Voltage Quality Improvement In Distribution Networks
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Voltage Quality Improvement In Distribution Networks as PDF for free.
More details
- Words: 3,786
- Pages: 6
Voltage Quality Improvement in Distribution Networks A case study
Zúñiga Reyes, Cristián Andrés Dept. Power Quality Chilectra S.A. Santiago, Chile [email protected]
Abstract— This paper is presented by Chilectra, a subsidiary of the Endesa Group Spain and the biggest electric utility in Chile. It sets out a methodology for estimating and determining control parameters for the voltage quality to be delivered to its end customers. The methodology is able to estimate the voltage at each end customer location in the low voltage (LV) distribution network, using measurements of voltage, current and power factor in real operating conditions at the feeder substation bus bars in the medium voltage (MV) distribution network. The available information sources, the assumptions used for analysis and the partial results at each stage of the methodology are described. A case study shows the main results of the methodology: the number of end customers with voltage quality problems and the geographical location of these customers. Finally, it shows the improvement obtained as a percentage decrease in end customers subject to bad voltage quality when applying the methodology. Keywords-component; power quality; distribution network; power flow methods.
I.
voltage
quality;
INTRODUCTION
The legal requirements for voltage quality in Chile include restrictions over the voltage levels at the end customer connection points. The requirements consider the nominal voltage at the end customer connection points (high, medium or low voltage). The requirements also consider that the voltage must be within a determined range at least 95% of the time over any seven day measurement period. The legally allowed ranges are presented in Table I.
TABLE I.
LEGALLY ALLOWED VOLTAGE RANGES IN CHILE
Voltag e level
Nominal voltage (kV)
Allowed range as a percentage of the nominal voltage
HV MV LV
110 – 220 12 – 23 0.38 (*)
5% 6% 7.5%
(*) In Chilectra’s electrical system, the nominal voltage in LV network is 400 (V)
Further more, the legal requirements consider the voltage quality evaluation by using two methods: •
Specific distribution network point evaluation. This evaluation determines the individual voltage quality.
•
Statistical evaluation of a group of distribution network points. This evaluation determines the global voltage quality. The group of network points must be selected by statistical methods in order to ensure valid results for all network points.
Currently, the regulating authority evaluates the voltage quality only when it receives an end customer complaint. The regulating authority has yet to begin a systematical evaluation of individual voltage quality. Nevertheless, Chilectra has developed a simulation tool in order to estimate voltage at each end customer connection point. The main objective is to anticipate the voltage quality
problems of end customers. In addition, this simulation tool enables Chilectra to know the current state of voltage quality in Chilectra´s network and to anticipate the results of future regulating authority evaluations. II.
DESCRIPTION OF CHILECTRA’S ELECTRICAL SYSTEM
Chilectra is a utility that delivers the public service of electricity supply to most of Santiago (Chile’s capital). Next Table II presents the main characteristics of Chilectra’s electrical system. III.
DESCRIPTION OF CHILECTRA’S ELECTRICAL SYSTEM
Due to the high number of Chilectra’s end customers and the impossibility of evaluating the voltage quality in all of them, Chilectra has developed a simulation tool to detect voltage quality problems at end customer connection points. This simulation tool estimates voltage at each end customer connection point, by using the operating conditions in the substation bus bars of medium voltage feeders and by using the electrical and topological configuration of the MV and LV distribution networks. The simulation tool uses power flow methods in the MV and LV voltage networks. Figure 1 describes the main stages of the simulation tool. TABLE II.
MAIN CHARACTERISTICS OF THE CHILECTRA’S ELECTRICAL SYSTEM
Characteristic
HV network kilometers HV/MV substations MV network kilometers MV/LV substations LV network kilometers MV feeders Company transformers Private transformers (MV end customers) LV end customers Power flow in the MV distribution network SCADA measurements
Transfer of the MV power flow results to the LV distribution network Voltage, current and power factor in the MV side of distribution transformers
Execution of power flow in the MV network
MV network topology
Transference of MV power flow results to the LV network
No.
