Proceedings of the ASME 2016 Power Conference POWER2016 ...

8
The Application of Smart, Connected Power Plant Assets for Enhanced Condition Monitoring and Improving Equipment Reliability Michael Reid Duke Energy Charlotte, NC, USA Bernie Cook Duke Energy Charlotte, NC, USA ABSTRACT The U.S. electric utility industry continues to undergo dramatic change due to a number of key trends and also prolonged uncertainty. These trends include: Increasing environmental regulations uncertainty Natural gas supply uncertainty and price Economic / decoupling of electricity demand growth from GDP Aging coal and nuclear generation fleet / coal retirements Aging workforce Increasing distributed energy resources Increasing customer expectations The transformation ultimately demands significant increases in power plant generation operating capabilities (e.g. flexibility, operating envelop, ramp rates, turn-down etc.) and higher levels of equipment reliability, while reducing O&M and capital budgets. Achieving higher levels of equipment reliability and flexibility, with such tightening budget and resource constraints, requires a very disciplined approach to maintenance and an optimized mix of the following maintenance practices: Reactive (run-to-failure) Preventive (time-based) Predictive (condition-based) Proactive (combination of 1, 2 and 3 + root cause failure analysis) Many U.S. electric utilities with fossil generation have adopted and implemented elements of an equipment reliability process consistent with Institute of Nuclear Power Operations (INPO) AP-913. The Electric Power Research Institute has created a guideline modeled from the learnings of AP-913, that consists of six key sub-processes [1]: 1. Scoping and identification of critical components (identifying system and component criticality) 2. Continuing equipment reliability improvement (establishing and continuously improving system and component maintenance bases) 3. Preventive Maintenance (PM) implementation (implementing the PM program effectively) 4. Performance monitoring (monitoring system and component performance) 5. Corrective action 6. Life cycle management (long-term asset management) A significant proportion of Duke Energy’s coal fleet is of an age where individual components have reached their design intent end-of-life thereby creating an increased need for performance monitoring. Until recent times this was largely performed by maintenance technicians with handheld devices. This approach does not allow regular data collection for trending and optimization of maintenance practices across the fleet. Significant and recent advances in sensor technology, microprocessors, data acquisition, data storage, communication technology, and software have enabled the transformation of critical power plant assets such as steam turbines, combustion turbines, generators, transformers, and large balance-of-plant equipment into smart, connected power plant assets. These enhanced assets, in conjunction with visualization software, provide a comprehensive conditioning monitoring solution that continuously acquires sensory data and performs real time analysis to provide information and insight. This advanced condition monitoring capability has been successfully applied Proceedings of the ASME 2016 Power Conference POWER2016 June 26-30, 2016, Charlotte, North Carolina POWER2016-59189 1 Copyright © 2016 by ASME

Transcript of Proceedings of the ASME 2016 Power Conference POWER2016 ...

Page 1: Proceedings of the ASME 2016 Power Conference POWER2016 ...

The Application of Smart, Connected Power Plant Assets for Enhanced Condition Monitoring and Improving Equipment Reliability

Michael Reid Duke Energy

Charlotte, NC, USA

Bernie Cook Duke Energy

Charlotte, NC, USA

ABSTRACT

The U.S. electric utility industry continues to undergo

dramatic change due to a number of key trends and also

prolonged uncertainty. These trends include:

Increasing environmental regulations uncertainty

Natural gas supply uncertainty and price

Economic / decoupling of electricity demand growth from

GDP

Aging coal and nuclear generation fleet / coal retirements

Aging workforce

Increasing distributed energy resources

Increasing customer expectations

The transformation ultimately demands significant

increases in power plant generation operating capabilities (e.g.

flexibility, operating envelop, ramp rates, turn-down etc.) and

higher levels of equipment reliability, while reducing O&M

and capital budgets. Achieving higher levels of equipment

reliability and flexibility, with such tightening budget and

resource constraints, requires a very disciplined approach to

maintenance and an optimized mix of the following

maintenance practices:

Reactive (run-to-failure)

Preventive (time-based)

Predictive (condition-based)

