sensors Article
Research on the Error Characteristics of a 110 kV Optical Voltage Transformer under Three Conditions: In the Laboratory, Off-Line in the Field and During On-Line Operation Xia Xiao 1, *, Haoliang Hu 2 , Yan Xu 1 , Min Lei 2 and Qianzhu Xiong 2 1 2
*
School of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan 430070, China; [email protected] State Grid Electric Power Research Institute, Wuhan 430070, China; [email protected] (H.H.); [email protected] (M.L.); [email protected] (Q.X.) Correspondence: [email protected]; Tel.: +86-136-2710-7240
Academic Editor: Jose Luis Santos Received: 12 June 2016; Accepted: 9 August 2016; Published: 16 August 2016
Abstract: Optical voltage transformers (OVTs) have been applied in power systems. When performing accuracy performance tests of OVTs large differences exist between the electromagnetic environment and the temperature variation in the laboratory and on-site. Therefore, OVTs may display different error characteristics under different conditions. In this paper, OVT prototypes with typical structures were selected to be tested for the error characteristics with the same testing equipment and testing method. The basic accuracy, the additional error caused by temperature and the adjacent phase in the laboratory, the accuracy in the field off-line, and the real-time monitoring error during on-line operation were tested. The error characteristics under the three conditions—laboratory, in the field off-line and during on-site operation—were compared and analyzed. The results showed that the effect of the transportation process, electromagnetic environment and the adjacent phase on the accuracy of OVTs could be ignored for level 0.2, but the error characteristics of OVTs are dependent on the environmental temperature and are sensitive to the temperature gradient. The temperature characteristics during on-line operation were significantly superior to those observed in the laboratory. Keywords: OVT; error characteristics; laboratory; field off-line; on-line operation
1. Introduction Optical Voltage Transformers (OVTs) based on the linear electro-optic effect and applied at a certain scale around the world are recognized as a current development trend in the power transformer sector because of their simpler insulation structure, greater security and anti-electromagnetic interference properties [1–3]. OVTs developed by Nxtphase (Vancouver, BC, CA) have been applied to voltage levels from 123 kV to 500 kV [4,5], and OVTs developed by NAE (Beijing, China) have been applied to voltage levels from 110 kV to 500 kV [6]. OVTs are one of the electric energy measurement data sources in the power grid, and their measurement performance determines the accuracy and fairness of electric energy metering, therefore, the accuracy calibration and the temperature characteristics of OVTs must be tested before they leave the factory, and the accuracy of OVTs should be calibrated again before being put into operation on-site [7,8]. However, OVTs can be influenced by the ambient temperature. In previous studies, the influence of temperature on the stability of OVTs has been well researched. Researchers have proposed Sensors 2016, 16, 1303; doi:10.3390/s16081303
www.mdpi.com/journal/sensors
Sensors 2016, 16, 1303
2 of 9
some methods to advance the temperature stability of OVS, such as temperature control, results compensation, self-healing detection and so on [9–12]. In [9], Lee improved the temperature stability of an OVT from ±7.0% to ±0.75% within −2~65 ◦ C with dual light path compensation. In [10], the accuracy of an OVT was reported to be improved to ±0.5% within the −40 ◦ C~+60 ◦ C range by introducing a reference voltage for comparison. The testing of OVTs based on the above research on temperature characteristics was conducted in the laboratory. The experimental methods usually complied with the relevant standards such as IEC60044-7 [13], which are very different from the actual temperature changes during operation. Many testing results indicate it would be difficult to meet the accuracy requirements of level 0.2 in a wide temperature range. The electromagnetic environment and temperature changes during on-site operation are different from the laboratory ones. The accuracy test is usually performed only for a single OVT under laboratory test conditions, whereas three-phase transformers run synchronously in field operation, where the existence of the adjacent phase may cause additional errors [14]. In addition, in the laboratory the temperature experiments are performed according to the testing method recommended in IEC60044-7, in which the temperature gradient is more than that experienced under actual operation and it is unknown whether the testing results in the laboratory can truly reflect the actual performance of field operation, if the on-site operation characteristics are not monitored. In this paper, we selected OVT prototypes from a company of which multiple OVT products have already been put into operation, and then tested the error characteristics under the three conditions of the laboratory, off-line in the field and during on-line operation. The basic accuracy calibration, temperature characteristic experiments, and the adjacent phase effect experiment were tested in the laboratory. Then the OVT prototypes were transported to the station. The off-line accuracy tests were conducted before operation. Finally, the operation of OVT prototypes was monitored continuously for more than 1 year. The comparison and analysis for the error characteristics of OVT under the three conditions of laboratory, off-line in the field and during on-line operation, are expected to help promote the application of OVTs. 2. Principle and Structure of OVT The sensing principle of OVT which has been put into operation is the linear electro-optic effect (Pockels effect), whereby under the action of an external electric field, the refractive index of a sensing crystal changes with the applied electric field, as shown in Figure 1. When linearly polarized light incides on a sensing crystal along some direction, the birefringence phase delay δ caused by the Pockels effect is proportional to the applied electric field intensity [15], and: δ=
πl 3 n0 γ41 Ek λ
(1)
where λ is the wavelength of the input light, n0 is the refractive index, γ41 is electro-optic coefficient of the crystal, l is the length of the path of the light through the crystal, and Ek is the applied electric field intensity. From Equation (1), the corresponding applied electric field could be obtained by measuring the birefringence phase retardation. The installation structure diagram of an OVT is shown in Figure 2. The voltage sensor was installed inside the bottom tank of the insulator. The voltage to be measured was applied to the voltage sensor through a high voltage electrode in the insulating sleeve. The insulation between the high voltage electrode and ground was reinforced by the insulating basin and SF6 gas. The external tank was grounded to shield the external stray electric field. The output light signal of the voltage sensor was led out from the bottom of the cavity, and the signal demodulation was carried out at the bottom grounded supporter or in the control room.
Sensors 2016, 16, 1303
3 of 10
Sensors2016, 2016,16, 16,1303 1303 Sensors
of 910 33 of
Figure 1. The principle of OVT. Figure 1. The principle of OVT. Figure 1. The principle of OVT.
Figure 2. The installation structure diagram of OVT. Figure 2. The installation structure diagram of OVT.
3. Testing Scheme Scheme 3. Testing
Figure 2. The installation structure diagram of OVT.
Three 3. Testing Scheme Three OVT OVT prototypes prototypes based based on the the structure structure of of Figure Figure 22 were were selected selected with with the the following following √on parameters: rated voltage of 110/ 3 kV, rated frequency 50 Hz, accuracy class 0.2, rated secondary parameters: rated voltage of 110/√3 kV, rated frequency 50 Hz, accuracy class 0.2, rated secondary Three OVT prototypes based on the structure of Figure 2 were selected with the following output digital quantity 2D41H. TheThe temperature tests tests (Figure 3) and adjacent phase effect tests (Figure 4) output digitalrated quantity 2D41H. temperature (Figure and adjacent phase tests parameters: voltage of 110/√3 kV, rated frequency 50 Hz, 3) accuracy class 0.2, ratedeffect secondary for the OVT prototypes were carried out in laboratory. Then the prototypes were transported to the (Figure for the OVT 2D41H. prototypes carried tests out in(Figure laboratory. the prototypes output 4) digital quantity Thewere temperature 3) andThen adjacent phase effectwere tests Dongshan substation in Hegang City (Heilongjiang Province, China) which is an extremely cold is area. transported to the Dongshan substation in Hegang City (Heilongjiang Province, China) which an (Figure 4) for the OVT prototypes were carried out in laboratory. Then the prototypes were The field off-line accuracy was tested (Figure 5) before being put into operation. The on-line test results extremely cold The field off-line accuracy was tested (Figure 5) before being put into which operation. transported to area. the Dongshan substation in Hegang City (Heilongjiang Province, China) is an in operation (Figure 6) were transferred to the monitoring system (as shown in Figure 7) located in The on-line test results in operation (Figure 6) were transferred to the monitoring system shown extremely cold area. The field off-line accuracy was tested (Figure 5) before being put into(as operation. Wuhan City. Details in of Wuhan the testsCity. are shown inofTable 1. are shown in Table 1. inThe Figure 7) located tests on-line test results in operationDetails (Figure 6)the were transferred to the monitoring system (as shown in Figure 7) located in Wuhan City. Details of the tests are shown in Table 1. Table 1. Testing cases list for OVTs. Table 1. Testing cases list for OVTs.
