dawn of grid synchronization

january/february 2008 IEEE power & energy magazine 491540-7977/08/$25.00©2008 IEEE

©D

IGIT

AL

VIS

ION

, PH

OT

OD

ISC

Digital Object Identifier 10.1109/MPE.2007.911811

RRECENT INCREASE IN WORLD-WIDE DISTURBANCES, CONGESTIONmanagement challenges, and increased complexity in operating the power gridhave emphasized the need for grid revitalization. Advanced applications inwide area monitoring, protection, and control (WAMPAC) systems offer acost-effective solution to improve system planning, operation, maintenance,and energy trading. Synchronized measurement technology and applicationsare an important element and enabler of WAMPAC.

Time synchronization is not a new concept or a new application in power sys-tems. As power system and telecommunication technologies have advanced, thetime frame of measured and communicated synchronized information has beensteadily reduced from minutes, to seconds, milliseconds, and now microseconds.At present, phasor measurement units (PMUs) are the most accurate andadvanced time-synchronized technology available to power engineers and systemoperators for WAMPAC applications. This technology has been made possibleby advancements in computer and processing technologies and availability ofGPS signals. We are approaching an era where all metering devices will be time-synchronized with high precision and accurate time tags as part of any measure-ment. To achieve the potential benefits, advancements in time synchronizationmust be matched by advancements in other areas. One example is data commu-nications, where communication channels have become faster and more reliablein streaming PMU data from remote sites to a central facility.

While advanced applications are in the development phase, PMU hardwareis based on proven technology. Phasor measurement technology was developed

near the end of the 1980s and the first products appeared onthe market in the early 1990s. Presently, a number of ven-dors are either offering or developing components, plat-forms, and tools for the synchronized measurement systems(such as PMUs, data concentrators, data acquisition andcommunication systems, EMS/SCADA, market operationssystems, etc.). Most of the PMU products are either based onexisting platforms or have PMU functionality added to theexisting products by simply adding hardware, such as a GPSreceiver, to achieve accurate time stamping, memory storage,and data streaming methods such as IEC-61850 interfaces.

Several concurrent large-scale deployment projects(resulting from the recent widespread blackouts that make the

integration of the synchronized measurement technologytimely) have been initiated in different parts of the world inrecent years. Examples include the North American SynchroPhasor Initiative (NASPI) led by NERC [an extension of heEastern Interconnection Phasor Project (EIPP) that has beensupported by Department of Energy (DOE)], the wide areamonitoring system in the Western Grid of the North America,the Brazilian phasor measurement system led by ONS (theBrazilian ISO), the wide area monitoring system in China,various developments in Europe, etc. These efforts haveshown the great potential of this technology to improve gridstability and outage avoidance, system analysis and model-ing, and congestion management.

While a number of applications based on GPS-synchro-nized data have already been developed, there is a need forvendors to continue improving existing and developing appli-cations using the latest visualization techniques, intelligentoperational tools using IEC-61850, advanced algorithms, andinformation semantics to improve grid reliability under com-plicated power system conditions. As business benefits forthose new applications are not easy to define, given the largenumber and immaturity of applications and its market, ven-dors need both system and application requirements fromusers to justify investments in new products. This is one of thereasons for an “industry business case” analysis to provideuseful insights to the expected commercial success and thesocietal and rate-payer value in deploying this promising tech-nology. This analysis also helps identify policy, economic, andtechnology barriers to commercial deployment.

This article evaluates business justification for investing indeploying the synchronized measurement technology throughassessing benefits of various applications for electricity con-

sumers, transmission owners, andother market participants. It alsoreviews technology, identifies presentimplementation gaps, and presents aroadmap for the deployment of PMUapplications in the near-term (one tothree years), mid-term (three to fiveyears), and longer-term (five to tenyears). Results should assist variousstakeholders (utilities, system opera-tors, regulators, and vendors) indefining next steps in large-scaledeployment of synchronized meas-urement systems with many usefulapplications.

Technology EnablesImproving ElectricalPower Grid Reliability There is a large number of existingand potential applications of the syn-chronized measurement technology.As the phasor technology is deployed

50 IEEE power & energy magazine january/february 2008

figure 1. Stages of development for key deployment factors.

figure 2. Key system voltage angles: WECC disturbance on 24 July 2006.

Path to Full Utilization

Dev

elop

men

t Mat

urity

Timeline

Hardware

Data Collection

ApplicationsBusiness

Processes

january/february 2008 IEEE power & energy magazine

and as users gain experience and additional tools are devel-oped, new applications will continue to be identified. Potentialeconomic benefits could result in avoiding major system dis-turbances and in minimizing widespread blackouts. Majorwide area outages cost consumers and power industry severalbillion dollars per major incident. Other benefits includereduction in congestion costs (estimated to be approximatelyUS$250 million per year in California alone), reducing costand time to analyze power system events, and providingmeans for quicker restorations following major grid outages.