358 41 5,035 27,006 9,586 374 21,626 5,380 1,426,445 Power flow in the LV distribution network Voltage, current and power factor in LV side of distribution transformers
Execution of power flow in the LV network
LV network topology Equivalent model of the distribution transformers
Voltage estimation at each end customer connection point
Figure 1. Main stages of the simulation tool
The simulation tool passes through three main stages to achieve results. Each one of these three stages is described below: A. Execution of power flow in the MV distribution network: The input data for this stage are measurements of voltage, current and power factor on the header of each MV feeder. These measurements are taken every 15 minutes, from Chilectra’s SCADA system and show the real value at that present time. Other input data for this stage is the MV network topology. This information is used to simulate the power flow in the MV network. The available topological information of the MV network includes the geographical location of the MV distribution network and of the protection and switching equipment. The length, cross section and electrical characteristics of the MV distribution network conductors are also included. The information about the MV distribution network is maintained and updated daily in an Informix Database. The power flow in the MV distribution network uses a program developed in Chilectra. The program was developed by using the FORTRAN language and it works in the UNIX operating system. The program calculates voltage, current and power factor on the MV side of all the distribution transformers in the MV distribution network. This power flow method uses an algorithm of aggregated power and it assumes that the power distribution at each transformer is proportional to its nominal power. As an outcome of this stage of the process, we can get an estimation of the voltage, current and power factor on the MV side in each distribution transformer. B. Transfer of MV power flow results to the LV distribution network: This stage considers the transfer of the power flow results obtained on the MV side of each distribution transformer in relation to their corresponding values on their LV side. The input data for this stage are voltage, current and power factor on the MV side at each distribution transformer. These results are obtained from the previous stage. Other input data for this stage is an equivalent model for the distribution transformers. This model considers the nominal voltage, the complex series impedance, the tap changer available positions and the connected tap position for each distribution transformer. The information regarding the complex series impedance (magnitude and angle) is obtained from standard impedance tables of Chilectra´s distribution transformers. The information regarding the connected tap position for each transformer is obtained from Maintenance Unit databases and from measurement programs concerned with the verification of voltage quality problems. A MS-Access database is used to provide the results of this stage: the voltage, the current and the power factor on the LV side at each distribution transformer.
IV.
CASE STUDY
The operating conditions of Chilectra’s electrical system vary throughout the year. The performance of the loads connected to the electrical system of which Chilectra’s system is connected and the performance of the loads connected to Chilectra’s electrical system are both important factors. Consequently, the voltage at the substation bus bars of the MV feeders and the voltage at each end customer connection point vary throughout the year. The main factors that determine the voltage level at end customer connection points are: •
Voltage in the MV bus bars.
•
Voltage drop along the MV feeders.
•
Distribution transformers connected tap positions.
•
Voltage drop along the LV distribution network.
Because the simulation tool developed considers all factors described, it was used to estimate the number of end customers with a voltage level out of the required range. The simulation tool was applied in the period when the variation voltage at the MV bus bars was highest. Figure 2 shows an example with the performance of voltage in the period when its variation is highest, in a MV bus bar of Chilectra’s electrical system. It shows the measurement of voltage on week days (Monday to Friday) for a month. In Figure 2, the performance of an “average voltage day” and the performance of a “minimum voltage day” are also shown. The “average voltage day” is the day showing the average voltage level for each hour of the day. This is calculated by recording the voltage levels every day, at 15 minute intervals for one month, and then taking the average over the month. Similarly, the “minimum voltage day” is the day showing the minimum voltage level for each hour of the day. This is calculated by recording the voltage levels every
11,5
11,0
10,5
Week days
Average voltage day
23:00
22:00
21:00
20:00
19:00
18:00
17:00
16:00
15:00
14:00
13:00
12:00
11:00
9:00
10:00
8:00
7:00
6:00
5:00
4:00
3:00
10,0 2:00
The result of this stage is an estimation of the voltage at each end customer connection point.
12,0
1:00
A program developed in Chilectra is used to calculate power flow results in the LV distribution network. This program uses an aggregated power algorithm to calculate power flow. The program was developed using the C++ language and it has a graphic interface to analyze the LV network.
12,5
0:00
Additionally, the LV network topology is used. Similarly, just as it does for the MV network, the LV network information includes the length and electrical characteristics of the LV network conductors. The geographical location of all end customers is also included. The database used to maintain and to update the information about the LV distribution network is the same as the one used for the MV network information.
day, at 15 minute intervals for one month, and then taking the minimum value over the month. These two simulations were used to obtain a range for the number of end customers with a voltage level out of the allowed range when the simulation tool considers the operating conditions for the voltage in a MV bus bar.
Bus bar voltage (kV)
C. Power flow in the LV distribution network: The input data for this stage are the voltage, the current and the power factor on the LV side at each distribution transformer obtained from the previous stage.
Minimun voltage day
Figure 2. Example with the voltage performance in the period when its variation is highest.
