IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012 1405Laboratory-Based Smart Power System, Part II:Control, Monitoring, and ProtectionVahid Salehi, Student Member, IEEE, Ahmed Mohamed, Student Member, IEEE,Ali Mazloomzadeh, Student Member, IEEE, and Osama A. Mohammed, Fellow, IEEEAbstractWide area monitoring (WAM), wide area protection(WAP), and wide area control (WAC) systems will enhance thefuture of smart grid operation in terms of reliability and security.In part I of this paper, a proposed architecture for a hybridac/dc smart grid hardware test-bed system was presented. Designdetails of the various components and their connectivity in theoverall system architecture were identied. In part II of the paper,the focus is on the design of monitoring, control, and protectionsystems and their integrated real-time operation. Various con-trol scenarios for system startup and continuous operation areexamined. We have developed a control system based on widearea measurements. The advanced measurement system basedon synchrophasors was also implemented using DAQs real-timesynchronous data. The developed system features a wide varietyof capabilities such as online system parameters calculation andonline voltage stability monitoring. These are implemented asan experimental case to enhance wide area monitoring systems.Moreover,the protection system was designed inside of thereal-time software environment to monitor the real-time wide areadata, and make a comprehensive and reliable coordination for thewhole system. Ideas related to the interaction of a dc microgridinvolving sustainable energy sources with the main ac grid havebeen also implemented and presented. The implemented systemis explicit and achievable in any research laboratory and forreal-time real-world smart grid applications.Index TermsOnline voltage stability monitoring, phasor mea-surement units (PMU), smart grid, wide area control (WAC), widearea monitoring (WAM), wide area protection (WAP).I.INTRODUCTIONPOWER AND energy systems researchers need to de-velop a platform for smart grid technology to study andidentify the issues involved in their operation. This approachshould include objectives such as visualization enhancementof the power system, real-time analysis for wide area network,model validation, developing strategies for wide area moni-toring, protection, and control system (WAMPAC). Moreover,concerns related to the use of renewable generation, energystorage,demand response,and electric vehicles introduceManuscript received September 26, 2011; revised February 02, 2012; ac-cepted March 24, 2012. Date of publication June 08, 2012; date of current ver-sion August 20, 2012. Part of this work was supported by grants from ONR andthe U.S. Department of Energy. Paper no. TSG-00558-2011.The authors are with the Electrical Engineering Department, Florida Interna-tional University, Miami, FL 33174 USA (e-mail: mohammed@u.edu).Color versions of one or more of the gures in this paper are available onlineat http://ieeexplore.ieee.org.Digital Object Identier 10.1109/TSG.2012.2194519further complexity to system operation and control [1]. Recenttechnologies enhanced the communication structure of powersystems by their high accuracy, security, and data transferspeed. Hence, the smart grid uses a modern communicationinfrastructure at wide area power system to improve grid reli-ability, reduce the price of electricity, and improve operationalefciency, security, and safety, [2][4]. The communicationinfrastructure is responsible for fast data exchange among thesystem components and control centers [5], [6]. Currently,phasor measurement units (PMUs) are the most accurate andadvanced time-synchronized technology available. They pro-vide voltage and current phasors and frequency information,synchronized to a common time reference with a high pre-cision. Recently, protection and control solutions based onsynchrophasors for wide area applications are being introducedas a preventive or self-healed system for frequency, voltage,and rotor angle instabilities. The system data are collected bythe PMUs and are sent to a phasor data concentrator (PDC).The PDCs, in turn, send the system data to a control center[7]. The PMU-based state measurement is expected to be moreefcient than the present state estimation since synchronizedphasor signals provide the state variables, in particular, thevoltage angles. A successful protection and control applicationrequires a carefully designed WAMPAC architecture, includingthoroughly considered latency (communication delay) [8]. It isexpected that WAMPAC system will, in the near future, reducethe number of cascade events and will improve the reliabilityand security of energy production, especially in power gridsintegrated with operational uncertainties.Our research is aimed at developing integrated real-time toolsrelated to wide area applications. Hence, after introducing thedeveloped laboratory-based smart grid test-bed in part I of thispaper, the focus of part II is on developing an integrated real-time monitoring, control, and protection system for the test-bedgrid as shown in Fig. 1. The system uses real-time data moni-toring for startup and continuous control of generation stations.The wide area monitoring system is implemented by the ap-plication of PMUs functions which are developed by data ac-quisition systems (DAQs) and implemented for all nodes andbranches of the ac grid. The real-time system parameter mea-surements and the tracking of online voltage stability indices arethe applications developed using the synchrophasor monitoringfeature. Another goal of developing such an experimental setupis to deploy simple, reliable, and safe automatic scheme for wide1949-3053/$31.00 2012 IEEE1406 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012Fig. 1. Overall implemented real-time system for integrated wide area monitoring, control, and protection systems.area protection system which are designed to apply and test newcontrol and protection strategies.II. GENERATION CONTROL SCHEMEA.System Start-UpThe generators are controlled solely by the real-time softwareby sending the control commands to their prime mover drivecontrollers