Multipurpose Platform for Power System Monitoring and Analysis With Sample Grid Applications

566 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 Multipurpose Platform for Power System Monitoring and Analysis With Sample Grid Applications Tevhid Atalik, Student Member, IEEE, Isik Çadirci, Member, IEEE, Turan Demirci, Muammer Ermis, Member, IEEE, Tolga Inan, Member, IEEE, Alper Sabri Kalaycioglu, and Özgül Salor, Member, IEEE (in alphabetical order) Abstract— This paper is devoted to the design and imple- mentation of a multipurpose platform (MPP) for power system monitoring and analysis. This MPP is a novel device, which combines the abilities of a power quality (PQ) analyzer, an event logger, a synchronized phasor measurement unit, and an interarea oscillation identifier, all in one device. The multiple functions of the proposed MPP can serve the needs of the power system operators (SOs) as a wide-area monitoring system to observe the states and stability of the power system, and as a PQ analyzer to monitor the PQ events and parameters, permanently. Furthermore, the algorithms of flicker and har- monic current contributions at a point of common coupling can be embedded on this MPP, to determine the individual contributions of different loads supplied from the same bus. The operational features of the developed MPP have been tested on the Turkish electricity transmission system and its interfaces with the distribution system by installing 450 MPPs and integrating them with a monitoring and control center. The proposed all-in- one MPP can therefore meet the multiple requirements of the power SOs to serve the needs of modern electricity markets and convert an ordinary electricity system to a smart grid. Index Terms— Industrial power system testing, monitoring, power quality (PQ), power system measurements, power system monitoring. Manuscript received December 26, 2012; revised July 5, 2013; accepted July 8, 2013. Date of publication September 26, 2013; date of current version February 5, 2014. This work was supported in part by the Public Institutions Research Projects Support Group (KAMAG), Scientific and Technological Research Council of Turkey (TÜBITAK) and in part by the Turkish Electricity Transmission Corporation. The Associate Editor coordinating the review process was Dr. Robert Gao. T. Atalik and T. Demirci are now with Energy Institute, TUBITAK Marmara Research Center, Ankara TR06800, Turkey (e-mail: [email protected]; [email protected]). I. Çadirci is with the Electronics Engineering Department, Hacettepe University, Ankara 06100, Turkey (e-mail: [email protected]). M. Ermis is with Electrical and Electronics Engineering Depart- ment, Middle East Technical University, Ankara 06800, Turkey (e-mail: [email protected]). T. Inan was with the Power Electronics Department, Tubitak Uzay Institute, Ankara 06490, Turkey. He is now with TED University, Ankara 03048, Turkey (e-mail: [email protected]). A. S. Kalaycioglu was with Power Electronics Department, Tubitak Uzay Institute, Ankara 06490, Turkey. He is now with BIG Vienna Company, Vienna 1300, Austria (e-mail: [email protected]). Ö. Salor is with the Electrical and Electronics Engineering Department, Gazi University, Ankara 06500, Turkey (e-mail: [email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TIM.2013.2281556 I. INTRODUCTION POWER SYSTEM monitoring and analysis in large elec-tricity transmission systems and at their interfaces with the distribution systems are a critical issue to observe not only the operation states and stability of the power system, but also power quality (PQ) parameters and events. For the reliability and quality of the electrical supply, necessary countermeasures can be taken against the observed disturbances through wide- area monitoring the power system. Nowadays, various devices and instruments with different technical features are commer- cially available in the market for power system monitoring and/or analysis. These are specialized in one or more than one of power and energy, synchronized phasor (synchrophasor), harmonics, light flicker, transient, event, and PQ measure- ments. Almost all of them use microcontroller and/or DSP- based hardware. Although they are quite user friendly, their hardware and software are rigid, and hence do not allow the user to modify them for satisfying the possible measurement needs in the future. Most of them are more suitable for the measurements in the distribution and utilization sides of the power system. Various system approaches and devices are proposed in the literature for power system monitoring and analysis [1]–[22]. Some of these have been developed for the trans- mission and distribution systems with particular emphasis on voltage quality and/or voltage harmonics. A distributed fault location detection system is proposed and implemented to operate in a pilot region on the electricity distribution system in Quebec [1]. A virtual instrument environment with a new network server architecture for the database and server systems has been developed in [2], to process PQ measurements and GPS signals. Another distributed system based on GPS synchronization for harmonic phasor measurements has been proposed in [3], and implemented on an experimental setup with two measurement units. A web-enabled national event recording system has been developed and used on the electric- ity distribution system as described in [4] and [5]. Monitoring of PQ in a wide area based on wireless general packet radio service (GPRS) has also been proposed for implemen- tation [6]. Another internet-based distributed transient distur- bance detection system for the electricity distribution system 0018-9456 © 2013 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. ATALIK et al.: MPP FOR POWER SYSTEM MONITORING AND ANALYSIS 567 has been designed and tested in an industrial plant with two workstations [7]. A distributed system has been proposed and tested on four low-voltage measurement points in Politecnico di Milano, as given in [8]. All these systems are uniquely dedicated to PQ analysis, and/or event recording. Only [4] and [5] report nationwide systems operating on the electric- ity distribution systems. On the other hand, a nationwide real-time monitoring system has been presented in [9] and [10] to monitor all the electrical quantities and PQ parameters of the electricity transmission network, including its interfaces with the generation and distribution systems. In addition to PQ and event analysis and monitoring, power system stability monitoring has also been a critical issue for system operators (SOs) to prevent brownouts and blackouts. Synchronized phasor measurement techniques in power systems have been reported in [11] and [12], and a related standard to built synchrophasor measurement units has been released by the IEEE [13]. Various power system oscillation detection, monitoring, and analyses techniques have also been reported in [14]–[20]. Power system stability has been monitored over the past two decades by wide-area measurement systems (WAMSs) employing synchrophasor measurement units. Wide-area detection of voltage instability from synchronized phasor measurements has been reported in [15], and voltage stability protection and control using a wide-area network of phasor measurements has been presented in [16]. A method for monitoring interarea oscillations by synchronized phasor measurements has been described in [17]. In addition, power system stabilizers are shown to receive an additional input from such phasor measurement units [18], to damp out the low-frequency oscillations. WAMS-based detection of low-frequency oscillations in large-scale power systems has been proposed in [19], and an internet-based, real-time GPS-synchronized, wide-area frequency monitoring network has been realized in [20]. Each of these systems has been developed for a unique purpose, either to assess the power system stability parameters, or to monitor and analyze the PQ problems, which occur in the power system. On the other hand, a vision of next generation monitoring, analysis, and control functions for future smart control centers has been proposed in [21], and smart transmission grid applications and their supporting infrastructure have been discussed in [22]. In this paper, a multipurpose platform (MPP) for power sys- tem monitoring and analysis has been proposed, which incor- porates the multiple functions of PQ analysis, event logging, raw data collection, synchrophasor measurements, interarea oscillation identification, and flicker and harmonic