casesSite list forSite OVTs. Test Time Remarks Test TimeTable 1. TestingTest Test Remarks 23 March 2014–3 Temperature tests Three Cases Test Time TestManufacturer Site Remarks Manufacturer Temperature tests of OVT 23 April March 2014 2014–3 April 2014 of OVT 23 March 2014–3 Laboratory Laboratory Manufacturer Temperature tests of OVT 17 April 2014–18 Grid Electric Power Adjacent phasephase effect tests Grid Electric Power Research Adjacent effect April2014–18 2014 April 2014StateState 17 April Laboratory April 2014 Research Institute, Wuhan on OVT Institute, Wuhan on OVT 17 April 2014–18 State Grid Electric Power Adjacenttests phase effect tests 6 May 2014–9 May Dongshan substation in Hegang The field off-line accuracy Dongshan substation in Hegang The field off-line April 2014 Research Institute, Wuhan on OVT Field off-line Field off-line 6 May 2014–9 May 2014 City of Heilongjiang Province accuracy for OVT 2014 City of Heilongjiang Province tests fortests OVT 6 May 2014–9 May Dongshan substation in Hegang The field off-line accuracy Field off-line On-line Dongshan substation in Hegang On-line Dongshan substation in Hegang The online operation 2014 tests for OVTof OVT 5 August 2014–2016 The online operation 5 August 2014–2016 City of Heilongjiang Province operation City of Heilongjiang Province of OVT operation City of Heilongjiang Province On-line Dongshan substation in Hegang 5 August 2014–2016 The online operation of OVT operation City of Heilongjiang Province Three Cases Three Cases
Sensors 2016, 16, 1303 Sensors 2016, 16, 1303
4 of 9 4 of 10
Sensors 2016, 16, 1303
4 of 10
Sensors 2016, 16, 1303 Sensors 2016, 16, 1303
4 of 10 4 of 10
Figure test in the laboratory. Figure 3.3.The The temperature testin inthe thelaboratory. laboratory. Figure3. The temperature temperature test Figure 3. The temperature test in the laboratory. Figure 3. The temperature test in the laboratory.
Figure4.4.The The adjacent adjacent phase effect test in the laboratory. Figure phase effecttest testin inthe the laboratory. Figure The effect Figure 4. Theadjacent adjacentphase phase effect laboratory. Figure 4. The adjacent phase effect test in the laboratory.
Figure5.5.Field Field off-line off-line accuracy test. Figure accuracy test. Figure accuracytest. test. Figure5.5.Field Field off-line off-line accuracy Figure 5. Field off-line accuracy test.
Figure 6. Site installation of the OVT. Figure6.6.Site Site installation of the OVT. Figure Figure 6. Site installation installationof ofthe theOVT. OVT.
Figure 6. Site installation of the OVT.
Sensors 2016, 16, 1303
5 of 10
Sensors 16, 1303 Sensors 2016,2016, 16, 1303
5 of 95 of 10
Figure 7. OVT on-line error monitoring system.
For the laboratory and field off-line accuracy tests, the calibration principle and circuit are shown Figure 7.consists OVT on-line monitoring system. in Figure 8. The calibration Figure system oferror a standard voltage 7. OVT on-line error monitoring system. transformer, high precision acquisition module, pulse forming circuit, industrial control computer, and control and calculation For laboratory the laboratory and field off-lineaccuracy accuracy tests, the calibration principle circuit are shown ForThe the and field off-line tests, theless calibration principle and circuit areless shown software. uncertainty of the calibration system was than 0.05 for and ratio error and than in Figure 8. The calibration system consists of a standard voltage transformer, high precision acquisition in Figure 8. The calibration system consists of a standard voltage transformer, high precision 2 min for phase error. For on-line operation, the accuracy results of OVT were obtained by comparing module, pulse forming circuit, industrial control computer, and control and calculation software. module, pulse forming circuit, industrial control computer, and control and calculation theacquisition output of OVT with the traditional voltage transformer. level of2traditional voltage The uncertainty of the calibration system was less than 0.05 for The ratio accuracy error and less than min for phase software. The uncertainty of the calibration system was less than 0.05 for ratio error and less than(2) transformer wason-line class operation, 0.2. The ratio error εresults and the phase error ϕe can definedthe byoutput Equations error. For the accuracy of OVT were obtained by be comparing of 2 min for phase error. For on-line operation, the accuracy results of OVT were obtained by comparing and (3),OVT respectively: with the traditional voltage transformer. The accuracy level of traditional voltage transformer was the output ofThe OVT with theε and traditional voltage The accuracy level of traditional voltage class 0.2. ratio error the phase error ϕtransformer. e can be defined by Equations (2) and (3), respectively: kU U s p transformer was class 0.2. The ratio error error ϕe can be defined by Equations(2) (2) 100% (%)ε and the phase kU − U U s p p and (3), respectively: ε (%) = × 100% (2) e U ps p (3) kU U (2) (%)ϕ = ϕs − pϕ 100% (3) e U where Us is the average value of the r.m.s. value insp 10 pperiods from the output of the OVT; k is the e of ins primary where Us is the average of the r.m.s.value value 10 periods from theaccording output of the k is the(3) rated transformation ratio; Uvalue p is the r.m.s. voltage to OVT; the output of the p rated transformation ratio; U is the r.m.s. value of primary voltage according to the output of the standard voltage transformer. ϕspis the initial phase of Us, and ϕp is the initial phase of Up. where Us is the average value ofϕsthe r.m.s. value inof10Usperiods the output the OVT; k is the standard voltage transformer. is the initial phase , and ϕp from is the initial phase of of U p. rated transformation ratio; Up is the r.m.s. value of primary voltage according to the output of the standard voltage transformer. ϕs is the initial phase of Us, and ϕp is the initial phase of Up.
Figure 8. The calibration scheme of OVT in the lab and off-line in the field.
Figure 8. The calibration scheme of OVT in the lab and off-line in the field. 4. Comparison of the OVT Error Characteristics under Three Conditions
Figure 8. The calibration scheme of OVT in the lab and off-line in the field. 4. Comparison of the OVT Error Characteristics under Three Conditions
In order to ensure the comparability of data, the testing method and the testing equipment in the laboratory exactly the Characteristics same as of those used in testing the off-line field tests. For online order to ensure the comparability data, the method and the testingoperation, equipment in 4. In Comparison of were the OVT Error under Three Conditions the output of a exactly traditional wasin compared as standard data with the same testing the laboratory were thevoltage same transformer as those used the off-line field tests. For online operation, the
In order to ensure the comparability of data, the testing method and the testing equipment in output of a traditional voltage transformer was compared as standard data with the same testing the laboratory were exactly the same as those used in the off-line field tests. For online operation, the method as in the laboratory. After the OVT prototypes were calibrated for the first time, no output of a traditional voltage transformer was compared as standard data with the same testing adjustment or compensation for error were done during the experiments. method as in the laboratory. After the OVT prototypes were calibrated for the first time, no adjustment or compensation for error were done during the experiments.
Sensors 2016, 16, 1303
6 of 9
Sensors 2016, 16, 1303
6 of 10
method as in theoflaboratory. After the OVT prototypes were calibrated for the first time, no adjustment 4.1. Comparison Error Characteristics of OVT in the Laboratory Sensors 2016, 16, 1303 6 of 10 or compensation for error were done during the experiments. The basic accuracy of the OVT was calibrated and tested under the rated voltage, and then the 4.1. Comparison of Error Characteristics OVT in temperature temperature characteristics were tested. For the 4.1. Comparison of Error Characteristics ofofOVT in the Laboratory Laboratorycharacteristic test, three OVT prototypes were placed in a temperature controlled cavity (Figure 3) andunder subjected to the ratedand voltage at room The basic accuracy OVTwas wascalibrated calibrated and thethe rated voltage, then then the The basic accuracy of of thethe OVT andtested tested under rated voltage, and the temperature (19.5 °C). The temperature changes with timecharacteristic are shown test, in Figure 9. The prototypes temperature characteristics were tested.For For the temperature temperature three OVT prototypes temperature characteristics were tested. the characteristic test, three OVT prototypes werewere warmed up 65 °C at a rate of 10 cavity °C every 20 min and kept atto65 for voltage 4 h, then cooled to placed in to a temperature controlled (Figure 3) and subjected the°C rated at room were placed in a temperature controlled cavity (Figure 3) and subjected to the rated voltage at room temperature (19.5 °C). The temperature changes with time are shown in Figure 9. The prototypes −40 °C with the same rate and kept at −40 °C for 4 h, and then warmed up to room temperature again temperature (19.5 ◦ C). The temperature changes with time are shown in Figure 9. The prototypes were warmed to kept 65 °Catatroom a ratetemperature of 10 °C everyfor 204min andtemperature kept at 65 °C characteristic for 4 h, then cooled with were the same rate◦up and h. The of thetoOVTs warmed upwith to 65 C at arate rateand of kept 10 ◦ C every 20 min andthen keptwarmed at 65 ◦ C for 4 h, then cooled to −40 ◦ C −40 °C the same at −40 °C for 4 h, and up to room temperature again was obtained in the temperature range from −40 °C to +65 °C, and phase B OVT testing results ◦ with with the same raterate andand kept at at −room 40 Ctemperature for 4 h, andfor then up to room temperature the same kept 4 h.warmed The temperature characteristic of theagain OVTs with are shown in Figure 10, where the ratio error and the phase error were calculated by the same rate and in kept room temperature for−40 4 h.°C The characteristic of theresults OVTs was was obtained theattemperature range from to temperature +65 °C, and phase B OVT testing Equations (2) and (3). ◦ ◦ are in shown in Figure 10, where thephase phaseB error were calculated obtained the temperature range fromthe −40ratio C toerror +65 and C, and OVT testing results areby shown Equations (2) and (3).ratio error and the phase error were calculated by Equations (2) and (3). in Figure 10, where the
Figure9.9.Temperature Temperature curve curvewith withtime. time. Figure time.
