scieee Open visual document viewer

Random Generation of Arbitrary Waveforms for Emulating Three-Phase Systems

Montaño Asquerino, Juan-Carlos; León de Mora, Carlos; García Delgado, Antonio; López Ojeda, Antonio; Monedero Goicoechea, Iñigo Luis; Personal Vázquez, Enrique

Abstract

This paper describes an apparatus for generating a signal representative of steady-state and transient disturbances in three-phase waveforms of an ac electrical system as described in IEEE Std 1159-09. It can be configured as a synthesizer of randomly distorted signals for different applications: for testing the effects of disturbed grid on equipment and to generate patterns of electrical disturbances for the training of artificial neural networks, which are used for measuring power quality tasks. For the first purpose, voltage and current amplifiers are added in the output stage, which allows the generation of disturbed signals at grid level.

Full text

Random Gene a ion o A bi a y Wa e o ms o Emula ing Th ee-Phase Sys ems Juan-Ca los Mon año, Senio Membe , IEEE, Ca los León, Senio Membe , IEEE, An onio Ga cía, An onio López, Iñigo Monede o, and En ique Pe sonal, S uden Membe , IEEE Abs ac —This pape desc ibes an appa a us o gene a ing a signal ep esen a i e o s eady-s a e and ansien dis u bances in h ee-phase wa e o ms o an ac elec ical sys em as desc ibed in IEEE S d 1159-09. I can be con igu ed as a syn hesize o andomly dis o ed signals o di e en applica ions: o es ing he e ec s o dis u bed g id on equipmen and o gene a e pa - e ns o elec ical dis u bances o he aining o a i icial neu al ne wo ks, which a e used o measu ing powe quali y asks. Fo he i s pu pose, ol age and cu en ampli ie s a e added in he ou pu s age, which allows he gene a ion o dis u bed signals a g id le el. Index Te ms—AC gene a o s, a i icial neu al ne wo k (ANN), educa ion, load modeling, powe ha dwa e in he loop (PHIL), powe quali y (PQ), powe sys em simula ion, signal syn hesis, es equipmen . I. INTRODUCTION MOST o oday’s ins umen a ion o h ee-phase sys ems used in he indus y is ocused and designed o es ing and measu ing powe supply sys ems and equipmen connec ed he e o [1]. Eme ging me hods o ad anced expe imen a ion exis , such as he powe ha dwa e-in- he-loop (PHIL) simula- ion [2], [3] whe eby a piece o powe ha dwa e, o example, an ene gy-powe me e o a powe elec onic d i e, is ope a ed om a i ual g id, simula ed in eal ime wi h he necessa y powe capabili ies h ough p ecision powe ampli ie s. PHIL simula ion me hod can be applied oo o esea ching he eal- ime beha io o con olle and p o ec ion equipmen . All o hese applica ions ha e a posi i e and economic impac on indus y. In ano he con ex , i is well known ha an e ec i e way o es ing equipmen seems o be ha o supplying hei ol - age and cu en inpu channels wi h p io unknown signals andomly gene a ed [4], [5]. I he es is epea ed a gi en numbe o imes and a new se o es signals is andomly gene a ed a each epe i ion, each new es shows di e en andomly gene a ed es wa e o ms, and he e o e, he beha io o equipmen unde es is likely o be e i ied in eal condi ions o e y close o he eal ones. This wo k was suppo ed in pa by he esea ch p ojec e . C03-056 and in pa by he esea ch p ojec s CICYT/FEDER DPI2006-15467-C02-01 and CICYT/FEDER DPI2006-15467-C02-02. The au ho s a e wi h he Elec onic Technology Depa men , Uni e - si y o Se ille, Escuela Poli ecnica Supe io , 41011 Se ille, Spain (e-mail: [email p o ec ed]; [email p o ec ed]; [email p o ec ed]; [email p o ec ed]; imonede o@ us.es; [email p o ec ed]). In his pape , we de eloped an appa a us capable o gen- e a ing ealis ic wa e o ms ep esen a i e o many indus ial powe applica ions. Con ol and con igu a ion o he uni is e y in ui i e owing o he use o sel -explana o y g aphical