Measurements of the Communications Environment in Medium Voltage Power Distribution Lines for Wide-Band Power Line Communications

Size: px

Start display at page:

Download "Measurements of the Communications Environment in Medium Voltage Power Distribution Lines for Wide-Band Power Line Communications"

Egbert Chase
5 years ago
Views:

1 Measuremets of the Commuicatios viromet i Medium Voltage Power Distributio Lies for Wide-Bad Power Lie Commuicatios Jae-Jo Lee *,Seug-Ji Choi *,Hui-Myoug Oh *, Wo-Tae Lee *, Kwa-Ho Kim * ad Dae-Youg Lee ** * KR(Korea lectrotechology Research stitute), Korea ** Departmet of lectrical ad lectroic gieerig, KyugHee Uiversity, Korea Tel: , -mail:jjlee@keri.re.kr Abstract Durig the last several years, iterest i broad-bad power lie commuicatios (PLC) has bee grow over medium voltage lies as well as low voltage lies. this paper itroduce a medium voltage PLC test field is set up i the suburbs of uiwag city i Korea. This test field could be used ot oly to test modems but also to measure commuicatio chael eviromet. This paper shows measuremets methods ad results of commuicatios chael parameters such as oise, lie impedace, ad impulse respose i the test field. 1. troductio Curretly, as iformatio techology is beig developed, demads for high data rate trasmissio systems have bee beig icreased. Also i the area of power lie commuicatios, ivestigatio o the usage of wide frequecy bad ad the high data rate commuicatios is doe. Actually, as it is well kow, several Mbps modems have already come to markets. However, practical test of modem products, such as performace test ad proof, system security ad improvemet should be doe i advace. For this purpose, a real practical PLC test field to test the developed PLC modem is ecessary. this paper, a medium voltage PLC test field is set up. This test field could be used ot oly to test ew modems but also to fid their problems. Usig the test field this paper shows measuremets of commuicatios chael parameters such as oise, lie impedace, ad impulse respose. t is expected that the results are very useful for establishig PLC subscriber etworks ad applyig to developig high data rate modems. As kow util ow, chael oise, multi-path propagatio ad impedace mismatch are mai impedimets of the power lie commuicatios [][4]. Geerally, oise measured i low voltage grids have bee itroduced as a sum of backgroud oise, impulsive oise, ad sychroouso-sychroous colored oise, primarily geerated by electrical appliaces [][4]. medium voltage lies, however, backgroud oise ad colored oise maily caused by radio frequecies, for example short-wave radio sigal, HAM were measured typically. That is to say, cotrary to low voltage grids, oise level does ot vary i the time domai cosiderably, because the measured positios are too far from electrical load. this reaso, we just tried to fid the tred of oise levels roughly with a spectrum aalyzer i the frequecy domai. Multi-path propagatio is oe of the critical parameters for improvig the performace of a high-speed power lie commuicatios system [4]. However, oly few articles described multi-path ad time-varyig ature of power lie chael, especially i the medium voltage lies. For impulse respose aalysis we used the correlatio method of PN code sequece. Ad the careful aalysis of measured results was made to give path loss, mea excess delay, RMS delay spread. Ad a aalysis of lie impedace is importat to trasfer maximum power to the receiver [1]. case of medium voltage lies, lie impedace is eve depedet o lie legths, the juctio of couplig uits, cut off switch rather tha electrical loads i low voltage sides. This paper is orgaized as follows: Sectio, we described the test field set up at Chuggae area of uiwag city. Ad i Sectio ad 4, chael oise ad impedace characteristics are described respectively. Ad the, i Sectio 5, we itroduced a measuremet system for a multi-path propagatio ad measuremet results i medium voltage lie. Fially, it preseted coclusios i Sectio 6.. Test Field Setup i Medium Voltage Lies The test field had bee costructed over several kilometers at Chuggae area located i the suburbs of uiwag city as show i Fig. 1. This test field has bee beig used efficietly for measuremet of M

eviromet, developmets of PLC subscriber etworks as well as chael aalyses discussed i this paper. Besides, Voice over teret Protocol (VoP) service is beig provided with local residets.