Figure 10. Temperature characteristic of OVTs in the laboratory.
Figure 10. OVTs in laboratory. Figure 10. Temperature Temperature laboratory. From Figure 10, the ratio error of the characteristic OVT is withinof−0.58% to the +2.43% in the temperature range from −40 °C to +65 °C, which is beyond the error limit value of level 0.2, whereas the phase error was From Figure 10, the ratio error of the OVT is within −0.58% to +2.43% in the temperature range withinFigure ±10’. 10, From the ratio error of the OVT is within −0.58% to +2.43% in the temperature range from −40 °C +65 three °C, which beyond the error limit value ofatlevel 0.2, whereas phase error was phaseisvoltage transformers are installed a certain distance, the adjacent phases ◦ Cto from −40When to the +65 ◦ C, which is beyond the error limit value of level 0.2, whereas the phase error was may interfere with each other. In order to investigate the extent of their mutual influence, three phase within ±10’. within ±10’. OVTs were set at 1phase m spacing intervals. The error are of the middle phase B was tested whenadjacent the voltage When the three voltage transformers installed at a certain distance, phases When the three phase voltage transformers are installed at a certain distance, adjacent phases of phase A and C was applied at zero volts and the rated voltage, respectively. The testingphase may interfere with each other. In order to investigate the extent of their mutual influence, three may interfere eachinother. In 11, order to investigate the extent of phase their mutualwere influence, threebyphase arewith shown Figure where ratiooferror and the OVTsresults were set at 1 m spacing intervals. Thethe error the middle phase B error was tested calculated when the voltage OVTsEquations were set (2) at 1and m (3). spacing intervals. The error of the middle phase B was tested when the voltage
of phase A and C was applied at zero volts and the rated voltage, respectively. The testing of phase A and C was applied at zero volts and the rated voltage, respectively. The testing results are results are shown in Figure 11, where the ratio error and the phase error were calculated by shown in Figure 11, where the ratio error and the phase error were calculated by Equations (2) and (3). Equations (2) and (3).
Sensors 2016, 16, 1303 Sensors 2016, 16, 1303
7 of 9
7 of 10
Sensors 2016, 16, 1303
7 of 10
(a)
(b)
(a)adjacent phases on OVT. (a) Influence of adjacent phases (b) Figure Influence ratio error; Figure 11.11. Influence ofofadjacent phases on OVT. (a) Influence of adjacent phasesononOVT’s OVT’s ratio error; (b) Influence of adjacent phases on OVT’s phase error. 11. Influence of adjacent phases on phase OVT. (a) Influence of adjacent phases on OVT’s ratio error; (b)Figure Influence of adjacent phases on OVT’s error. (b) Influence of adjacent phases on OVT’s phase error.
From Figure 11, it is shown that the influence of the adjacent phase on OVT is small; the From Figure it is shown that the influence of the adjacent phase onless OVT is small; additional additional ratio11, error than 0.02% and influence the additional phase error phase is than 1’. isthe From Figure 11, is it less is shown that the of the adjacent on OVT small; the ratio error is less than 0.02% and the additional phase error is less than 1’. additional ratio error is less than 0.02% and the additional phase error is less than 1’. 4.2. Comparison of Error Characteristics of OVT under the Three Conditions 4.2. Comparison of Error Characteristics of OVT under the Three Conditions 4.2. Comparison of Error Characteristics of OVT underand the Three Conditions After testing the temperature characteristics adjacent phase influence in the laboratory, the After testing the temperature characteristics and adjacentThe phase influence in the laboratory, the OVT prototypes were transported to the Hegang substation. accuracy was before being After testing the temperature characteristics and adjacent phase influence in tested the laboratory, the OVT prototypes were transported to the Hegang substation. The accuracy was tested before being put put into operation. The error curves in thesubstation. laboratory,The in off-line state the field and being in the OVT prototypes were transported to of theOVT Hegang accuracy wasintested before into operation. The error of OVT in theinlaboratory, inthe off-line state in theinfield and thein on-line are shown in Figure 12. order to exclude effect of temperature, the in tests inon-line the put into state operation. The curves error curves of In OVT the laboratory, in off-line state the field and the state are shown in Figure 12. In order to exclude the effect of temperature, the tests in the laboratory laboratory and the field off-line tests were carried out at ambient temperature (20 °C). The online on-line state are shown in Figure 12. In order to exclude the effect of temperature, the tests in the operation wastests collected during August to 8 August 2014, when was to and the fielddata off-line were carried out atcarried ambient (20 ◦the C).temperature The online data laboratory and the field off-line tests5were outtemperature at ambient temperature (20 °C).operation The close online ◦ 20 °C. In Figure 12, the ratio error and the phase error were calculated by Equations (2) and (3). was collecteddata during August during to 8 August 2014,towhen the temperature close to 20 C. Inclose Figure operation was 5collected 5 August 8 August 2014, when was the temperature was to 12, the20ratio error and12, the phase calculated by Equations (2) and °C. In Figure the ratio error error were and the phase error were calculated by (3). Equations (2) and (3).
(a)
(b)
(a) Figure 12. Error comparison of OVT under the three conditions of lab, in the(b) field off-line and during on-line operation. (a) The ratio error comparison of OVT; (b) The phase error comparison of OVT. Figure Errorcomparison comparisonof of OVT OVT under under the field off-line and during Figure 12.12.Error the three threeconditions conditionsofoflab, lab,ininthe the field off-line and during on-line operation. (a) The ratio error comparison of OVT; (b) The phase error comparison of OVT. on-line operation. (a) The ratio error comparison of OVT; (b) The phase error comparison of OVT. From Figure 12, it can be seen that the errors of OVT in the laboratory and off-line in the field wereFrom nearly consistent. Thebeerror between on-line operation and and the laboratory is 0.05%, Figure 12, it can seendifference that the errors of OVT in the laboratory off-line in the field as the testing results of on-line operation were compared with a traditional voltage transformer. This From Figure 12, it can be seen that the errors of OVT in the laboratory and off-line in the field were nearly consistent. The error difference between on-line operation and the laboratory is 0.05%, indicated that the effect of the transportation process and the external electric field caused by field were nearly consistent. The error difference between on-line operation and the laboratory is 0.05%, as the testing results of on-line operation were compared with a traditional voltage transformer. This as wiring the the OVT could neglected. theindicated testingonresults ofeffect on-line operation were compared with traditional voltage This that ofbe the transportation process and theaexternal electric field transformer. caused by field indicated that effect ofbe the transportation process and the external electric field caused by field wiring on thethe OVT could neglected. 4.3. Error Characteristics in On-Site Operation wiring on the OVT could of beOVT neglected. 4.3. Error of OVT On-Site Operationin August 2014, then the data from September 2014 The Characteristics OVT prototypes werein put into operation 4.3.toError Characteristics of OVT in On-Site Operation October 2015 were selected as analysis object. in The operation the twofrom summer and winter The OVT prototypes were put into operation August 2014,during then the data September 2014 seasons which display the greatest temperature difference, were used to illustrate the operational The OVT2015 prototypes were put into operation in August 2014, then the the two datasummer from September 2014 to October were selected as analysis object. The operation during and winter characteristics of theselected OVTs. Theanalysis data shown Figure 13 were correspond toillustrate August 2015 when the which display the greatest temperature used tothe the operational to seasons October 2015 were as object.indifference, The operation during two summer and winter temperatures ranged from 17The °C data totemperature 38 shown °C and in December 2014, when theto ranged from characteristics of the OVTs. Figure 13 correspond August 2015 when the seasons which display the greatest difference, were used totemperature illustrate the operational temperatures of ranged from 17The °C to 38 °C and December when the temperature ranged fromthe characteristics the OVTs. data shown in Figure 2014, 13 correspond to August 2015 when temperatures ranged from 17 ◦ C to 38 ◦ C and December 2014, when the temperature ranged from
Sensors 2016, 16, 1303
8 of 9
Sensors 2016, 16, 1303
8 of 10
◦ −20 ◦ C −20 to − InIn Figure ratioerror errorand and phase calculated °C1 toC, −1respectively. °C, respectively. Figure10, 10, the the ratio thethe phase errorerror were were calculated using using Equations and (3). Equations (2) and(2)(3).