in e ace. Ins ead o compe ing wi h comme cial sys ems, i is designed o p o ide a e y low cos al e na i e o eal- ime ha dwa e-in- he-loop applica ions while main aining ac- cep able speci ica ions. In he li e a u e, he e a e some examples o equipmen simila o he andom gene a o desc ibed in his pape , bu in some cases only is i possible o gene a e a limi ed numbe o dis u bances [6] o he g aphical in e ace is no use - iendly and has low con igu a ion possibili ies [7]. The ollowing sec ions will discuss he p oposed p ocedu e and p o ide he de ails o a labo a o y p o o ype speci ically de eloped: 1) wi hou ou pu ampli ie s: a) a bi a y and andom gene - a ion o aining pa e ns o op imal design o a i icial neu al ne wo ks (ANNs) [8] and b) a bi a y gene a ion o low- ol age and low-cu en le els (scaling-down e - sion o h ee-phase powe sys ems) in sepa a ed channels o aining o eaching; 2) wi h ou pu ampli ie s: c) a bi a y and andom gene a- ion o elec ical pa e ns a high-powe le els o es ing he e ec s o powe -line dis u bances on equipmen [5] and d) a bi a y gene a ion o high- ol age and high- cu en le els in sepa a ed channels o ep esen ing dy- namic beha io s o a powe sys em. II. AWG Acco ding o he conside a ions in he p e ious sec ion, he p oposed solu ion o es ing equipmen unde common wo k- ing condi ions o o aining ANNs is based on he gene a ion o andomly dis o ed wa e o ms o he ol age and cu en signals. This can be done in a ela i ely simple way employing a digi al- o-analog con e sion (DAC) boa d connec ed o a p og ammable de ice, such as a digi al signal p ocesso o PC. A. So wa e Based on his concep , LabVIEW and Tes S and so wa e (Na ional Ins umen ) we e used o p og am an au oma ic signal gene a o ha allows comple e con igu a ion o he pa e ns e- qui ed. Fi s , a i ual ins umen (VI), ha we called Pando a, was implemen ed in LabVIEW [9]. This VI mainly gene a es a pa e n based on he pa ame e s de ined by he use ; hus, Fig. 1. F on panel o Pando a 3ph. s eady-s a e o ansien -s a e dis u bances we e modeled as desc ibed he eina e . In a i s s age, he sys em was designed o be used manually, so he use needed o speci y he pa- ame e s o he pa e n o e e y simula ion. Then, Tes S and was included in o de o p o ide au oma ion. This so wa e ool can help he use o de elop au oma ed es and alida ion sys ems, i.e., i can de elop es sequences ha in eg a e code modules w i en in any es p og amming language. Sequences also speci y execu ion low, epo ing, da abase logging, and connec i i y o o he en e p ise sys ems. Finally, es sys ems o p oduc ion wi h easy- o-use ope a o in e aces can be o ga- nized [10]. Se e al algo i hms we e p og ammed in Tes S and o launch he Pando a VI acco ding o di e en goals ( andom ha monic con en s pa e ns, noisy signals, ansien dis u bances, e c.). Gene a ed pa e ns can be sa ed as ex iles o be used o line o ANN aining o ins ance o sen o he DAC boa d in li e expe imen s (as in es pu poses). The on panel o h ee-phase Pando a VI is shown in Fig. 1. A ypical sequence consis s basically o a se up ope a ion, whe e all he a iables ese o hei de aul alues, and he main ope a ion whe e pa ame e s and inpu s ha de ine he new ex- pe imen a e modi ied and Pando a VI is execu ed. Then, wi h all hese alues, epo s o he esul s a e gene a ed. Finally, he cleanup ope a ion is execu ed, whe e he a iables a e ese again o de aul alues. The e o e, he Pando a VI consis s o he main p ocesses o Fig. 2. Fi s , inpu s and pa ame e s a e collec ed om manual use o Tes S and au oma ic sequences. Second, he signal is buil acco ding o hese speci ica ions, and hi d, he pa e n is sen o a ile o o he DAC boa d. The main ea u e o he de eloped a chi ec u e is ha i is a laye -based model. I dis inguished h ee dis inc pa s o laye s p og ammed in LabVIEW ha ca y ou asks ha a e essen ially di e en . 