whenever needed. They keep the last commands astheir operating setting until they receive a new one. These com-mands are the prime mover start/stop and change the output fre-quency or torque to control the generators frequency and outputactive power. Hence, the generators can be set as a slack gener-ator to maintain frequency, or as a constant output voltage-activepower, i.e., in PV control mode. Fig. 2 shows the control vir-tual instrument (VI) inside the LabVIEW environment to con-trol and visualize the input/output information as data, curves,or indicators. Here, G1 is used as a slack generator to control themicro grid frequency whereas the other generators are in the PVcontrol mode. While the grid is energized by G1, generator G2should be synchronized with the grid in order to be connected toit safely. This should be done by checking synchronizing con-ditions and then use the implemented automatic controller toincrease/decrease G2s torque until all the synchronizing con-ditions are satised. Fig. 3 shows the automatic controller at-tempt to synchronize and connect the generators to the grid. Themain VI also presents the used capacity (KVA) of each generatorand the torque change buttons to change the generators activepower during the systemoperation. In addition, a dynamic breakhas been designed in order to synchronize generators to the gridfaster than the case when the synchronization is approached byjust changing torque commands. Different methods for soft syn-chronizing of micro generators by design procedure and exper-imental results are provided in [9].SALEHI et al.: LABORATORY-BASED SMART POWER SYSTEM, PART II: CONTROL, MONITORING, AND PROTECTION 1407Fig. 2. Overall view of real-time generation monitoring, communication, and controller VI.B. System Continuous OperationThe steady state parameters in the real-time monitoringsoftware determine the system components loading, over/undervoltage situation, frequency drop, active and reactive powerow, losses, and so on. The main generator control VI alsopresents the generators loadings in order to share the genera-tion level optimally when one of them encounters an overloadsituation. The constant active power-voltage magnitude gen-erators known as PV generators, G2, G3, and G4, shouldparticipate in generation according to the control commandsafter the system start-up. Hence, their active power is increasedto a reasonable value according to the system total load andthe slack generator loading. This procedure was implementedmanually by entering a proper torque command to the primemover as shown in Fig. 4. The systems active power load isincreased in steps of 300 W for each load in different times (L1,L2, L3, and L4). Without any change in generation, the slackgenerator is responsible to maintain the system frequency at 60Hz. Therefore, the total load change leads to an increase in the1408 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012Fig. 3. Synchronizers controller actions in order to connect generators to grid.active power generation of the slack generator, G1. Increasingthe active power of the generators will alleviate the total gener-ation of G1. The active and reactive power changes of all fourgenerators are shown in Fig. 4. Practically, the reactive powersreturn back to their initial values since the total reactive loadwas constant during the experiment.A real-time power controller is implemented to monitor theactive power of the load buses and specify the same amountof power to the nearest generator. This is the automatic powersharing process presented at Fig. 1 in order to show the effec-tiveness of the integrated real-time monitoring and control inwide area system. Fig. 5 shows the active and reactive powerchange of every generator which is controlled automaticallyby the real-time software. Similar to previous experiments, thesame increase in load pattern is considered in this case, and testresults of this controller were compared with the manual gen-eration control. Whenever a load is increased, the monitoringand control system recognized it. The system also changes thetorque control command of the generator connected close to thatload to change the amount of generated power. This procedureis clear as presented in Fig. 5. Since the monitoring system hasa time latency of about 1 s, the control system will follow theload variation by a step change within a same or twice timedelay. The automatic control scheme is fast enough to operatethe system in a smart manner and to share the load among thegeneration stations. The range of variation of the active and re-active power of each generator in the automatic mode is lessthan manual control because of the fast response of the controlFig. 4. (a) Active power and (b) reactive power of generators during load in-crease in manual control mode.Fig. 5. (a) Active power and (b) reactive power of generators during load in-crease in automatic control mode.system. Fig. 6 shows the benets of automatic generation con-trol system versus manual control through the voltage and fre-quency changes for the slack generator.The automatic control system for the generators should becomprehensive considering system conditions such as loadingSALEHI et al.: LABORATORY-BASED SMART POWER SYSTEM, PART II: CONTROL, MONITORING, AND PROTECTION 1409Fig. 6. (a) Voltage and (b) frequency of generators during load increase inmanual (upper plot) and automatic (lower plot) control mode.of equipment, voltage, and generation limitations as well as sta-bility issues. The use of the real-time software makes it pos-sible to monitor all system conditions and create a proper powersharing algorithm such as optimal power ow (OPF).III. REAL-TIME WIDE AREA MONITORINGThe evolution of power systems creates major changes intheir operational procedures, especially the monitoring net-works in real time. We need to use advanced, smart monitoringtools to quickly and reliably estimate the real-time state ofthe systems. One of the most promising technologies in thiseld is the system monitoring based on phasor measurementsunits (PMUs). In power engineering, these systems are alsocommonly referred to as synchrophasors and are consideredone of the most important