A well-planned, system-wide PMU deployment, imple-menting optimal system architecture, is necessary to take full

advantage of the technology. Once the adequate PMU systemis built, incremental costs of adding applications are minimalin comparison to the added values received.

Two key application categories that benefit from the syn-chronized measurement technology are:

1) Avoidance and analysis of outages that may haveextreme manifestation in blackouts:

✔ Due to its accuracy and wide area coverage, synchro-nized measurement technology is a paradigm shiftenabling unique tracking of power system dynamics.

✔ Synchronized measurement applications enable trueearly warning systems to detect conditions that lead to

51

figure 3. Voltage plots for 500-kV substations with dc power injection at Celilo dc terminal.

figure 4. FFT analysis at Big Eddy substation with dc injection.

−40.00000

−30.00000

−20.00000

−10.00000

0.00000

10.00000

4.482422Frequency (Hz)

Celilo 3Celilo 4Dalles PH 3Dalles PH 4Dalles PH 5Dalles PH 6Banks 2&5Midway 1Troutdale 1

Most Dominant Mode is 0.217 Hz ; Damping (%) = 14 ; Time Const = 5.3 s

08/07/03 Event at 11:20 Pacific Time (08/07/03 at 18:20 GMT ) Bandwidth 0.05 - 4.5

MW

(s)

[dB

]

4.394531E-02 0.7836914 1.523438 2.263184 3.00293 3.742676

522.0

530.0

538.0

546.0

554.0

562.0

11:25:00.00

Pacific Time

Grand Coulee 1Grand Coulee 2John DayMalin NColstripBig Eddy 500Keeler 500 kVCapt Jack N. BusCapt Jack S. BusOlinda 500 kVSummer Lake N. BusSummer Lake S. BusSlatt W. BusSlatt E. BusVincentDevers 500 kV

08/07/03 Event at 11:20 Pacific Time (08/07/03 at 18:20 GMT )

kV

11:20:00.00 11:20:50.00 11:21:40.00 11:22:30.00 11:23:20.00 11:24:10.00

catastrophic events, help with restoration, and improvethe quality of data for event analysis.

2) Market and system operations✔ Congestion mitigation through better system margin

management is facilitated through synchronized meas-urement applications. Those applications allow real-time knowledge of actual system conditions asopposed to conditions defined by system models.

✔ State estimation solutions could be improved for vari-ous uses, such as locational marginal pricing calcula-tions, thereby improving overall accuracy ofcalculations and associated energy clearing charges.

As more applications are identified or the research anddevelopment efforts lead the industry to more groundbreak-ing applications, system requirements and deployment strate-gies will further change. The need for more applications willfurther drive the development of new products and tools tosupport large-scale implementation. Other deployment chal-lenges facing the industry include:

✔ design of a system that meets many diverse requirements✔ ensuring the consistent performance of PMUs acquired

from multiple vendors and installed, operated, andmaintained by different entities; some of the major ben-efits result from the system-wide applications requiringPMUs to be connected across utility boundaries.

Figure 1 shows various stages for key deployment factors.Hardware development is ahead of the data collection, appli-cations, and business processes. While technology for datacollection and required system architecture exists, many of thecommercially available products need enhanced capabilitiesfor synchronized measurement applications as users are look-ing for ways to quickly present the collected data in the formof actionable intelligence. Particular effort is required indeveloping standard application and system integration prod-ucts as well as investment by users in setting up operationaland business processes.

Implementation of synchronized measurement technologyrequires investment and commitment by regulators, powercompanies, and vendors. The necessary investments includeapplication studies, system architecture design, equipment pur-chase and upgrade, development of tools that provide action-able intelligence, maintenance, resource allocation, andtraining. It is also required to identify major requirements forthe overall system and select major applications that wouldbenefit both the individual systems and the interconnected grid.

Benefits of Using Wide Area Monitoring,Protection, and ControlThis section describes a summary of 11 major applicationareas with focus on benefits and implementation gaps (Table


figure 5. Display of modal frequencies and damping forthe dc injection test.

figure 6. Voltage oscillations during WECC system disturbance on 4 August 2004.

500.0

512.0

524.0

536.0

548.0

560.0

Pacific Time

Grand CouleeJohn DayMalin NColstripBig Eddy 500Keeler 500 kVVincentDevers 500 kV

08/04/00 Event at 12:55 Pacific Time (08/04/00 at 19:55 GMT )

12:56:13.00 12:56:27.93 12:56:42.87 12:56:57.80 12:57:12.73 12:57:27.67 12:57:42.60

kV


1). The list is as comprehensive as possible, given today’sPMU deployment status and applications identified by theindustry. As time progresses, these areas refined and extend-ed while additional applications identified as PMU systemswill become more widely deployed. At present time, only afew of these areas have been implemented and used on-line;examples are presented in Figure 2 through Figure 7.

