Design, Implementation and Simulation of Non-Intrusive Sensor for On-Line Condition Monitoring of MV Electrical Components

Transcription

1 Engineering, 2014, 6, Published Online Otober 2014 in SiRes. Design, Implementation and Simulation of Non-Intrusive Sensor for On-Line Condition Monitoring of MV Eletrial Components Muhammad Shafiq 1, Matti Lehtonen 1, Lauri Kutt 1, Muzamir Isa 2 1 Department of Eletrial Engineering, Shool of Siene and Tehnology, Aalto University, Espoo, Finland 2 Shool of Eletrial Systems Engineering, Universiti Malaysia Perlis, Malaysia muhammad.shafiq@aalto.fi, matti.lehtonen@aalto.fi, lauri.kutt@aalto.fi, Smuzamir@unimap.edu.my Reeived 6 August 2014; revised 29 August 2014; aepted 16 September 2014 Copyright 2014 by authors and Sientifi Researh Publishing In. This work is liensed under the Creative Commons Attribution International Liense (CC BY). Abstrat Non-intrusive measurement tehnology is of great interest for the eletrial utilities in order to avoid an interruption in the normal operation of the supply network during diagnostis measurements and inspetions. Indutively oupled eletromagneti sensing provides a possibility of non-intrusive measurements for online ondition monitoring of the eletrial omponents in a Medium Voltage (MV) distribution network. This is aomplished by employing Partial Disharge (PD) ativity monitoring, one of the suessful methods to assess the working ondition of MV omponents but often requires speialized equipment for arrying out the measurements. In this paper, Rogowski oil sensor is presented as a robust solution for non-intrusive measurements of PD signals. A high frequeny prototype of Rogowski oil is designed in the laboratory. Step-by-step approah of onstruting the sensor system is presented and performane of its omponents (oil head, damping omponent, integrator and data aquisition system) is evaluated using pratial and simulated environments. Alternative Transient Program-Eletromagneti Transient Program (ATP-EMTP) is used to analyze the designed model of the Rogowski oil. Real and simulated models of the oil are used to investigate the behavior of Rogowski oil sensor at its different stages of development from a transduer oil to a omplete measuring devie. Both models are ompared to evaluate their auray for PD appliations. Due to simple design, flexible hardware, and low ost of Rogowski oil, it an be onsidered as an effiient urrent measuring devie for integrated monitoring appliations where a large number of sensors are required to develop an automated online ondition monitoring system for a distribution network. Keywords Non-Intrusive, Sensors, Condition Monitoring, Partial Disharge, Rogowski Coil, ATP-EMTP How to ite this paper: Shafiq, M., Lehtonen, M., Kutt, L. and Isa, M. (2014) Design, Implementation and Simulation of Non-Intrusive Sensor for On-Line Condition Monitoring of MV Eletrial Components. Engineering, 6,

2 1. Introdution The major deriving fores to modernize the urrent power grid inlude inreasing requirements for reliability, effiieny and safety of power grid along with optimization of apital assets while minimizing the operation and maintenane ost. Some of the greatest effets on reliability of a power distribution network are provided by various types of failures due to environmental and operational stresses [1]. Insulation degradation is one of the most frequent auses of the failure in the ritial and expensive power omponents suh as, motors, generators, transformers, swithgear, and power lines. One of the methods to predit the inoming insulation faults is to perform on-line ondition monitoring of the network omponents. Partial Disharge (PD) diagnosti is a wellknown tehnique for insulation ondition monitoring. PD is the proess of loalized dieletri breakdown of a small portion (avities, voids, raks or inlusions) of a solid or liquid eletrial insulation part whih is under high voltage stress during operation [2]. The eletrial stress due to applied voltage auses disharges within the defetive portion of the insulation. There are different phenomena whih appear during disharges and give rise to respetive detetion indiators suh as eletromagneti radiation, sound or noise, thermal radiation, gas pressure, hemial formation, and eletromagneti impulses [3]-[6]. The PD measurement sensor