Proactive (combination of 1, 2 and 3 + root cause failure

analysis)

Many U.S. electric utilities with fossil generation have

adopted and implemented elements of an equipment reliability

process consistent with Institute of Nuclear Power Operations

(INPO) AP-913. The Electric Power Research Institute has

created a guideline modeled from the learnings of AP-913, that

consists of six key sub-processes [1]:

1. Scoping and identification of critical components

(identifying system and component criticality)

2. Continuing equipment reliability improvement (establishing

and continuously improving system and component

maintenance bases)

3. Preventive Maintenance (PM) implementation

(implementing the PM program effectively)

4. Performance monitoring (monitoring system and

component performance)

5. Corrective action

6. Life cycle management (long-term asset management)

A significant proportion of Duke Energy’s coal fleet is of an

age where individual components have reached their design

intent end-of-life thereby creating an increased need for

performance monitoring. Until recent times this was largely

performed by maintenance technicians with handheld devices.

This approach does not allow regular data collection for

trending and optimization of maintenance practices across the

fleet.

Significant and recent advances in sensor technology,

microprocessors, data acquisition, data storage, communication

technology, and software have enabled the transformation of

critical power plant assets such as steam turbines, combustion

turbines, generators, transformers, and large balance-of-plant

equipment into smart, connected power plant assets. These

enhanced assets, in conjunction with visualization software,

provide a comprehensive conditioning monitoring solution that

continuously acquires sensory data and performs real time

analysis to provide information and insight. This advanced

condition monitoring capability has been successfully applied

Proceedings of the ASME 2016 Power Conference POWER2016

June 26-30, 2016, Charlotte, North Carolina

POWER2016-59189

1 Copyright © 2016 by ASME

Page 2: Proceedings of the ASME 2016 Power Conference POWER2016 ...

to obtain earlier detection of equipment issues and failures and

is key to improving overall equipment reliability.

This paper describes an approach by Duke Energy to create

and apply smart, connected power plant assets to greatly

enhance its fossil generation continuous condition monitoring

capabilities. It will discuss the value that is currently being

realized and also look at future possibilities to apply big data

and analytics to enhance information, insight, and actionable

intelligence. INTRODUCTION

The “True North” of Duke Energy’s Fossil Hydro

Operations (FHO) is Event Free, Reliable, and Cost Effective

operations. This is becoming increasingly more challenging due

to transformation of the electric utility industry that Utility

Dive summarized recently [2]:

1. Utility business models are changing

2. Utilities becoming more customer-centric

3. Utilities buying into storage

4. Utilities are modernizing the grid

5. Debates over rate design reforms and value of DERs are

heating up

6. Utilities getting in on the solar game

7. Utilities face growing load defection

8. Renewables reaching grid parity

9. Natural gas is growing fast

10. Coal power in decline

The impacts to FHO are far ranging, and probably not yet

fully understood. In the near-term FHO is seeing an aging

fossil fleet →1 and numerous coal plant retirements →2.

Layered on top is an aging workforce and limited O&M, capital

and staff resources.

Operationally, FHO is seeing significant and increasing

challenge too, driven by abundant, cheap natural gas and

increasing amounts of intermittent distributed energy resources.

Today, the coal fleet is being dispatched differently, with many

units seeing increased cycling, lower capacity factors, and a

need for flexible operation.

Across all industries there is considerable technology

advancement. This involves nine foundational technologies [3]:

Autonomous robots, simulation, horizontal and vertical system

integration, the industrial internet of things, cybersecurity, the

cloud, additive manufacturing, augmented reality, and big data

and analytics.

This fundamental change in technology presents an

opportunity to overcome present and future industry challenges

and remain relevant in the energy arena. The opportunity

discussed in this paper is its application to asset condition

monitoring, an essential component of an effective maintenance

strategy and equipment reliability program.