contribution measurements all in a single device. This approach lowers the initial investment cost of WAMS in comparison with the system based on various application specific monitoring and analysis devices. Furthermore, since some measurements such as the flicker contribution and the harmonic contribution measurements at a point of common coupling (PCC), and raw data collection should be carried out temporarily, the design approach presented in this paper lowers also the running cost of WAMS, by eliminating the need for the circulation of some application-specific devices within the power system. On the other hand, anyone of the application-specific devices Fig. 1. Illustration of the operation modes of the multipurpose power system monitoring and analysis platform. such as the flicker-contribution meter, harmonic-contribution meter, and the interarea oscillation identifier for the oscillation frequency range from 0.1 to 0.2 Hz is not available on the shelf, unfortunately. Novel algorithms for flicker contribution, harmonic current contribution, and interarea oscillation iden- tification are then developed within the scope of the National Power Quality Project of Turkey. The corresponding software and the necessary calibration coefficients for harmonics can be remotely uploaded to the chosen MPPs. The proposed MPP has the additional feature of being reconfigured by software and programmed remotely via Inter- net or the existing fiber network of SO. MPP hardware is based on a minicomputer, where different algorithms of various operating functions have been implemented; 450 of those developed MPPs are now in use both at the transmission system level and the interface of the transmission and dis- tribution network of the Turkish Power System. The field experience has shown that the MPPs are very effective tools to evolve an ordinary electricity system toward a smart grid, when integrated with a monitoring and control center. II. SYSTEM DESCRIPTION The MPP for power system monitoring can operate in one or more than one of the following modes simultaneously according to the standards listed next to each mode: 1) PQ monitoring and analysis (IEC 61000-4-30 [23], IEC 61000-4-7 [24], IEC 61000-4-15 [25], and IEEE Std. 1159.1995 [26]); 2) event logging (IEC 61000-4-30 [23]); 3) synchrophasor measurement (IEEE C37.118.2005 [27]); 4) power system oscillation monitoring; 5) raw data collection; 6) flicker contribution measurement (IEC 61000-4-15 [25]); 7) harmonic contribution measurement (IEC 61000-4-7 [24]). 568 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 TABLE I OPERATION MODES AND USAGE FOR MPPS Fig. 1 shows the operational modes of the MPP. This plat- form performs online operation in modes 1)–4) and for all of these modes, simultaneous operation can take place. However, for modes 6) and 7), offline calculations are separately carried out. One or more than one operation modes of the platform among 1)–5) can be selected remotely. For modes 6) and 7), more than one platforms are required depending on the electricity system topology and offline calculations are performed after their data are transferred to a monitoring and control center. Operation modes and usage of MPPs are summarized in Table I. Such a monitoring and control center for either the transmission SO (TSO) or distribution SO (DSO) has been introduced in [9]. The overall system operation of the proposed MPPs in communication with the TSO or DSO monitoring and control center and other possible monitoring and control center(s) in cooperation is shown in Fig. 2. MPPs communicate over internet via ADSL, fiber optic network of the SOs or 3G/4G GPRS connection. MPPs can also be used as isolated platforms for local power system monitoring applications. All the operation modes in Fig. 1 necessitate raw data collection and usage. For modes 1), 4), 6), and 7), raw data are removed after they are processed; however, for modes 2), 3), and 5), raw data are stored in the device memory until they are transferred to an external medium. III. MPP DESIGN In this section, the hardware and software designs of the proposed MPP are described. Fig. 2. Illustration of the operation for the monitoring and control center together with multiple MPPs and possible other monitoring and control center(s) in cooperation. SSL: secure socket layer. (a) (b) Fig. 3. Developed MPP for power system monitoring and analysis. (a) General front view; is the logo of the National Power Quality Project. (b) Rear top view/2× three-phase current transducers are shown disconnected. A. Hardware Design The MPP in Fig. 3 is a device employing a mini-ITX motherboard whose block diagram is as shown in Fig. 4. It is composed of: 1) the mini-ITX motherboard; 2) a data acquisition (DAQ) unit; 3) analog signal conditioning circuitry (ASCC); ATALIK et al.: MPP FOR POWER SYSTEM MONITORING AND ANALYSIS 569 Fig. 4. Hardware block diagram of the implemented MPP. 4) digital I/O functionality (circuits); 5) GPS-based synchronization circuitry; 6) RS485 interface circuitry; 7) various power supplies; 8) human–machine interface (HMI) circuitry. 1) Mini-ITX Motherboard: The mini-ITX mainboard is an ultracompact motherboard form factor with a new level of integration for the new generation of small footprint value personal computers (PCs), information stations, and digital PC appliances [28]. This topology has been preferred in the development of MPP owing to its: 1) high computational capability with 1.5-GHz dual-core CPU and 2-GB DDR2 800-MHz RAM; 2) availability of on-board communication peripherals such as Ethernet port, RS232 communication ports, parallel ports, and USB ports; 3) executability of commonly used operating systems such as Linux and Windows, which allow the usage of high-level programming languages, and predeveloped open-source libraries, reducing the effort for code development and testing; 4) usability of solid-state drive, which supplies high-speed, high-capacity, and robust data storage; 5) easy upgradability to satisfy the needs of the advances in computer hardware technology. 2) DAQ Unit: A DAQ unit with the following minimum requirements is to be chosen. 1) A minimum number of 20 single-ended input ports to measure six single-ended voltage plus two neutral lines, and six differential current channels are required to perform measurements at two different three-phase measurement points. 2) Total sampling rate of the device should be at least 307.2 ksamples/s for 50-Hz utility grids to provide 512 samples/cycle for each of the 12 channels. 3) The device is capable of acquiring short and long dura- tion variations according to the IEEE Std. 1159.3-2002 [26]. Low- and medium-frequency oscillatory transients can also be recorded for one channel only, using a sampling rate of 500 ksamples/s. 4) A 16-bit analog-to-digital converter (ADC) is needed for sufficiently high accuracy in the voltage and cur- rent measurements (≤0.02%), and power computations (≤0.05%). 5) Sequential, configurable sampling property for DAQ unit is preferred for cost competitiveness at the expense of 570 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 more difficult software development for accurate power computations. 6) Adjustable input voltage range is a must in such applications, using high-quality programmable gain amplifiers. 7) The DAQ unit should be capable of operating in both internal clock mode for stand-alone mode operation and external clock mode for GPS-based synchronization. 8) Direct memory access minimizing the CPU clock time is required for acquisition and double-buffer mode acqui- sition should be used to avoid possible loss of data samples. 9) Programmable TTL digital I/Os are required to commu- nicate with external devices. 3) Analog Signal Conditioning Circuitry: ASCC has been designed to convert the measured voltage and current signals into six single-ended voltage and six differential current sig- nals, to the levels compatible with the analog input of the DAQ unit. Properties of ASCC are as follows. 