tehnology is based on the type of the energy exhange whih takes plae during above mentioned disharge phenomena. In this paper eletromagneti impulses are onsidered for assessment of the PD ativity. Due to rapid displaement of the harges during disharge event, voltage and urrent transient appears in the form of eletromagneti waves. These signals travel away from the site of origin along the power lines and an be measured by resistive, apaitive or indutive methods [7]. A variety of sensors are available for measurement of eletromagneti signals. However, onsidering pratial aspets suh as low ost, high bandwidth, good sensitivity, saturation harateristis, high linearity, and wide operating temperature, Rogowski oil has been regarded as a favorite tool for high-frequeny urrent-sensing purposes [7]. Its flexible design provides the possibility to install it in a variety of physial loations espeially in the tight spaes that may be inaessible with typial iron-ore urrent transformers. For MV overhead overed ondutor lines, it an diretly be installed around the CC line. For an MV able network and able aessories (joints and terminations), swithgears, and transformers, the oil an easily be installed around the earth straps to detet the fault urrent. Moreover it an also be used for normal urrent measurements at the substation. Researhers and engineers have been using Rogowski oil for high-amplitude sinusoidal urrents and transients as well as low-amplitude and high-frequeny signals in power systems and power eletronis appliations [8]-[11]. Nowadays it is also being used for detetion and loalization of insulation and short iruit faults while its appliation for relay protetion is impressive as well [12]. Obtaining simpliity in the design, better auray for a wide range of amplitude and frequeny of measured signal, and low ost are the hallenges during development of a sensor. In this paper, the main features, design and onstrution features of Rogowski oil have been explored. Rogowski oil sensor has been identified as a omposite of four essential and sequential omponents: Rogowski oil head, damping omponent, integrator and Data Aquisition System (DAS). Step-by-step implementation of Rogowski oil is desribed and performane of eah stage is evaluated for low-amplitude and high-frequeny transient urrent signals. The operation of the Rogowski oil is simulated in ATP-EMTP simulation software environment whih provides an in-depth analysis, the verifiation of the oil prototype and the possibility of developing further PD diagnosti tehniques for distribution lines. 2. Components of Rogowski Coil Sensor A measuring sensor transats a omplete measuring funtion from initial detetion to final indiation of the measured quantity as desribed in Figure 1. Considering Rogowski oil sensor, the initial detetion is done by a urrent sensing oil whih ats as an interfae between primary and measured urrent. Intermediate signal Damping omponent Integrator Sensing element Data aquisition system Rogowski oil head Terminating resistor Digital integrator Figure 1. Essential elements of a PD measuring sensor. Display of reording 681

3 proessing onsists of signal onditioning based on the operational features of the oil, its physial and eletrial harateristis, and the purpose (requirement) of measurements and diagnostis. This is aomplished by damping omponent and integrator. Final indiation is displayed or reorded with the help of a suitable Data Aquisition System (DAS). This reorded data an be transported to personal omputers to be used for further investigations. Figure 1 represents the sequential onstrution and operation of Rogowski oil measuring sensor while its implementation is depited in Figure 2. The detailed desription and implementation of eah omponent is explained further in this setion Coil Head The Rogowski oil s main sensing part is omposed of toroidal wound oil with n number of turns, on an air-ore (dieletri) former of onstant ross-setional area. The air-ore is split in one loation to allow assembly of the oil head unit. As shown in Figure 3, the wire is wound suh that the winding start from the first end, progresses towards the other end and returning through the entre of the oil bak to the first end, so that both terminals are at the same end of the oil. Number of turns ( n ), outer