Electric Power Quality Monitoring and Analysis at a Tri-generation Plant under Development

Electric Power Quality Monitoring and Analysis at a Tri-generation Plant under Development IOANA PISICĂ, LAURENŢIU CONSTANTIN LIPAN, PETRU POSTOLACHE, CORNEL TOADER Department of Power Systems University Politehnica of Bucharest Splaiul Independentei 1 ROMANIA ioanapisica@gmail.com http://el.poweng.pub.ro GHEORGHE COMĂNESCU Department of Electric Power Generation and Use University Politehnica of Bucharest Splaiul Independentei 1 ROMANIA gheorghe.comanescu@gmail.com http://cceei.energ.pub.ro DAN CONSTANTIN DAŞU OMV Petrom SA ROMANIA dan_dasu@yahoo.com Abstract: - The aim of the study presented within this paper is the monitoring and analysis of electric power quality supplied to a tri-generation plant. The measurements were made before commissioning the plant, while investment work was still in progress. The main objective of the study was to certify that the voltage values are kept within the limits imposed by contract between the user and the supplier. In order to accomplish this objective, the electrical power supplied at the location of interest was monitored at two measurement points, one in the 10 kv grid, at the interface with the supply network and one in the 20 kv substation owned by the beneficiary. The analysis was based on a complete set of parameters that were monitored alongside voltage values: power factors, harmonics, flicker, voltage and current unbalances and others. Key-Words: - power quality, voltage measurement, power system harmonics, power system monitoring 1 Introduction The site is supplied from a 10 kv substation through two underground power lines. Another substation, owned by the beneficiary, is connected to the 10 kv substation through a short underground power line. This substation has only two bays: one dedicated to the connection with the supplier substation and one equipped with a 10/20 kv transformer. Both substations are placed at the same location. The transformer supplies the 20kV substation owned by the beneficiary and has the following characteristics: Rated power 2500 kva; Turns ratio 20 kv şi ± 2,5 %/ 10000 V; Connexion type Yyn0; Short circuit voltage 7 %. The supplier electricity billing meter is located on the connection between the two substations. The following measurement points were established, where class A power quality monitoring devices were installed: - one measurement point at 10 kv in the secondary winding of the potential transformer (turn ratio 10000 100 V) connected to the bus bar from beneficiaryowned substation and in the secondary winding of the current transformer (turn ratio 75/5 A) from the 10/20 kv transformer; - one measurement point in the secondary winding of the potential transformer (turn ratio 20000 100 V) from the bus bar of the 20 kv beneficiary-owned substation and in the secondary winding of the current transformer (turn ratio 150/1 A) from the 20 kv side of the 10/20 kv ISSN: 1792-5924 / ISSN: 1792-5940 446 ISBN: 978-960-474-27-0

transformer. For space reasons, only data collected from the first measurement point (10 kv) are presented in this paper. 2 Measured Data Analysis Power quality has been a puzzling problem over the last years, mainly due to the use of power electronics, but also because of sensitive equipment that is being used. Power quality distortion phenomena can be subdivided into the following categories [1]: - harmonic distortion - transients - voltage dips - voltage swells - flicker - voltage asymmetry (unbalance) - frequency variations. 2.1 Analysis of voltage values Voltage values measured throughout a working day are presented in Fig. 1, in order to check whether voltage limits are violated. Fig.1. Daily variation of line voltage values - 09.02.2010. The chart in Fig. 1 shows that two of the line voltages are quasi-equal, while the third is always lower than the others. Therefore, a voltage unbalance can be observed. A full analysis for the 9 days of measurements reveals a maximum difference between the phase voltage values of 61 V and of 10 V between the highest and lowest line voltage values. These differences are lower than 1% of the rated voltage. The first conclusion is that the voltage unbalance between the three phases is placed within acceptable limits. In order to check the extent to which the voltage values are kept within admissible limits set by contract, the chart in Fig. 2 presents the variation of the line voltage with the highest values. The variation is expressed in percents of rated voltage (10000 V). Fig.2. Daily voltage variation, Uca [%] The