

� MaintenanceOpt indication of equipment degradation through empiricalmodels of on-line signals as well as potential problems detected throughmanual inspection

� Some problems that would escape notice through one or the other


� Some problems that would escape notice through one or the othermethodology become obvious when viewed through both

� MaintenanceOpt also surfaces problems detected by NeuCo’s closed-loop optimizers

� PerformanceOpt informs closed-loop optimizers though virtual on-lineanalyzers indicating important boiler performance parameters not directlymeasured� Coal quality, boiler efficiency and heat rate for CombustionOpt� Boiler cleanliness and heat rate for SootOpt

Condenser Problem

� Unit: APS Four Corners Unit #2, 190 MW

� Usage: During a check of PerformanceOpt andMaintenanceOpt by NeuCo, we noticed a largechange in condenser performance. Plant was notifiedvia email of the problem.

� Incident: During routine maintenance on thed th i t h d i d t tlcondenser, the maintenance crew had inadvertently

disconnected both vacuum pumps causing a largeincrease in backpressure

� Fix: The maintenance crew was notified of theproblem and then properly connected a vacuumpump. Th.e performance returned to normal

M’Opt and P’Opt In Action� Routine review of triggers show alerts on condenser backpressure and

cleanliness have triggered. Backpressure is at 3.05 when expected to be2.2, and cleanliness is at 0.55 when expected at 0.88.

BackpressureAlert from M’Opt


CleanlinessAlert from P’Opt

Condenser Cleanliness

� Condenser Cleanliness significantly changes fromexpected at around 10 am.


Condenser Backpressure

� Condenser backpressure also significantly changesat the same time.


Diagnosis

� Low condenser cleanliness and high condenser backpressureindicates inadequate air removal equipment performance.


Resolution� Maintenance crew reconnects the vacuum pump and

performance returns to normal.


Questions?


neuco's maintenanceopt great catches

Business

type of equipment

confidential model

optimizer alerts

following types ofequipment

unnecessary outages

forair heaters

actual performance

remedythe underlying