MAINTENANCE STRATEGY AND EQUIPMENT RELIABILITY PROGRAM

Duke Energy uses a holistic approach for effective asset

management that is focused around three key elements: People,

Processes and Technology →3. Duke Energy’s FHO

maintenance strategy, a key component of effective asset

management, seeks an optimal balance between under

maintaining and over maintaining its plant assets, that results in

lowest operating and maintenance costs. A Condition-Based

Maintenance approach →4 is applied that seeks to optimize a

mix of failure-based, preventive, predictive, and proactive

maintenance.

1 Duke Energy’s Fossil Fleet Commission/Retirement History

2 Duke Energy’s Coal Retirement Landscape

Note: Each block within a bar represents a coal plant

2 Copyright © 2016 by ASME

Page 3: Proceedings of the ASME 2016 Power Conference POWER2016 ...

3 Holistic Approach to Asset Improvement

4 Condition-Based Maintenance Approach

The FHO equipment reliability program is applied to

execute the maintenance strategy and is based on EPRI process

guidelines [1] that have been developed over several years in

partnership with the fossil power generation industry.

The programs brings together six key sub-processes →5:

1. Scoping and identification of critical components

(identifying system and component criticality)

2. Continuing equipment reliability improvement (establishing

and continuously improving system and component

maintenance bases)

3. Preventive Maintenance (PM) implementation

(implementing the PM program effectively)

4. Performance monitoring (monitoring system and

component performance)

5. Corrective action

6. Life cycle management (long-term asset management)

5 EPRI’s Equipment Reliability Process [1]

The foundation of the maintenance strategy is to define and

apply system and component criticality criteria such that each

unique system or component is assigned to one of three

classifications:

Classification Description

Critical Function is so vital that all efforts are made to prevent all failures that are known to occur.

Important

Non-critical components that can impact operations and cost-effective methods are used to maintain component health; typically time-based or condition-based monitoring tasks.

Run-to-Failure These components do not impact event free operations and it is therefore acceptable to run to failure.

Another key element to the maintenance strategy is

monitoring system and component health. This is done by a

variety of processes and technologies:

Equipment Assessments (collection of data and analysis

such as vibration, oil analysis, flow, pressure etc.)

Equipment Inspections (major plant equipment such as

boilers, turbines, large balance-of-plant)

Condition monitoring technology applied to directly

measure mechanical, electric or thermodynamic parameters)

On-line Condition monitoring (continuous monitoring of

asset performance through sensor data)

Developments with on-line condition monitoring

technology are making this technology much more affordable

and therefore accessible. This offers the opportunity to replace

manual collection of data with greater focus on analysis →6.

AssetsCritical Station Assets: Turbines / GeneratorsBoilers / Combustion

BOP: Electrical / SwitchyardBOP: Other

People

Processes Technology

3 Copyright © 2016 by ASME

Page 4: Proceedings of the ASME 2016 Power Conference POWER2016 ...

6 Monitoring System and Component Health

ENHANCED CONDITION MONITORING

As mentioned earlier, this is a time of significant

technology advancement. It is referred to by some as Industry

4.0 [4,5] and is described as a fourth wave of technological

transformation, essentially, the fourth industrial revolution →7. 7 Industry 1.0 to Industry 4.0

Industry 4.0 can be summarized by advances in materials

(including nanotechnology), advanced manufacturing

(including additive manufacturing) and learning ability (due to

advances in computing and communications technology).

Schwab describes it thus, "It is characterized by a much more

ubiquitous and mobile internet, by smaller and more powerful

sensors that have become cheaper, and by artificial intelligence

and machine learning" [6].

For the power generation industry the opportunity now

exists to combine sensors, advances in computing, data

acquisitions, data storage and software with critical power plant

assets such as boilers, steam turbines, combustion turbines

generators, transformers, and critical balance-of-plant

equipment. The resulting “smart, connected power plant assets”

→8 have intelligence and connectivity that enable an entirely

new set of functions and capabilities [7].

The technology advances discussed above have created

microelectromechanical systems (MEMS), that are routinely

used in sensors resulting in cheaper and more reliable designs

with advanced capabilities. This offers great opportunity to

expand on-line monitoring capabilities to support condition-

based maintenance, enhance safety and reliability, and improve

asset performance and utilization. A summary of on-line

condition monitoring technologies applied to FHO assets is

presented in →9.