diameter of oil ( d ) of oil ( i ) o, and inner diameter d are the major physial parameters of oil head (see Table 1). Its physial parameters are seleted based on the appliation requirements of the oil. Spae for installation, bandwidth, and sensitivity of the oil are the main onerns required to initiate the design Damping Component Every piee of wire has resistane and indutane while every two wires present some apaitane during an eletrial appliation [13]. Response of the Rogowski oil to primary urrent signal an be well explained by developing eletrial model of this eletromagneti devie. Lumped parameter model is used in this work where R, L and C are struture-based inherent self-resistane, self-indutane and self-apaitane of the Rogowski oil. During urrent measuring operation, the sensed signal passes through RLC iruit of the oil whih Figure 2. Stages of onstrutions of Rogowski oil measuring sensor. Figure 3. Physial model of the Rogowski oil. Table 1. Physial parameters of Rogowski oil head. Name of the parameter Symbol Speifiation Number of turns n 30 Outer diameter of oil d o 16.1 m Inner diameter of oil d i 14.1 m Core diameter d r 1.96 m 682

4 introdues osillations in the output voltage and urrent. The rate of these osillations is determined by the resonant frequeny of the oil. Suh osillations an be damped by properly terminating the response of the oil with a suitable resistane whih is normally onneted aross the output terminal of the oil Integrator The output voltage of Rogowski oil whih is indued by the variable urrent passing through the primary ondutor is proportional to the derivative of primary urrent. The main operating priniple is thus desribed with Farady s law. Due to very nature of the eletrial iruit of Rogowski oil, osillations are introdued due to seond order equivalent iruit harateristis. These fators affet the output of the Rogowski oil sensor (with referene to primary signal). Reliability of the measurement depends on how aurately the primary signal is aptured. In order to rereate the original waveform of the measured signal, first step was to remove sensor s osillations using damping omponent (desribed above). An integration operation is needed to proess differential output to provide output waveform (measurement result) to math the measured primary urrent waveform. Rogowski oil is an air-ore indution sensor and relatively low number of turns are required to guarantee the high-frequeny operation. Therefore the urrent transfer ratio (output urrent to input urrent) is very low; even minor resistive losses an signifiantly redue the amplitude of the measured signal in the sensor s output. The speifi alibration of the sensor transfer rate an thus be required to ahieve the aurate amplitude of the measured signal Data Aquisition System The output of Rogowski oil sensor is an analogue signal. For more detailed analysis and reording of the measured transients patterns, the signal is passed on to a DAS system and digitized. This inorporates in the first stage an Analog-to-Digital Converter (ADC) with very high sampling rate, in order to apture the extremely short-time transients of the PDs ourring on the power line. The Nyquist riterion states that the sampling frequeny of a system needs to be at least twie of the frequeny needed to be aptured. The bandwidth needed to apture the PD traes would reah several tens MHz, thus the sampling frequeny required from the DAS would be in tens of millions of samples per seond (MS/s) also. The ADC would need to have at least 8-bit resolution (better higher), therefore having multiple sensors in a substation would mean very high data transfer bandwidth. The seond stage of the DAS would be initial proessing of the results to provide redution of the data to be proessed. The PD transients have speifi waveforms and reognition/detetion of suh waveforms an be then used for triggering the further storage and proessing operations. As a result of suh event filtering, the data stored will be limited in a window of some thousands of samples at eah deteted PD ourrene, or even less. The data bandwidth is redued for easy data storage or transmission to data enters for further analysis. Suitable equipment for establishing the DAS would be for example Digital Storage Osillosopes (DSO) in the laboratory measurements or on-hip data logger systems an be used for on-site appliations. 