1) Voltage inputs of ASCC are allowed up to 100 V rms line-to-line voltage for direct connection with the sec- ondaries of the measurement-type voltage transformers (VTs) existing at the transformer substations (TSs). Since DAQ card permits voltage signals in the range either of ±1.25 or ±2.5 V peak, the voltage signals that are differentially taken as phase-to-neutral voltage signals are reduced to the desired level by the use of active voltage dividers and then input as single-ended signals to the DAQ card. 2) Current inputs of ASCC are allowed up to 2.5-mA rms level. This signal level is achieved by reducing 5- or 1-A rms rated secondary currents of the exist- ing measurement-type current transformers at the TSs using ferrite-core, through-hole, high-frequency custom- design current transformers with a turns ratio of 1:2000 and a 100-� burden located inside the ASCC. Dif- ferential current signals are converted into differential voltage signals inside the ASCC and transferred to the DAQ card. 3) Low-offset, low-input-bias-current, high-precision linear operational amplifiers are required to satisfy high accu- racy. 4) Ultrahigh precision, improved-thermal, and long-term stability types of resistors and capacitors are to be used for accurate measurements. 5) Crosstalk among input channels is avoided using differ- ent operational amplifiers, one for each channel. 6) To suppress the signals above Nyquist frequency of 12.8 kHz, antialiasing filters are employed, one for each channel. 7) Surge protection facilities are included at the inputs of the ASCC. 4) Digital Input and Output Functionality: Digital inputs are designed to monitor the status of circuit breakers and disconnectors at the desired measurement points in the TSs. Digital outputs are used for control and alarm purposes. In the current MPP, 14 isolated digital inputs with 5–24 V dc input voltage range, and six isolated open-collector digital outputs, capable of driving relays with an external power supply are used to meet the needs of the measurement points at the typical TSs, thus providing an infrastructure to convert an ordinary electricity system to a smart grid. 5) GPS Synchronization Functionality: An MPP constitutes the building block of a wide-area monitoring system for use at different points of the transmission system and its interfaces with the distribution system. Synchronized measurements of voltages and currents are therefore a must for all the operation modes of the MPP shown in Fig. 1. Fully synchronized operation can only be achieved by: 1) simultaneous start of each 1-s period of sampling; 2) generation of synchronous triggering sampling signals inside each 1-s period; 3) synchronizing the operating system clock of each MPP with the universal time coordinate (UTC). Simultaneous start of each 1-s sampling period of all the MPPs can be achieved using pulse per second (pps) signal of a GPS receiver, which identifies the start of a second with respect to UTC. Synchronous triggering signals for all the MPPs are generated using the frequency output of the GPS receiver. pps signal and frequency output of the GPS receiver are given as input to a complex programmable logic device to generate a pulse train of 307 200 pps (12 channels × 25 600 samples/s). These pulses are then used to trigger the DAQ unit externally. GPS receiver is able to transmit information about date, time, and location to mini-ITX every second through RS232 port. This information is acquired by the operating system from RS232 serial communication and used by network time protocol (NTP) to keep the operating system time synchronized to UTC. Precision of time synchronization is further improved by applying the pps signal to data terminal ready input of the RS232 port. This signal generates an interrupt on serial port of the Mini-ITX motherboard and is fed into NTP control block with low delay and jitter. GPS starts to generate pps and frequency signals when its 3-D position is fixed by at least four satellites. Once the 3-D position of the stationary MPP is determined, accuracy of the pps signals can be maintained even with a single satellite. GPS receiver of all the MPPs can produce a pps signal that is locked to the UTC with a maximum rms error of 100 ns and the user can configure the frequency output of the GPS from 10 Hz to 10 MHz. 6) RS485 Communication Circuit: The isolated RS485 cir- cuit integrated into each MPP is used to export data to and col- lect data from power system measuring and protection devices such as relays, energy analyzers, and energy meters existing at the associated TSs. RS485 circuitry of the current MPPs is capable of transmitting and receiving data at a maximum rate of 115 200 kb/s. Half-duplex master or slave modbus protocol can be chosen as the communication protocol. 7) Power Supplies: All the MPPs should be capable of being powered from the uninterruptable dc and/or ac sources available at TSs. For this purpose, input voltage ratings of 100–375 V dc and 70–264 V ac at 50/60 Hz are recommended in the design of built-in MPP power supplies. ATALIK et al.: MPP FOR POWER SYSTEM MONITORING AND ANALYSIS 571 Fig. 5. Data flow diagram of the MPP software common to all the operation modes. 8) HMI Circuitry: The HMI circuitry of the MPP should allow the users to monitor the basic measurements of the MPP through LCD screens and tactile buttons. The HMI can also be used to change the settings of the device, such as voltage or current transformer ratios, local IP address, and measurement point ID of the MPP by the authorized operators. B. Software Design MPP software is designed specifically to perform one or more of the operation modes shown in Fig. 1. The general data flow diagram of the developed software, which is common to all the operation modes of the MPP, is shown in Fig. 5. MPP is designed to collect three-phase, analog current, and voltage waveforms obtained from two different measurement points. The analog data are then converted to a suitable range by means of the ASCC, to be input to the ADC. These data are then sampled at a rate of 25.6 kHz/channel (30.72 kHz/channel), which corresponds to 512 samples/cycle of the 50-Hz (60 Hz) power frequency. The sampled data are processed using 3-s buffers. Each buffer is analyzed by the application-specific signal analysis block of the corresponding operation mode. Analyses results are then served to the SO (TSO/DSO), or to the control system of a FACTS device, or they can be stored into a suitable medium. Signal analysis block of each operating mode is described in the following sections. 1) PQ Monitoring and Analysis Software: PQ monitoring and analysis software is designed to comply with the IEEE Std. 1159 [26], IEC61000-4-30 [23], IEC61000-4-7 [24], and IEC61000-4-15 [25] standards. All the PQ parameters defined in the IEC61000-4-30, and some additional electrical quantities are computed, as shown in Fig. 6. The wave- form magnitude, harmonics, and interharmonics are computed for every 10 cycles of both voltage and current channels. Unbalance is computed only for the voltage channels, every 10 cycles as recommended in [23]. Active, reactive, and apparent powers and the power factor are also computed every 10 cycles. Although these parameters are not included in PQ parameters in [23], they are essential to power SOs. Fig. 6. PQ monitoring and analysis software flowchart. The 10-cycle computations are then aggregated first to 3 s, then to 10 min for reducing the data size to be stored as recommended in [23]. The implemented MPP, however, can also be configured to serve 10-cycle or 3-s data if required. The frequency measurement is achieved every 10-s with the fundamental frequency output being the ratio of number of integral cycles counted during the 10-s time clock interval divided by the cumulative duration of the integer cycles. Before each assessment, a low-pass filter is used to filter out harmonics and interharmonics, and zero-crossings are used to determine integer cycles, as recommended in [23]. The cutoff frequency of this low-pass filter is chosen as 65 Hz. In addition, a high-pass filter with a cut-off frequency 40 Hz is also applied before zero-crossing detection to eliminate frequencies below the fundamental frequency. Hence, the zero- crossings of the filtered voltage signal are estimated much more accurately. Short- and long-term flicker values (Pst and Plt) are com- puted with Pst every 10 min, and from these Pst values, Plt is computed every 2 h for each voltage channel according to [25]. 2) Synchrophasor Measurement Software: Synchrophasor measurements are carried out according to IEEE Std. C37.118-2005 [27]. The frequency, and voltage amplitude and phase angle are computed as 10 values/s for each voltage chan- nel. Synchrophasor measurement packets are then obtained according to [27], as shown in Fig. 7. 