Table III describes the input data (voltage, current and power factor) concerned with the simulations explained previously. In Figure 2, the measurements of voltage are received every 15 minutes, between 17:00 hours and 23:00 hours, and this gives 24 simulation intervals for each case. In summary, for each one of the 24 simulation intervals, the simulation tool uses the input data (voltage, current and power factor in MV bus bars) corresponding to the “average voltage day” and corresponding to the “minimum voltage day” in order to obtain estimated numbers of end customers with a voltage level outside the allowed range. The input data for the MV power flow method, the transference of MV power flow results to the LV side of distribution transformers and the determination of the number of end customers with a voltage level out of the allowed range, was developed by using a MS-Access Database. The execution of the MV power flow was automated by using the Arc-Info Macro Language (AML) in the UNIX operating system. The LV power flow was executed manually in each simulation interval. TABLE III. Input data
Voltage Current and power factor
INPUT DATA FOR SIMULATION CASES Simulation cases
Average voltage day
Minimum voltage day
Monthly average Values used are those measured simultaneously with the voltage reading nearest to average voltage
Monthly minimum Values used are those measured simultaneously with the minimum voltage
Figure 3 also shows that the number of end customers with a voltage level out of the allowed range increases when the MV bus bar decreases. This was expected (in this case, the voltage level outside the allowed range must be lower than 203.5 (V)). For example, if the voltage level on the selected MV bus bar at 19:15 hours is 11.7 (kV), then the number of end customers with voltage level out of the required range is between 430 y 1,200 customers. The simulation tool results enable us to identify those end customers with a voltage level outside the allowed range and to identify the distribution transformers of these customers. In Figure 4, the geographical location of the distribution transformers with end customers with voltage level out of the allowed range is shown. In Figure 4, the distribution transformers are represented by different dots: the grey dots are transformers where there are no voltage quality problems. The other dots represent the severity of the voltage quality problem. The severity of the voltage quality problem in a distribution transformer is measured according to the number of simulation intervals in which a problem occurred. The higher the number of intervals experiencing difficulties the more severe the problem. The maximum severity is 48, which corresponds to a transformer whose end customers have voltage quality problems in each one of the simulation intervals for the “average voltage day” and for the “minimum voltage day”. There were a total of 25 distribution transformers with end customers experiencing voltage quality problems.
5.000
12,5
4.500
12,0
4.000
11,5
3.500
11,0
3.000
10,5
2.500
10,0
2.000
9,5
1.500
9,0
1.000
8,5
500
8,0
-
Number of end customers
13,0
17 :1 17 5 :3 17 0 :4 18 5 :0 18 0 :1 18 5 :3 18 0 :4 19 5 :0 19 0 :1 19 5 :3 19 0 :4 20 5 :0 20 0 :1 20 5 :3 20 0 :4 21 5 :0 21 0 :1 21 5 :3 21 0 :4 22 5 :0 22 0 :1 22 5 :3 22 0 :4 23 5 :0 0
Figure 3 shows the results obtained by using the simulation tool (voltage values related to the two simulation cases are also showed in Figure 2). Additionally, Figure 3 shows the number of end customers with a voltage level outside the required range (number of end customers with voltage lower than 203.5 (V) or voltage higher than 236.5 (V)).
Bus bar voltage (kV )
A case study using a selected MV bus bar of the Chilectra’s electrical system was developed. The selected MV bus bar has a connected load with the characteristics presented in Table IV.
Average voltage Minimum voltage Number of end customers at average voltage Number of end customers at minimum voltage
Figure 3. Results obtained by the simulation in the selected MV bus bar
By identifying the distribution transformers of end customers with voltage quality problems, the corresponding LV distributions networks and their characteristics are also identified. This, added to the verification of the conditions on the ground, allows the utility to take actions in order to improve the voltage quality. These actions could be: change the connected tap position of the distribution transformer to increase the voltage level, or redistribute the loads connected to the distribution transformers in order to decrease the voltage drop along the LV network. These actions are implemented easily and aren’t expensive. Other actions may be considered, if the aforementioned do not prove successful. This case study presents the improvement in the voltage quality of the end customers obtained by changing the connected tap position of the transformers. From Figure 4, a group of 10 distribution transformers whose end customers have voltage quality problems was selected using the following considerations: . !
! ?! ! @ ! ! !!! ! ! !!! ! ! ! ! ! ? ! !! ! (! ! ! ! !! ! ! !! ! > !!! . !! !! ! ?! ! ! ! ?! ! ! ! ! ? ! ! !!! !!! ! ! ! ! !! ! ! ! > ! ! ! ! ! ! ! ! ! > ! !! ! ! ! ! !! ! !!! .! !! ! ! !! ! ! ! ! ! ! > ! ! !! @! !!! ! ! ! ! ! ! ! ! ( !! (! ! ! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !
! ? ?! !
TABLE IV.
CHARACTERISTICS OF THE LOAD CONNECTED TO SELECTED MV BUS BAR Characteristic
Number of MV feeders Number of company transformers Number of company end customers Number of private transformers (number of private customers)
No.
3 193 17,534 39
#
> !
! !!