3. Stage-Wise Design and Implementation of Components of Rogowski Coil Sensor 3.1. Physial and Eletrial Model of the Coil Head When the oil is plaed around the ondutor arrying alternating or transient urrent ip ( t ) to be measured (see Figure 2), the voltage Vr ( t ) indued within the oil is expressed as dip ( t) Vr ( t) = M, (1) dt where M is mutual indutane or sensitivity of the oil, expressed in V A at a speifi frequeny. Mutual Indutane depends on the number of turns of the oil, ross setional area of the ore, and the diameter of the toroid (to determine the radial distane of the oil winding from the urrent arrying ondutor plaed at the enter of the oil). The physial parameters of the Rogowski oil head are given in Table 1. The value of eletrial parameter depends on the physial design of the oil. Table 2 represents the measured eletrial parameters of the Rogowski oil head. The details of the parameters measuring methodology has been desribed by the authors in [14]. The lumped parameters (RLC) equivalent eletrial model is shown in Figure

5 Table 2. Eletrial parameters of Rogowski oil head. Name of the parameter Symbol Speifiation Self-resistane of oil R 0.71 Ω Indutane of oil system L 1.19 μh Capaitane of oil system C 14.7 pf Measuring resistane R m 1 MΩ Figure 4. Eletrial model of the Rogowski oil head. Equation (1) presented the basi operation of a Rogowski oil by providing the indued voltage Vr ( ) in the oil due to primary signal ip ( t ). The voltage olleted at the output terminals of the oil is Vo ( ) t witht as shown in Figure 4 inludes also effets of self-indutane L, self-apaitane C and ative input resistane (the DSO has R m = 1 MΩ ). It has be pointed out that the Rogowski oil sensor system inludes the apaitane of the onnetion ables and the ative input apaitane, whih have been ounted for in the total C value. At lower frequeny range, the parameters L and C of the oil do not have onsiderable effet and therefore V t V t V t an be expressed by the transfer funtion as o ( ) ( ). However at higher frequeny range ( ) r o 1 V t V tv s LC V s ( ) ( ) ( ) = ( ) o r o r 2 1 L 1 R s + + RC s+ + 1 LC Rm LC Rm During high frequeny measurements, the interation of sensed signal with RLC parameters results in signifiant energy exhange between the oil s self-indutane and self-apaitane, and auses osillations with a resonant frequeny ω whih an be alulated as 1 ω = = 37.6 MHz (3) LC To ensure the pratial response of Rogowski oil, a laboratory test is made. A PD pulse is injeted from a PD alibrator into a simple test iruit as shown in Figure 5. The primary urrent pulse in the test line is measured by a ommerial High Frequeny Current Transformer (HFCT) and is shown in red olor in Figure 6(a). This urrent pulse will be used as referene primary signal further in this paper. The aptured output voltage of Rogowski oil is shown as time-domain plot in blue olor in Figure 6(a). This osillating signal is the measured V t whih needs to be proessed during the inoming stages of the sensor development in order to get signal o ( ) the primary signal i ( ) p t. Fast Fourier Transform (FFT) of the aptured response shown in Figure 6(b) represents the frequeny response of the oil s output voltage. The resonant frequeny of measured response is determined as 37.6 MHz. The resonant frequeny reflets the value of the LC parameters of an indution sensor. Therefore, exat math of alulated and pratially measured resonant frequeny validates the auray of the measured eletrial parameters Coil Head Along with Damping Component In the previous, the output voltage V ( ) o t was measured by the DSO in the presene of a very large ative (2) 684