The display shown in Figure 2 is actually a dynamic angledisplay that could be used for both postdisturbance analysisand as an operational tool to help during the disturbance.

The following example shows the voltage magnitude signals(Figure 3), FFT analysis at one of the substations (Figure 4),and a table showing the dominant modal frequencies and damp-ing parameters important for analysis of inter-area oscillations(Figure 5). Figure 3 shows the dc injection at 0.25 Hz squarewave. The odd harmonics of the 0.25 Hz injected signal can beseen clearly in Figure 4.

Our next example shows how the PMU system enabled post-disturbance analysis and provided information on what operatedcorrectly that would otherwise not be possible. Figure 6 showsthe operation of a capacitor bank near the Keeler substation that

provided the reactive power and enabled a system to damp outthe oscillations, which otherwise could have resulted in systembreak-up. Also, note the step change in voltage at the Keeler sub-station resulting from the capacitor bank.

The following example is one using PMUs for analyzingrestoration. PMU measurements were very useful during adisturbance in Europe on 4 November 2006. Figure 7 showsPMU readings for reclosing attempts between two areas,including the final successful reclosing between those twoareas and, consequently, with the third area. Although opera-tors did not use PMU data during their reclosing attempts,PMU measurements were very valuable for postdisturbanceanalysis. Using the PMU data during restoration could havehelped avoid unsuccessful reclosing attempts.

Business Justification for Synchronized MeasurementsTo bring the synchronized-measurement technology closer tothe commercial operation and integrate into daily decisions oractions, it is necessary that a systematic process be followedto plan for the infrastructure design, to delineate benefits, and

53

figure 7. PMU measurements from three areas during reclosing attempts: UCTE disturbance 4 November 2006.

figure 8. The five-phase process for business justification of synchronized measurement technology.

50,6

50,4

50,2

50

49,804-11-2006

22:33:0004-11-2006

22:34:5304-11-2006

22:36:4604-11-2006

22:38:4004-11-2006

22:40:3304-11-2006

22:42:2604-11-2006

22:44:2004-11-2006

22:46:1304-11-2006

22:48:0704-11-2006

22:50:00

WAMS Measurements

12 3

4 5

6

7

8

Area 2

Area 1

Area 3

Identify Areas forAnalysis

AnalyzeOpportunities forImprovement

IdentifyStakeholders

EstimateDeployment Plan

and Costs

Perform FinancialAnalysis

table 1. Status of wide-area monitoring, protection, and control applications.

ApplicationReal-Time

Monitoring andControl

State Estimation (SE)

Real-Time CongestionManagement

Post-Disturbance Analysis

Benchmarking,System Model,Validation, andFine-Tuning

Benefits and approachEarly indication of grid problems

(abnormal angle difference; inter-areaoscillations; voltage stability); enablingoperators to assess stress on the grid,and take timely actions.

Figure 2 shows a typical screen available to the System Operators in Californiaas a part of the real-time tool.

In general, phase angle differences or the rate of change of the phase anglecould be displayed as numbers, vectoror bar graphs (using a reference angle),and/or angle-time curves. In addition,the phase angle display may be stand-alone or superimposed onto a networkdiagram.

Three complementary approaches provide benefits:

— “Evolutionary” solution: Improvementsachieved by adding synchronizedmeasurements to existing SEmeasurement set and applying “meterplacement” methods to determinemost beneficial PMU locations.

— “Revolutionary” (next generationSCADA): State measurement solutionwith synchronized measurementsinstead of estimation for full systemobservability.

— “Equivalent” solution: For ISO/RTO SEapplications to represent “boundaryconditions” for the utility SE.

Synchronized measurements make it possible to operate the grid accordingto true dynamic limits, notconservative limits that are derivedbased on off-line studies for worst-casescenarios.

Offers improved visibility of flowgates with improved detection of angularstability, voltage stability, low-frequency oscillations, and thermalconstraints.

Quick troubleshooting of power-outage events.

PMUs are a must-have resulting in savings in troubleshooting time (several ordersof magnitude) and resources.Expediting outage analysis alsoenables faster restoration and, as PMUcompresses data, more efficient datastorage and transmission. Sinceoutages in 1996 and 2003, it has beendemonstrated that the outageinvestigations could be significantlyreduced.

Better model parameters (based on PMU data) allow for more accuratecomputation of control actions.Identify errors in system modeling dataand for fine-tuning power systemmodels for both online and offlineapplications (power flow, stability,short circuit, OPF, security assessment,modal frequency response, etc.).

Portable PMUs could be used.

Beneficiaries/StatusRate payers ISOs, utilities, and

neighboring systems.Phase-angle and oscillation

monitoring are currently inthe initial deploymentphase by some majorutilities. The information isyet to be linked to well-defined follow-on actionsand procedures.

Utilities, ISOs and rate payers.Major EMS vendors are

incorporating PMUmeasurements into theirSEs. A number of utilitiesare having pilot projects.