1) Laye 3: This module is esponsible o managing he use in e ace ca ying ou he display o he windows on each case. On he o he hand, i allows he use o Fig. 2. Gene al diag am o gene a ion wi h Pando a. en e all he in o ma ion necessa y o he edi ion o he desi ed signal. The e o e, his laye is ocused on ga he ing in o ma ion ha comes om he p og am’s con ols, g ouped acco ding o some da a s uc u e, and ans e s i o laye 2. 2) Laye 2: I is he eal engine o he applica ion. F om he da a en e ed by he use and ha e p e iously been man- aged h ough laye 3, laye 2 pe o ms all calcula ions necessa y o gene a e he da a signal o be edi ed. This is whe e he a ious models a e implemen ed in each o he dis u bances ha we s udied p e iously. The ou pu o laye 2 is a se o nume ical da a co esponding o he edi ed signal. These alues a e ans e ed o laye 1. 3) Laye 1: This block pe o ms all he communica ions wi h he sc een, iles, and da a ca d. In his block, we p ope ly ake ca e o displaying he use in e ace. On he o he hand, i condenses he implemen a ion o all he unc ions necessa y o expo he da a signal ecei ed om laye 2 o a da a ile ha ollows a s anda d o ma . Mo eo e , i pe o ms all he communica ion wi h he ha dwa e ins alled o pe o m he physical gene a ion o he edi ed signal h ough he da a ca d gene a ion. The scheme o he laye -based model p og ammed in Lab- VIEW is shown in Fig. 3. 1) Cons ain s o Random Gene a ion: Using his sys em, i is possible o syn hesize any kind o wa e o ms in he equency ange o in e es o he emula ion o ac ual powe sys ems. O cou se, a pu ely andom gene a ion o he dis o ion com- ponen s is no likely o ep esen he eal ope a ing condi ions o ac ual powe sys ems. The sys em is designed o p og am dis u bances as desc ibed in powe quali y (PQ) s anda ds (i.e., IEEE S d 1159-09). The e o e, a numbe o cons ain s mus be conside ed by he gene a ion algo i hm. In pa icula , he ollowing cons ain s ha e been conside ed. 1) Highes ha monic o de is gi en o he syn hesize as an inpu pa ame e so ha he desi ed bandwid h is no exceeded. 2) Maximum allowed dis o ion is imposed as an inpu pa ame e in e ms o maximum allowable o al ha monic dis o ion (THD) ac o , sepa a ely o he ol age and cu en signals. 3) Fundamen al componen o he ol age signal, U1, can be se o any alue in he ange [0.9U1,1.1U1], whe e U1is he a ed ol age [1]. 4) Fundamen al componen o he cu en signal can be se o any desi ed pe cen age o he a ed cu en , and i s phase shi wi h espec o he undamen al componen o Fig. 3. LabVIEW a chi ec u e o Pando a. he ol age signal can be se o any alue, ei he leading o lagging. 5) Fundamen al equency can be se o any alue in he ange [0.98 n,1.02 n], whe e nis he a ed equency [1] (whe e n=50o 60 Hz). Taking in o accoun he a o emen ioned cons ain s, a an- dom numbe o s eady-s a e and ansien -s a e dis u bances can be emula ed. 2) S eady-S a e Dis u bances: • Ha monics Ha monic componen s, andomly dis ibu ed be ween he undamen al componen and he maximum desi ed ha monic o de , can be gene a ed. The cu en wa e o m always has he same ha monic compo- nen s as he ol age wa e o m. I also has addi ional ha monic componen s ha , when p o ided by he andom gene a ion, accoun o he p esence o nonlinea loads. Ampli ude and phase shi a e andomly gene a ed o each o he ha monic componen s de e mined in he p e ious s ep. A check is pe o med so ha he maximum desi ed THD ac o is no exceeded. I a sui able numbe o di e en andomly gene a ed dis- o ed signals a e used o supply he uni unde es (UUT), i is possible o assume ha a signi ican numbe o possible wo king condi ions ha e been ep oduced so ha he UUT can be hypo he ically es ed in i s eal wo king condi ions. The ollowing ma hema ical model was implemen ed: C( )=A+ N  