measuring devices in the future ofsmart operation of power systems [10], [11]. They providepositive sequence voltage/current and frequency measurementssynchronized within a microsecond. In addition, they may becustomized to measure harmonics, negative and zero sequencequantities, as well as individual phase voltages and currents.Signicant improvements in system operations can be achievedby utilizing the synchronized measurement technology in theenhancement of state estimation, real-time load management,real-time angular and voltage stability analysis and design andenhancement of an adaptive protection and control system [12].In this laboratory-based smart power system, the focusis on the real-time calculation of power system parametersusing PMUs and their applications in power system studies.The PMUs implementation was carried out by the DAQsfor research purposes. According to Fig. 7, different DAQswere used to get the whole system data. Every DAQ dataare synchronized and hence can simulate different PMUsdata concentrating in one system. For instance, all generatorsvoltages and currents are measured in a DAQ and then thevoltage phasors are presented in the same reference frame.In this case, the reference signal for this DAQ is phase A ofFig. 7. Synchronized phasor measurement system developed by real-timesoftware.generator G1 according to Fig. 7. As shown in this gure, thepositive sequence of the voltages and currents are in the samereference and with a 12-kHz sampling rate. It should be pointedout that 10 data packages can be sent per second. Other DAQsare connected to several buses which are shown with the samereference arrows. A challenge will be synchronizing all DAQstogether. A possible solution would be the use of one DAQin common with at least one measured signal with the otherDAQs, similarly to the generators connected DAQ.The implemented PMUconcept is described in details in [13].In addition, a comparison between the implemented real-timePMU and the available measurement devices also were doneand the results were veried. With an acceptable accuracy, wecan conclude that the power system measurements by this PMUdesign is explicit, inexpensive, and achievable and applicablefor many real-time studies. Two different application cases werestudied using the real-time monitoring system by developedPMUs.A. Real-Time Networked Model Parameters MeasurementThe implementation of PMUs meets the requirements forreal-time estimation of unknown parameters or for updatingnetwork model parameters. For instant critical branches in thepower system can be monitored if two PMUs are located atboth sides of the line terminals as presented in Fig. 8. Thisconguration provides reliable information about the powerangle differences between adjacent buses and hence determinesthe power transferred through the line. It is possible to directlycalculate the actual line parameters if the voltages and currentsat both side of the line are measured in a synchronized form.The following equations give the real-time line pi-model pa-rameters calculation using synchrophasors:(1)(2)(3)1410 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012Fig. 8. PMU placement for a transmission line and related pi-model.Fig. 9. Voltages and current magnitude of receiving and sending of the lineduring load increase.For measuring the line parameters, an experiment was de-signed to calculate pi-model line parameters. For one of thelines connected between the generation bus and the load bus,the measured real-time voltage and currents are identied forboth the receiving and sending ends as in Fig. 9. The activeand reactive power load starts to increase in different time stepsaccording to Fig. 10 and lead to step changes in the values inthis gure. Using the phasor calculation in real-time, and inser-tion into (2) and (3), the real line parameters can be achievedaccording to Fig. 11. For an ideal transmission line pi-model,circuit parameters should be constant during the load increase.This experiment however shows that the developed line has pa-rameters which are dependent on the line current. For a healthyline, the calculated line parameter changes give some informa-tion about ambient condition and the line situation. With fastreal-time PMUmeasurements, the parameter monitoring can in-dicate any kind of faults inside the line which may have consid-erable variations.B. Real-Time Voltage Stability MonitoringThe voltage instability problem is always distinguished bythe system voltage prole, heavily reactive line ows, inade-quate reactive power support, and heavily loaded power sys-tems in general. Voltage collapse typically occurs right after anFig. 10. Active and reactive power of the load at receiving end side.Fig. 11. Measured line pi-model parameters during the experiment.event which causes one of the conditions, and it may last in timeframes of a few seconds to several minutes, but rarely hours[14]. Many indices were developed to detect the voltage stabilityin a static security-stability assessment manner [15][24]. Manyof these indices are based on the proximity of the power systemoperating state to its collapse point which can be achieved bywell-known PV curve. A voltage stability margin, as an index,determines the distance of the power system operating point tothe instability point. Furthermore, it can be used to recognizeweak buses and the highly risky areas involved in the voltage in-stability. In this study, several voltage stability indices which aredeveloped in literature [15][24] are presented in Table I. Theyare used to monitor online voltage stability margins in powersystem using PMUs. The real-time application was developedto monitor the behavior of these indices under system dynamicchanges. By comparison and verication of these indices in areal power system, a better view can be achieved to a specicsecurity margin and therefore a proper remedial action is to betaken. These indices were achieved for a transmission line withtwo end bus power systemusing the PMUs data. This is becauseall indices need the data from both sides of the line.Fig. 12 presents the indices variation under the active load in-crease at different time steps similar to