chart shows that line voltages (for the sample day) do not exceed 106% from the rated voltage, and therefore the voltage values are limited within the admissible strip of U n ± 10 %. A full analysis based on data recorded for 9 days reveals the following minimum and maximum values: o For phase voltages - minimum value: 576,58 V, respectively 99,71 % of V n ; - maximum value: 618,18 V, respectively 106, % of V n ; o For line voltages - minimum value 9957,40 V, respectively 99,57 % of U n ; - maximum value: 1069,4 V, respectively 106,9 % of U n. The second conclusion that can be drawn is that the supplied voltages during the entire time frame of measurements are maintained within 99.5% and +107% of the rated voltage. These values define a strip between -0.5% and +7%. A graphical representation of all line voltages for the full time frame of measurements would become unreadable. In order to obtain a chart easy to follow, only the variation of voltage Uca is presented in figure, for the full monitored period, with 1296 sets of measurements. Fig.. Variation of line voltage Uca ISSN: 1792-5924 / ISSN: 1792-5940 447 ISBN: 978-960-474-27-0

Fig. highlights the daily voltage peaks, which are observed during night hours (between 2 and 4 AM) and the minimum voltage values, recorded during evening hours (19-21). The currents on the three phases have similar values, showing a small current unbalance. The currents on each phase are presented in figure 8, for the full monitored time frame. 2.2 Anlysis of other collected data The equipment used for power quality monitoring during this study also records: currents on each phase, active, reactive and apparent power, power factor for each phase, total current harmonic distortion factor (THD-I), total voltage harmonic distortion factor (THD- U), current and voltage harmonics, flicker and unbalances [2-5]. Taking into consideration the large amount of measured and computed values, some conclusions regarding the supplied power quality are presented in the following. The first analysis concerns the powers absorbed by the beneficiary equipment at the measurement point.. Figures 4 through 7 present, respectively, the variations of apparent powers on each phase, total active, reactive and apparent powers and active (P) and reactive (Q) powers on each phase, during the full monitored period. Fig. 6. Active powers P [kw] variations on each phase. Fig. 7. Variation of reactive powers Q [kvar] on each phase. Fig. 4. Apparent powers on each phase. Fig. 8. Variation of currents at 10 kv busbar. Fig. 9 presents the variation of currents during a working day. Fig. 5. Total active, reactive and apparent powers. During the period of study, the energy consumption is low, and therefore the reactive power values are low. Most of the time, the reactive powers are injected into the supplier network, suggesting the characteristics of a capacitive load. Fig. 9. Variation of currents during a working day ISSN: 1792-5924 / ISSN: 1792-5940 448 ISBN: 978-960-474-27-0

Data presented in figures 10 and 11 reveal low flicker levels, within acceptable limits imposed by norms. Higher voltage variations can be observed on phase b, alongside higher flicker levels. Pst and Plt curves at the point of measurement have similar trends. Busbar voltages are quasi-sinusoidal, having a total harmonic distortion factor lower than the maximum acceptable limit (THD acc = 6%). The total harmonic distortion factor was during measurements between 1,5 and.7% ( Fig. 12). generated by connecting loads that produce distortions, like power electronics. The average values are limited to 0-40%, which represent also high values. Fig. 1.Current total harmonic distortion factor (THD I). Fig.10 Short-term flicker levels (Pst). a. Fig. 11. Long-term flicker levels (Plt). b. Fig. 14. Current harmonic spectra for the three phases. c. Fig. 12. Voltage total harmonic distortion factor measured on phases a, b, c. The current total harmonic distortion factor (THD I) is presented in Fig. 1, showing high values that highlight the existence of important current harmonic components. The current harmonic spectra for the three phases are presented in Fig. 14,a,b,c. Harmonic components of order 5 and 7 reach up to 70-80%. These high values can be High values of harmonics of order 5 and 7 and therefore high values of total harmonic distortion factor (THD-I) indicate the existence of distorting loads. The beneficiary should consider installing filters for current waveforms at the connection point when the consumption levels increase. The power factor variation for each phase during the measurements time frame, when the monitoring device was in service, is presented in Fig. 15. As it can be ISSN: 1792-5924 / ISSN: 1792-5940 449 ISBN: 978-960-474-27-0