2 eviromet, developmets of PLC subscriber etworks as well as chael aalyses discussed i this paper. Besides, Voice over teret Protocol (VoP) service is beig provided with local residets. Medium Voltage Power Lies,.9kV C.O.S. Couplig Uit Couplig Uit C.O.S. Couplig Capacitor Coaxial Cable Measuremet quipmet Drai Coil Sigal Lie Fig. terface betwee measuremet equipmet ad.9kv lies by a couplig uit. Fig. 1 Test field set up at Chugge area. The followig Fig. shows the simplified drawig of the test field. We istalled couplig uits that cosist of a couplig capacitor ad a drai coil at four poits, i.e., poit B, C, D, ad. The chael aalyses of oise, impedace, ad impulse respose were carried out through these poits. Ad all power lies betwee B ad are overhead lies, except for a part of the sectio betwee B ad C. The legth of this udergroud lie is 16m as show i Fig.. ; O8 : : : 8 ;<9^ ^ :; ;9 ^ :9^ ;^ :;^ ;: ; : 9;^ Q:: <^ ;:^ 8. Noise low voltage grids, oise characteristic is maily affected by the coected load such as ballast, vacuum cleaer ad so o. Ad geerally the oise is divided ito impulsive oise, sychroouso-sychroous colored oise, ad backgroud oise [][]. medium voltage lies, however, backgroud oise ad colored oise maily caused by radio frequecies, for example short-wave radio sigal, HAM were measured typically. That is to say, cotrary to low voltage grids, oise level does ot vary i the time domai cosiderably, because the measurig positios are too far from electrical load. this reaso, we just tried to fid the tred of oise levels roughly with a spectrum aalyzer i frequecy domais. Fig. 4 shows the iterpolated result of oise measured from the test bed, ad it is compared with other oise levels which measured from apartmet ad houses. D N<; :^ <;^ <: : Fig. Simplified drawig of the test field. Fig. shows the iterfaces betwee measuremet equipmet ad medium voltage lies by a couplig uit. order for protectig electrical devices like PLC modems ad measuremet equipmet agaist electrical faults, a cut-off-switch (COS) ad drai coil are iserted. additio, a isolatio trasformer is used to prevet rushig curret i case of ay faults. The couplig capacitor which does ot pass a high voltage but commuicatios sigals has bad-pass characteristics from 1.7MHz to 5MHz. Fig. 4 Compariso of oise levels betwee medium voltage power lies ad low voltage subscribers.

3 Accordig to this compariso, we could kow that the oise level i medium voltage lies is higher tha low voltage sides, approximately -db i a frequecy from 1MHz to MHz. Through these medium voltage oise data, we represeted a simple yet effective mathematical model of chael oise with just a backgroud oise ad a arrowbad oise []. A backgroud oise BN ( f ) ad a arrowbad oise NBN ( f ) are represeted as follow: BN( f ) = 5+ 15e f 8 = A B( f i + a [ dbmv], (1) NBN ( f ) f ) [ db]. () As a result, the total oise is described by TN ( f ) = BN( f ) + NBN( f ) [ db], () this figure a couplig uit symbolized as a dashed rectagle is placed betwee the.9kv power lies ad the measuremet equipmet. These -parameters i this T-equivalet etwork could be obtaied by measurig S- parameters with a etwork aalyzer. Usig S measured i medium voltage lies with a etwork aalyzer as show Fig. 6, the iput impedace could be described by 1 + S = = 1 + ( + L ) 5 1 S where is the characteristic impedace as a value of 5 Ω. S-matrix of coupler where: f is frequecy i MHz, a (,4) Normal radom variable i db, A -th bad oise amplitude, f B ( f ) si c ( ) for -.<f<. ad for otherwise, 6 f -th bad ceter frequecy i MHz. S [S] Fig.6 llustratio of impedace measuremets i medium voltage power lies. L 4. mpedace 4.1 mpedace Aalysis Method Ad the L is represeted by this paper we used the reflectio coefficiet S of the medium voltage power lie chaels to fid the lie impedace. Fig. 5 shows the simplified block diagram for ivestigatig the impedace of.9kv power lies. T-equivalet etwork of coupler = 1 Fig. 5 Total equivalet etworks usig T-shaped equivalet circuit. L L = (1 S) + ( + )( 1 S( 1 + (( + )( S 1) + ( S + 1)) 4. Measuremet Results 1 )) 7 We coected the measuremet equipmet, HP 875D vector etwork aalyzer, with medium voltage lies by couplig uits. Ad we performed measuremets of the S-parameters for medium voltage lies at those poits from B to preseted i Fig.. From the measuremet results, we calculated the average impedace value ad the showed the shape of polyomial curve fittigs i Fig. 7. Ad we foud that the average impedace value betwee 1MHz ad MHz is about 1~Ω by mea of averagig the curvig fittig results. particular, the impedace has periodic characteristic i frequecy domais due to reflectios i couplig juctios. However this problem is supposed to be improved by more careful studies.

4 5 total impedace Fit Values The measuremet system ca resolve the pulse duratio of 5 s, the time resolutio of the auto correlatio of MHz PN sequeces.