6 PD alibrator HFCT Rogowski oil DSO Rs= 50 Ω (a) Figure 5. Experimental setup for measurement of PD urrent pulse. (a) Ciruit model; and (b) Laboratory implementation. (b) (a) Figure 6. Measured response of Rogowski oil. (a) Time domain plot; (b) Frequeny domain plot. measurement devie input resistane R m. With this measuring setup, Rogowski oil operates in an un-damped mode. In order to damp the osillation, a suitable value of terminating resistane is required whih should be a speifi ratio of the harateristis impedane of the Rogowski oil as shown in Figure 7. Based on the terminal loading, the Rogowski oil an operate in over damped, undamped, and ritially damped modes, as shown in Figure 8. Suitability of the operating mode depends upon the measured waveform of the output whih aims at to be the waveform of the primary signal. The output voltage an be expressed as The response Vo ( ) 1) V ( ) ξω 2 ( ) ( ) ( ) e sin ( 1 ) Ct ω ξ V t = V t + V t t (4) o r r t of the oil an be divided into two parts: t, the fored response due to primary signal and; r ξω 2 2) V ( ) e sin ( 1 ) Ct r t ω ξ t, the natural response due to sensor s operation. where ξ is the damping oeffiient. Natural response reflets oil s iruit properties whih hides the information of the original measured signal. As visible from Equation (4), damping oeffiient an be effiiently used to ontrol the effet of osillations in the output voltage of Rogowski oil. Considering Equation (2), the lassial 2 nd order behavior of the oil s eletrial model an be represented by the harateristi equation written as s + 2ξω s+ ω = 0 (5) 2 2 Comparing with the transfer funtion, the damping oeffiient ξ for oil model without damping omponent an be an be expressed as 1 L ξ = + RC (6) 2 ωlc Rm whereas the damping oeffiient with terminating resistane ξ t an be written as 1 L ξt = + RC. (7) 2 ωlc ( Rt Rm) (b) 685

7 Figure 7. Eletrial model of the Rogowski oil. Figure 8. Modes of Rogowski oil operation based on the value of damping. Here it should be noted that damping oeffiient depends on all the four parameters ( L, C, R and R m ) of the oil. The parameters L, C and R are fixed for a ertain physial design of Rogowski oil. Therefore one of the effiient ways to speify value of ξ is by variation in the R m. The oil operation properties in different modes (for different values of terminating resistane) an be desribed in the frequeny plot as shown in Figure 8. The peak transfer value at 37.6 MHz for undamped operation indiates the presene of high amplitude of osillation. The amplitude of osillations is redued for the ritially damped mode while the magnitude transfer rate has been onsiderably redued for overdamped mode of operation, not only for the osillations at single resonant point. The redution or eliminations of osillations an also be visualized in the poles zero plot for the designed oil. The solution of (4) results in a pair of poles as P ξω ω ξ 1. (8) P = 2 1 = ξω ω ξ 1 For different values of terminating resistane, the loation of poles provides a quantitative view of presene of osillations within the output of Rogowski oil as shown in Figure 9. Loation of poles from the real axis quantifies the presene of osillations. It an be seen that for Rt > Z the poles are away from real axis showing higher osillations. It has been identified in [15] that for 1 Z R = t 2ω C = 2 n Here Z is the harateristi impedane of the oil. Using suh damping omponent value, the poles still lie on real axis, while their magnitude has the highest value. The poles an have even higher magnitude values, however in suh ase the poles will also have an imaginary part. The removal of the osillations from the output is benefiial for the sensor appliation, as removal of natural osillatory response modifies (4) as ( t) dip Vo ( t) Vr ( t) = M. (9) dt Figure 10 shows the transformation of the signal from the output of stage 1 to the output of stage 2. The ompensation of the effet of time differential and mutual indutane an be arried out by integration and alibration. 686

8 Figure 9. Effet of damping on the loation of poles for different value of damping resistane. Figure 10. Signal onditioning from (a) undammped mode to (b) ritially damped mode Coil Head Along with Damping Component and Integrator Integration of the output of Rogowski oil an be performed by one of two ommon means, (1) by use of an eletrial or eletroni integrator, or (2) by using numerial integration in software after the oil output voltage is digitized [16]. In order to avoid omplex iruit and expensive omponents, the numerial integration is proposed to be done by the built-in numerial funtions of the DSO. The sampling frequeny is seleted based on the resonant frequeny (onsidering Nyquist riterion) of the Rogowski oil. The digital integration an be expressed as 1 i( N) = i( N 1) + V ( N 1 ) + V ( N) N > 0, f (10) where i is the urrent obtained by the digital integration