Figure Error comparison of of OVT andand winter. Figure 13. 13. Error comparison OVTininsummer summer winter.
From Figure 13, it can be seen that the ratio error of OVT was affected by the temperature. During
From Figure it canwhich be seen thatthe the ratio temperature error of OVT was affected byerror the range temperature. summer and13, winter, display biggest difference, the ratio was from During summer−0.25% and winter, display the was biggest temperature difference, the ratio errorwas range was from to +0.3%,which and the phase error within ± 4'. In summer, the ratio error difference 0.55%, is higher that in winter. −0.25%which to +0.3%, andthan the phase error was within ± 40 . In summer, the ratio error difference was 0.55%, Compared with in thewinter. temperature characteristics in the laboratory, the temperature characteristics which is higher than that in the field are much better. When the temperature gradient is different, the thermal stress formed in Compared with the temperature characteristics in the laboratory, the temperature characteristics the sensing crystal has a great difference [16,17]. The greater the temperature gradient is, the greater in the field are much better. When the temperature gradient thecharacteristics thermal stress formed in the thermal stress is, and the greater the additional error is. is Fordifferent, temperature in the the sensing crystal a great difference [16,17]. greater temperature gradientgradient is, the greater laboratory thehas temperature variation rate was 10 The degrees everythe 20 min, but the temperature in thestress actual is, runtime environment faradditional less. Therefore, theis. on-line operation characteristics of the in the the thermal and the greater is the error For temperature characteristics OVTs were better than the laboratory temperature testing results. laboratory the temperature variation rate was 10 degrees every 20 min, but the temperature gradient in the actual runtime environment is far less. Therefore, the on-line operation characteristics of the 5. Discussion OVTs were better than the laboratory temperature testing results. 5.1. Effect of the Adjacent Phase
5. Discussion
The effect of the adjacent phase on a 110 kV OVT with reasonable structure was small. The additional error was within 0.02% and could be ignored for the OVTs of level 0.2.
5.1. Effect of the Adjacent Phase
Differences of theadjacent Laboratory phase and On-Site TestkV Results The5.2.effect of the on Field a 110 OVT with reasonable structure was small. The additional within 0.02% could were be ignored for the OVTs levelobtained 0.2. in the The error testingwas results of OVT in theand laboratory in good consistency withofthose field, and the additional error was less than 0.05%. On the one hand, the electric field in the OVT
5.2. Differences ofconcentrated, the Laboratory andaccuracy On-Sitewas Field Results sensors is so the notTest sensitive to the wire connection mode; on the other hand, the OVTs were not affected by the vibration of the transportation process.
The testing results of OVT in the laboratory were in good consistency with those obtained in the field, and additional error was less than 0.05%. On the one hand, the electric field in the OVT 5.3.the Differences of Temperature Characteristics sensors is concentrated, theofaccuracy was not sensitive to the connection mode; range on the other The ratio error so range OVT in the laboratory was −0.58% to wire +2.43% in the temperature hand, the OVTs were not affected by the vibration of the transportation process. from −40 °C to 65 °C, and −0.25% to +0.3% during on-site operation in a temperature range from −20 °C to 40 °C. The temperature characteristics of OVT could not meet the 0.2 level accuracy
5.3. Differences of Temperature requirements. However,Characteristics the temperature characteristics during on-site operation were better than those in the laboratory, as the temperature gradient was larger for the laboratory tests
Thethan ratio error range of OVT in the laboratory was −0.58% to +2.43% in the temperature range during on-site operation. The temperature characteristics of OVTs were dependent on the ◦ C to 65 ◦ C, and −0.25% to +0.3% during on-site operation in a temperature range from from −40 temperature gradient. −20 ◦ C to 40 ◦ C. The temperature characteristics of OVT could not meet the 0.2 level accuracy requirements. However, the temperature characteristics during on-site operation were better than those in the laboratory, as the temperature gradient was larger for the laboratory tests than during on-site operation. The temperature characteristics of OVTs were dependent on the temperature gradient. Acknowledgments: We would like to express our gratitude for financial support from the National Natural Science Foundation of China under Grant 51307066 and State Grid Corporation of science and technology foundation under Grant JL71-16-005. Author Contributions: X.X., Y.X. and M.L. conceived and designed the experiments; H.H. and Q.X. performed the experiments; X.X. and H.H analyzed the data; X.X. wrote the paper.
Sensors 2016, 16, 1303
9 of 9
Conflicts of Interest: The authors declare no conflict of interest.
Abbreviations OVT MDPI DOAJ TLA LD
Optical Voltage Transformer Multidisciplinary Digital Publishing Institute Directory of Open Access Journals Three letter acronym linear dichroism
References 1.
2.
3. 4.
5.
6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.
Rahmatian, F.; Ortega, A. Applications of optical current and voltage sensors in high-voltage systems. In Proceedings of the 2006 IEEE/PES Transmission and Distribution Conference and Exposition: Latin America, Caracas, Venezuela, 15–18 August 2006; pp. 1–4. Marchese, S.V.; Bohnert, K.; Wildermuth, S.; van Mechelen, J.L.M.; Steiger, O.; Rodoni, L.C.; Eriksson, G.; Czyzewski, J. Electro-optic voltage sensor based on BGO for air-insulated high voltage substations. In Proceedings of the 2013 IEEE Photonics Conference, Bellevue, WA, USA, 8–12 September 2013; pp. 608–609. Li, H.; Cui, J.; Shi, M. Application of optical voltage transformer in digital substation. Chin. Electr. Power Technol. Ed. 2012, 11, 460–462. Sanders, G.A.; Blake, J.N.; Rose, A.H.; Rahmatian, F.; Herdman, C. Commercialization of fiber-optic current and voltage sensors at NxtPhase. In Proceedings of the 15th Optical Fiber Sensors Conference, Portland, OR, USA, 10 May 2002; pp. 31–34. Rahmatian, F.; Chavez, P.P.; Jaeger, N.A. 138 kV and 345 kV wide-band SF 6-free optical voltage transducers. In Proceedings of the Power Engineering Society Winter Meeting, New York, NY, USA, 27–31 January 2002; pp. 1472–1477. Li, J.; Zheng, Y.; Gu, S.; Xu, L. Application of Electronic Instrument Transformer in Digital Substation. Autom. Electr. Power Syst. 2007, 31, 94–98. Liu, B.; Ye, G.; Guo, K.; Tong, Y.; Hu, B.; Wan, G. Quality Test and Problem Analysis of Electronic Transformer. High Voltage Eng. 2012, 38, 2972–2980. Ding, T.; Xu, E.; Wu, H.; Wang, X.; Yang, N.; Liu, Z.; Jiao, J. Measurement and Problem’s Analysis of the On-Site Error for Electronic Transformer. Electr. Meas. Instrum. 2011, 48, 36–39. Lee, K.S. Electrooptic Voltage Sensor: Birefringence Effects and Compensation Methods. Appl. Opt. 1990, 29, 4453–4461. [CrossRef] [PubMed] Wang, H. Research of Capacitor Divider Optical Voltage Transducer. Ph.D. Thesis, Harbin Institute of Technology, Harbin, China, 2010. Arpaia, P.; Daponte, P.; Grimaldi, D.; Michaeli, L. ANN-based error reduction for experimentally modeled sensors. IEEE Trans. Instrum. Meas. 2002, 51, 23–30. [CrossRef] Filippov, V.N.; Starodumov, A.N.; Minkovich, V.P.; Lecona, F.G.P. Fiber sensor for simultaneous measurement of voltage and temperature. IEEE Photonics Technol. Lett. 2000, 12, 1543–1545. [CrossRef] International Electrotechnical Commission. Instrument Transformers—Part 7: Electronic Voltage Transformers; International Electrotechnical Commission: Geneva, Switzerland, 1999. Wu, W.; Xu, Y.; Xiao, X.; Hu, H. Research on Proximity Effect in Measuring Error of Active Electronic Voltage Transformers. IEEE Trans. Instrum. Meas. 2016, 65, 78–87. [CrossRef] Chen, G.; Liao, L.; Hao, W. Fundermentials of Crystal Physics, 2nd ed.; Science Press: Beijing, China, 2007. Xiao, X.; Xu, Y.; Dong, Z. Thermodynamic modeling and analysis of an optical electric-field sensor. Sensors 2015, 15, 7125–7135. [CrossRef] [PubMed] Xiao, X.; Xu, Y.; Xu, K.; Ye, M. Influence of temperature and stress field on optical voltage sensor. In Proceedings of the 2009 Symposium on Photonics and Optoelectronics, Wuhan, China, 14–16 August 2009; pp. 1–4. © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
Research on the Error Characteristics of a 110 kV Optical Voltage Transformer under Three Conditions: In the Laboratory, Off-Line in the Field and During On-Line Operation Xia Xiao 1, *, Haoliang Hu 2 , Yan Xu 1 , Min Lei 2 and Qianzhu Xiong 2 1 2
*
School of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan 430070, China; [email protected] State Grid Electric Power Research Institute, Wuhan 430070, China; [email protected] (H.H.); [email protected] (M.L.); [email protected] (Q.X.) Correspondence: [email protected]; Tel.: +86-136-2710-7240
Academic Editor: Jose Luis Santos Received: 12 June 2016; Accepted: 9 August 2016; Published: 16 August 2016