3) Event-Logger Software: Event-logger software detects three types of events defined in [23] every half cycle of the 50-Hz power frequency, as recommended in [23]. One-cycle- 572 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 Fig. 7. Synchrophasor measurement software flowchart. Fig. 8. Event-logger software flowchart. long data windows with half-cycle overlaps are used, which helps not to miss very short-duration events. Detected events are voltage sags, voltage swells, and voltage interruptions, defined in [23], and no events are defined for current channels. Events are detected separately for each channel and when an event is detected in any one of the voltage channels, 3-s raw data of the event, 0.5 s before the event start, and 2.5 s after are stored both for all voltage and current channels. If the event takes longer than 3 s, 3-s raw data is labeled as event beginning and a search is activated to detect the end of this event to make another 3-s data record of the end of the event. The MPP software sends the raw data of any event as soon as it is detected to a TSO or DSO monitoring and control center, or it stores the data on a storage medium, as shown in Fig. 8. During a voltage, sag, swell, or an interruption, measured PQ parameters are flagged since the measurement algorithms may produce unreliable values, as required in [23]. Flagging is inherently included in the MPP algorithm due to time stamping of raw data collection during events. Fig. 9. Power system oscillation identifier software flowchart. 4) Oscillation Identifier Software: Oscillation identification software operates on the frequency waveform of one voltage channel, unlike synchrophasor or power measurements. As shown in Fig. 9, the consecutive frequency values obtained every 10 ms construct the frequency waveform against time. The frequency waveform is then filtered out to eliminate the dc and noise components. The output of the filter gives the frequency deviation with respect to power frequency (50 Hz) in time, since the mean value of the frequency around 50 Hz is removed. Discrete Fourier transform (DFT) with 50-mHz resolution is applied on this frequency deviation to attain the oscillation components existing in the power system. DFT components in the predefined oscillation band, deter- mined according to the simulations achieved on the power system model, are compared with the deviation threshold values of the oscillation identifier system. Whenever the fre- quency oscillation is beyond the peak-to-peak preset threshold value (e.g., 10-mHz peak-to-peak), the output of the identifier is set to either −1 or +1 logic level, for the positive and negative half-cycles of the frequency deviation, respectively. The identifier output is zero whenever the frequency deviation within the predefined oscillation band is less than the preset threshold value. The corresponding identifier output signal is then sent as a reference signal to the FACTS device, as shown in Fig. 9. When the identifier output is zero, no action is taken by the FACTS device. For the identifier output set to +1 level, the FACTS device is controlled in the inductive reactive power mode to reduce the bus voltage and hence, the active power ATALIK et al.: MPP FOR POWER SYSTEM MONITORING AND ANALYSIS 573 Fig. 10. Flicker contribution measurement flowchart. demand, whereas for the −1 level, it is controlled in the capacitive reactive power mode, to increase the bus voltage and hence the active power demand. This action is taken to damp out the oscillations by modulating the corresponding PCC voltage, and hence active power demand in time syn- chronization with the unwanted oscillations. The success of the damping action depends on the number of FACTS devices operating in this mode, their location in the power system, their installed MVAr capacity allocated for this purpose, and limitations imposed on PCC voltage magnitude. 5) Flicker Contribution Measurement: MPPs are able to compute the individual flicker contributions of the loads supplied from a PCC in addition to the background flicker contribution coming from the supply side according to the method described in detail in [29]. Flicker contribution of an individual load means what the IEC flickermeter measurement would be if it were the only load supplied from the common bus with a constant flicker-free source. Therefore, this func- tionality of the MPP is crucial for power SOs to take necessary countermeasures against flicker sources. The standard IEC flickermeter, whose operation principles are defined in the IEC Standard 61000-4-15 [25], computes the flicker using the voltage measurement at a PCC. Hence, it is not possible to directly discriminate the individual flicker contributions of various loads supplied from the same PCC, using the IEC flickermeter. It has been shown in [29] that individual flicker contri- butions of the loads are mainly caused by the variations of their reactive current components and it is possible to decou- ple the flicker contribution of the interconnected electricity Fig. 11. Harmonic current contribution measurement flowchart. system from the flicker contributions of the loads. The flicker contribution-meter operation mode of the MPP uses reactive components of the load currents and the short-circuit system impedance, either provided by the utility or computed from the current and voltage measurements at the PCC. The voltage drop on the short-circuit system impedance caused by the reactive current variation of each load is then fed to the IEC flickermeter for providing the individual flicker contribution of the corresponding load. The block diagram of the flicker contribution measurement method of the MPPs is shown in Fig. 10. 6) Harmonic Current Contribution Measurement: The har- monic current contributions of the loads supplied from a PCC can be computed using multiple MPPs measuring load feeders as shown in Fig. 11. This method defines the harmonic current contribution of a load supplied from a PCC as the amount of harmonic current injected by that load if it were the only harmonic generating load, i.e., if there were no harmonic contributions coming from upstream loads. Harmonic contributions of both the individual loads sup- plied from the PCC and the upstream loads at the utility side can be determined based on identifying harmonic cur- rent sources in the system analytically through the harmonic Norton’s equivalent circuit model of the system constructed on the basis of the sample by sample, time-synchronized field measurements by calculating utility and load impedances from the field measurements of load currents and PCC voltage [30]. Harmonic computations are performed directly on raw data using 10-cycle time windows, which coincide with the IEC harmonic subgroup computation standard [24]. In this method, the load impedances on an electrical system are estimated using fundamental components of the time-synchronized cur- rent and voltage measurements obtained from each load feeder. Then, the utility system impedance is attained using the Power System Simulator for Engineering load flow and shortcircuit simulations. After obtaining all the load and utility impedances, the electrical system is analyzed analytically as 574 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 TABLE II PERFORMANCE SPECIFICATIONS OF MPP AND ACCURACY COMPARISON WITH SOME COMMERCIALLY AVAILABLE PQ ANALYZERS shown in Fig. 11, using the superposition principle through the harmonic Norton equivalent circuit model of the system. Multiple MPP measurements should be available for such an analysis to be made in the monitoring and control centers. C. Measurement Range and Accuracy of the MPP The measurement accuracy of the MPPs has been certified to comply with the standards IEC 61000-4-30 [23], IEC 61000-4-7 [24], and IEC61000-4-15 [25] by an indepen- dent institution, the National Physical Laboratory (NPL) in U.K. [31]. A summary of the measurement range and accuracy of the MPP certified by NPL is shown in Table II. It is seen from Table II that the measurement accuracy of the MPP outperforms the Class A limits specified in the IEC 61000-4-30 Standard. It should be noted that according to IEC 61000-4-30 Standard, the measurement transducers and their associated measurement uncertainties are not considered in the norma- tive part of the instruments. Nevertheless, it is not possible to measure accurately the harmonic voltage components at the transmission voltage levels, by the use of conventional voltage measurement transformers unless a calibration work based on measurements with some special equipment such as resistive-capacitive voltage transducers (RCVTs) or fully optical type transducers is carried out. In addition, magnitudes of harmonic voltage components are quite low as compared with the fundamental component, which makes