Utilities, ISOs, power producers, and rate payers.

No commercial-grade application in real-time.Applications are in placefor monitoring andcomparing results againstcurrent practices.

Utilities, regulators, ISOs.Very precise synchronization

has already shownsignificant benefits.

An example of using PMUs for post-disturbance analysisis shown in Figure 6.

Utilities and ISOs.Although already in use,

applications (particularlyfor dynamic and oscillatorymodes) need to be furtherdeveloped.

Implementation costs/gapsBasic PMU set-up. Current lack of: commercial-

grade tools; establishedprocess in control centers; drillscenarios for operator training,the operator training; “to-do”list when a certain observation(such as oscillations) isdetected.

Basic PMU set-up.Interface issues between PMU

measurements and legacySCADA.

Revolutionary approach requires massive PMU deployment(30%-40% of buses). It enablesmore accurate and frequentcalculations and foundationfor “closed loop” control.Strategy should include waysto transition additional PMUsas they are installed withoutmajor redesign of the tools.

High requirements for data communications and real-timedata processing. Requiresadequate system visibility andtools to comparemeasurements with evaluatedcontingencies to keep dynamicmargins and make adjustments.

The gap is in industry and staff adoption of new rules andprocedures.

Deployment is low cost (use local storage instead of real-time data communications).

Supporting software tools required to assist in dataanalysis are still lacking.

Low-cost deployment.Need is to develop a

systematized approach formodel validation and parameterestimation (PE), as well asmethods that integratemeasurements into the PE.Actual field data needed formodel development and the PE.


table 1. (continued)

ApplicationPower System

Restoration

Protection & ControlApplications forDistributedGeneration

Overload Monitoring andDynamic Rating

Adaptive Protection

Real-Tme AutomatedControl

System Integrity ProtectionScheme(SIPS)

Benefits and ApproachTimely and proper decision to bring

equipment back into service withoutrisking stability or unsuccessfulreclosing attempts.

Ability to directly measure system conditions, e.g. operator knows if it isfeasible to reclose the tie line orreconnect substation. This is valuabletool for operator who is under pressureto reenergize the grid—increasesconfidence level for the decision.

Highly precise detection of islanding.PMU facilitated coordination can allow

a micro-grid to continue to operate inisland mode until the utility griddisturbance is resolved, reducedblackout likelihood.

Tracking the line impedance in real time helps improve any applicationthat makes use of line-impedancedata:

— Calculate the impedance of the line in real time to estimate the averagetemperature over the length of theconductor and track line loading. Thisapplication only provides the averagetemperature of the whole line —hotspots, conductor sags or criticalspans are not visible.

— Actual on-line line parameter data for accurate fault location: fasterrestoration for permanent faults andbetter detection of week spots fortemporary faults.

Improving existing relay algorithms by making certain functions/parametersself-adjustable based on changingsystem conditions. Some potentialapplications:

— Adaptive security & dependability toavoid cascading.

— Improved out-of-step protectionschemes (incl. multi-machineinstability).

— Improved backup protection.— Intelligent load shedding.

Major benefits for automated prevention of angular stability, voltage stability,low-frequency oscillations, andthermal constraints.

Support optimized and integrated control with FACTS, SVC, HVDC.

Improved planned separation of power system into islands when instabilityoccurs.

— More accurate detection whether apower system is heading to anunstable state and if a networkseparation is necessary to avoid acatastrophic failure.

— Dynamically determine islandingboundaries according to theprevailing system conditions (e.g.among which groups of generatorsthe loss of stability is imminent andhow to optimally balance load andgeneration in each island).

Beneficiaries/StatusGeneral public, utilities,

ISOs, power producers.Key element is phase-angle

monitoring, which iscommercially available.

Field experience is still lacking.

Independent power producers, utilities,regulators.

Pilot installations are underway.

Ratepayers, utilities, ISOs.Overload monitoring

commercially available butnot used in any decision-making process.

Utilities, ISOs.

Rate payers, utilities, ISOs, power producers.

In RD&D phase.

Utilities, ISOs, power producers.

Under investigation by a few utilities.

Implementation Costs/GapsBasic PMU set-up.The barrier is in procedures and

guidelines for real-time dataintegration, as well as to getoperator trained and confident.

Simulators needed to provide trainee with feedback signalsthat simulate directmeasurements.

Cost of using PMUs remains a high percentage of DG projectcost.

Modest implementation cost: Two PMUs at ends of the tieline and communications to aprocessing facility.

Slow data rate is ok, but time stamping has to be precise.

Instrumentation errors may significantly impact results.

Barriers include: dedicated data high-speed communications,field experience, industryacceptance, cost.

A very fast and accurate system requires a high price.