of a voltage signal V, is the order number of sample. The digital integration rereates the urrent as m 2 s ( ) ( )d o f s is sampling frequeny and N i t = V t t (11) The amplitude of the indued voltage is redued by a fator M. Similarly resistive or stray losses an further redue the amplitude of the measured pulse. Calibration of the measured signal an be done by omparing the measured signal im ( t ) of the Rogowski oil with the referene signal. The alibration fator K al an be alulated as ip ( t) = Kal (12) i t where m ( ) K al depends on the M and the stray losses st ipm t an be obtained as i t = i t K (13) K. Thus the ( ) pm ( ) m ( ) al The overall measurement senario of the primary urrent i ( ) to provide pm ( ) p t proessed by the desribed stages of sensor i t as the final outome is shown in Figure 11 and Figure Data Capturing Using Suitable Data Aquisition System PD transients are high frequeny signals therefore a high sampling rate is required to apture the signals reliably without losing any information. During laboratory measurements, a sampling rate of 2.5 GS/s is used with the help of DSO. However it is important to take into fator the fator of sampling rate beause it affets the eonomial parameters in terms of ost of high frequeny proessors and large memory for data storage. For real 687

9 Figure 11. Complete eletrial equivalent iruit of Rogowski oil measuring sensor. (a) (b) () (d) (e) Figure 12. Signal wave-shapes at different stages of sensor design. (a) Primary urrent pulse i p (t) to be measured; (b) Signal sensed by oil head; () Damped signal; (d) Integrated signal and (e) Calibrated signal as final outome primary measured signal i pm (t). appliations, suh high sampling rate (2.5 GS/s) is not neessary. A suitable range of required sampling rate an be estimated by the resonant frequeny of the measuring sensor. The resonant frequeny of the Rogowski oil used in this work is 37.6 MHz. Therefore, onsidering Nyquist riterion, 76 MHz or more, is a suitable sampling frequeny for apturing the PD data. 4. Simulation of Rogowski Coil Sensor Using ATP-EMTP Rogwoski oil is simulated using ATP-EMTP power system transients simulation software. In literature, the pulse used in different simulation software is mostly based on the waveforms haraterized by its mathematial model desribed as [13] α1t α2t ipulse ( t) = A( e e ) (14) where A is the peak value of the pulse, α t is the rise time and 1 α t 2 is half of the fall time. In this work, the data of pratially aptured PD urrent pulse is imported in ATP to be measured by the simulated model of Rogowski oil. Both mathematial and pratial urrent pulses are shown in Figure 13(a). It an be learly seen that pratial urrent pulse ontains fast variations during rising and fall slopes of the pulse. This kind of variations or distortions in the pulse, signifiantly affet the output of the Rogowski oil. To avoid any loss of high frequeny omponents present in the pratial pulse, the simulated model of Rogowski oil is assigned to measure the same pratial primary urrent. This ensures the orret assessment for evaluation of the identified parameters of the oil to reate an aurate model. The measured parameters (Table 2) of the oil are used for simulation of model. The urrent arrying line used in the experimental setup is shown as test line in the simulation. Two types of ATP models are shown in Figure 13(b) and Figure 13(). Blok B1 shown in shemati Figure 13(b) represents the PD urrent pulse soure. The oil senses the urrent through the test iruit. Blok B2 represents the time-derivative of the primary urrent. Blok B3 shows the RLC equivalent of Rogowski oil. In the first model, the sensor is eletrially onneted with the primary line. The arrangement works well as long as the oil iruit does not provide any disturbane to the primary urrent. However, if any refletions our within the oil iruit, they may ondut towards the primary (test) line. This phenomenon may affet the harateristis of PD signal. In real pratie, the Rogowski oil has no eletrial onnetion with the primary line. Therefore, an improved model is developed in Figure 13() in order to isolate the sensor form the line. The Transient Analysis of Control Systems (TACS) shown as blok B2 senses the urrent of the primary line and eliminates the possibility of onduting any refletions from seondary side towards the primary side. In this work, first model has been onsidered for further analysis and development. The above shown shematis represent the simulation of the Rogowski oil head. Further stages are implemented in Figure