them difficult to measure accurately. The frequency response characteristics of conventional CVTs show that conventional CVTs are not able to measure voltage harmonics accurately because their lowest frequency resonance peaks appear at frequencies ATALIK et al.: MPP FOR POWER SYSTEM MONITORING AND ANALYSIS 575 Fig. 12. Some measurement points of MPPs on the electricity transmission system (Turkey extends >1600 km from west to east). electricity transmission system in Fig. 12, on which 450 MPPs are currently in operation at various locations distributed all over the country. Data collected at the databases of the TSO monitoring center can be analyzed by the developed PQ mon- itoring interfaces as explained in detail in [9]. Snapshots from those interfaces are presented in this section for each operation mode of the MPP described in Section II and Fig. 1. Flicker- and harmonic-current contribution measurement modes are carried out temporarily whenever they are needed at different points of the transmission system; thereby the processed data collected by the TSO monitoring center are reported later by a host computer. A. Power Monitoring Turkish electricity transmission system and European network of electricity transmission are interconnected via Babaeski (Turkey)—Neo Santa (Greece) 400kV O/H line and Hamitabat (Turkey)—Maritsa (Bulgaria) 400kV double-circuit O/H line [37], [38]. The associated measurement points are marked, respectively, by MP1 and MP2 in Fig. 12. Power transfer between the two transmission systems has crucial importance for both the Turkish TSO and European network of TSOs of electricity (ENTSO-E) and should therefore be monitored continuously. Fig. 13 shows a snapshot of the power monitoring screen in which Bulgaria-1 O/H line circuit is out of service and the net power transfer in the intertie is given in the last row. Since the interconnection is viewed from the Turkish TSO side, negative P in Fig. 13 corresponds to import of electricity from ENTSO-E. B. PQ Problem Identification Severe PQ problems can be identified using data collected and processed by MPPs. One of those problems arises from Fig. 13. Power flow at the interconnection lines between Turkish elec- tricity transmission system and European network of electricity transmission (1-s averaged data). dense agricultural irrigation in the Southeastern parts of Ana- tolia during late spring, summer, and early autumn seasons. Total power demand of the irrigation area, marked on Fig. 12, may exceed 2 GW from the transmission system during the irrigation season. For this purpose, thousands of submersible water pumps are installed in the area. These are supplied from the low-voltage sides of individual 34.5/0.4-kV transformers connected to a weak distribution system. The mean well depth is 200 m and each pump is equipped with a Wye/Delta starter. When more and more submersible water pump motors are energized, potential drop in the lines of the distribution system, on the leakage impedance of the pump transformer 576 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 Fig. 14. PQ monitoring of severe voltage sag problem in 380-kV Sanliurfa TS busbar arising from dense agricultural irrigation (3-s averaged data). and on the six cables feeding the three-phase induction motor increase markedly, resulting in a significant drop in the applied stator voltage of the pump motors. Since the electromechanical torque produced by the three-phase induction motor is directly proportional to the applied voltage, the drastic drop in the applied stator voltage causes the movement of the operating point toward the pull-out torque point on the useful portion of induction motor torque-speed characteristic. This phenomenon leads to a continuously growing increase in the motor current and hence the line current in the distribu- tion and the transmission systems. Since the reactive power demands of series inductive reactances of motor windings, transformers, and O/H lines are directly proportional to the square of current (Q = I 2 X), movement of induction motor operating points toward the pull-out torque increases the reac- tive power demand from the distribution system and hence the transmission system, e.g., when the current is doubled, reactive power demand is quadrupled. Huge amounts of reactive power demands from the distribution and transmission systems cause further voltage drops at the distribution and transmission system busbars, which in turn results in increased line currents and hence reactive power demands. This positive feedback effect goes on until all the pump motors are collectively disconnected from the grid by their overload relays, when their currents exceed twice the rated value. A part of this irrigation area is supplied from 380-kV feeder of Sanliurfa TS. A typical snapshot of the querying screen taken at this point (MP3 in Fig. 12) is shown in Fig. 14. As can be understood from this figure, collective and automatic disconnection of pump motors from, and then their reconnection to the grid repeatedly occurs several times a day during the irrigation season. The drastic increase in reactive power demand may reach 370 MVAr resulting in 40-kV drop in the 380-kV busbar voltage. C. Synchrophasor Measurements Synchrophasor measurement mode of the MPPs is used for wide-area monitoring of the synchronized voltage phasors and frequency of the electricity transmission system. MPPs have been used effectively to monitor the status of the Turkish electricity system during the interconnection with the European network of electricity transmission. Two sample MPPs, which are used in synchrophasor mode, are shown as MP4 and MP5 on the map in Fig. 12. These MPs are selected from the southeastern (Elbistan-B lignite power plant) and the western (Hamitabat natural gas power plant) parts of the electricity transmission system shown in Fig. 12. A sample snapshot from the synchrophasor monitoring screen is shown in Fig. 15. It can be observed from Fig. 15 that active power flows from Elbistan-B TS toward Hamitabat TS in the observation period, since the phase difference between the associated voltages (Feeder1 subtracted from Feeder2) is positive and around 20°, as shown on the last row of the snapshot. D. Event Logging Event logging mode of the MPPs is able to collect raw data of the detected events and to send it immediately to the TSO monitoring center. An example of a short voltage interruption detected at the Esenboga TS, showed as MP6 on the map in Fig. 12, is shown in Fig. 16. The cause of this event is a single- line-to-ground fault in one of the feeders of the distribution ATALIK et al.: MPP FOR POWER SYSTEM MONITORING AND ANALYSIS 577 Fig. 15. Synchrophasor monitoring screen (100-ms averaged data). Fig. 16. Event logging at the interface of transmission and distribution systems. A single-line-to-ground fault followed by double-line-to-ground fault in one of the distribution feeders. system. It is followed by a double-line-to-ground fault until the fault is cleared. The associated fault periods are marked on Fig. 16. The effectiveness of the relay coordination could be checked by such records. Moreover, data collected from several MPPs could give information about the cause of the event and its penetration into the whole transmission system. E. Interarea Oscillation Monitoring and Damping Power system simulations carried out have shown that after the interconnection of the Turkish power system with Fig. 17. Interarea oscillation monitoring (100-ms averages). the European network of electricity transmission, interarea oscillations at ∼0.15 Hz on weak tie lines (Babaeski-Neo Santa, Hamitabat-Maritsa) are expected to occur under weak system conditions [39], [40]. To maintain the overall system stability against interarea oscillations, 154 kV, ±50 MVAr T-STATCOM operating at Sincan TS (MP7 on the map in Fig. 12) [41], [42] and four large static VAr compensators (SVC) operating at various iron and steel plants are equipped with additional control facilities to damp out the oscillations, and power system stabilizers of large power plants in the Turkish power system are retuned [40]. MPPs working in interarea oscillation monitoring mode identify the oscillations based on frequency variation 578 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 Fig. 18. Part of the interconnected system where flicker contribution measurements are carried out. measurements and create the necessary control signals to mod- ulate the reactive power output of T-STATCOM and SVCs, synchronously with the oscillation frequency. A