n=1 Ansin(2πn 1 +ϕn)(1) whe e Ais he dc e m (V o A), Anis he ampli ude o he n h ha monic o he signal (V o A), 1is he undamen al equency (Hz), ϕnis he phase o he n h ha monic ( ad), and nis he ha monic o de (n=1,...,N). • Flicke Flicke is conside ed an ampli ude modula ion o he ca ied signal C( ), which changes in unc ion o he modula ing signal F( )[11]. The modula ing signal has sinusoidal o m wi h p e- ixed andom ampli ude, equency (usually a ound 30 Hz), and an ini ial phase. The equa ion which de ines he ma hema ical model implemen ed by he emula o is Z( )=F( )C( )=[1+A ksin(2π k +ϕ k)] C( )(2) whe e A k, k, and ϕ k a e he licke ampli ude, licke e- quency, and licke phase, espec i ely. • Unbalance: S eady-s a e symme ical componen s The me hod o symme ical componen analysis akes i s o igin om [12]. I p esen s a ma hema ical app oach o he analysis o an asymme ical polyphase sys em (unde s eady- s a e condi ions) by ans o ming i in o a se o symme ical se- quence ne wo ks called posi i e-, nega i e-, and ze o-sequence ne wo ks. I Fa,Fb, and Fcdeno e he co esponding o iginal phaso s o phases a, b, and c o a h ee-phase sys em, hen ⎡ ⎣ Fa Fb Fc ⎤ ⎦=⎡ ⎣ F+ a F+ b F+ c ⎤ ⎦+⎡ ⎣ F− a F− b F− c ⎤ ⎦+⎡ ⎣ F0 a F0 b F0 c ⎤ ⎦(3) whe e +,−, and 0 deno e posi i e-, nega i e-, and ze o- sequence componen s, espec i ely. Since each se o sequence componen s is balanced, one can deduce [13] ⎡ ⎣ F0 a F+ a F− a ⎤ ⎦=1 3 ⎡ ⎣ 11 1 1ββ 2 1β2β ⎤ ⎦⎡ ⎣ Fa Fb Fc ⎤ ⎦(4) whe e β=exp(j2π/3). 3) T ansien -S a e Dis u bances: • O e ol age, swell, unde ol age, and sag In hese kinds o dis u bances, he ampli ude o he signal ises (o e ol age o swell) o alls (unde ol age and sag) a ce ain alue along a ime in e al. In he de elopmen o he dis u bance gene a o , a ape- zoidal model o he ampli ude e olu ion (lineal slope) was conside ed. The model makes i possible o app oxima e he ampli ude dis u bances mos equen ly encoun e ed in powe sys ems. Fig. 4 shows a g aphic o he model used o o e ol - ages o swells (in e se apeze o unde ol ages and sags); he ollowing pa ame e s a e de ined: pini ial sample o he dis u bance; pip slope o he ini ial amp; p p slope o he inal amp; n1numbe o samples o he ini ial amp; n2numbe o samples o he inal amp; bnumbe o samples a he bo om; M o al numbe o samples. • Oscilla o y ansien s Fig. 4. Sag model. The elec ical pa e n gene a o (EPG) models ansien as a damped sine h ough a supe posed exponen ial unc ion, which is added o C( )a a ce ain poin . The implemen ed ma hema ical model is exp essed as T( )=e−a A sin(2π +ϕ )(5) whe e ais he oscilla o y- ansien exponen , A is he ampli- ude o he ipple (V o A), is he equency o he ipple (Hz), and ϕ is he ini ial phase o he ipple ( ad). •Noise The gene a o makes i possible o add an addi i e whi e Gaussian noise in o de o simula e mo e ealis ic signals o he powe line. B. Ha dwa e The emula o design is based on gene a ing h ee-phase ol ages and ou line cu en s o emula e a low- ol age powe sys em. Gene a ed da a se s a e ob ained om a hos PC in he o m o da a iles wi h Ame ican S anda d Code o In o ma ion In e change o ma compa ible wi h he mos popula da a- analysis ools (Ma lab, Ma hcad, e c.). The hos PC is equipped wi h a NI PXI 6733 boa d wi h eigh analog ou pu s a up o 1 MS/s, 16-b esolu ion, and ±10-V ou pu ange. Se en ol age signals a e a ailable: h ee p opo ional o he h ee phase-neu al ol ages (Va,Vb, and Vc) and ou p opo ional o he h ee line cu en s (Ia,Ib, and Ic) and neu al cu en (In). A he inal s age, he ampli ie sec ion b ings he ol age signals o he g id le el. The numbe and ype o ampli ie s depend on he applica ion o he sys em. The ol age ac oss he load and/o he cu en lowing h ough he load o UUT can be simula ed. Fig. 5 shows he simpli ied h ee-phase powe emula ion ci cui . 