the previous experiment.SALEHI et al.: LABORATORY-BASED SMART POWER SYSTEM, PART II: CONTROL, MONITORING, AND PROTECTION 1411TABLE IREAL-TIME VOLTAGE STABILITY INDICES: Voltage on sending and receiving buses; : Active and reactivepower on the sending bus; : Active and reactive power on thereceiving bus; , : Voltage angle on sending and receiving buses; :Line resistance and reactance; : Line suceptance and conductance;: angle difference between sending and receiving buses;: line series impedance angle.All the indices show negative changes in voltage stability mar-gins expected during load increasing with different sensitivities.Here, and present the knee point of well-knownPV curve for voltage stability and hence the available powermargin can be achieved after subtracting from the loadactive power. Simulation results and analytical comparison waspresented in [25].IV. SYSTEM PROTECTION SCHEMEIn a self-healed power system, unplanned outages and abnor-malities could be prevented through better prediction, analysis,and control. Recognition and diagnosis of fault conditions hasa signicant role that may prevent disturbances extension andspreading to other healthy parts of the power system [26]. Sucha capability requires timely based coordination between moni-toring, analysis, and control of power system in different scales.It would be interesting to implement a protection system whichFig. 12. Experimental results of online voltage stability indices monitoring.can offer various types of advantages for controlling and healingin the power system.The wide area protection system (WAPS) is an implementa-tion of existing protection systems equipped with a fast bidi-rectional communication link with a control center. Most of thereal-time softwares have the capabilities to implement mathe-matical and logical built in functions to model available relayfunctions and settings inside their environment. DAQs are usedto transfer the measurements from the secondary part of the CTsand PTs to the real-time application software in a synchronousway. Hence, these synchronous data are used as PMU data in-side the real-time environment in order to mimic physical relayfunctions.The speed of communication guarantees the accuracy of op-eration of this real-time system to predict relays operation inreal power system, and simulate the system behavior as fast assystem topology and parameters vary. The application of thissetup, consist of the control and self-healing of the power systemis presented here.1412 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012Fig. 13. Digital generic relay model implemented at real-time software.To build an intelligent and adaptive protection system, thewide area protection system which is applicable in powersystem control centers is implemented. This implementation isnot only for investigation of the coordination between realisticrelays, but is also for the prediction of cascaded events in powersystem and devising self-healing procedures. This system usesPMUs for monitoring very fast real-time data with a rate of atleast 10 data packets per second for 60-Hz system accordingto IEEE Std C37.118.1-2011 [27]. The PMUs are particularlyeffective in improving the protection functions which haverelatively slow response times. For such protection functions,the latency of remote measurements is not a signicant issue.For example, back-up protection functions of distance relaysand protection functions concerned with managing angular orvoltage stability of networks can benet from remote measure-ments with propagation delays with latencies of up to severalhundred milliseconds [28]. In this laboratory implementation,the generic form of digital relay with different capabilities forprotecting the power system elements were developed in thereal-time software environment and using them as a protectionrelay. The developed relay system has the following functions: over/under voltage; rate of change of voltage (ROCOV); over/under frequency; rate of change of frequency (ROCOF); time over-current; MHO distance relay; polygonal distance relay; inverse power ow; voltage unbalance function; current unbalance function.The whole set of functions were developed in the real-timesoftware which can receive all voltages and currents fromPMUs, and then make a decision for breaker action and pro-tection scenario. These relays were tested in the developedtest-bed. Furthermore, it was operated by the real-time data andat the speed of the communication system [29]. In the labora-tory test system, the relay speed is as fast as the DAQs speedused to gather the data from the secondary side of the PTs andCTs and send data packets to the real-time software, and nallysend the trip command to the switches. Fig. 13 represents thefront panel of this digital generic relay with settings for relayfunctions and fault indicators. By the same way, a systemdesigner can add the functional model of any realistic relay andmonitor the protection system behavior by applying the PMUssynchronous data in the real-time environment. Moreover,these synchronous data provide additional information suchas voltage phase angle difference between buses for wide areaapplications.SALEHI et al.: LABORATORY-BASED SMART POWER SYSTEM, PART II: CONTROL, MONITORING, AND PROTECTION 1413Fig. 14. (a) Active power and (b) frequency of all generators during G1 outage.Since all the switching control and relay models were imple-mented inside the real-time environment, the wide area protec-tion system was designed to test the new protection schemes forwide area systems.An experiment was performed involving the outage of theslack generator G1. The event is the disconnection of the slackgenerator, G1, which will cause frequency drop of the wholesystem since generation and demand balance is no longer main-tained. The frequency functions of all generators relays are ac-tivated. For G2 and G3 frequency settings considered 59 Hzand for G4 is 58 Hz. So, whenever the frequency drops, themonitoring system will calculate the frequency and the relayfunction will detect the situation and will send the trip signal tothe breaker right away. Here, G1 is disconnected atand G2, G3 and G4 relays are detected