sample snapshot of reactive power and frequency vari- ations on 400/154-kV autotransformer secondary at Sincan TS (MP7 in Fig. 12) and sum of power flows on weak tie lines (sum of MP1 and MP2 in Fig. 12) is shown in Fig. 17. Interarea oscillations are apparent on active power and frequency. This snaphot is a result of the query dated at April 23, 2011, when the Easter Holiday in Europe and the National Independence and Children’s Day in Turkey coincided. Since this date corresponds to minimum load conditions in both systems, persistent oscillations at ∼0.15 Hz are observed, but they were successfully damped out. Whenever the 0.15-Hz component of grid frequency variations exceeds ±10 mHz (middle record in Fig. 17), the associated MPP detects the occurrence of an interarea oscillation and processes the filtered data for three consecutive periods of interarea oscillation to make a decision whether the oscillation is growing or not. In the case of a growing oscillation, the associated MPP generates a control signal for the T-STATCOM or SVC, which is synchronized with the 0.15-Hz component of the grid frequency variations. That is, in the positive half-cycle of the 0.15-Hz frequency component, the control system forces T-STATCOM to generate the preset value of MVArs in the capacitive region, while in the negative half-cycle, in the inductive region. The reactive power variations in Fig. 17 (top record) rep- resent the combined response of T-STATCOM, passive loads, natural gas plant, and other overhead lines connected to the same 154-kV bus during damping action of T-STATCOM. F. Flicker Contribution Monitoring at PCC In the case where IEC flickermeter measurements at a certain PCC exceeds the limits, with more than one flicker sources connected to that PCC, individual effects of each flicker source can be determined by MPPs, using a novel Fig. 19. Flicker contributions of various industrial plants in Fig. 18, supplied from Iskenderun-2 TS (MP8 in Fig. 12). (a) All loads’ contributions. (b) Background flicker of interconnected system. (c) Contribution of rolling mill stands and small industry [29]. method based on reactive current components of individual loads [29] instead of busbar voltages. Background flicker, which is the upstream flicker component coming from the source side, can also be computed. A part of the region of the Turkish electricity transmis- sion system monitored by MPPs operating in the flicker- contribution-meter mode (MP8 in Fig. 12) is shown in Fig. 18. The PCC in Fig. 18 supplies three different downstream loads, D (dc arc furnace load), A (ac arc furnace load), and R + In (rolling mill and small industry). A snapshot of 360 min of the MPP measurements is shown in Fig. 19. Fig. 19(a) shows a comparison of the short-term ATALIK et al.: MPP FOR POWER SYSTEM MONITORING AND ANALYSIS 579 Fig. 20. Flicker contribution of the induction melting furnace (Im in Fig. 18) operating in a small steel melt shop [29]. flicker (Pst) measurement of the MPP at the PCC according to IEC flickermeter standard [25], and the total contributions of the loads using the novel method in [29]. It is observed that the total flicker contributions of the loads exceed the penalty limit defined by the related regulations in a significant portion of the measurement period, however, contribution of the upstream loads cannot be disregarded. For example, when the total flicker contributions of the loads is relatively low from 30th to 80th min, Pst at the PCC is still much above the penalty limit, which shows that there is a significant amount of flicker contribution from the upstream loads on the source side of the PCC. Flicker contribution of the upstream loads is plotted together with the flicker at the PCC in Fig. 19(b). A comparison of the of Fig. 19(a) and (b) shows that flicker measurement at the PCC is not simply the algebraic sum of the flicker contributions of the loads A, D, R + In, and those of the upstream loads. Both contributions can be high from time to time as observed from measurements of 90th to 120th min. Contributions can attenuate or amplify the flicker at the PCC depending on the phases of the voltage envelope variations. Finally, Fig. 19(c) shows the flicker contribution of the load R + In, which is lower than the penalty limit during >99% of the measurement period. This is an expected result since rolling mills and small industry usually make negligible flicker contributions due to their nature of operation. Fig. 20 shows another snapshot of 800 min illustrating the flicker contribution of the load indicated as Im on the map in Fig. 18. Im is an induction melting furnace load operating in a small steel melt shop, whose operating characteristics have been presented in detail in [43]. It is observed from Fig. 20 that although Pst measurements at the PCC to which the Im is connected is much higher than the penalty limits of the related regulations, contribution of the induction furnace load is quite below the limits for a significant amount of the measurement period. G. Harmonic Current Contribution Monitoring at PCC The harmonic current contribution of a load supplied from a given PCC should be considered as the amount of harmonic Fig. 21. Results obtained by harmonic current contribution mode of MPPs at the PCC in Fig. 18. (a) I2, I5, and TDD measurements at the PCC measured by PQ analyzer mode of the MPP. (b) I2 contributions of loads A, D, and R + In measured by the harmonic contribution measurement mode of the MPP. (c) I5 contributions of loads A, D, and R + In measured by the harmonic contribution measurement mode of the MPP. (d) Comparison of the fifth harmonic contribution of Plant D and the fifth harmonic current measurement obtained from the feeder directly. current injected by that load if it was the only harmonic generating load at that PCC. Since the direct harmonic current measurements may provide misleading information due to the interaction of harmonic sources with neighboring shunt harmonic filters and the possibility of harmonic current com- ponents sunk by the equivalent load impedances, harmonic 580 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 Fig. 22. Comparison of (a) amplitude errors and (b) phase shifts of RCVT and conventional voltage measurement transformers [45]. current contribution of each plant is to be computed individ- ually. Using the proposed method in [30], harmonic current contributions of loads are decoupled from the upstream effects. The TSO has carried out PQ measurements at 154-kV PCC and at the feeders of the mentioned loads in Fig. 18, using the PQ analyzer functionality of the MPPs. As can be understood from snapshots shown in Fig. 21(a), the measured values of TDD, I2, and I5 exceed the limits in the associated harmonic standards. To determine the plants, which are responsible from this undesirable situation, the MPPs are connected to 154-kV feeders of the plants A, D, and R + In, and operated in harmonic current contribution measurement mode. Computed contributions of these plants to I2 and I5 are given in Fig. 21(b) and (c), respectively. The source of the second harmonic current component problem is the iron and steel plant having a large ac arc furnace (Plant A), as expected [44]. However, the source of the fifth harmonic problem is identified to be the iron and steel plant having a large dc arc furnace (Plant D), as an unexpected result. This result is attributed to the nonidealities and the unbalanced operation of the two six-pulse ac/dc converters of 12-pulse dc arc furnace rectifier. Plants marked by R + In do not lead to any harmonic problem to the transmission system. A comparison of the fifth harmonic contribution of Plant D and the fifth harmonic current measurement obtained from the feeder directly is given in Fig. 21(d). It is observed that the contribution of Plant D is higher than the measurement of I5 obtained by the PQ measurement mode of the MPP at that feeder according to [24], upon the application of the proposed harmonic contri- bution algorithm. This is because the effect of fifth harmonic current owing to the upstream loads in the supply side and Fig. 23. Various views of field measurements by mobile RCVT and optical VCT type transducer infrastructure ( is the logo of the National Power Quality Project of Turkey). (a) Custom-design, mobile measurement and recording system for Transient Events and Voltage/Current Harmonics. 