Adding PMU measurements to the existing SIPS is within thescope of the technology. Moredemanding applications mayrequire a large number ofsynchronized data points anddedicated fiber-optic channelsso that data latency can belimited to less than 50 ms.Coherency detectionalgorithms and self-sufficientisland identification algorithmswould need to be furtherdeveloped and tested.

january/february 2008 IEEE power & energy magazine 55

to justify the expenses that might be incurred. As with anynew technology, the following steps (Figure 8) need to takeninto consideration:

✔ Phase I involves determining the desired state basedon the current and projected operating environment.The focus is on understanding performance gapswith the existing technologies and practices and inrecognizing the value of PMU in bridging the gaps.The key objective is to identify an organization,function, activity and/or process to improve and todefine the criteria for success.

✔ Phase II includes the delineation of issues specific tothe organization that can be addressed by the PMUtechnology. Key activities in this phase include: focus-ing on the problem area for analysis and collectingdata/information to quantify the benefits.

✔ Phase III identifies the stakeholders and the benefitsthat PMU may mean to them. This is important in thecase that the investment must be made by more thanone organization. Understanding the benefits to eachstakeholder can help articulate the “sale” of the tech-nology to that particular stakeholder.

✔ Phase IV provides an expected plan for PMU deploy-ment to be integrated into the annual capital investmentby the power companies. The deployment typicallytakes several years.

✔ Phase V compares the projected benefits over a timehorizon with the initial investment costs and recurring(annual) costs. A number of project valuation tech-niques can be used to arrive at a decision of whetherthe project should start or not.

A multiple-purpose deployment is the means to reapmajor benefits from the PMU technology. This is because thesame capital investment can be used by different subjectareas, stacking up the benefits. This kind of deployment,however, requires a careful analysis and planning as the capi-tal investment is high.

A comparable analogy in investment versus benefits maybe in the area of integration of protection and control usingthe latest commercially available technology. Cost justifica-tion for comprehensive upgrade programs comes from acombination of various benefits. Those benefits do not resultonly from upgrading equipment but deploying an integratedsystem using an enterprise information technology architec-ture to enable collecting required data and processing anddistributing information when and where required for use byvarious applications and users.

In a multipurpose deployment, the PMU sites are in“shared use” for a number of purposes. As some PMUapplications might still be in experimental stages, a timelapse parameter, before the benefit is actually realized,needs to be accounted for. The time lapse is due to issuessuch as personnel training, experience gathering, incorpo-ration into existing technical and business processes, soft-ware and hardware debugging, etc. In a typicalorganization, all requirements need to be deployed overseveral budget cycles. The next step in the process is todevelop an optimal deployment plan, i.e., which PMUsare to be deployed first so that the budget allocation foreach year is met, while the benefits are to be realized asearly as possible.

A concrete example to illustrate financial benefits for atypical utility company (~80 GWh, 15-k transmission miles)is based on the following assumptions of the PMU systembenefits:

✔ preventing blackouts (from “one event in five years” to“one event in ten years”)

✔ reducing system disturbances (from “three events/yearto one event/year”)

✔ reducing generator tripping (from “six to twoevents/year”)

✔ reducing postmortem analysis by 50% and speeding upsystem recovery

✔ accurate congestion margin assessment reducing costsby US$2.4M/year.

The above benefits and their financial impact, PMU systemcomponents to achieve those benefits, and costs of the overallPMU systems, are based on information from this particular util-ity, as well as from interviewing NASPI leadership. For a fullydeployed system, even with conservative assumptions (withoutconsidering all potential future benefits) the total net presentvalue (NPV) (ten years) was calculated at over US$44M.

Path to Deploying Wide Area Monitoring,Protection, and Control Implementing a large-scale PMU system presents someunique challenges. Some challenges are technical, as PMUsystems need to transmit, process, and store vast amounts ofdata. An ideal PMU system architecture should properlyaddress the following issues:

✔ Scalability: As the number of installed PMUs and IEDswith integrated PMU functions increase gradually, thesystem architecture must be designed so that it can keepup with this trend.


Potential economic benefits could result in avoiding widespread blackouts and reducing congestion costs.


✔ Flexibility: As many of the system components will beacquired, installed, operated, and maintained by differententities, the system architecture should be flexible toaccommodate the diverse requirements of various entities.

✔ Communications bandwidth and latency: In the newparadigm, on-going communications cost (leased fromservice providers or inter-company network) couldbecome the main cost item of a PMU system. Reducingthe bandwidth requirement will help to reduce the on-going cost of PMU applications. Minimizing the com-munication bandwidth requirement will also help toreduce the latency of the PMU data transferring. Forreal-time applications, reducing the communicationlatency is a must.

Other challenges include coordinated strategy and collec-tive actions among stakeholders. To facilitate a large-scaledeployment of PMUs, the following process is proposed tothe industry to speed up deployment and minimize costs:

✔ Roadmap Development: Each PMU user in the gridshould develop a near-, mid-, and long-term application/technology deployment roadmap. This roadmap wouldinclude application requirements that would guidePMU installations and system architecture needs local-ly and regionally.