1) Ampli ie Sec ion: A e y c i ical issue in he design o he es sys em is he ampli ie sec ion since i mus ensu e accu- a e and cons an gain and phase shi on he whole bandwid h equi ed. On he o he hand, he ampli ie ype depends on he selec ed applica ion. Two op ions ha e been conside ed. Op ion 1. A powe ampli ie o es ol ages a g id le el In his ype o applica ion, dis u bed ol age signals a g id le el a e gene a ed o es ing equipmen unde eal condi ions Fig. 5. Simula ion o an ac ual ol age sou ce (Op ion 1): Vol age gene a o wi h p og ammed dis u bances and powe ampli ie . Fig. 6. Simula ion o a comple e sys em (Op ion 2): (a) Powe sou ce wi h p og ammable dis u bances applied o an a bi a y load and (b) ol age and cu en gene a o wi h p og ammable dis u bances and ampli ied ou pu s. (see Fig. 5). In his case, gene a ed ol age signals can be con- igu ed o single-phase o h ee-phase sys ems. The ampli ie sec ion consis s o one o h ee ol age ampli ie s wi h he powe capabili y necessa y o supply he UUT. A labo a o y p o o ype has been implemen ed o es ing single-phase sys ems. The Paci ic Powe Sou ce Model 320 was used as a powe ampli ie o ul ill all he p oposed equi emen s: 1) ou pu ol age up o ±600 peak ol s; 2) maximum ou pu powe : 1.2 kVA; 3) bandwid h (30–5 kHz) a ull powe ; 4) THD <0.2%. Op ion 2. A high- ol age ampli ie and a high-cu en anscon- duc ance ampli ie o es ol ages and load cu en s a g id le el (see Fig. 6) . In his second ype o applica ions, i is necessa y o gene a e bo h ol ages and cu en s a g id le els o simula e a comple e sys em wi h high- ol age sinusoidal supply sou ce wi h added dis u bances and a bi a y load. In hese applica ions, common UUT (powe me e s, powe analyze s, p o ec ion elays, e c.) Fig. 7. Labo a o y p o o ype o he p oposed es sys em wi h ol age and cu en ampli ie s. measu es he ol ages and/o cu en s h ough independen inpu s. These inpu s ypically ha e high-inpu impedance o ol age measu emen s and low-inpu impedance o cu en measu e- men s, so i is no necessa y o ampli y he ou pu signals wi h high-powe capabili ies. Vol age ampli ie s (A )ha e been designed using Apex PA05 high- ol age low-powe ope a ional ampli ie s wi h some impo an speci ica ions: 1) ou pu ol age up o ±600 peak ol s; 2) maximum ou pu cu en =5mA; 3) bandwid h (dc o 50 kHz); 4) THD <0.1%. Cu en ampli ie s (Ai)ha e been designed using ope a ional ampli ie s and disc e e ansis o s as high-cu en low-powe ansconduc ance ampli ie s. Thei main cha ac e is ics a e as ollows: 1) ou pu cu en up o ±50-A peak; 2) maximum ou pu ol age =±8V; 3) bandwid h (dc o 50 kHz); 4) THD <0.2%. Fig. 7 shows he implemen ed labo a o y p o o ype o he p oposed es sys em, wi h he wo ypes o ampli ie s. III. APPLICATIONS The single-phase o h ee-phase powe a bi a y wa e o m gene a o (AWG), wi hou o wi h he ampli ie s age con- igu ed acco ding o Op ion 1 o Op ion 2, allows a se o applica ions equi ing he use o high-powe o low-powe sou ces. A. T aining Pa e ns o Op imal Design o ANNs The AWG has been de eloped as an auxilia y ool o gene a e elec ic pa e ns o ANNs, i.e., we ha e con igu ed he AWG as an EPG. Pa e n ecogni ion in ANNs gene ally equi es p ep ocessing o da a, ea u e ex ac ion, and inal classi ica ion [16]–[19]. One o he mos impo an asks in he design and de elopmen p ocess o an ANN is o gene a e an adequa e numbe o aining pa e ns in o de o app oxima e u u e inpu s. Some imes, an op imal design o he ANN is ound, bu he limi ed numbe o aining pa e ns does no gi e good esul s. In pa icula , in PQ measu emen , a g ea numbe o elec ical pa e ns is necessa y due o he mul iple combina ions o di e en dis u bances which can coincide in one o a ious samples. Thus, in o de o ex ac he signal ea u es, ANNs a e usu- ally combined wi h ma hema ical analysis, such as Fou