170 kV or 400 kV optical or RCVT type transducers can be mounted on the custom-designed trailer. Each transducer can be tilted by a hydraulic mechanism. (b) Each transducer can be lifted up and down by hydraulic controlled scissors. Horizontal span between two neighboring transducers is also adjustable. (c) Measurement of switching transients in one of the TSs of Electricity Transmission Company of Turkey (400-kV combined, optical voltage–current transducers (VCTs) on the left and 400-kV RCVT on the right). (d) Measurement of voltage harmonics by a single RCVT in one of the 170-kV TSs of Electricity Transmission Company of Turkey. the actual fifth harmonic current contribution of the Plant D itself can either be additive or subtractive, depending on the phase angle between the corresponding phasors. In the case of no harmonic contribution from upstream loads, the measured harmonic current values and calculated contributions would be the same in both magnitude and phase angle. V. CONCLUSION The MPP presented in this paper is a compact, robust, and flexible environment, which is able to satisfy various needs of power system analysis and monitoring in one device. The wide-area power system monitoring and analysis employing MPPs can be used to improve the efficiency, reliability, quality, and sustainability of electricity services. Since MPP hardware is upgradable and software is reconfigurable, it may reach much more powerful computation and analysis capacity in the future, with the advents in minicomputer technology and improvements in algorithms. The device can be easily con- verted to a low-cost distribution system monitoring device, by increasing the number of input channels and making necessary changes in the hardware and software. The proposed MPPs can also constitute the building blocks of next generation SCADA systems in the future with high design flexibility. When integrated with a monitoring and control center, the MPPs are very effective tools to evolve an ordinary electrical system toward a smart grid. ATALIK et al.: MPP FOR POWER SYSTEM MONITORING AND ANALYSIS 581 APPENDIX A. Frequency Response of Measurement Transformers Amplitude errors and phase shifts against frequency of RCVTs are compared with conventional VTs in Fig. 22(a) and (b) [45]. B. Harmonic Measurement Infrastructure for Transmission System Custom-designed harmonic measurement and transient recording infrastructure, which is developed within the scope of the National Power Quality Project of Turkey [10], is shown in Fig. 23. ACKNOWLEDGMENT The core technologies presented in this paper have been developed within the scope of the National Power Quality Project of Turkey. The authors gratefully acknowledge Turkish Electricity Transmission Corporation authorities for the effort allocated to spread the MPPs over the whole country. Special thanks to NPL of U.K., for the careful work and valuable effort spent during measurement accuracy tests. REFERENCES [1] M. Tremblay, R. Pater, F. Zadova, D. Valiquette, G. Simard, R. Daniel, M. Germain, and F. Bergeron, “Accurate fault-location technique based on distributed power-quality measurements,” in Proc. 19th Int. Conf. Electr. Distrib., Vienna, Austria, May 2007, pp. 1–4. [2] S. J. S. Tsai and C. C. Luo, “Synchronized power-quality measure- ment network with LAMP,” IEEE Trans Power Del., vol. 24, no. 1, pp. 484–485, Jan. 2009. [3] A. Carta, N. Locci, and C. Muscas, “GPS-based system for the measure- ment of synchronized harmonic phasors,” IEEE Trans. Instrum. Meas., vol. 58, no. 3, pp. 586–593, Mar. 2009. [4] D. Divan, G. A. Luckjiff, W. E. Brumsickle, J. Freeborg, and A. Bhadkamkar, “A grid information resource for nationwide real-time power monitoring,” IEEE Trans. Ind. Appl., vol. 40, no. 2, pp. 699–704, Mar./Apr. 2004. [5] W. E. Brumsickle, D. M. Divan, G. A. Luckjiff, J. W. Freeborg, and R. L. Hayes, “Operational experience with a nationwide power quality and reliability monitoring system,” in Proc. IEEE Ind. Appl. Soc. Annu. Meeting, Oct. 2003, pp. 1063–1067. [6] P. K. Lee and L. L. Lai, “A practical approach to wireless power quality, energy and facilities monitoring system,” in Proc. IEEE Power Energy Soc. General Meeting, Convers. Del. Electr. Energy 21st Century, Jul. 2008, pp. 1–3. [7] P. Daponte, M. D. Penta, and G. Mercurio, “TransientMeter: A distrib- uted measurement system for power quality monitoring,” IEEE Trans. Power Del., vol. 19, no. 2, pp. 456–463, Apr. 2004. [8] L. Cristaldi, A. Ferrero, and S. Salicone, “A distributed system for electric power quality measurement,” IEEE Trans. Instrum. Meas., vol. 51, no. 4, pp. 776–781, Aug. 2002. [9] T. Demirci, A. Kalaycioglu, D. Kucuk, O. Salor, M. Guderl, S. Pakhuylu, T. Atalik, T. Inan, I. Cadirci, Y. Akkaya, S. Bilgen, and M. Ermis, “Nationwide real-time monitoring system for electrical quantities and power quality of the electricity transmission system,” IET Generat. Transmiss. Distrib., vol. 5, no. 5, pp. 540–550, May 2011. [10] (2012, Jun.). National power quality project of Turkey [Online]. Avail- able: http://www.guckalitesi.gen.tr [11] A. G. Phadke, “Synchronized phasor measurements in power systems,” IEEE Comput. Appl. Power, vol. 6, no. 2, pp. 10–15, Apr. 1993. [12] R. O. Burnett, Jr., M. M. Butts, T. W. Cease, V. Centeno, G. Michel, R. J. Murphy, and A. G. Phadke, “Synchronized phasor measurements of a power system event,” IEEE Trans. Power Syst., vol. 9, no. 3, pp. 1643–1650, Nov. 1994. [13] K. E. Martin, G. Benmouyal, M. G. Adamiak, M. Begovic, R. O. Burnett, Jr., K. R. Carr, A. Cobb, J. A. Kusters, S. H. Horowitz, G. R. Jensen, G. L. Michel, R. J. Murphy, A. G. Phadke, M. S. Sachdev, and J. S. Thorp, “IEEE standard for synchrophasors for power systems,” IEEE Trans. Power Del., vol. 13, no. 1, pp. 73–77, Jan. 1998. [14] D. Karlsson, M. Hemmingsson, and S. Lindahl, “Wide area system monitoring and control,” IEEE Power Energy Mag., vol. 2, no. 5, pp. 68–76, Sep./Oct. 2004. [15] M. Glavic and T. Van Cutsem, “Wide-area detection of voltage instability from synchronized phasor measurements. Part II: Simulation results,” IEEE Trans. Power Syst., vol. 24, no. 3, pp. 1417–1425, Aug. 2009. [16] B. Milosevic and M. Begovic, “Voltage-stability protection and control using a wide-area network of phasor measurements,” IEEE Trans. Power Syst., vol. 18, no. 1, pp. 121–127, Feb. 2003. [17] N. Kakimoto, M. Sugumi, T. Makino, and K. Tomiyama, “Monitoring of interarea oscillation mode by synchronized phasor measurement,” IEEE Trans. Power Syst., vol. 21, no. 1, pp. 260–268, Feb. 2006. [18] I. Kamwa, R. Grondin, and Y. Hébert, “Wide-area measurement based stabilizing control of large power systems—A decentralized/hierarchical approach,” IEEE Trans. Power Syst., vol. 16, no. 1, pp. 136–153, Feb. 2001. [19] S. Zhang, X. Xie, and J. Wu, “WAMS based detection and early- warning of low frequency oscillations in large-scale power systems,” Electr. Power Syst. Res., vol. 78, no. 5, pp. 897–906, 2008. [20] Z. Zhong, C. Xu, B. J. Billian, L. Zhang, S.-J. S. Tsai, R. W. Conners, V. A. Centeno, A. G. Phadke, and Y. Liu, “Power system frequency monitoring network (FNET) implementation,” IEEE Trans. Power Syst., vol. 20, no. 4, pp. 1914–1921, Nov. 2005. [21] P. Zhang, F. Li, and N. Bhatt, “Next generation monitoring, analysis, and control for the future smart control center,” IEEE Trans. Smart Grids, vol. 1, no. 2, pp. 186–192, Sep. 2010. [22] A. Bose, “Smart transmission grid applications and their supporting infrastructure,” IEEE Trans. Smart Grids, vol. 1, no. 1, pp. 11–19, Jun. 2010. [23] Electromagnetic Compatibility (EMC)—Part 4-30: Testing and Measure- ment Techniques—Power Quality Measurement Methods, IEC Interna- tional Standard 61000-4-30, Oct. 2008. [24] Testing and Measurement Techniques—General Guide on Harmonics and Interharmonics Measurements and Instrumentation, for Power Supply Systems and Equipment Connected Thereto, IEC Standard 61000-4-7, 2003. [25] Testing and Measurement Techniques—Flickermeter—Functional and Design Specifications, IEC Standard 61000-4-15, 2002. [26] IEEE Recommended Practice for the Transfer of Power Quality Data, IEEE Standard 1159.3-2002, 2002. [27] IEEE Standard for Synchrophasors for Power Systems, IEEE Standard C37.118, 2005. [28] VIA Technologies. (2013, Mar.). Mini-ITX Mainboard Specification White Paper [Online]. Available: http://www.viatech.com [29] E. Altıntas, O. Salor, I. Cadirci, and M. Ermis, “A new flicker contribu- tion tracing method based on individual reactive current components of multiple EAFs at PCC,” IEEE Trans. Ind. Appl., vol. 46, no. 5, pp. 1746–1754, Sep./Oct. 2010. [30] O. Unsar, O. Salor, I. Cadirci, and M. Ermis, “Determination of harmonic current contributions at points of common coupling of