✔ Overall Infrastructure Architecture: Organizations suchas NERC ERO, NASPI and regional councils such asthe Western Electricity Coordinating Council (WECC)should champion required data exchange and thedevelopment of the overall system infrastructure.Based on individual user requirements, it is necessaryto develop system architecture designs, specifications,and deployment plans. All users connecting to theoverall architecture would need to fulfill key integra-tion requirements (hardware and software interoper-ability, data quality, etc). It is also beneficial toprioritize applications from the grid perspective.

✔ Network Protocols, Software and Firmware Upgrades,and Data Access Security: The consistent and accurateperformance of all PMUs is the key to the overall per-formance of the system. Develop uniform require-ments and protocols for data collection,communications, interoperability, and security throughstandards (NERC, IEEE, WECC, NASPI) and testingprocedures. Assure ease of PMU upgrades (particularlyconsidering the novelty of the IEEE C37.118 standard)and calibration.

✔ Regulatory Considerations: Regulators at both federaland state levels need to provide incentives for technol-ogy deployment considering significant benefits forrate payers and transmission system reliability.

✔ Company Process: Each user should set up operationaland business processes for installations, operations,maintenance, and benefits sharing. This would be com-prised of creating projects with defined deliverablesand deadlines; identifying asset owners, managers, and

service providers; setting up procedures and rules; edu-cating and training users; and facilitating culturechange.

✔ Research and Development: Continue investing inRD&D (DOE, PIER, vendors, users, etc.) and promotedeveloping and sharing test cases to develop new appli-cations. Also, continue using a proven approach ofpilot projects to gain experience and confidence.

Full financial and reliability benefits of this promisingtechnology can be realized by a collective process amongall stakeholders including the general public and theratepayers. A significant market penetration of this technol-ogy is possible through a collaborative process to engagesuppliers and manufacturers in developing the requiredtools and products.

Deployment Roadmap and RecommendationsGiven the nature of PMU implementation requiring participa-tion of a broad base of users, the “overall industry roadmap” isan important step in designing and deploying large-scale PMUsystems. Based on an interview process with industry expertsand users, the roadmap has been developed to serve as a basefor development of individual deployment roadmaps and guid-ance to the vendors to prioritize their efforts. This roadmap,shown in Figure 9, is based on applications’ business needs,commercial availability and cost, and complexity with deploy-ing those applications.

First, industry needs are identified (critical, moderate,unknown) regardless of technology. Second, the value ofthe PMU technology, for each identified application, hasbeen mapped related to importance in serving industry asnecessary, offering additional benefit, or requiring moreinvestigation. Third, deployment challenges have beenmapped for each application (low, medium, high). Thedeployment challenges are defined based on technology(communications and hardware and software requirementsand development status) and applications status (commer-cially available, pilot installation, in the research phase, notdeveloped yet). Business case examples described in moredetail in the phasor measurement application study by Cali-fornia Energy Commission provided data to create theinformation in Figure 9.

The following applications have a major improvementimpact with PMUs: angle/frequency monitoring, postmortemanalysis, model benchmarking, outage prevention (includingplanned power system separation), state measurement, andreal-time control. As for the rest of the applications, non-PMU technologies are available; however, the deployment ofPMUs allows the same measurements to be used to realizeadditional benefits from the same investment (integratedapproach). The matrix provides the basis to create the near-,mid-, and long-term deployment roadmap. The resultingroadmap is shown in Figure 10, where the applications aregrouped into near term (one to three years), medium term

57


(three to five years), or long term (more than five years). Thisroadmap differs from the RD&D roadmap as it focuses onbusiness and reliability needs to commercialize and deployPMU technology and applications.

Note that the list of applications in Figure 9 and Figure 10appears to be larger than the 11 groups under the Benefitssection. This is because some groups were too broad andneeded to be subdivided to address specific utility problems.For example, real-time monitoring and control is subdividedinto angle/frequency monitoring, voltage stability monitoring,real-time control, and wide area stabilization.

Applications in the near-term group reflect the immediateneeds. For this group, the applications are commercially avail-able (either at present time or will be soon offered by a vendorbased on the status of the working prototypes). They also reflectthe fact that the deployment can be achieved rather quickly dueto factors such as the applications can be used for specific spotson the grid and the infrastructure requirements are relativelymodest. These applications can be termed “low-hanging fruit.”

The most obvious “low-hanging fruit” for which PMUs providemajor benefits are angle/frequency monitoring and postmortemanalysis (including compliance monitoring).

Applications in the medium-term group largely reflect thateven though the needs are great, the commercial prospect isstill far off as no working prototypes are known to exist.

Applications in the long-term group indicate a combina-tion of distant commercial status, extensive infrastructurerequirements (and thus costs), and/or that lengthy field trialsare required to gain acceptance by end users. For example,the wide area stabilization application, even though commer-cially ready, remains to show its superiority to the conven-tional power system stabilizers.