ie and wa ele ans o ms, o he gene a ion o signal ea u es which se e as inpu s o he ne wo k [8]. Thus, wi h he help o hese ma hema ic ools, he de ec ion o he elec ical dis u bances has ended o be easy, bu hei classi ica ion is s ill a di icul ask in which ANNs play an impo an ole [14]–[16]. Ano he addi ional p oblem wi h ANNs applied o PQ is he impossibili y o ob aining eal use ul aining pa e ns di ec ly om he powe g id due o he i egula appa i ion o hese dis u bances and he di icul y o cap u e hem. In u n, i is e y di icul o ge access o da abases o elec ical dis u bances and, once achie ed, o adjus he ea u es o he ob ained elec- ical pa e ns o he equi emen s o he sys em o be designed (in pa ame e s such as sample ime and ol age ange de ined o each ype o dis u bance). Thus, o he ask o aining ANNs o he de ec ion and classi ica ion o elec ical dis u bances, he EPG gene a es an unlimi ed numbe o pa e ns o be used by a classi ica ion sys em. The ypes o dis u bances include o e ol age, swell, unde ol age, sag, oscilla o y ansien s, ha monics, licke , unbalance, equency a ia ions, and in e up ions. Fo he applica ion desc ibed in [8], o e 27 000 signal iles ha e been gene a ed, including one-dis u bance signals and wo-dis u bance signals, and ied o sweep all he ypes o dis u bances. B. T aining o Teaching Tool Ano he impo an applica ion o he emula o sys em is as eaching and/o aining ool. Th ee-phase powe can be augh in a sa e labo a o y en i onmen using he h ee-phase powe emula o wi hou exposing he s uden o dange ous high ol ages and cu en s. Ano he impo an conside a ion o he emula ion sys em is he low cos o build and main ain. An a o dable low-cos emula ion sys em can be e y help ul o schools wan ing o imp o e hei labo a o y equipmen and eaching echniques. The s uden s can lea n abou phase-angle ela ionships by edi ing he ou pu wa e o ms, which a e dis- played on he sc een ia he moni o ing p og am. C. Tes ing Equipmen Using Op ion 1, i is possible o es di e en ypes o sys- ems in he p esence o supply ol age dis u bances a g id le el. The ou pu ol ages should be used o d i e he UUT wi hin he powe limi s o he powe ampli ie [14], [15]. In his kind o applica ions, i is possible o es he equipmen unde ol age dis u bances [20], [21], esea ching he eal- ime beha io o con olle and p o ec ion equipmen [4], [22] o es ing and Fig. 8. Labo a o y se up o es ing a powe me e . compa ing sys ems and/o echniques o dis u bance analysis [23]–[26]. This ype o applica ion has been implemen ed in he case s udy desc ibed in Sec ion IV o his pape . D. Scaling-Down Ve sion o Th ee-Phase Powe Sys ems Finally, sys em designe s and manu ac u e s can use he h ee-phase powe emula o o build and es a scaled-down e sion o he sys em being conside ed. S udying scaled-down e sions o he p oposed equipmen can g ea ly aid he designe in es ing and pe o ming sys em al e a ions by sa ing ime and cos . T oubleshoo ing can be con olled and easy o pe o m wi h smalle manageable sys ems. Once he small-scale model is ully unc ional, he sys em is eady o ull-scale assembly. IV. EXAMPLE OF REAL APPLICATION Fig. 8 shows an example o a eal expe imen al se up o es ing an ene gy-powe me e (powe line communica ion) unde condi ions o g id dis u bances. The accu acy in he measu emen o ol age, cu en , ac i e and eac i e powe , and powe ac o , wi h di e en kinds o load (linea Ro RL and nonlinea ) and pe iodical dis u bances ( licke , ha monics, combined, ansien , e c.), has been analyzed. Pando a sys em combined wi h a comme cial powe ol age ampli ie has been used o gene a e a single-phase ol age sig- nal a g id le el wi h an added se o common dis u bances. This signal had been applied o a digi al ene gy/powe me e (UUT) wi h di e en loads: linea ( eal o complex) and nonlinea (swi ched powe