Abstract: Optical voltage transformers (OVTs) have been applied in power systems. When performing accuracy performance tests of OVTs large differences exist between the electromagnetic environment and the temperature variation in the laboratory and on-site. Therefore, OVTs may display different error characteristics under different conditions. In this paper, OVT prototypes with typical structures were selected to be tested for the error characteristics with the same testing equipment and testing method. The basic accuracy, the additional error caused by temperature and the adjacent phase in the laboratory, the accuracy in the field off-line, and the real-time monitoring error during on-line operation were tested. The error characteristics under the three conditions—laboratory, in the field off-line and during on-site operation—were compared and analyzed. The results showed that the effect of the transportation process, electromagnetic environment and the adjacent phase on the accuracy of OVTs could be ignored for level 0.2, but the error characteristics of OVTs are dependent on the environmental temperature and are sensitive to the temperature gradient. The temperature characteristics during on-line operation were significantly superior to those observed in the laboratory. Keywords: OVT; error characteristics; laboratory; field off-line; on-line operation
1. Introduction Optical Voltage Transformers (OVTs) based on the linear electro-optic effect and applied at a certain scale around the world are recognized as a current development trend in the power transformer sector because of their simpler insulation structure, greater security and anti-electromagnetic interference properties [1–3]. OVTs developed by Nxtphase (Vancouver, BC, CA) have been applied to voltage levels from 123 kV to 500 kV [4,5], and OVTs developed by NAE (Beijing, China) have been applied to voltage levels from 110 kV to 500 kV [6]. OVTs are one of the electric energy measurement data sources in the power grid, and their measurement performance determines the accuracy and fairness of electric energy metering, therefore, the accuracy calibration and the temperature characteristics of OVTs must be tested before they leave the factory, and the accuracy of OVTs should be calibrated again before being put into operation on-site [7,8]. However, OVTs can be influenced by the ambient temperature. In previous studies, the influence of temperature on the stability of OVTs has been well researched. Researchers have proposed Sensors 2016, 16, 1303; doi:10.3390/s16081303
www.mdpi.com/journal/sensors
Sensors 2016, 16, 1303
2 of 9
some methods to advance the temperature stability of OVS, such as temperature control, results compensation, self-healing detection and so on [9–12]. In [9], Lee improved the temperature stability of an OVT from ±7.0% to ±0.75% within −2~65 ◦ C with dual light path compensation. In [10], the accuracy of an OVT was reported to be improved to ±0.5% within the −40 ◦ C~+60 ◦ C range by introducing a reference voltage for comparison. The testing of OVTs based on the above research on temperature characteristics was conducted in the laboratory. The experimental methods usually complied with the relevant standards such as IEC60044-7 [13], which are very different from the actual temperature changes during operation. Many testing results indicate it would be difficult to meet the accuracy requirements of level 0.2 in a wide temperature range. The electromagnetic environment and temperature changes during on-site operation are different from the laboratory ones. The accuracy test is usually performed only for a single OVT under laboratory test conditions, whereas three-phase transformers run synchronously in field operation, where the existence of the adjacent phase may cause additional errors [14]. In addition, in the laboratory the temperature experiments are performed according to the testing method recommended in IEC60044-7, in which the temperature gradient is more than that experienced under actual operation and it is unknown whether the testing results in the laboratory can truly reflect the actual performance of field operation, if the on-site operation characteristics are not monitored. In this paper, we selected OVT prototypes from a company of which multiple OVT products have already been put into operation, and then tested the error characteristics under the three conditions of the laboratory, off-line in the field and during on-line operation. The basic accuracy calibration, temperature characteristic experiments, and the adjacent phase effect experiment were tested in the laboratory. Then the OVT prototypes were transported to the station. The off-line accuracy tests were conducted before operation. Finally, the operation of OVT prototypes was monitored continuously for more than 1 year. The comparison and analysis for the error characteristics of OVT under the three conditions of laboratory, off-line in the field and during on-line operation, are expected to help promote the application of OVTs. 2. Principle and Structure of OVT The sensing principle of OVT which has been put into operation is the linear electro-optic effect (Pockels effect), whereby under the action of an external electric field, the refractive index of a sensing crystal changes with the applied electric field, as shown in Figure 1. When linearly polarized light incides on a sensing crystal along some direction, the birefringence phase delay δ caused by the Pockels effect is proportional to the applied electric field intensity [15], and: δ=
πl 3 n0 γ41 Ek λ
(1)
where λ is the wavelength of the input light, n0 is the refractive index, γ41 is electro-optic coefficient of the crystal, l is the length of the path of the light through the crystal, and Ek is the applied electric field intensity. From Equation (1), the corresponding applied electric field could be obtained by measuring the birefringence phase retardation. The installation structure diagram of an OVT is shown in Figure 2. The voltage sensor was installed inside the bottom tank of the insulator. The voltage to be measured was applied to the voltage sensor through a high voltage electrode in the insulating sleeve. The insulation between the high voltage electrode and ground was reinforced by the insulating basin and SF6 gas. The external tank was grounded to shield the external stray electric field. The output light signal of the voltage sensor was led out from the bottom of the cavity, and the signal demodulation was carried out at the bottom grounded supporter or in the control room.
Sensors 2016, 16, 1303
3 of 10
Sensors2016, 2016,16, 16,1303 1303 Sensors
of 910 33 of
Figure 1. The principle of OVT. Figure 1. The principle of OVT. Figure 1. The principle of OVT.
Figure 2. The installation structure diagram of OVT. Figure 2. The installation structure diagram of OVT.
3. Testing Scheme Scheme 3. Testing
Figure 2. The installation structure diagram of OVT.
Three 3. Testing Scheme Three OVT OVT prototypes prototypes based based on the the structure structure of of Figure Figure 22 were were selected selected with with the the following following √on parameters: rated voltage of 110/ 3 kV, rated frequency 50 Hz, accuracy class 0.2, rated secondary parameters: rated voltage of 110/√3 kV, rated frequency 50 Hz, accuracy class 0.2, rated secondary Three OVT prototypes based on the structure of Figure 2 were selected with the following output digital quantity 2D41H. TheThe temperature tests tests (Figure 3) and adjacent phase effect tests (Figure 4) output digitalrated quantity 2D41H. temperature (Figure and adjacent phase tests parameters: voltage of 110/√3 kV, rated frequency 50 Hz, 3) accuracy class 0.2, ratedeffect secondary for the OVT prototypes were carried out in laboratory. Then the prototypes were transported to the (Figure for the OVT 2D41H. prototypes carried tests out in(Figure laboratory. the prototypes output 4) digital quantity Thewere temperature 3) andThen adjacent phase effectwere tests Dongshan substation in Hegang City (Heilongjiang Province, China) which is an extremely cold is area. transported to the Dongshan substation in Hegang City (Heilongjiang Province, China) which an (Figure 4) for the OVT prototypes were carried out in laboratory. Then the prototypes were The field off-line accuracy was tested (Figure 5) before being put into operation. The on-line test results extremely cold The field off-line accuracy was tested (Figure 5) before being put into which operation. transported to area. the Dongshan substation in Hegang City (Heilongjiang Province, China) is an in operation (Figure 6) were transferred to the monitoring system (as shown in Figure 7) located in The on-line test results in operation (Figure 6) were transferred to the monitoring system shown extremely cold area. The field off-line accuracy was tested (Figure 5) before being put into(as operation. Wuhan City. Details in of Wuhan the testsCity. are shown inofTable 1. are shown in Table 1. inThe Figure 7) located tests on-line test results in operationDetails (Figure 6)the were transferred to the monitoring system (as shown in Figure 7) located in Wuhan City. Details of the tests are shown in Table 1. Table 1. Testing cases list for OVTs. Table 1. Testing cases list for OVTs.