Three complementary approaches exist in using PMUtechnology with state estimation: conventional SE improve-ment (evolutionary), boundary conditions SE, and state meas-urement (revolutionary). These concepts are considered to beelements of short- to long-term PMU deployment strategy asincreasing numbers of PMUs are planned locally and

The most obvious “low-hanging fruit” for which PMUs providemajor benefits are angle/frequency monitoring and postmortemanalysis (including compliance monitoring).

figure 9. Synchronized measurements and industry needs.

NecessaryOffers AdditionalBenefit

Requires MoreInvestigation

Critical

IndustryNeeds

Moderate

Minor

Value of Synchronized Measurement

12 35

8

1211

13

Deployment Challenge

• Number of PMUs• Communication Req.• Application Needs

LOW MED

9

166

7

HI

4 14

15

10

Reviewed with Key Implementers

1 Angle/Freq. Monitoring 2 Voltage Stability Monitoring 3 Thermal Overload Monitoring 4 Real-Time Control 5 State Estimation (Improvement) State Estimation (Boundary 6 Conditions) 7 State Measurement (Linear) 8 WA Stabilization (WA-PSS) 9 Adaptive Protection10 Congestion Management11 Power-System Restoration Post-Mortem Analysis (Including12 Compliance Monitoring) Model Benchmarking; Parameter13 Estimation (Steady-State) Model Benchmarking; Parameter14 Estimation (Dynamic)15 Planned Power-System Separation16 DG/IPP Applications


regionally. In fact, the revolutionary case is a natural extensionof the evolutionary approach as numbers of PMUs installedincrease over time. Use of PMUs for representing boundaryconditions will stem from system-wide regional deployment.

ConclusionsPMU applications offer large reliability and financial benefitsfor customers/society and the electrical grid when imple-mented across the interconnected grid. As measurements arereported 20 to 60 times per second, PMUs are well suited totrack grid dynamics in real time. Compared to current EMSmonitoring tools that use information from state estimationand SCADA over several-second intervals, time-synchro-nized PMUs introduce the possibility of directly measuringthe system state instead of estimating it based on systemmodels and telemetry data. Implemented properly, the tech-nology also allows for providing data integrity validationwithout added cost. This technology is instrumental for:

✔ improving WAMPAC in real time including earlywarning systems, system integrity protection scheme,detecting and analyzing system stability, and enablingfaster system restoration

✔ faster and more accurate analysis of a vast number ofdata during transient events

✔ validation and development of power system models. Power companies could realize financial benefits if sever-

al integrated applications are deployed using basic PMUsystem infrastructure. The above conclusions have beenachieved through comprehensive analysis of various applica-tions and related financial benefits using concrete data onPMU system related costs and industry experiences withPMU implementation obtained through a number of inter-views and literature search.

Synchronized measurement capability has advanced tech-nologically to the point that commercial implementation ofselected applications is both possible and warranted,

59

The wide-area stabilization application, even though commercially ready, remains to show its superiority to the conventional power system stabilizers.

figure 10. Roadmap for deploying PMU applications.

1–3 Years 3–5 Years

10. Congestion Management

1. Angle/Freq. Monitoring1. Angle/Freq. Monitoring1. Angle/Freq. Monitoring

12. Post-mortem Analysis (incl. Compliance Monitoring)12. Post-mortem Analysis (incl. Compliance Monitoring)12. Post-Mortem Analysis (IncludingCompliance Monitoring)

2. Voltage Stability Monitoring

3. Thermal Overload Monitoring

5. State Estimation (Improv.)

13. Model Benchmarking;Parameter Estim. (Steady-State)

16. DG/IPP Applications

11. Power-System Restoration14. Model Benchmarking; Parameter Estimation (Dynamic)14. Model Benchmarking; Parameter Estimation (Dynamic)14. Model Benchmarking; ParameterEstimation (Dynamic)

15. Planned Power-System Separation – Special Protection 15. Planned Power-System Separation – Special Protection 15. Planned Power-SystemSeparation – Special Protection

6. State Estimation (BoundaryConditions)

4. Real-Time Control4. Real-Time Control4. Real-Time Control

7. State Measurement (linear)7. State Measurement (linear)7. State Measurement (Linear)

9. Adaptive Protection

8. WA stabilization (WA-PSS)8. WA stabilization (WA-PSS)8. WA Stabilization (WA-PSS)

Deployment NeedsDepend on Regional

and Utility Requirements

Necessary

Additional Benefits

More Investigation

Role of PMUs

> 5 Years

1 Angle/Freq. Monitoring 2 Voltage Stability Monitoring 3 Thermal Overload Monitoring 4 Real-Time Control 5 State Estimation (Improvement) State Estimation (Boundary 6 Conditions) 7 State Measurement (Linear) 8 WA Stabilization (WA-PSS) 9 Adaptive Protection10 Congestion Management11 Power-System Restoration Post-Mortem Analysis (Including12 Compliance Monitoring) Model Benchmarking; Parameter13 Estimation (Steady-State) Model Benchmarking; Parameter14 Estimation (Dynamic)15 Planned Power-System Separation16 DG/IPP Applications