egula o ). The ou pu da a alues o he UUT ha e been compa ed, o di e en loads and dis u bances, o a e e ence powe -analysis sys em consis ing in a calib a ed senso ca d equipped wi h ol age and cu en senso s (bo h TABLE I EXPERIMENTAL RESULTS based on Hall e ec ) and a PC equipped wi h an analog- o- digi al acquisi ion boa d and con olled by da a-analysis so - wa e de eloped in LabVIEW. Fig. 9. Example 1—Ha monic dis u bance (Tes ID 17). (a) Pando a’s on panel. (b) Pando a’s ha monic se up. (c) Measu es om he e e ence powe - analysis sys em ( he measu ed cu en , whi e signal, is scaled by 20). Table I summa izes he esul s o es s in ac i e powe measu emen s. Because he measu es a e loca ed in he lowe ange o he UUT (<1kW), he p ac ical ou pu esolu ion is limi ed o wo digi s. Wi h hese p ecision limi s, we conclude ha he UUT is p ac ically no a ec ed, in he measu e o he ac i e powe , by he dis u bances es ed. Howe e , he e a e signi ican di e ences in he eac i e powe , and powe ac o , measu ed when he cu en has high- equency ha monics. This may be caused by a swi ching load o by he p esence o high- equency ha monics in ol age applied on a esis i e load. Figs. 9–11 show he on panel o Pando a and he co e- sponding sc een o he powe -analysis sys em o h ee ypes o dis u bance: ha monics, licke , and ansien . Fig. 10. Example 1—Flicke dis u bance (Tes ID 6). (a) Pando a’s on panel. (b) Measu es om he e e ence powe -analysis sys em ( he measu ed cu en , whi e signal, is scaled by 20). Fig. 11. Example 3—Gene a ion o a ansien dis u bance (Tes ID 25). (a) Pando a’s on panel. (b) Measu es om he e e ence powe -analysis sys em ( he measu ed cu en , whi e signal, is scaled by 20). This eal applica ion demons a es he lexibili y and ease o use ha he sys em Pando a in ol ed in his ype o applica ion. V. C ONCLUSION This pape has desc ibed an AWG o signals ep esen a i e o s eady-s a e and ansien dis u bances in single- o h ee- phase wa e o ms o elec ical sys ems, as desc ibed in IEEE S d 1159-09. The sys em has di e en po en ial applica ions: aining pa e n gene a ion o ANN design, aining o eaching ool, equipmen es ing, o scaling-down e sion o h ee-phase powe sys ems. Depending on he applica ion, i is possible o use he sys em wi h o wi hou an ou pu ampli ie . In ha case, he e a e wo possible op ions. The i s one consis s in a powe ampli ie o es a g id- ol age le el. The second op ion combines a low- powe high- ol age ampli ie and a high-cu en ansconduc- ance ampli ie o es g id- ol age and load-cu en le els. An example o a eal equipmen es ing applica ion is de- sc ibed in o de o demons a e he sys em lexibili y and ease o use. The sys em is a aluable al e na i e o comme cial a bi a y gene a o s in e ms o use - iendly in e ace, low cos , and lexibili y o dis u bance combina ion and con igu a ion. REFERENCES [1] Tes ing and Measu emen Techniques. Gene al Guide on Ha monics and In e ha monics Measu emen s and Ins umen a ion, o Powe Supply Sys ems and Equipmen Connec ed The e o, IEC 61000-4-7, 2004. [2] M. S eu e , C. S. Ed ing on, M. Slode becj, W. Rein, and J. Langs on, “A megawa -scale powe ha dwa e-in- he-loop simula ion se up o mo- o d i es,” IEEE T ans. Ind. Elec on., ol. 57, no. 4, pp. 1254–1260, Ap . 2010. [3] C. León de Mo a, Í. Monede o Goicoechea, A. Ga cía Delgado, A. López Ojeda, J. M. Elena O ega, and J. C. Mon año Asque ino, “Gene ado de pa ones eléc icos,” Spain, Pa en P200502457, Oc . 3, 2005. [4] B. Lu, X. Wu, H. Figue oa, and A. Mon i, “A low-cos eal- ime ha dwa e- in- he-loop es ing app oach o powe elec onics con ols,” IEEE T ans. Ind. Elec on., ol. 54, no. 2, pp. 919–931, Ap . 2007. [5] A. Fe e o and C. Muscas, “On he selec ion o he bes es wa e o m o calib a ing elec ical ins umen s unde nonsinusoidal condi ions,” IEEE T ans. Ins um. Meas., ol. 49, no. 2, pp. 382–387, Ap . 