casesSite list forSite OVTs. Test Time Remarks Test TimeTable 1. TestingTest Test Remarks 23 March 2014–3 Temperature tests Three Cases Test Time TestManufacturer Site Remarks Manufacturer Temperature tests of OVT 23 April March 2014 2014–3 April 2014 of OVT 23 March 2014–3 Laboratory Laboratory Manufacturer Temperature tests of OVT 17 April 2014–18 Grid Electric Power Adjacent phasephase effect tests Grid Electric Power Research Adjacent effect April2014–18 2014 April 2014StateState 17 April Laboratory April 2014 Research Institute, Wuhan on OVT Institute, Wuhan on OVT 17 April 2014–18 State Grid Electric Power Adjacenttests phase effect tests 6 May 2014–9 May Dongshan substation in Hegang The field off-line accuracy Dongshan substation in Hegang The field off-line April 2014 Research Institute, Wuhan on OVT Field off-line Field off-line 6 May 2014–9 May 2014 City of Heilongjiang Province accuracy for OVT 2014 City of Heilongjiang Province tests fortests OVT 6 May 2014–9 May Dongshan substation in Hegang The field off-line accuracy Field off-line On-line Dongshan substation in Hegang On-line Dongshan substation in Hegang The online operation 2014 tests for OVTof OVT 5 August 2014–2016 The online operation 5 August 2014–2016 City of Heilongjiang Province operation City of Heilongjiang Province of OVT operation City of Heilongjiang Province On-line Dongshan substation in Hegang 5 August 2014–2016 The online operation of OVT operation City of Heilongjiang Province Three Cases Three Cases
Sensors 2016, 16, 1303 Sensors 2016, 16, 1303
4 of 9 4 of 10
Sensors 2016, 16, 1303
4 of 10
Sensors 2016, 16, 1303 Sensors 2016, 16, 1303
4 of 10 4 of 10
Figure test in the laboratory. Figure 3.3.The The temperature testin inthe thelaboratory. laboratory. Figure3. The temperature temperature test Figure 3. The temperature test in the laboratory. Figure 3. The temperature test in the laboratory.
Figure4.4.The The adjacent adjacent phase effect test in the laboratory. Figure phase effecttest testin inthe the laboratory. Figure The effect Figure 4. Theadjacent adjacentphase phase effect laboratory. Figure 4. The adjacent phase effect test in the laboratory.
Figure5.5.Field Field off-line off-line accuracy test. Figure accuracy test. Figure accuracytest. test. Figure5.5.Field Field off-line off-line accuracy Figure 5. Field off-line accuracy test.
Figure 6. Site installation of the OVT. Figure6.6.Site Site installation of the OVT. Figure Figure 6. Site installation installationof ofthe theOVT. OVT.
Figure 6. Site installation of the OVT.
Sensors 2016, 16, 1303
5 of 10
Sensors 16, 1303 Sensors 2016,2016, 16, 1303
5 of 95 of 10
Figure 7. OVT on-line error monitoring system.
For the laboratory and field off-line accuracy tests, the calibration principle and circuit are shown Figure 7.consists OVT on-line monitoring system. in Figure 8. The calibration Figure system oferror a standard voltage 7. OVT on-line error monitoring system. transformer, high precision acquisition module, pulse forming circuit, industrial control computer, and control and calculation For laboratory the laboratory and field off-lineaccuracy accuracy tests, the calibration principle circuit are shown ForThe the and field off-line tests, theless calibration principle and circuit areless shown software. uncertainty of the calibration system was than 0.05 for and ratio error and than in Figure 8. The calibration system consists of a standard voltage transformer, high precision acquisition in Figure 8. The calibration system consists of a standard voltage transformer, high precision 2 min for phase error. For on-line operation, the accuracy results of OVT were obtained by comparing module, pulse forming circuit, industrial control computer, and control and calculation software. module, pulse forming circuit, industrial control computer, and control and calculation theacquisition output of OVT with the traditional voltage transformer. level of2traditional voltage The uncertainty of the calibration system was less than 0.05 for The ratio accuracy error and less than min for phase software. The uncertainty of the calibration system was less than 0.05 for ratio error and less than(2) transformer wason-line class operation, 0.2. The ratio error εresults and the phase error ϕe can definedthe byoutput Equations error. For the accuracy of OVT were obtained by be comparing of 2 min for phase error. For on-line operation, the accuracy results of OVT were obtained by comparing and (3),OVT respectively: with the traditional voltage transformer. The accuracy level of traditional voltage transformer was the output ofThe OVT with theε and traditional voltage The accuracy level of traditional voltage class 0.2. ratio error the phase error ϕtransformer. e can be defined by Equations (2) and (3), respectively: kU U s p transformer was class 0.2. The ratio error error ϕe can be defined by Equations(2) (2) 100% (%)ε and the phase kU − U U s p p and (3), respectively: ε (%) = × 100% (2) e U ps p (3) kU U (2) (%)ϕ = ϕs − pϕ 100% (3) e U where Us is the average value of the r.m.s. value insp 10 pperiods from the output of the OVT; k is the e of ins primary where Us is the average of the r.m.s.value value 10 periods from theaccording output of the k is the(3) rated transformation ratio; Uvalue p is the r.m.s. voltage to OVT; the output of the p rated transformation ratio; U is the r.m.s. value of primary voltage according to the output of the standard voltage transformer. ϕspis the initial phase of Us, and ϕp is the initial phase of Up. where Us is the average value ofϕsthe r.m.s. value inof10Usperiods the output the OVT; k is the standard voltage transformer. is the initial phase , and ϕp from is the initial phase of of U p. rated transformation ratio; Up is the r.m.s. value of primary voltage according to the output of the standard voltage transformer. ϕs is the initial phase of Us, and ϕp is the initial phase of Up.
Figure 8. The calibration scheme of OVT in the lab and off-line in the field.
Figure 8. The calibration scheme of OVT in the lab and off-line in the field. 4. Comparison of the OVT Error Characteristics under Three Conditions
Figure 8. The calibration scheme of OVT in the lab and off-line in the field. 4. Comparison of the OVT Error Characteristics under Three Conditions
In order to ensure the comparability of data, the testing method and the testing equipment in the laboratory exactly the Characteristics same as of those used in testing the off-line field tests. For online order to ensure the comparability data, the method and the testingoperation, equipment in 4. In Comparison of were the OVT Error under Three Conditions the output of a exactly traditional wasin compared as standard data with the same testing the laboratory were thevoltage same transformer as those used the off-line field tests. For online operation, the
In order to ensure the comparability of data, the testing method and the testing equipment in output of a traditional voltage transformer was compared as standard data with the same testing the laboratory were exactly the same as those used in the off-line field tests. For online operation, the method as in the laboratory. After the OVT prototypes were calibrated for the first time, no output of a traditional voltage transformer was compared as standard data with the same testing adjustment or compensation for error were done during the experiments. method as in the laboratory. After the OVT prototypes were calibrated for the first time, no adjustment or compensation for error were done during the experiments.
Sensors 2016, 16, 1303
6 of 9
Sensors 2016, 16, 1303
6 of 10
method as in theoflaboratory. After the OVT prototypes were calibrated for the first time, no adjustment 4.1. Comparison Error Characteristics of OVT in the Laboratory Sensors 2016, 16, 1303 6 of 10 or compensation for error were done during the experiments. The basic accuracy of the OVT was calibrated and tested under the rated voltage, and then the 4.1. Comparison of Error Characteristics OVT in temperature temperature characteristics were tested. For the 4.1. Comparison of Error Characteristics ofofOVT in the Laboratory Laboratorycharacteristic test, three OVT prototypes were placed in a temperature controlled cavity (Figure 3) andunder subjected to the ratedand voltage at room The basic accuracy OVTwas wascalibrated calibrated and thethe rated voltage, then then the The basic accuracy of of thethe OVT andtested tested under rated voltage, and the temperature (19.5 °C). The temperature changes with timecharacteristic are shown test, in Figure 9. The prototypes temperature characteristics were tested.For For the temperature temperature three OVT prototypes temperature characteristics were tested. the characteristic test, three OVT prototypes werewere warmed up 65 °C at a rate of 10 cavity °C every 20 min and kept atto65 for voltage 4 h, then cooled to placed in to a temperature controlled (Figure 3) and subjected the°C rated at room were placed in a temperature controlled cavity (Figure 3) and subjected to the rated voltage at room temperature (19.5 °C). The temperature changes with time are shown in Figure 9. The prototypes −40 °C with the same rate and kept at −40 °C for 4 h, and then warmed up to room temperature again temperature (19.5 ◦ C). The temperature changes with time are shown in Figure 9. The prototypes were warmed to kept 65 °Catatroom a ratetemperature of 10 °C everyfor 204min andtemperature kept at 65 °C characteristic for 4 h, then cooled with were the same rate◦up and h. The of thetoOVTs warmed upwith to 65 C at arate rateand of kept 10 ◦ C every 20 min andthen keptwarmed at 65 ◦ C for 4 h, then cooled to −40 ◦ C −40 °C the same at −40 °C for 4 h, and up to room temperature again was obtained in the temperature range from −40 °C to +65 °C, and phase B OVT testing results ◦ with with the same raterate andand kept at at −room 40 Ctemperature for 4 h, andfor then up to room temperature the same kept 4 h.warmed The temperature characteristic of theagain OVTs with are shown in Figure 10, where the ratio error and the phase error were calculated by the same rate and in kept room temperature for−40 4 h.°C The characteristic of theresults OVTs was was obtained theattemperature range from to temperature +65 °C, and phase B OVT testing Equations (2) and (3). ◦ ◦ are in shown in Figure 10, where thephase phaseB error were calculated obtained the temperature range fromthe −40ratio C toerror +65 and C, and OVT testing results areby shown Equations (2) and (3).ratio error and the phase error were calculated by Equations (2) and (3). in Figure 10, where the
Figure9.9.Temperature Temperature curve curvewith withtime. time. Figure time.