representing prudent investment. Furthermore, the implemen-tation and use of this technology is necessary for the levels ofgrid operational management that are required for efficient useof the infrastructure currently in place as well as for infra-structure enhancements of the future. To gain the benefitsoffered by this technology, a coordinated effort among utili-ties, ISOs and other entities must be undertaken, with a coor-dination level beyond the present activities. Without asystem-wide approach, the capabilities and associated benefitswill not be achieved in the manner possible. This requires aneffort that includes a bottom-up approach from utilities indefining the needs and PMU applications and a top-downapproach from the system regulators and coordinators todefine an integrated infrastructure to optimize the benefitsoffered by the technology. Also, regulators at both federal andstate levels need to provide incentives for technology deploy-ment, particularly considering significant benefits for rate-payers and transmission system reliability.

There is a need for vendors to fully productize applica-tions presently in an RD&D phase and to develop new, prom-ising applications. Vendors need both common systemarchitecture requirements and common application prioritiesfrom users to be able to justify investments in new products.Provided common application priorities and deployment suc-cess factors help guide deployment priorities for individualusers and development priorities for vendors, as well as sup-port regulators and coordinators in prioritizing their efforts.

Well-planned, system-wide synchronized measurementdeployment infrastructure is necessary to take full advantageof the technology. It requires organizations such as NERCERO, NASPI, and WECC to facilitate required data exchange.They also need to facilitate certain system-wide deploymentof “low-hanging fruit” applications to speed up achieving keybenefits of synchronized measurement technology.

In summary, the findings described in the article shouldserve as a base for deployment of individual roadmaps by theusers and guide vendors to prioritize their development to sup-port the grid revitalization and reliability for the 21st century.

AcknowledgmentsResults in this article are partially based on “Business CaseStudy on Applying Phasor Measurement Technology andApplications in the WECC/California Grid” funded by Califor-nia Energy Commission (CEC), California Institute for Energyand Environment (CIEE), and the Department of Energy(DOE). The CIEE work has been a collaborative effort of dis-tinguished experts including representatives from Bonneville

Power Administration, Pacific Gas and Electric Company,Sempra, Southern California Edison, California IndependentSystem Operators, Tennessee Valley Authority, Entergy; NorthAmerica Electric Reliability Corporation, Virginia Tech, Geor-gia Tech, KEMA, and InfraSource. A final thanks to our indus-try colleagues and friends who have shared their ideas andhave inspired us in pursuit of knowledge.

For Further ReadingCalifornia Institute for Energy and Environment (CIEE),“Phasor Measurement Application Study,” final report, 2006.Available upon request.

D. Novosel and V. Madani, “Blackout Prevention,” inMcGraw Hill Yearbook on Science and Technology. New York:McGraw Hill, 2005.Available at https://reports.energy.gov/

V. Madani and D. Novosel, “Getting a grip on the grid,”IEEE Spectrum, vol. 42, no. 12, pp. 42–47, Dec. 2005.

A.G. Phadke, “Synchronized phasor measurement inpower systems,” IEEE Comput. Appl. Power, vol. 6, no. 2,pp. 10–15, Apr. 1993.

Y. Hu, V. Madani, R.M. Moraes, and D. Novosel,“Requirements of large-scale wide area monitoring, protec-tion and control systems,” in Proc. Fault and DisturbanceAnalysis Conference, Atlanta, GA, Apr. 2007.

V. Madani, G. Duru, B. Tatera, M. Jones, and B. Pace, “Aparadigm shift to meet the protection and control challengesof the 21st century,” in Proc. Georgia Tech Protective Relay-ing Conference, Atlanta, GA, May 2007.

J.W. Ballance, B. Bhargava, and G.D. Rodriguez, “Use ofsynchronized phasor measurement system for enhancing ac-dc power system transmission reliability and capability,”CIGRE C1-210 Session, Paris 2004.

U.S.-Canada Power System Outage Task Force, “FinalReport on the August 14, 2003 Blackout in the United Statesand Canada: Causes and Recommendations,” Apr. 2004.

UCTE (Feb. 2007), “System disturbance on November 4,2006,” [Online]. Available: http://www.ucte.com

About the authorsDamir Novosel is president of the technology division atQuanta Technology.

Vahid Madani is a principal protection engineer at PG&E.Bharat Bhargava is a consulting engineer at Southern

California Edison Company.Khoi Vu is a consultant at KEMA Inc.Jim Cole is a senior advisor at California Institute for

Energy and Environment.


Time-synchronized PMUs introduce the possibility of directlymeasuring the system state instead of estimating it based on system models and telemetry data.

p&e

dawn of grid synchronization

Documents