2000. [6] A. on Jouanne, E. Ma heson, and A. Wallace, “A powe quali y es pla - o m based on a 120 kVA p og ammable sou ce including expe imen al demons a ions,” Elec . Powe Compon. Sys ., ol. 31, no. 6, pp. 535–551, 2003. [7] K. O anpi oj, S. P em udeep eechacha n, M. Ngoudech, W. Muangjai, K. Yingkayun, and T. Boonsai, “The 3-phase 4-wi e ol age sag gene a o based on h ee dimensions space ec o modula ion in abc coo dina es,” in P oc. IEEE In . Symp. Ind. Elec on., 2009, pp. 275–280. [8] I. Monede o, C. León, J. Rope o, A. Ga cía, J. M. Elena, and J. C. Mon año, “Classi ica ion o elec ical dis u bances in eal ime using neu al ne wo ks,” IEEE T ans. Powe Del., ol. 22, no. 3, pp. 1288–1296, Jul. 2007. [9] NIC, Aus in, TX, LabView Analysis 8.2, Re e ence Manual, 2007. [10] NIC, Aus in, Tes S and, Using Tes S and, 2003. [11] B. Bha ga a, “A c u nace licke measu emen s and con ol,” IEEE T ans. Powe Del., ol. 8, no. 1, pp. 409–423, Jan. 1993. [12] C. L. Fo escue, “Me hod o symme ical co-o dina es applied o he solu ion o polyphase ne wo ks,” AIEE T ans., ol. 37, pp. 1027–1140, 1918. [13] U. K. Rao, M. K. Mish a, and A. Ghosh, “Con ol s a egies o load compensa ion using ins an aneous symme ical componen heo y un- de di e en supply ol ages,” IEEE T ans. Powe Del., ol. 23, no. 4, pp. 2310–2317, Oc . 2008. [14] K. P. B and, J. Kopainsky, and F. Wi we , “Th ee-phase measu emen e alua ion using a high-speed p ocesso wi h snapsho acili y,” IEEE T ans. Powe Del., ol. 3, no. 3, pp. 867–873, Jul. 1988. [15] D. Chu, “Phase digi izing sha pens iming measu emen s,” IEEE Spec ., ol. 25, no. 7, pp. 28–32, Jul. 1988. [16] P.K.Dash,S.K.Panda,A.C.Liew,B.Mish a,andR.K.Jena,“New app oach o moni o ing elec ic powe quali y,” Elec . Powe Sys . Res., ol. 46, no. 1, pp. 11–20, Jul. 1998. [17] J. V. Wijayakulasoo iya, G. A. Pu us, and P. D. Minns, “Elec ic powe quali y dis u bance classi ica ion using sel -adap ing a i icial neu al ne - wo k,” P oc. Ins . Elec . Eng.—Gene . T ansm. Dis ib., ol. 149, no. 1, pp. 98–101, Jan. 2002. [18] A. K. Ghosh and D. L. Lubkeman, “The classi ica ion o powe sys em dis u bance wa e o ms using a neu al ne wo k app oach,” IEEE T ans. Powe Del., ol. 10, no. 1, pp. 109–115, Jan. 1995. [19] D. Bo ás, M. Cas illa, N. Mo eno, and J. C. Mon año, “Wa ele and neu al s uc u e: A new ool o diagnos ic o powe sys em dis u bances,” IEEE T ans. Ind. Appl., ol. 37, no. 1, pp. 184–190, Jan./Feb. 2001. [20] F. V. Topalis, I. F. Gonos, and G. A. Vokas, “A bi a y wa e o m gene a o o ha monic dis o ion on compac luo escen lamps,” Measu emen , ol. 30, no. 4, pp. 257–267, Dec. 2001. [21] L. M. Ande son and K. B. Bowes, “The e ec s o powe -line dis u bances on consume elec onic equipmen ,” IEEE T ans. Powe Del., ol. 5, no. 2, pp. 1062–1065, Ap . 1990. [22] G. Pa ma and V. Dina ahi, “Real- ime digi al ha dwa e simula ion o powe elec onics and d i es,” IEEE T ans. Powe Del., ol. 22, no. 2, pp. 1235–1246, Ap . 2007. [23] A. Ca alio i and V. Cosen ino, “A ime-domain s a egy o he measu e- men o IEEE S anda d 1459-2000 powe quan i ies in h ee-phase and single-phase sys ems,” IEEE T ans. Powe Del., ol. 23, no. 4, pp. 2113– 2123, Oc . 2008. [24] E. Pé ez and J. Ba os, “A p oposal o on-line de ec ion and classi ica ion o ol age e en s in powe sys ems,” IEEE T ans. Powe Del., ol. 23, no. 4, pp. 2132–2138, Oc . 2008. [25] D. Gallo, C. Landi, and N. Rignano, “Real- ime digi al mul i unc ion in- s umen o powe quali y in eg a ed indexes measu emen ,” IEEE T ans. Ins um. Meas., ol. 57, no. 12, pp. 2769–2776, Dec. 2008. [26] A. S. Ce quei a, D. D. Fe ei a, M. V. Ribei o, and C. A. Duque, “Powe quali y e en s ecogni ion using a SVM-based me hod,” Elec . Powe Sys . Res., ol. 78, no. 9, pp. 1546–1552, Sep. 2008.