observed, the chart highlights reactive power flows from the installations at the beneficiary location towards the supplier network. In general, the power factor has values close to 1. Fig. 15. Power factor variation for phases a, b and c. Fig. 16 shows the variation of the voltage unbalance factor. It can be observed that the negative unbalance factor: U k s (1) U where U is the negative sequence voltage, and U + is the positive sequence voltage, has values lower than the maximum admissible negative unbalance factor (<2%). The variation of current unbalances is presented in Fig 17. The negative unbalance factor was computed by using (2): I I k s (2) where I is the negative sequence current component, and I + positive sequence current component. Conclusion.1 Regarding voltage levels During the entire monitoring time, the voltages supplied at the 10 kv busbar were maintained within the limits 99.5 % 107 % of rated voltage. This leads to a voltage strip of -0.5% - +7%. The supplied voltage levels supplied were therefore conforming to the values stipulated by contract. The voltage variations at the 20 kv busbar are shifted up with almost one percent, meaning that the voltage strip was between 100% and 109%. This is most probably due to reactive power generated by underground cables while the active energy consumption was low, as the site was under construction. As all loads of the beneficiary at the point of interest are connected to the 20 kv bus bar, it would be advisable to change the plot of the 10/20 kv transformer to -2.5%, for the moment..2 Regarding voltage unbalances Two types of voltage unbalances were studies. First, a difference between the lowest and the highest voltage values were computed and resulted into values lower than 1% for both line and phase voltages. These values were determined for the 10 kv and the 20 kv bus bars. Secondly, the unbalance was computed by the monitoring device, based on negative and positive sequence components. The resulted values are between 1.% and 1.5%. The conclusion is that the operating state is normal and no further action is required. Fig. 16. Negative voltage unbalance factor (k s ).. Regarding the voltage waveform The analysis of voltage harmonic components led to similar results in both cases (10kV and 20 kv). The total harmonic distortion factor for voltage, THD-U, was found have values placed between 1.5% and.5 %, below the admissible value of 6%. From this point of view, we can conclude that the power quality is satisfactory..4 Regarding the flicker level Flicker levels are a measure of fast voltage fluctuations that can produce harmful effects upon devices, especially artificial lighting. Data analysis of recorded values for the two measurement points shows low flicker levels, of around 0.%- 0.5 %. The monitored values are below the admissible level, of 0.9%. The power quality is therefore satisfactory from this point of view. Fig. 17. Negative current unbalance factor (k s ). ISSN: 1792-5924 / ISSN: 1792-5940 450 ISBN: 978-960-474-27-0

.5 Regarding currents Currents on the three phases were monitored at both points of measurement. The measured current values are low (14-18 A) at 10 kv and half of these at 20 kv. However, this is irrelevant, as it is only the load at this stage. The computed power quality indicator was the unbalance, showing whether the phases are equally loaded or not. The unbalance factor, computed as a fraction between the negative and positive sequence components, resulted in values between % and 9%, values larger than the ones for voltages. These values are high, but a perfect balancing of the currents requires a perfect distribution of load over the phases, which is impossible to obtain in practice. Furthermore, the installations are already put in place and it would be very hard to re-arrange them. The most important aspect regarding currents is the total harmonic distortion factor, which reaches frequently values up to 60%-70%. In other words, the current waveform is very distorted. The harmonic components of highest values have the orders 5 and 7. This problem should be solved sooner or later, as it leads to unwanted effects in the network. A solution at first glance would be to install a harmonic filter at the 20 kv substation. It is possible for the supplier to bill the distortions injected into the network..6 Regarding powers A first important aspect concerns the reactive power. In comparison with the