8 Smart, Connected Power Plant Assets

Plant Assets Smart Power Plant

Assets Smart, Connected

Power Plant Assets

Physical components e.g. boilers, steam

turbines, combustion turbines generators,

transformers, and critical balance-of-plant equipment: pumps and

motors etc.

Sensors, microprocessors, data

acquisitions, data storage, controls,

software, embedded operating system,

enhanced user interface etc.

Ports, antennae, protocols enabling

wired / wireless connections with plant asset: one-to-one, one-

to-many, many-to-many.

FHO is currently in the midst of an initiative, working with

EPRI and National Instruments, to establish a platform of

smart, connected critical power plant assets. The architecture of

this platform is outlined in →10. A key piece to the platform is

utilization of wireless communication technology to minimize

cost.

9 On-Line Condition Monitoring Technologies

4 Copyright © 2016 by ASME

Page 5: Proceedings of the ASME 2016 Power Conference POWER2016 ...

10 Smart, Connected Power Plant Asset Platform

MONITORING & DIAGNOSTICS (M&D) CENTER PERFORMANCE

The FHO Monitoring & Diagnostics (M&D) Center is a key

component to the equipment reliability program. Almost all

equipment gives off early warning signals before it fails. These

warning signals can be detected with condition monitoring

technologies and provides time to plan, schedule and make

repairs. This greatly reduces the probability of significant

failure, and also the cost of equipment failure.

On a daily basis the FHO M&D Center receives continuous

data on asset performance and health. Thousands of Advanced

Pattern Recognition (APR) models scan the data and provide

alarm notification when conditions on an asset deviate from

expected behavior. The M&D Center conducts preliminary

investigation, and if the alarm condition checks out, a

Notification is made to the station. A Notification is where an

abnormal condition is detected and there is interaction between

the station and the Monitoring & Diagnostics center to

investigate further. An investigated Notification, that identifies

an equipment issue that requires corrective action, is referred to

as a Find.

Advanced notification of equipment issues through

detection of potential failure symptoms, has substantial benefit,

allowing prevention of a full or partial functional failure, where

the asset fails to perform a required function. This also

potentially allows optimal repair scheduling and minimizes

operational impacts and repair costs.

The smart, connected power plant platform was initiated in

2012 and began providing on-line condition monitoring

capability at the start of 2013. The number of Notifications and

Finds, and the associated trend is outlined →11.

There has been a steady exponential growth in Finds over a

three year period as more sensors have been added to the smart,

connected power plant platform →12.

Cost avoidance savings are a result of early detection of

equipment failure and avoidance of sudden failure that typically

carry a high cost to repair, lost electricity generation and also

potentially leading to safety and/or environmental-related

events.

Assuming equipment failure is discovered at a point P that

occurs before point F on the P-F curve →13, it is possible to

estimate failure cost avoidance by estimating the results of

better managing the risk. In this example provided, if the failure

is discovered at point P2, it can be seen that there is potentially

significant cost difference (Δ$) between early detection and

corrective action, compared to allowing the component run to

failure.

5 Copyright © 2016 by ASME

Page 6: Proceedings of the ASME 2016 Power Conference POWER2016 ...

11 M&D Notifications vs. Finds

12 M&D Find History Trend

More accurate estimation of cost avoidance savings also

considers the use of a Risk Grid →14 to consider both the

probability and impact of likely failure scenarios. The approach

taken uses elements from the methods described by Cook and

Muiter [8] and EPRI [9]. Failure cost avoidance is estimated for three failure

scenarios that span across a likely range of scenarios: minimal,

significant and catastrophic. Probability of occurrence is

assigned to the three scenarios and is based on consideration of

the likely outcome if the M&D Center did not find the problem.

Historical information about specific equipment, impacts of

failure and repair costs are used to provide system and cost

impact information. Once known, the actual cost of the

corrective action is subtracted from the total of the three “most

likely” scenarios.