4 4 5 total impedace Fit Values The measuremet system ca resolve the pulse duratio of 5 s, the time resolutio of the auto correlatio of MHz PN sequeces. total [ohm] Frequecy[MHz] Fig. 7 Average impedace ad its iterpolated results. 5. mpulse Respose 5.1 Measuremet System 5. Measuremet Results We measured the impulse respose betwee two poits: B-C, C-B, C-D, D-C, D-, ad -D. For example, i case of B-C, the trasmitter ad the receiver were fixed respectively at poit B ad poit C show i Fig.. this maer, we set up the PN sequece geerator ad the sampled the trasmitted sigal with a digital oscilloscope for about secods i each sectio. The Fig. 9 ad Fig. 1 show the amplitude of impulse resposes i time domais for oe sectio, poit B to poit C ad poit C to poit B respectively. ach figure shows the results for 5 chael soudigs. Measuremets were carried out by the pseudo-radom oise correlatio method to obtai chael impulse respose. Ad the PN geerator was realized by FPGA board ad the PN sequeces geerated with the chip rate of MHz are multiplied by the carrier frequecy of 1 MHz. The block diagram of impulse respose measuremet system is show i Fig. 8. The carrier sigal modulated by PN sequeces are amplified by power amplifier ad ijected ito the power lie by broadbad lie coupler. order to achieve the high sesitivity of the receiver system, a low oise amplifier (LNA) is iserted after the coupler of receiver. The received sigals followed by BPF are over-sampled by the digital oscilloscope (DO), ad the the sampled data are saved at PC. After completig the measuremet process, the over-sampled sigals are dow coverted ad cross-correlated with the PN sequeces o PC to get chael impulse resposes. Fig. 9 mpulse respose from B to C. Mixer Power Amp Coupler PN Geerator Sigal Geerator P O W R L N PC DO BPF LNA Coupler Fig. 8 Block diagram of power lie chael impulse respose measuremet system. Fig. 1 mpulse respose from C to B.

5 additio to the represeted results, we worked out measuremets i every sectio ad aalyzed the acquired data. Summarizig the results briefly, the umber of multi-paths is icreased as the legth of measured two poits is exteded. There were multi-paths betwee B ad C, 5 multi-paths betwee C ad D, ad 6 multi-paths betwee D ad. Also, atteuatio ad mea excess delay are icreased proportio to the distace betwee trasmitter ad receiver. The atteuatio of the first paths is about -45dB, -5dB, ad -6dB for each sectio. Ad mea excess delay is approximately.15s,.9s, ad.1s for each sectio. ad Modelig for -House Power Lie Commuicatio, Proc. of teratioal Symposium o Power-Lie Commuicatios ad ts Applicatios, March. 6. Coclusio the previous sectios, we described the test bed set up at medium voltage lies cosidered broadbad trasmissio chaels ad the experimetal results from those power lies. Ad we reported the chael characteristics which divided ito oise, impedace, ad impulse respose. First of all, the oise level i medium voltage lies is higher tha other residetial buildigs ad affected much more by radio frequecies rather tha electric machies. the secod part, the value of the lie impedace was varied depedig o frequecies ad especially it had the periodic characteristics due to reflectios i couplig juctios. Ad the average impedace value i frequecy rage betwee 1MHz ad MHz was about 1-Ω. Fially, accordig to impulse respose aalysis we could get critical parameters such as the umber of multi-paths, atteuatio of each critical path, mea excess delay, ad so o. Ad i the future, researches o medium voltage lies should be carried out for more effective applicatios, so we would like to use this test field lastigly for uderstadig medium voltage power lie as broadbad commuicatios chaels. Referece [1] Yu Jug-Hu, et al., "mpedace Measuremet ad Matchig Techique for Medium-Voltage Powerlie Commucatio", teratioal Coferece o lectrical gieerig, C-188,. [] Hui-Myug Oh, Kwa-Ho Kim, Wo-Tae Lee, ad Jae-Jo Lee, A Fudametal Study for stablishmet of Chael Data Base i Power-Lie Commuicatios, Tras. o K, Vol. 5D, No., pp.17-1, Feb.. [] M. immerma ad K. Dostert, Aalysis ad modelig of impulsive oise i broad-bad powerlie commuicatios," Tras. o lectromagetic Compatibility, vol. 44, o. 1, pp.49-58, Feb.. [4] Yog-Hwa Kim, Hak-Hoo Sog, Jog-Ho Lee, ad Seog-Cheol Kim, Widebad Chael Measuremets

A New Space-Repetition Code Based on One Bit Feedback Compared to Alamouti Space-Time Code

A New Space-Repetition Code Based on One Bit Feedback Compared to Alamouti Space-Time Code Proceedigs of the 4th WSEAS It. Coferece o Electromagetics, Wireless ad Optical Commuicatios, Veice, Italy, November 0-, 006 107 A New Space-Repetitio Code Based o Oe Bit Feedback Compared to Alamouti