at, and , re-spectively. Because of the frequency setting for G4, it detectsthe event with a larger delay. Fig. 15 presents the active powerchange as well as frequency drop for this event for all gener-ators. The generators G1 frequency is returned to the no loadfrequency after disconnection from the grid and the other gen-erators frequency started to drop until the relays trip them fromthe grid. The remaining generators take the load share of thedisconnected generators. For instance, after disconnection of G1and then G2 and G3, suddenly, G4 which was producing 250-Win normal condition takes the active power load of whole system(1300 Watt) and then leads to large drop of frequency (48-Hz).A wide area protection algorithm is also implemented for thiscase, which is the disconnection of all generators when G1 isdisconnected from the grid right away. In this case we do notneed to worry about the frequency relays settings and their op-eration delay. Consequently, all the power systemdata are avail-able by the implemented PMUs for monitoring the system andhence any kind of intelligent and smart wide area protectionFig. 15. A ow chart of one of the real-time algorithm used to manage thecharge/discharge process of the batteries.system integrated with system controllers can be developed inreal-time to enhance power system operation.V. SYSTEM REAL-TIME OPERATIONA. DC Microgrid OperationIn the dc microgrid described in Part I of this paper, the su-pervisor controller is the master controller, it is responsible forsending the dc bus voltage reference to the controlled rectierand distributing power references to the local (slave) controllersof the other source converters. Moreover, this supervisor con-troller communicates with the main ac grid and is the placeto execute various real-time energy management algorithms inorder to manage the power sharing among the various sustain-able energy sources, the interaction with the main ac grid andthe battery charging and discharging. For instance, the real-timebattery charging/discharging algorithm presented in Fig. 15 isimplemented. This algorithm aims mainly at shifting the powerdemanded from the main ac grid to off-peak time by managingthe charging/discharging process of the battery such that: If there is a surplus in power, i.e., the total power gener-ated locally from the PV and fuel cells exceeds the demandon the dc microgrid, the charging rate of the battery is in-creased during off-peak time since the energy tariff is rela-tively low so it is more worthy to charge the battery than tosell power to the grid. On the other hand, the charging rateis decreased during peak time to charge the battery later onwhen the energy tariff drops. If there is a deciency in power meaning that the renew-able energy sources are not capable of satisfying the loaddemanded on the microgrid, the discharging rate of the bat-tery is increased during peak time to increase the savingby reducing the energy bought from the main grid withthe high tariff. Moreover, the discharging rate is decreasedduring off-peak time since the tariff is low and the avail-able energy in the battery can be more effectively utilizedduring the coming peak period. This algorithm can resultin an annual saving of around 7%9% [30]. The developedlaboratory-based smart power system can be used to testmore complex real-time energy management algorithmsinvolving online prediction and modeling of renewable en-ergy sources output power uncertainty, load forecasting,1414 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012and pulsed load mitigation such as the algorithms devel-oped by the authors in [30], [31].In order to investigate the real-time operation of the dc mi-crogrid, Fig. 16 shows the performance of the various compo-nents of the dc side corresponding to several step changes inthe reference power. The curves shown (from top to bottom)are corresponding to the dc bus voltage , the rectier cur-rent , the bidirectional current , the photovoltaic (PV)output current , the fuel cells output current and thebattery current , respectively. The case study is describedas follows. The supervisor controller sends reference signals to thevarious converters in the dc microgrid. The bidirectional converter is set to transfer 1 A from themain ac grid to the dc microgrid. is initially set to 2 A, then a step change from 2 to 0.5A is applied after 0.1 s. is set to 0.25 A along the whole interval shown inFig. 16. The battery is operating in the discharging mode, isinitially set to 0.2 A then the discharging current referenceis changed to 0.7 A after 1.1 s. The total loads connected to the dc bus are increased by75% after 1.6 s.As can be seen in the gure, the dc bus voltage is hardly af-fected by the step changes in the PV or battery power. However,the big step change in the load creates some uctuation in thebus voltage of around 1%, which is denitely below all the stan-dard limits. This transient period lasts for about 0.5 s.The rectier current is free to maintain the power balancein the system. Hence, when the PV power decreases from 2 Ato 0.5 A, the rectier current increases by the same amount.Moreover, when the battery current increases after 1.1 s, therectier current decreases to compensate for this increase.The bidirectional converter is capable of maintaining its refer-ence current of 1 A during the step changes of other converters.However, a uctuation of the output current can be seen at 1.6s when the sudden change in the load takes place. This is due tothe uctuation in the output voltage.The dc-dc converter interfacing the PV successfully suppliesthe desired current with the some uctuation during the suddenload change. In addition, the dc-dc converter interfacing the fuelcell is able to maintain the desired output current during thevarious states of the system with some ripple in the output whenthe load is suddenly changed.Finally, the battery charger/discharger successfully suppliesthe desired amount of current from the battery, which helpsimplementing any kind of real-time operation algorithms in-volving the battery.B. Integrated Hybrid AC/DC OperationIn order to evaluate the integration of the ac/dc system andstudy the system performance during load variations, an exper-iment was conducted to show the system behavior and ac/dcpower sharing control while the load variations take place inactive and reactive power. In this experiment, the dc micro-grid quanties its dc power availability during the upcomingFig. 