In the example provided below →15 high vibration was

detected on two bearings of a low pressure steam turbine. The

risk of catastrophic failure is very low because operations

would be alerted to high vibration issues once vibration levels

got to the alarm threshold limits. The likely scenarios in this

case range from minor bearing damage to complete destruction

of the low pressure rotor. In the end, early detection resulted in

a balance shot being applied during the next outage opportunity

that incurred minimal actual cost. 13 P-F Curve: Interval from Potential Failure to Functional Failure

P0 Pending failure detected (Potential Failure)

P1 1-9 months, PdM (predictive maintenance); ultrasound, vibration etc.

P2 1-6 months, PdM; oil analysis

P3 3-12 months, PdM; thermography

P4 5-8 weeks, PM (preventive maintenance)

P5 1-4 weeks, audible noise

P6 1-5 days, heat by touch

P7 0-0 days, smoke

F Actual Failure

Δ$ Differential cost (impact and repair) between equipment functional failure and early detection of failure at P2, and ability to take timely corrective action.

2013

2014

2015

6 Copyright © 2016 by ASME

Page 7: Proceedings of the ASME 2016 Power Conference POWER2016 ...

14 Risk Grid

15 Cost Avoidance Example for Steam Turbine Find

CONCLUSION

The smart, connected power plant platform installed on

Duke Energy’s Fossil Hydro Operations fleet is starting to

show strong promise of value →16, with significant numbers of

Notifications and Finds that have led to significant cost

avoidance →17.

This is very encouraging since there is still much of the

vision →18 to be realized. The vision is to generate and collect

much more data than was previously possible. There will be

much less manual collection of data and analysis will be

automated where possible. FHO is currently transitioning from

installation of hardware to deployment of National Instruments

NI InsightCM™ Enterprise for conditioning monitoring. This is

a software solution that allows maintenance specialist to

manage and visualize data and have critical information in their

hands when they need it. The M&D Center, system owners,

and subject matter experts will have considerably more insight

into the health of critical equipment and can use this to

optimize machine operational and maintenance performance, as

well as increased safety.

16 Value of Smart, Connected Power Plant Assets

Smart, Connected Power Plant Assets Supports FHO "True North"

Event Free Operations

Improved safety through replacing manual collection of data in hazardous areas / conditions.

Increased environmental monitoring capability and frequency in difficult to access areas.

Early detection of equipment issues can avoid significant or catastrophic equipment failure.

Reliable

Early detection of equipment issues allows for optimized time of repair and minimized downtime.

Much more effective identification and troubleshooting of equipment issues due to increased availability of data and enhanced analysis capability. This helps increase output and reduces downtime. Also reduces need to derate units whilst waiting on analysis.

Cost Effective

Cost avoidance through early identification of equipment failure and abnormal operating conditions.

Remote diagnostic analysis saves time and reduces cost.

Increased output and fuel savings due to ability to effectively identify performance related issues.

Increased online monitoring capability without resource / O&M increase.

Reduction of insurance premiums due to enhanced on-line monitoring capabilities.

Workforce Strategy

Predictive maintenance employees spent much less time on manual collection of data and increased time on analysis.

Increasing pressure on resources impresses need for using data and analysis for enhanced decision making.

17 Cost Avoidance History

Minimal Significant Catastrophic Actual

Probability of Occurrence 94.5% 5.0% 0.5%

SYSTEM IMPACT

Replacement Power Cost $/MWh 7 7 7 7

Reduction in Output MW 1,200 1,200 1,200 0

Time Offline Hrs 48 168 1,440 0

Avoided System Costs 381,024 70,560 60,480 0

Total $

O&M IMPACT

Material Cost $ 20,000 250,000 30,000,000 0

Repair Cost $ 0

Labor Cost $ 0

Other $ 5,000 20,000 1,000,000 0

Avoided O&M Costs 23,625 13,500 155,000 0

Total $

Total Avoided Costs $

512,064

704,189

192,125

7 Copyright © 2016 by ASME

Page 8: Proceedings of the ASME 2016 Power Conference POWER2016 ...