the electrical system based on field measurements,” in Proc. 20th IEEE Signal Process. Commun. Appl. Conf., Apr. 2012, pp. 1–4. [31] (2013, Sep.). National Physical Laboratory, U.K. [Online]. Available: http://www.npl.co.uk [32] (2013, May). Overview of Arbiter Products [Online]. Available: http://www.arbiter.com/catalog [33] (2013, May). Fluke 430 Series Power Quality Analyzer [Online]. Available: http://www.fluke.com/fluke/usen/power-quality-tools/three- phase/fluke-430-series.htm?pid=56078 [34] (2013, May). Dranetz 658-Power Quality Ana- lyzer [Online]. Available: http://dranetz.com/old/658- powerqualityanalyzer [35] Recommended Practices and Requirements for Harmonic Control in Electrical Power Systems, IEEE Standard 519-1992, 1992. [36] Instrument Transformers-Part 7: Electronic Voltage Transformers, IEC Standatd 60044-7, 1999. [37] TEIAS Co., “Complementary studies for the synchronization of the turkish power system with the UCTE power system,” Final Report— Dynamic Studies, Ankara, Turkey, May 2007. [38] H. Weber and I. Nassar, “Electrical interconnection between Turkey and Europe: Problems and solutions,” in Proc. 13th MEPCON, Dec. 2009, pp. 20–23. [39] O. B. Tor, C. Gencoglu, O. Yilmaz, E. Cebeci, and N. Guven, “Damping measures against prospective oscillations between turkish grid and ENTSO-E system,” in Proc. IEEE Int. Conf. Power Syst. Technol., Oct. 2010, pp. 1–7. 582 IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, VOL. 63, NO. 3, MARCH 2014 [40] TEIAS Co., “Preventive measures for effective damping of inter area oscillations after synchronous interconnection of the Turkish power system with the UCTE power system,” Final Report, Ankara, Turkey, May 2009. [41] B. Gultekin, C. O. Gercek, T. Atalik, M. Deniz, N. Bicer, M. Ermis, K. N. Kose, C. Ermis, E. Koc, I. Cadirci, A. Acik, Y. Akkaya, H. Toygar, and S. Bideci, “Design and implementation of a 154-kV ±50-Mvar transmission STATCOM based on 21-level cascaded multi- level converter,” IEEE Trans. Ind. Appl., vol. 48, no. 3, pp. 1030–1045, May/Jun. 2012. [42] T. Atalik, M. Deniz, E. Koc, C. O. Gercek, B. Gultekin, M. Ermis, and I. Cadirci, “Multi-DSP and -FPGA-based fully-digital control system for cascaded multilevel converters used in FACTS applications,” IEEE Trans. Ind. Inf., vol. 8, no. 3, pp. 511–527, Aug. 2012. [43] I. Yilmaz, M. Ermis, and I. Cadirci, “Medium frequency induction melting furnace as a load on the power system,” IEEE Trans. Ind. Appl., vol. 48, no. 4, pp. 1203–1214, Jul./Aug. 2012. [44] O. Salor, B. Gultekin, S. Buhan, B. Boyrazoglu, T. Inan, T. Atalik, A. Ak, A. Terciyanl, O. Unsar, E. Altntas, Y. Akkaya, E. Ozdemirci, I. Cadrc, and M. Ermis, “Electrical power quality of iron and steel industry in Turkey,” IEEE Trans. Ind. Appl., vol. 46, no. 1, pp. 60–80, Jan./Feb. 2012. [45] (2013, Feb.). Trench Group Switzerland [Online]. Available: http://www.trenchgroup.com/ Tevhid Atalik (S’10) received the B.Sc. degree in electronics engineering from Uludag University, Bursa, Turkey, in 2003, and the M.Sc. degree in elec- trical and electronics engineering from Hacettepe University, Ankara, Turkey, in 2007. He is currently pursuing the Ph.D. degree in electrical engineering with Baskent University, Ankara. He was with the Power Electronics Department in Space Technologies Research Institute, the Scientific and Technological Research Council of Turkey, as a Senior Researcher, until 2012. He is now with Energy Institute, TUBITAK Marmara Research Center, Ankara. His current research interests include analog and digital control circuit design, instrumen- tation, and power quality. Mr. Atalik was a recipient of the Outstanding Paper Award from the Metals Industry Committee of the IEEE Industry Applications Society in 2009. Isik Çadirci (M’98) received the B.Sc., M.Sc., and Ph.D. degrees in electrical and electronics engineer- ing from Middle East Technical University, Ankara, Turkey, in 1987, 1988, and 1994, respectively. She is currently a Professor of electrical engi- neering with Hacettepe University, Ankara. She was the Head of the Power Electronics Department, Space Technologies Research Institute, Scientific and Technological Research Council of Turkey, from 2004 to 2011. Her current research interests include power quality, electric motor drives, and switch- mode power supplies. Dr. Çadirci was a recipient of the Committee Prize Paper Award from the Power Systems Engineering Committee of the IEEE Industry Applications Society in 2000, the IEEE Industry Applications Magazine Prize Paper Award, Third Prize, in 2007, and the Outstanding Paper Award from the Metals Industry Committee of the IEEE Industry Applications Society in 2009. Turan Demirci received the B.Sc., M.Sc., and Ph.D. degrees in electrical engineering from Middle East Technical University, Ankara, Turkey, in 2000, 2002, and 2010, respectively. He was with the Power Electronics Department in Space Technologies Research Institute, the Scientific and Technological Research Council of Turkey, as a Chief Senior Researcher, until 2012. He is now with Energy Institute, TUBITAK Marmara Research Cen- ter, Ankara. His current research interests include power quality monitoring and wind power forecast- ing. Muammer Ermis (M’99) received the B.Sc., M.Sc., and Ph.D. degrees in electrical engineering from Middle East Technical University (METU), Ankara, Turkey, in 1972, 1976, and 1982, respectively, and the M.B.A. degree in production management from the Ankara Academy of Economic and Commercial Sciences, Ankara, in 1974. He is currently a Professor of electrical engineer- ing with the Department of Electrical and Electron- ics Engineering, METU. He was the Manager of the National Power Quality Project of Turkey. His current research interests include electric power quality. Dr. Ermis received the Overseas Premium Paper Award from the Institution of Electrical Engineers, U.K., in 1992, and the 2000 Committee Prize Paper Award from the Power Systems Engineering Committee of the IEEE Industry Applications Society. He was the recipient of the 2003 IEEE Power Engineer- ing Society Chapter Outstanding Engineer Award and the Outstanding Paper Award from the Metal Industry Committee of the IEEE Industry Applications Society in 2009. Tolga Inan (M’12) received the B.Sc., M.Sc., and Ph.D. degrees from the Department of Electrical and Electronics Engineering, Middle East Technical University (METU), Ankara, Turkey, in 2000, 2003, and 2011, respectively. He was a Research Assistant with the Department of Electrical-Electronics Engineering, METU, from August 2000 to January 2007, and as a Senior Researcher with the Power Electronics Group, Space Technologies Research Institute, Scientific and Tech- nological Research Council of Turkey, Ankara, from January 2007 to January 2012. He is currently with TED University, Ankara, where is an Assistant Professor. Alper Sabri Kalaycioglu received the B.Sc. degree in computer engineering from Bilkent University, Ankara, Turkey, in 2005, and the M.Sc. degree in software engineering from Middle East Technical University, Ankara, in 2009. He was a Researcher with the Power Electron- ics Group, Space Technologies Research Institute, Scientific and Technological Research Council of Turkey, Ankara, from 2005 to 2010. Currently, he is a Product Manager and Senior Software Engineer with BIG Vienna Company, Vienna, Austria. Özgül Salor (S’98–M’05) received the B.Sc., M.Sc., and Ph.D. degrees in electrical and electronics engi- neering from Middle East Technical University, Ankara, Turkey, in 1997, 1999, and 2005, respec- tively. She was a Professional Researcher with the Uni- versity of Colorado, Boulder, CO, USA, from 2001 to 2003. From 2006 to 2012, she has been with the Power Electronics Group, Space Technolo- gies Research Institute, Scientific and Technological Research Council of Turkey, Ankara, and a part-time Lecturer and an Associate Professor with the Department of Electrical and Electronics Engineering, Gazi University, Ankara, where she has been an Associate Professor since 2012. Her current research interests include speech- signal processing and signal processing for power quality. Dr. Salor received the Outstanding Paper Award from the Metal Industry Committee of the IEEE Industry Applications Society in 2009. /ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 600 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages false /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 400 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False /Description > >> setdistillerparams > setpagedevice