Figure 10. Temperature characteristic of OVTs in the laboratory.
Figure 10. OVTs in laboratory. Figure 10. Temperature Temperature laboratory. From Figure 10, the ratio error of the characteristic OVT is withinof−0.58% to the +2.43% in the temperature range from −40 °C to +65 °C, which is beyond the error limit value of level 0.2, whereas the phase error was From Figure 10, the ratio error of the OVT is within −0.58% to +2.43% in the temperature range withinFigure ±10’. 10, From the ratio error of the OVT is within −0.58% to +2.43% in the temperature range from −40 °C +65 three °C, which beyond the error limit value ofatlevel 0.2, whereas phase error was phaseisvoltage transformers are installed a certain distance, the adjacent phases ◦ Cto from −40When to the +65 ◦ C, which is beyond the error limit value of level 0.2, whereas the phase error was may interfere with each other. In order to investigate the extent of their mutual influence, three phase within ±10’. within ±10’. OVTs were set at 1phase m spacing intervals. The error are of the middle phase B was tested whenadjacent the voltage When the three voltage transformers installed at a certain distance, phases When the three phase voltage transformers are installed at a certain distance, adjacent phases of phase A and C was applied at zero volts and the rated voltage, respectively. The testingphase may interfere with each other. In order to investigate the extent of their mutual influence, three may interfere eachinother. In 11, order to investigate the extent of phase their mutualwere influence, threebyphase arewith shown Figure where ratiooferror and the OVTsresults were set at 1 m spacing intervals. Thethe error the middle phase B error was tested calculated when the voltage OVTsEquations were set (2) at 1and m (3). spacing intervals. The error of the middle phase B was tested when the voltage
of phase A and C was applied at zero volts and the rated voltage, respectively. The testing of phase A and C was applied at zero volts and the rated voltage, respectively. The testing results are results are shown in Figure 11, where the ratio error and the phase error were calculated by shown in Figure 11, where the ratio error and the phase error were calculated by Equations (2) and (3). Equations (2) and (3).
Sensors 2016, 16, 1303 Sensors 2016, 16, 1303
7 of 9
7 of 10
Sensors 2016, 16, 1303
7 of 10
(a)
(b)
(a)adjacent phases on OVT. (a) Influence of adjacent phases (b) Figure Influence ratio error; Figure 11.11. Influence ofofadjacent phases on OVT. (a) Influence of adjacent phasesononOVT’s OVT’s ratio error; (b) Influence of adjacent phases on OVT’s phase error. 11. Influence of adjacent phases on phase OVT. (a) Influence of adjacent phases on OVT’s ratio error; (b)Figure Influence of adjacent phases on OVT’s error. (b) Influence of adjacent phases on OVT’s phase error.
From Figure 11, it is shown that the influence of the adjacent phase on OVT is small; the From Figure it is shown that the influence of the adjacent phase onless OVT is small; additional additional ratio11, error than 0.02% and influence the additional phase error phase is than 1’. isthe From Figure 11, is it less is shown that the of the adjacent on OVT small; the ratio error is less than 0.02% and the additional phase error is less than 1’. additional ratio error is less than 0.02% and the additional phase error is less than 1’. 4.2. Comparison of Error Characteristics of OVT under the Three Conditions 4.2. Comparison of Error Characteristics of OVT under the Three Conditions 4.2. Comparison of Error Characteristics of OVT underand the Three Conditions After testing the temperature characteristics adjacent phase influence in the laboratory, the After testing the temperature characteristics and adjacentThe phase influence in the laboratory, the OVT prototypes were transported to the Hegang substation. accuracy was before being After testing the temperature characteristics and adjacent phase influence in tested the laboratory, the OVT prototypes were transported to the Hegang substation. The accuracy was tested before being put put into operation. The error curves in thesubstation. laboratory,The in off-line state the field and being in the OVT prototypes were transported to of theOVT Hegang accuracy wasintested before into operation. The error of OVT in theinlaboratory, inthe off-line state in theinfield and thein on-line are shown in Figure 12. order to exclude effect of temperature, the in tests inon-line the put into state operation. The curves error curves of In OVT the laboratory, in off-line state the field and the state are shown in Figure 12. In order to exclude the effect of temperature, the tests in the laboratory laboratory and the field off-line tests were carried out at ambient temperature (20 °C). The online on-line state are shown in Figure 12. In order to exclude the effect of temperature, the tests in the operation wastests collected during August to 8 August 2014, when was to and the fielddata off-line were carried out atcarried ambient (20 ◦the C).temperature The online data laboratory and the field off-line tests5were outtemperature at ambient temperature (20 °C).operation The close online ◦ 20 °C. In Figure 12, the ratio error and the phase error were calculated by Equations (2) and (3). was collecteddata during August during to 8 August 2014,towhen the temperature close to 20 C. Inclose Figure operation was 5collected 5 August 8 August 2014, when was the temperature was to 12, the20ratio error and12, the phase calculated by Equations (2) and °C. In Figure the ratio error error were and the phase error were calculated by (3). Equations (2) and (3).
(a)
(b)
(a) Figure 12. Error comparison of OVT under the three conditions of lab, in the(b) field off-line and during on-line operation. (a) The ratio error comparison of OVT; (b) The phase error comparison of OVT. Figure Errorcomparison comparisonof of OVT OVT under under the field off-line and during Figure 12.12.Error the three threeconditions conditionsofoflab, lab,ininthe the field off-line and during on-line operation. (a) The ratio error comparison of OVT; (b) The phase error comparison of OVT. on-line operation. (a) The ratio error comparison of OVT; (b) The phase error comparison of OVT. From Figure 12, it can be seen that the errors of OVT in the laboratory and off-line in the field wereFrom nearly consistent. Thebeerror between on-line operation and and the laboratory is 0.05%, Figure 12, it can seendifference that the errors of OVT in the laboratory off-line in the field as the testing results of on-line operation were compared with a traditional voltage transformer. This From Figure 12, it can be seen that the errors of OVT in the laboratory and off-line in the field were nearly consistent. The error difference between on-line operation and the laboratory is 0.05%, indicated that the effect of the transportation process and the external electric field caused by field were nearly consistent. The error difference between on-line operation and the laboratory is 0.05%, as the testing results of on-line operation were compared with a traditional voltage transformer. This as wiring the the OVT could neglected. theindicated testingonresults ofeffect on-line operation were compared with traditional voltage This that ofbe the transportation process and theaexternal electric field transformer. caused by field indicated that effect ofbe the transportation process and the external electric field caused by field wiring on thethe OVT could neglected. 4.3. Error Characteristics in On-Site Operation wiring on the OVT could of beOVT neglected. 4.3. Error of OVT On-Site Operationin August 2014, then the data from September 2014 The Characteristics OVT prototypes werein put into operation 4.3.toError Characteristics of OVT in On-Site Operation October 2015 were selected as analysis object. in The operation the twofrom summer and winter The OVT prototypes were put into operation August 2014,during then the data September 2014 seasons which display the greatest temperature difference, were used to illustrate the operational The OVT2015 prototypes were put into operation in August 2014, then the the two datasummer from September 2014 to October were selected as analysis object. The operation during and winter characteristics of theselected OVTs. Theanalysis data shown Figure 13 were correspond toillustrate August 2015 when the which display the greatest temperature used tothe the operational to seasons October 2015 were as object.indifference, The operation during two summer and winter temperatures ranged from 17The °C data totemperature 38 shown °C and in December 2014, when theto ranged from characteristics of the OVTs. Figure 13 correspond August 2015 when the seasons which display the greatest difference, were used totemperature illustrate the operational temperatures of ranged from 17The °C to 38 °C and December when the temperature ranged fromthe characteristics the OVTs. data shown in Figure 2014, 13 correspond to August 2015 when temperatures ranged from 17 ◦ C to 38 ◦ C and December 2014, when the temperature ranged from
Sensors 2016, 16, 1303
8 of 9
Sensors 2016, 16, 1303
8 of 10
◦ −20 ◦ C −20 to − InIn Figure ratioerror errorand and phase calculated °C1 toC, −1respectively. °C, respectively. Figure10, 10, the the ratio thethe phase errorerror were were calculated using using Equations and (3). Equations (2) and(2)(3).