18 Data → Information → Insight → Actionable Intelligence

It is clear that the integration of critical power plant assets,

data, processes, and people will provide understanding not

previously possible, and will advance informed and improved

decision making. However, at this time the path to information

and insight (hindsight) is much clearer than the path to

actionable intelligence (foresight) that involves predicting the

probability of future outcomes and taking action based upon

predicted outcomes [10].

REFERENCES [1] EPRI, “Developing an Equipment Reliability Program

Model”, EPRI 3002001348, February 2015.

[2] G. Bade, “The Top 10 Trends Transforming the Electric

Power Sector”, Utility Dive, September 17, 2015.

[3] M. Rüßmann, M. Lorenz, P. Gerbert, M. Waldner, J.

Justus, P. Engel, M. Harnisch, “Industry 4.0 The Future of

Productivity and Growth in Manufacturing Industries”,

The Boston Consulting Group, April 2015.

[4] M. Hermann, T. Pentek, B. Otto, “Design Principles for

Industrie 4.0 Scenarios: A Literature Review”, Working

Paper No. 01 / 2015.

[5] M Krueger, R. Drath, H. Koziolek, Z. Ouertani, “A New

Era”, ABB Review 4|14, pp. 70-75, 2015.

[6] K Schwab, "The 4th Industrial Revolution", January 2016.

[7] M. Porter, J. Heppelmann, “How Smart, Connected

Products Are Transforming Competition”, Harvard

Business Review, November 2014.

[8] M. Cook, M. Muiter, “Estimating Failure Avoidance

Costs”, uptime oct/nov 11, 2011.

[9] EPRI, "Predictive Maintenance Program Development and

Implementation", EPRI TR-108936, January 1998.

[10] J. Cutts, “Ghost in the Machine: The Predictive Power of

Big Data Analytics”, Technology Trends to Watch 2015,

pp. 5-10, Consumer Electronics Association, 2015.

ACKNOWLEDGEMENTS

The authors greatly appreciate the dedication and technical

expertise of the members of Duke Energy's Maintenance &

Diagnostics Team. Also, the sponsorship of Charlie Gates,

senior vice-president and chief fossil-hydo officer. Collectively,

this has enabled the SmartGen vision (Smart, Connected Power

Plant Assets) become reality.

Duke Energy would like to acknowledge the contributions

of the Electric Power Research Institute (EPRI), as well as

several other key industry partners in the development and

application of processes and technology necessary to enable

these smart, connected assets in the Duke fleet of bulk power

generation assets. EPRI has led an industry consortium in

defining and developing the equipment reliability process

outlined in this paper, and has been instrumental in supporting

the application of this equipment reliability process across the

existing Duke Energy FHO fleet. Additionally, EPRI has

provided guidance and oversight around the development of the

strategic enterprise platform for online, real-time condition

monitoring now in place at Duke Energy. This includes things

such as identification of new sensor opportunities, guidelines

on signal processing and signal assessment, monitoring and

alarming strategies, cybersecurity implications, data

management, and other related activities – all of which falls

under EPRI’s Integrating Information for Insight and

Intelligence for Generation (I4GEN) initiative aimed at shaping

and realizing the future of power generation. Finally, special thanks to National Instruments for being the

ideal technology partner for the entire SmartGen journey. Early

in the project Duke recognized a critical gap in the market

offerings for enterprise condition monitoring software that was

based on accepted industrial standards while being both open

and scalable. National Instruments stepped up to the challenge

and quickly brought InsightCM™ to market to address Duke's

challenges. Innovations such as this are not new to NI - for

almost 40 years, NI has helped scientists, engineers, and

companies like Duke Energy understand their world and make

better business decisions though quantitative discovery. The NI

technology platform referenced for smart, connected power

plant assets is a prime example of how the Industrial Internet of

Things (IIoT) is improving operational efficiency through

sensor data, edge processing, analytics, and enterprise software.

Solutions such as this, built on open, off the shelf platforms,

help technology keep up with evolving challenges and

empower domain experts to innovate. Open platforms are

essential for keeping the costs appropriate in brownfield

installations and are becoming increasingly more important as

IIoT solutions roll out to a variety of industries.

8 Copyright © 2016 by ASME