16. Performance of the various components of the dc microgrid corre-sponding to step change in the power reference. (a) shows the dc bus voltage, (b) the rectier current , (c) the bidirectional current , (d) thePV output current , (e) the fuel cells output current and (f) the bat-tery current .interval of time. Hence, it will inject any amount of active/re-active power demanded by the ac side, through the WAMPACconsidering all constraints.The dc system is connected to Bus-0050 through a bidi-rectional converter and injects any specied amount of activeand/or reactive power to the ac point of common coupling(PCC). In this experiment the dc microgrid is used to regulatethe voltage at the PCC.As shown in Fig. 17, a unity power factor load of 700-Wis initially connected to Bus-0050. The dc microgrid is com-manded to receive 100 W and zero Vars. Hence, ac grid takesthe responsibility of supplying both ac load and dc microgrid de-mand. The steady state voltage at PCC in this situation is 0.94p.u. whereas the voltage on the dc bus is 1 p.u. After 20 s, thedc microgrid is commanded by WAMPAC system to inject theSALEHI et al.: LABORATORY-BASED SMART POWER SYSTEM, PART II: CONTROL, MONITORING, AND PROTECTION 1415Fig. 17. Performance of the integrated hybrid ac/dc microgrid correspondingto step change in the load demand reference. (a) shows the load, dc, and ac activepower share, (b) the load, dc, and ac reactive power share, (c) the frequency ofthe ac bus, (d) the voltage of the ac and dc buses.total amount of demanded power on the ac side. Therefore, thevoltage amplitude is increased to 1.02 p.u. The controlled recti-er regulating the voltage on the dc bus maintains a voltage of1 p.u. after a transient period of around 6 s with an overshootof 0.02 p.u. A reactive load of 450 VARs is increased to PCCafter 43 s. Consequently the voltage amplitude drops to around0.95 p.u. The dc microgrid is then commanded by WAMPAC toinject 300-VARs to the ac grid. Hence, the voltage at the PCCin-creases to 0.98 p.u. The dc bus voltage is hardly affected by thischange in its reactive power reference. The frequency variationsare also shown in this gure. A maximum of 0.2 Hz deviationfrom 60 Hz is depicted in the measured frequency.This experiment shows that, using integrated controllers forhybrid ac/dc microgrids can enhance the performance of thesystem. Moreover, the developed test-bed is capable of beingeffectively used to test this kind of research ideas.VI. DISCUSSION AND CONCLUSIONSWe developed an integrated wide area system on the labo-ratory-scale smart grid test-bed with capabilities for WAMS,WACS, and WAPS. The implementation of this system in thereal-time software creates an environment for studying and ver-ifying new control and protection schemes for the whole powersystem. Moreover, it is very essential for power system studentsto experience, handle, and interact with smart grid componentsand its innovative operational aspects. Verication experimentswere presented to show the implemented system performanceand capabilities.This system was used for studying integrated wide area con-trol and protection system to monitor the system status for ab-normalities such as over/under voltage, overloads of equipment,and any other conditions. In the implemented example, when theloadings of a power system line increases, the relay indicatedthe system moving condition to the fault situation which may bedisconnected by available protection devices. Hence by settingthe protection relay under the settings of the physical relay, thecontrol scheme can retrieve the normal status by proper control-ling action. This case may be the changing of system topologyor power dispatching in alarms in a control center, self healingstrategies can maintain the system continuous operation withappropriate control scheme.This system can be used to monitor real-time system stabilityand security margin. The indices formulation and their imple-mentation on wide area monitoring and control centers werepresented. This provides an applicable view of system stabilityand security margin using PMUs in wide area networks. Thevoltage stability indices were measured during the operation ofthe power system when the load changes take places. A widearea monitoring system with high data resolution rate was de-veloped. This system was designed to have capabilities such asmaintaining system normal operation and take a proper reme-dial action when encountered by unexpected circumstances bymonitoring critical states in wide area system. As a result, thesystem operator will have proper knowledge and visualizationabout the power systems current situation and the distance ofstability margin.The developed system can be used for cascaded failures de-tection and applying proper remedies on the power system. Fol-lowing a disturbance, one or more components overload andhence fail. The equilibrium of the load ow will consequentlychange and the load will then be redistributed to other normalcomponents and this makes additional load transfer to other el-ements. Thus, a cascading failure is triggered by the overloadfailures and it cause networks collapse resulting in a blackout.The PMUs data can be used to follow the phase angle of eachbus to detect the system failures which may cause cascadedevents. In this process, we need to detect upcoming faults bynetwork data and analyze the network reaction to this outage bysome algorithms such State Estimation, N-1 Contingency, Faultcalculation, OPF, etc. Therefore, the online calculation softwarewas used to estimate system states following any circumstancesto apply self-healing reaction.This system was used for applying online setting of protec-tion devices. The protection coordination settings in a large areanetwork are completely dependent on the topology and systemstatus which are varying frequently. For example the applicationof distributed generation, whose generation