Figure Error comparison of of OVT andand winter. Figure 13. 13. Error comparison OVTininsummer summer winter.
From Figure 13, it can be seen that the ratio error of OVT was affected by the temperature. During
From Figure it canwhich be seen thatthe the ratio temperature error of OVT was affected byerror the range temperature. summer and13, winter, display biggest difference, the ratio was from During summer−0.25% and winter, display the was biggest temperature difference, the ratio errorwas range was from to +0.3%,which and the phase error within ± 4'. In summer, the ratio error difference 0.55%, is higher that in winter. −0.25%which to +0.3%, andthan the phase error was within ± 40 . In summer, the ratio error difference was 0.55%, Compared with in thewinter. temperature characteristics in the laboratory, the temperature characteristics which is higher than that in the field are much better. When the temperature gradient is different, the thermal stress formed in Compared with the temperature characteristics in the laboratory, the temperature characteristics the sensing crystal has a great difference [16,17]. The greater the temperature gradient is, the greater in the field are much better. When the temperature gradient thecharacteristics thermal stress formed in the thermal stress is, and the greater the additional error is. is Fordifferent, temperature in the the sensing crystal a great difference [16,17]. greater temperature gradientgradient is, the greater laboratory thehas temperature variation rate was 10 The degrees everythe 20 min, but the temperature in thestress actual is, runtime environment faradditional less. Therefore, theis. on-line operation characteristics of the in the the thermal and the greater is the error For temperature characteristics OVTs were better than the laboratory temperature testing results. laboratory the temperature variation rate was 10 degrees every 20 min, but the temperature gradient in the actual runtime environment is far less. Therefore, the on-line operation characteristics of the 5. Discussion OVTs were better than the laboratory temperature testing results. 5.1. Effect of the Adjacent Phase
5. Discussion
The effect of the adjacent phase on a 110 kV OVT with reasonable structure was small. The additional error was within 0.02% and could be ignored for the OVTs of level 0.2.
5.1. Effect of the Adjacent Phase
Differences of theadjacent Laboratory phase and On-Site TestkV Results The5.2.effect of the on Field a 110 OVT with reasonable structure was small. The additional within 0.02% could were be ignored for the OVTs levelobtained 0.2. in the The error testingwas results of OVT in theand laboratory in good consistency withofthose field, and the additional error was less than 0.05%. On the one hand, the electric field in the OVT
5.2. Differences ofconcentrated, the Laboratory andaccuracy On-Sitewas Field Results sensors is so the notTest sensitive to the wire connection mode; on the other hand, the OVTs were not affected by the vibration of the transportation process.
The testing results of OVT in the laboratory were in good consistency with those obtained in the field, and additional error was less than 0.05%. On the one hand, the electric field in the OVT 5.3.the Differences of Temperature Characteristics sensors is concentrated, theofaccuracy was not sensitive to the connection mode; range on the other The ratio error so range OVT in the laboratory was −0.58% to wire +2.43% in the temperature hand, the OVTs were not affected by the vibration of the transportation process. from −40 °C to 65 °C, and −0.25% to +0.3% during on-site operation in a temperature range from −20 °C to 40 °C. The temperature characteristics of OVT could not meet the 0.2 level accuracy
5.3. Differences of Temperature requirements. However,Characteristics the temperature characteristics during on-site operation were better than those in the laboratory, as the temperature gradient was larger for the laboratory tests
Thethan ratio error range of OVT in the laboratory was −0.58% to +2.43% in the temperature range during on-site operation. The temperature characteristics of OVTs were dependent on the ◦ C to 65 ◦ C, and −0.25% to +0.3% during on-site operation in a temperature range from from −40 temperature gradient. −20 ◦ C to 40 ◦ C. The temperature characteristics of OVT could not meet the 0.2 level accuracy requirements. However, the temperature characteristics during on-site operation were better than those in the laboratory, as the temperature gradient was larger for the laboratory tests than during on-site operation. The temperature characteristics of OVTs were dependent on the temperature gradient. Acknowledgments: We would like to express our gratitude for financial support from the National Natural Science Foundation of China under Grant 51307066 and State Grid Corporation of science and technology foundation under Grant JL71-16-005. Author Contributions: X.X., Y.X. and M.L. conceived and designed the experiments; H.H. and Q.X. performed the experiments; X.X. and H.H analyzed the data; X.X. wrote the paper.
Sensors 2016, 16, 1303
9 of 9
Conflicts of Interest: The authors declare no conflict of interest.
Abbreviations OVT MDPI DOAJ TLA LD
Optical Voltage Transformer Multidisciplinary Digital Publishing Institute Directory of Open Access Journals Three letter acronym linear dichroism
References 1.
2.
3. 4.
5.
6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.
Rahmatian, F.; Ortega, A. Applications of optical current and voltage sensors in high-voltage systems. In Proceedings of the 2006 IEEE/PES Transmission and Distribution Conference and Exposition: Latin America, Caracas, Venezuela, 15–18 August 2006; pp. 1–4. Marchese, S.V.; Bohnert, K.; Wildermuth, S.; van Mechelen, J.L.M.; Steiger, O.; Rodoni, L.C.; Eriksson, G.; Czyzewski, J. Electro-optic voltage sensor based on BGO for air-insulated high voltage substations. In Proceedings of the 2013 IEEE Photonics Conference, Bellevue, WA, USA, 8–12 September 2013; pp. 608–609. Li, H.; Cui, J.; Shi, M. Application of optical voltage transformer in digital substation. Chin. Electr. Power Technol. Ed. 2012, 11, 460–462. Sanders, G.A.; Blake, J.N.; Rose, A.H.; Rahmatian, F.; Herdman, C. Commercialization of fiber-optic current and voltage sensors at NxtPhase. In Proceedings of the 15th Optical Fiber Sensors Conference, Portland, OR, USA, 10 May 2002; pp. 31–34. Rahmatian, F.; Chavez, P.P.; Jaeger, N.A. 138 kV and 345 kV wide-band SF 6-free optical voltage transducers. In Proceedings of the Power Engineering Society Winter Meeting, New York, NY, USA, 27–31 January 2002; pp. 1472–1477. Li, J.; Zheng, Y.; Gu, S.; Xu, L. Application of Electronic Instrument Transformer in Digital Substation. Autom. Electr. Power Syst. 2007, 31, 94–98. Liu, B.; Ye, G.; Guo, K.; Tong, Y.; Hu, B.; Wan, G. Quality Test and Problem Analysis of Electronic Transformer. High Voltage Eng. 2012, 38, 2972–2980. Ding, T.; Xu, E.; Wu, H.; Wang, X.; Yang, N.; Liu, Z.; Jiao, J. Measurement and Problem’s Analysis of the On-Site Error for Electronic Transformer. Electr. Meas. Instrum. 2011, 48, 36–39. Lee, K.S. Electrooptic Voltage Sensor: Birefringence Effects and Compensation Methods. Appl. Opt. 1990, 29, 4453–4461. [CrossRef] [PubMed] Wang, H. Research of Capacitor Divider Optical Voltage Transducer. Ph.D. Thesis, Harbin Institute of Technology, Harbin, China, 2010. Arpaia, P.; Daponte, P.; Grimaldi, D.; Michaeli, L. ANN-based error reduction for experimentally modeled sensors. IEEE Trans. Instrum. Meas. 2002, 51, 23–30. [CrossRef] Filippov, V.N.; Starodumov, A.N.; Minkovich, V.P.; Lecona, F.G.P. Fiber sensor for simultaneous measurement of voltage and temperature. IEEE Photonics Technol. Lett. 2000, 12, 1543–1545. [CrossRef] International Electrotechnical Commission. Instrument Transformers—Part 7: Electronic Voltage Transformers; International Electrotechnical Commission: Geneva, Switzerland, 1999. Wu, W.; Xu, Y.; Xiao, X.; Hu, H. Research on Proximity Effect in Measuring Error of Active Electronic Voltage Transformers. IEEE Trans. Instrum. Meas. 2016, 65, 78–87. [CrossRef] Chen, G.; Liao, L.; Hao, W. Fundermentials of Crystal Physics, 2nd ed.; Science Press: Beijing, China, 2007. Xiao, X.; Xu, Y.; Dong, Z. Thermodynamic modeling and analysis of an optical electric-field sensor. Sensors 2015, 15, 7125–7135. [CrossRef] [PubMed] Xiao, X.; Xu, Y.; Xu, K.; Ye, M. Influence of temperature and stress field on optical voltage sensor. In Proceedings of the 2009 Symposium on Photonics and Optoelectronics, Wuhan, China, 14–16 August 2009; pp. 1–4. © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