10 20 x 10-3 Simulated urrent pulse Pratial urrent pulse 15 Amplitude ( Amps) Time (se) x 10-6 (a) (b) () Figure 13. (a) Comparison of mathematial and pratial urrent pulses; (b) ATP simulation model of Rogowski oil head using RLC equivalent iruit. (a) (b) () (d) (e) (f) Figure 14. (a), (b), and () shows stage 1, stage 2, and stage 3 respetively; (d), (e), and (f) shows the output of Rogowski oil at stage 1, stage 2, and stage 3 respetively. i pm,real (t) is primary urrent while i mp,sim (t) is the measured urrent of simulated oil. o t is measured with the voltage probe. A omparison is arried out onsidering the time domain performane and FFT analysis of the measured and simulated outputs. The omparison has proven a good math. This not only ensures the orret identifiation of the parameters of Rogowski oil but also validates the simulated model of the oil head with enough auray for further analysis. It must be emphasized that the sampling rate of the simulation setup must be set as same as that of the pratial measurements in order to have a valid omparison between the signals of real and simulated model. The stage-wise simulated models and their outputs are shown in Figure 14. Extending the initial model, Figure 14 shows the models inluding oil head added with terminating resistor, and omplete Rogowski oil unit inluding oil head, terminating resistane, integration, and alibration. The measured waveforms of the simulated model have been shown in Figures 14(d)-(f), under their respetive stage of model. The performane of the simulated models has a good math with the pratially measured response. A omparison of the measured and simulated output response of the oil models is shown in Figure 14(f). The waveform and amplitude of the measured urrent niely mathes and validates the designed model of Rogowski oil PD measurements. The output voltage V ( ) 689

11 5. Conlusions For employing suessful on-line monitoring of the distribution lines using PD monitoring, the tasks to be arried out inlude PD detetion, loalization, and measurement whih onsequently provide the information about the presene of PD ativity, where the deterioration is happening and extent of the damage that has been done. Suh information determines the exeution of the required (repair, replaement or standby) tasks. This paper presents design of high-frequeny Rogowski oil for PD diagnostis. Rogowski oil is designed in four asaded stages. First stage is the design of Rogowski oil head, whih is used to sense the primary urrent as an indution sensor. In this form, Rogowski oil an be used to detet the presene and polarity of PD signals whih are suffiient to detet and loalize the PD fault within the eletrial omponents [17]-[19]. PD measurement task requires quantifiation of the disharged phenomenon whih needs the information of the wave shape of atual PD signals. The area of the PD transient waveform is used to alulate the amount of harge released during PD event and hene the extent of PD defet an be estimated. For this purpose the implementations of the seond, third, and fourth stages are neessary. Rogowski oil head along with damping omponents is analogue while integration, alibration, and data reording are done by the digital omponent (DSO). Nowadays many of the DSOs have numerial integration funtion as a built-in feature and make the integration possible without any additional omplex iruits and ost. Similarly alibration onstant an be applied numerially. Due to limitations of ost, weight, size, and vast sope of appliations for measurements, this devie (DSO) is generally used for laboratory or ertain onsite appliations. However the appliations where a large number of oil sensors are integrated into the network, the digital part of the Rogowski oil sensor an be developed using on-hip omponents and analog-to-digital onverters with required sampling rate having programmable signal proessor units and data logging and storage funtions. For reating more omplex monitoring systems, the ATP-EMTP simulation software has been shown to provide good results