active power, the values of reactive powers are low. This leads to a power factor near to the value of 1, which is desirable. On the other hand, there is a frequent excess of reactive power at the 20 kv substation that is injected into the supplier network. This is the cause for higher voltage levels at the 20 kv substation, but also at the 10 kv one. A detailed analysis for implementing a reactive power management system including harmonic filters for 5 and 7 th order harmonics should be undertaken only after starting the operation at full capacity. It is possible that the problem will be resolved by self when all consumers and generators will be in service. References: [1] Baggini, Angelo: Handbook of power quality. West Sussex, John Wiley and Sons, 2008 [2]Stanescu, Carmen, Vatra, F., Poida, Ana, Postolache P., Power quality in romanian electricity market, 9th International Conference on Electric Power Quality and Utilisation, EPQU-2007, Barcelona, oct. 2007. []Chicco G., Postolache P., Toader C., Analysis of Three-Phase Systems With Neutral Under Distorted and Unbalanced Conditions in the Symmetrical Component-Based Framework, IEEE Transactions on Power Delivery, Volume 22, Issue 1, Jan. 2007, pag. 674 68. [4] Spain K. Strunz. Developing Benchmark Models for Studying the Integration of Distributed Energy Resources. Power Engineering Society General Meeting, June 2006. [5] IEEE 519-1992 Recommended Practices and Requirements for Harmonic Control in Electrical Power Systems. Biographies Ioana Pisică. Born in Bucharest (Romania), on March 5 th, 1984. Graduated from University Politehnica of Bucharest, Faculty of Power Engineering in 2007. She has begun her final year of Ph.D studies in Electrical Power Engineering. Present position: Teaching assistant in Electrical Engineering Politehnica University of Bucharest, Department of Electric Power Systems. Fields of activity include: Teaching (and research) Electrical Networks, Numerical Methods, Artificial Intelligence Techniques, Electronic Devices and SCADA systems in Power Engineering. Laurenţiu Constantin Lipan. Born in Hunedoara (Romania), on June 28, 1975. Graduated from the Politehnica University of Bucharest, Power Engineering Faculty in 2001. Ph.D in Electrical Power Field in 2009. Present position: Lecturer in Electrical Engineering Politehnica University of Bucharest, Department of Electric Power Systems. Fields of activity include: Teaching (and research) the Electricity Use in Industry, Management of Electricity, Energy Efficiency, Large Industrial Consumers Supply, Modern Electrical Technologies. He is currently carrying out Post-Doctoral research studies, and his contributions to this paper were financially supported by the POSDRU contract POSDRU/89/1.5/S/62557. Dan Constantin Daşu. Born on March 29, 1970. He graduated from University of Petrol and Gas, Ploiesti, specialization Automation and Industrial Informatics and is pursuing his Master degree in Advanced Automation. He is currently with OMV Petrom as head of operators department at the trigeneration site Petrom City, Bucharest. Gheorghe Comănescu. Born in Novaci (Romania), on June 9, 1948. Graduated from the Politehnica University of Bucharest, Power Engineering Faculty in 1971. Ph.D in Electrical Power Field in 1988. Present position: Professor in Electrical Engineering Politehnica University of Bucharest, Department of Energy Generation and Use. Didactic activities on Electrical Part of Power Plants and Electrical Substations. Research activities on Design, Operation and Economical Optimization of Electricity Distribution Networks and Electrical Switchgear for Power Plants and Substations. Petru Postolache. Born June 28, 1940 in Chisinau. Graduated from University Politehnica of Bucharest in 1966 and obtained his Ph.D in 1991. Member of IEEE, CIRED, ASRO, IRE, SIER and other scientific organizations. Current position: Professor in Electrical Engineering University Politehnica of Bucharest, Department of Electric Power Systems. His fields of interest include: Power Quality, Sustainable Development, Renewable Energy Sources, Energy Efficiency. Cornel Toader. Born July 17, 1945. Graduated from University Politehnica of Bucharest in 1968. Ph.D in Electrical Power Systems in 1989. Member of IEEE, ASRO, IRE, SIER and other scientific organizations. Current position: Professor in Electrical Engineering University Politehnica of Bucharest, Department of Electric Power Systems. His fields of interest include: Power Quality, Energy Efficiency, Energy Audits, Energy Management. ISSN: 1792-5924 / ISSN: 1792-5940 451 ISBN: 978-960-474-27-0