depends on energyavailability such as photovoltaic and wind power, may causedifferent operational settings for relays connected to the samenetwork. This system could detect the grid status and will runprotection coordination software to achieve proper settings forrelays. Finally, it will apply new settings by real-time softwareand communication networks to the selected system protectiondevices.This paper is supplementary to a companion paper [32],which presents the design and implementation procedure1416 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 3, SEPTEMBER 2012for laboratory-scaled smart power system.Combining theachievements in both papers, the smart grid concepts can beimplemented in terms of experiments and operational sights.The authors attempted to design a microgrid power systemtest-bed lab in order to apply novel and innovative ideas forsmart grid applications using this hardware/software test-bedsetup.REFERENCES[1] P. Zhang, F. Li, and N. Bhatt, Next-generation monitoring, analysis,and control for the future smart control center, IEEE Trans. SmartGrid, vol. 1, no. 2, pp. 186192, 2010.[2] K. Seethalekshmi, S. N. Singh, and S. C. Srivastava, A synchrophasorassisted frequency and voltage stability based load shedding schemefor self-healing of power system, IEEE Trans. Smart Grid, vol. 2, no.2, pp. 221230, Jun. 2011.[3] M. McGranaghan, D. Von Dollen, P. Myrda, and E. Gunther, Utilityexperience with developing a smart grid roadmap, in Proc. PowerEnergy Soc. Gen. Meet.Convers. Del. Electr. Energy 21st Century,Jul. 2024, 2008, pp. 15.[4] P. P. Varaiya, F. F. Wu, and J. W. Bialek, Smart operation of smartgrid: Risk-limiting dispatch, Proc. IEEE, vol. 99, no. 1, pp. 4057,Jan. 2011.[5] A. Bose, Smart transmission grid applications and their supportinginfrastructure, IEEE Trans. Smart Grid, vol. 1, no. 1, pp. 1119, Jun.2010.[6] Z. Yingchen, P. Markham, X. Tao, C. Lang, Y. Yanzhu, W. Zhongyu,Y. Zhiyong, W. Lei, J. Bank, J. Burgett, R. W. Conners, and L. Yilu,Wide-area frequency monitoring network (FNET) architecture andapplications, IEEE Trans. Smart Grid, vol. 1, no. 2, pp. 159167, Sep.2010.[7] A. Armenia and J. H. Chow, A exible phasor data concentrator de-sign leveraging existing software technologies, IEEE Trans. SmartGrid, vol. 1, no. 1, pp. 7381, Jun. 2010.[8] V. Terzija, G. Valverde, C. Deyu, P. Regulski, V. Madani, J. Fitch, S.Skok, M. M. Begovic, and A. Phadke, Wide-area monitoring, protec-tion, and control of future electric power networks, Proc. IEEE, vol.99, no. 1, pp. 8093, Jan. 2011.[9] A. Mazloomzadeh, V. Salehi, and O. Mohammed, Soft synchroniza-tion of dispersed generators to micro grids for smart grid applications,in Proc. Innov. Smart Grid Technol. (ISGT), Jan. 1620, 2012.[10] G. Missout, J. Beland, and G. Bedard, Dynamic measurement ofthe absolute voltage angle on long transmission Lines, IEEE Trans.Power App. Syst., vol. PAS-100, no. 11, pp. 44284434, Nov. 1981.[11] P. Bonanomi, Phase angle measurements with synchronized clocks-principles and applications, IEEE Trans. Power App. Syst., vol. PAS-100, no. 11, pp. 50365043, Nov. 1981.[12] A. G. Phadke and J. S. Thorp, History and Applications of phasorMeasurements, in Proc. Power Syst. Conf. Expo., Oct. 29Nov. 2006,pp. 331335.[13] V. Salehi, A. Mazloomzadeh, and O. Mohammed, Development andimplementation of a phasor measurement unit for real-time monitoring,control and protection of power systems, in Proc. IEEE Power EnergySoc. Gen. Meet., Jul. 2429, 2011, pp. 17.[14] J. A. Momoh, X. Yan, and G. Boswell, Voltage stability enhance-ment using phasor measurement unit (PMU) technology, in Proc. 40thNorth Amer. Power Symposium (NAPS), Sep. 2830, 2008, pp. 16.[15] P. Kundur, J. Paserba, V. Ajjarapu, G. Andersson, A. Bose, C.Canizares, N. Hatziargyriou, D. Hill, A. Stankovic, C. Taylor, T.Van Custem, and V. Vittal, Denition and classication of powersystem stability IEEE/CIGRE joint task force on stability terms anddenitions, IEEE Trans. Power Syst., vol. 19, no. 3, pp. 13871410,Aug. 2004.[16] M. Moghavvami and F. M. Omar, Technique for contingency mon-itoring and voltage collapse prediction, IEE Proc. Gener., Transm.,Distrib., vol. 145, no. 6, pp. 634640, Nov. 1998.[17] I. Musirin and T. K. Abdul Rahman, Online voltage stability basedcontingency ranking using fast voltage stability index (FVSI), in Proc.IEEE/PES Transm. Distrib. Conf. Exhib.: Asia Pac., Oct. 2002, vol. 2,pp. 11181123.[18] M. V. Suganyadevia and C. K. Babulal, Estimating of loadabilitymargin of a power system by comparing voltage stability indices,in Proc. Int. Conf. Control, Autom., Commun., Energy Conserv. (IN-CACEC), Jun. 46, 2009, pp. 14.[19] M. M. El-Kateb, S. Abdelkadar, and M. S. Kandil, Linear indicatorfor voltage collapse in power systems, IEE Proc. Gener., Transm.,Distrib., vol. 144, no. 2, pp. 139146, Mar. 1997.[20] N. Sudipta and P. Shrabani, A novel technique for determination ofvoltage stability margin, in Proc. 5th IET Int. Conf. Syst. Safety, Oct.1820, 2010, pp. 15.[21] Y. Gong, N. Schulz, and A. Guzman, Synchrophasor-based real-timevoltage stability index, in Proc. 2006 IEEE PES Power Syst. Conf.Expo. (PSCE), Oct. 29Nov. 1 2006, pp. 10291036.[22] D. Thukaram, B. Ravikumr, V. Seshadri Sravan Kumar, Y. Prasad Rao,S. Surendra, and S. R. Kolla, Real-time monitoring of critical nodeswith minimal number of phasor measurement units, in Proc. Int. Conf.Power Syst. (ICPS), Dec. 2729, 2009, pp. 16.[23] Y. G. Zeng, G. Berizzi, and P. Marannino, Voltage stability analysisconsidering dynamic load model, in Proc. 4th Int. Conf. Adv. PowerSyst. Control, Oper. Manage. (APSCOM), Nov. 1114, 1997, vol. 1,pp. 396401, Conf. Publ. No. 450.[24] Y. Nguegan, Real-Time identication and monitoring of the voltagestability margin in electric power transmission systems using synchro-nized phasor measurements, Ph.D. thesis, Dept. Elect. Eng., KasselUniv., Kassel, Germany, 2009.[25] V. Salehi and O. Mohammed, Real-time voltage stability monitoringand evaluation using synchorophasors, in Proc. North Amer. PowerSymp. (NAPS), Aug. 46, 2011.[26] Y. Haibo, V. Vittal, and Y. Zhong, Self-healing in power systems:An approach using islanding and rate of frequency decline-based loadshedding, IEEE Trans. Power Syst., vol. 18, pp. 174181, Feb. 2003.