in the transient analysis. The design input is simplified by ATP-EMTP as it provides graphial and mouse-driven preproessor, where the user an onstrut a wide range of power system iruits. A variety of line models (taking into aount the high frequeny effets) an be developed by input of the geometrial design parameter whih provides an opportunity to further develop and/or analyze diagnosti tehniques for PDs travelling over the power lines. Aknowledgements The researh was supported by Smart Control Arhiteture for Smart Grids SAGA projet at Aalto University. The authors highly appreiate the valuable and interesting disussions with Dr. Petri Hyvönen, Aalto University. Referenes [1] Mahmoudi, M., Barkany, A. and Khalfi, A. (2014) A Maintenane Optimization Poliy for an Eletri Power Distribution System: Case of the HV/MV Substations. Engineering, 6, [2] Eletro Tehnial Commission (IEC) 60270, High-Voltage Test Tehniques. [3] Moore, P.J., Portugues, I.E. and Glover, I.A. (2005) Radiometri Loation of Partial Disharge Soures on Energized High-Voltage Plant. IEEE Transations on Power Delivery, 20, [4] Muhr, M., Strehl, T., Gulski, E., Feser, K., Gokenbah, E., Haushild, W. and Lemke, E. (2006) Sensors and Sensing Used for Non-Conventional PD Detetion. CIGRE, Paris, 1-7. [5] Shwarz, R., Muhr, M. and Pak, S. (2005) Evaluation of Partial Disharge Impulses with Optial and Conventional Detetion Systems. Proeedings of the 24 th International Symposium on High Voltage Engineering, Beijing. [6] Lundgaard, L.E. (1992) Partial Disharge. XIV. Aousti Partial Disharge Detetion-Pratial Appliation. IEEE Eletrial Insulation Magazine, 8, [7] Zhao, X.L., Asada, T., Odendaal, W.G. and Van Wyk, J.D. (2003) An Overview of Integratable Current Sensor Tehnologies. Industry Appliations Conferene, 38th IAS Annual Meeting, 2, [8] Robles, G., Argueso, M., Sanz, J., Giannetti, R. and Tellini, B. (2007) Identifiation of Parameters in a Rogowski Coil Used for the Measurement of Partial Disharges. IEEE Instrumentation and Measurement Tehnology Conferene (IMTC) Proeedings, Warsaw, 1-3 May 2007, 1-4. [9] Jalebi, A. and MMahon, R. (2007) High-Performane Low-Cost Rogowski Transduers and Aompanying Ciruitry. 690

12 IEEE Transations on Instrumentation and Measurement, 3, [10] Ponelas, O., Rosero, J.A., Cusido, J., Ortega, J.A. and Romeral, L. (2008) Design and Appliation of Rogowski Coil Current Sensor without Integrator for Fault Detetion in Indution Motors. IEEE International Symposium on Industrial Eletronis, Cambridge, 30 June-2 July 2008, [11] Zhu, J., Yang, L., Jia, J. and Zhang, Q. (2005) The Design of Rogowski Coil with Wide Band Using for Partial Disharge Measurements. Proeeding of International Symposium on Eletrial Insulating Materials, 5-9 June 2005, [12] Kojovi, L. (1997) Rogowski Coils Suit Relay Protetion and Measurement. IEEE Computer Appliations in Power, 10, [13] Knight, D.W. (2010) The Self-Resonane and Self-Capaitane of Solenoid Coils. Version 0.01 (Provisional). [14] Shafiq, M., Kütt, L., Lehtonen, M., Nieminen, T. and Hashmi, M. (2013) Parameters Identifiation and Modeling of High Frequeny Current Transduer for Partial Disharge Measurements. IEEE Sensors Journal, 13, [15] Shafiq, M., Lethonen, M., Kutt, L., Hussain, G.A. and Hashmi, M. (2013) Effet of Terminating Resistane on High Frequeny Behavior of Rogowski Coil for Transient Measurements. Journal of Eletronis and Eletrial Engineering (Elektronika Ir Elektrotehnika), 19, [16] Lane, D.S. (2004) An Aelerometer Based Approah to Measuring Displaement of a Vehile Body. University of Mihiga, Dearborn. [17] Shafiq, M., Kütt, L., Isa, M., Hashmi, M. and Lehtonen, M. (2012) Diretional Calibration of Rogowski Coil for Loalization of Partial Disharges in Smart Distribution Networks. International Review of Eletrial Engineering, 7, [18] Shafiq, M., Lehtonen, M., Isa, M. and Kutt, L. (2013) Online Partial Disharge Diagnostis in Medium Voltage Branhed Cable Networks. Proeedings of IEEE International Conferene on Power Engineering, Energy and Eletrial Drives (POWERENG 2013), Istanbul, May 2013, [19] Kane, C. and Golubev, A. (2010) Field Results of Monitoring Partial Disharges on In-Servie Large Power Transformers. Proeedings of the Conferene Reord of the IEEE International Symposium on Eletrial Insulation, San Diego, 6-9 June 2010,

13