[27] IEEE Standard for Synchrophasor Measurements for Power Systems,IEEE Std. C37.118.1-2011, Dec. 28, 2011, (Revision of IEEE Std C37.118-2005), pp. 161.[28] A. G. Phadke and J. S. Thorp, Synchronized Phasor Measurements andTheir Applications. New York: Springer, 2008, p. 197.[29] V. Salehi and O. Mohammed, Developing virtual protection systemfor control and self-healing of power system, in Proc. IEEE Ind. Appl.Soc. Annu. Meet. (IAS), Oct. 913, 2011, pp. 17.[30] A. Mohamed, M. Elshaer, and O. Mohammed, Dynamic energy com-mitment scheme for hybrid renewable energy systems in smart grid ap-plications, Energy Syst., 2011, MS. no. ENSY-D-11-00038, submittedfor publication.[31] A. Mohamed, V. Salehi, and O. Mohammed, Real-time energy man-agement algorithm for mitigation of pulse loads in hybrid AC/DC mi-crogrids, IEEE Trans. Smart Grid, 2011, submitted for publication.[32] V. Salehi, A. Mazloomzadeh, A. Mohamed, and O. A. Mohammed,Laboratory-based smart power system, Part I: Design and system de-velopment, IEEE Trans. Smart Grid, 2011, submitted for publication.Vahid Salehi (S09) was born in Tabriz, Iran, in1980.He received the B.S.degree in electricalengineering from the University of Tabriz in 2003and the M.Sc. degree in power system engineeringfrom the University of Tehran, Iran, in 2006. He iscurrently working toward the Ph.D. degree at FloridaInternational University, Miami.During 2003 to 2008 he was working in the EnergyResearch Institute (MATN), Tehran. His research in-terests include power system studies, smart grid, re-newable energy integration and energy conversion inpower systems, distributed energy resource integration, dynamic modeling ofpower systems, power system stability, protection, wide area monitoring, con-trol, and protection of power systems. His dissertation related to developmentand verication of control and protection strategies in wide area power systemsfor smart grid applications.Ahmed Mohamed (El-Tallawy) (S09) was born inMinia, Egypt, in 1984. He received the B.S. degreefrom the College of Engineering, Minia University,and the M.S. degree from the Faculty of Engineering,Minia University, in 2006 and 2009, respectively. Heis working toward the Ph.D. degree and a ResearchAssistant in the Electrical and Computer EngineeringDepartment, College of Engineering and Computing,Florida International University, Miami.From 2006 to 2009, he was a Research/TeachingAssistant in the College of Engineering, Minia Uni-versity. His current research interests are smart grids, renewable energy systems,hybrid ac/dc power systems, and sensorless control of electric machines.SALEHI et al.: LABORATORY-BASED SMART POWER SYSTEM, PART II: CONTROL, MONITORING, AND PROTECTION 1417Ali Mazloomzadeh (S09) was born in Tehran, Iran,in 1983. He received the B.S. degree in electricalengineering from Islamic Azad University, Tehran,in 2005 and the M.Sc. degree in electrical engi-neering from Amirkabir University of Technology(Tehran Polytechnic), Tehran, in 2009. He is cur-rently working toward the Ph.D. degree at FloridaInternational University, Miami.His research interests include power and energysystems, renewable energy sources, smart grid ap-plications, monitoring and control of power systems,power quality, and real-time analysis of power systems. Furthermore, he alsohas interest in articial intelligence, signal processing, and communication sys-tems for applications in the energy systems area.Osama A.Mohammed (S79SM84F94)re-ceived the M.S. and Ph.D. degrees in electricalengineering from Virginia Polytechnic Institute andState University, Blacksburg.He is a Professor of Electrical and Computer Engi-neering and the Director of the Energy Systems Re-search Laboratory at Florida International University,Miami. He published numerous journal articles overthe past 30 years in areas relating to power systems,electric machines and drives, computational electro-magnetics, and in design optimization of electromag-netic devices, articial intelligence applications to energy systems. He authoredand coauthored more than 300 technical papers in the archival literature. He hasconducted research work for government and research laboratories in shipboardpower conversion systems and integrated motor drives. He is also interested inthe application of communication and wide area networks for the distributedcontrol of smart power grids. He has been successful in obtaining a number ofresearch contracts and grants from industries and Federal government agenciesfor projects related to these areas. He has also published several book chapters,including Chapter 8 on direct current machinery in the Standard Handbook forElectrical Engineers, 15th Edition (McGraw-Hill, 2007) and a chapter entitledOptimal Design of Magnetostatic Devices: the Genetic Algorithm Approachand System Optimization Strategies, in Electromagnetic Optimization by Ge-netic Algorithms (Wiley, 1999).Prof. Mohammed is the recipient of the IEEE PES 2010 Cyril Veinott Electro-mechanical Energy Conversion Award. He is also a Fellow of the Applied Com-putational Electromagnetic Society. He is Editor of IEEE TRANSACTIONS ONENERGY CONVERSION, IEEE TRANSACTIONS ON MAGNETICS, , Power Engi-neering Letters, and also an Editor of COMPEL. He is the past President of theApplied Computational Electromagnetic Society (ACES). He received manyawards for excellence in research, teaching, and service to the profession and hasdelivered numerous invited lectures at scientic organizations around the world.He has been the general chair of several international conferences including;ACES 2006, IEEE-CEFC 2006, IEEE-IEMDC 2009, IEEE-ISAP 1996, andCOMPUMAG-1993. He has also chaired technical programs for other majorinternational conferences including IEEE-CEFC 2010, IEEE-CEFC-2000, andthe 2004 IEEE Nanoscale Devices and System Integration. He also organizedand taught many short courses on power systems, electromagnetics, and intel-ligent systems in the United States and abroad. He has served ACES in variouscapacities for many years. He also serves IEEE in various boards, committees,and working groups at the national and international levels.