! ! ! !! ! ! ! !! ! ! ( ! !! ! @ ! ! ! ! ! ! ! !! ! ! !! ! ! ( !! ! !! ! !! ! ! ! ( ! ! ! ! !! ! ! !!! ! ! !! ! ! ! ! !! ! ! !
! !
!
!
? !
! ! !
! ! ! ! ! ! ! !
Severity of voltage quality problems !
! (
1-5
! @ ? ! . !
8 - 10
> !
#
12 - 14 16 - 19 32 - 46 MV/LV substation
Figure 4. Geographical location of end customers with voltage level outside the required range
•
Severity of the voltage quality problems
The selected group of 10 distribution transformers and the actions taken to improve the voltage quality problems are shown in Table V. Additionally, Table V also presents actions to complement the change of the connected tap position of the distribution transformers. These complementary actions are to change the distribution transformer and redistribute the loads connected to the transformers. Figure 5 shows the improvement in voltage quality in relation to the execution of all the actions indicated in Table V. From Figure 5, we can obtain the percentage of end customers whose modified voltage is now in the allowed range in each simulation interval. The average improvement for the “average voltage day” and for the “minimum voltage day” is 17%. The improvement obtained by changing the connected tap position of the distribution transformers, is related to the transformer tap changer (number of available positions and voltage variation for each consecutive position) and is related to the voltage drop along the LV distribution network.
TABLE V.
SELECTED GROUP OF 10 DISTRIBUTION TRANSFORMERS AND THE ACTIONS TAKEN TO IMPROVE THE VOLTAGE QUALITY PROBLEMS
Distribu tion transfor mer ID
Actual tap positio n
Modified tap position
55660
12540
11880
53511 57700 5641 10362 50798
12000 12000 12300 12000 12000
11700 11700 12000 11700 11700
54268 7517
12600 11880
12300 11700
8757
12540
12000
6541
11880
11880
Other recommended actions
(*) Change the transformer for another with a tap changer in line with Chilectra’s requirements
Redistribute the loads connected to transformers (important voltage drop along the LV distribution network) (*) Change the transformer for another with a tap changer in line with Chilectra’s requirements (*) Change the transformer for another with a tap changer in line with Chilectra’s requirements Redistribute the loads connected to transformers (important voltage drop along the LV distribution network)
(*)Actually, the Chilectra’s requirement for new distribution transformers indicates that their tap changer must have the following available positions: 12,600-12,300-12,000-11,700-11,400 (V). This tap changer was selected in order to decrease the voltage variation for each consecutive position.
2000
1500
1000
500
0 17 :1 5 17 :3 0 17 :4 5 18 :0 0 18 :1 5 18 :3 0 18 :4 5 19 :0 0 19 :1 5 19 :3 0 19 :4 5 20 :0 0 20 :1 5 20 :3 0 20 :4 5 21 :0 0 21 :1 5 21 :3 0 21 :4 5 22 :0 0 22 :1 5 22 :3 0 22 :4 5 23 :0 0
Number of end customers with voltage quality problems.
Number of end customers
•
Number of end customers at average voltage Number of end customers at minimum voltage Modified number of end customers at average voltage Modified number of end customers at minimum voltage
Figure 5. Improvement obtained by excuting the actions indicated in Table V
Further more, the voltage level at the higher end must be considered in order to ensure that a change to the connected tap position to resolve a low voltage problem doesn’t produce a high voltage problem for customers. If this occurs, action must be taken to redistribute the loads connected to the transformer. This is the case of transformer ID 6541: the next available position for the connected tap position. It increases the voltage by 5% and this increase produces a high voltage problem at end customers connected near to the distribution transformer. In this case, the improvement must be to change the distribution transformer for one with a tap changer with a lower voltage variation for each consecutive position. By considering the factors previously discussed, the methodology for solving voltage quality problems can be applied to all MV bus bars in Chilectra’s electrical system. V.
CONCLUSIONS
This paper presented a simulation tool in order to estimate the voltage at each end customer connection point in Chilectra’s electrical system. The simulation tool uses power flow methods and geographical information about the MV and LV distribution networks. Each one of the three stages of the methodology has been described. The description has included their input data, calculation methods and results. In the case study, the simulation tool has been applied to two simulation cases concerned with voltage variations in MV bus bars. The number of end customer connection points with voltage level outside the required range was obtained by applying the simulation tool using the “average voltage day” and the “minimum voltage day”. This, together with the verification of conditions on the ground related to the simulation, enables Chilectra to take actions in order to solve voltage quality problems. A 17% decrease in the number of end customers with a voltage level outside the required range was obtained by applying the simulation tool in the case study. This improvement was obtained just by changing the connected tap position of 10 distribution transformers whose end customers have voltage quality problems in most of simulation intervals. The quantity of transformers whose end customers have voltage quality problems was 25 .
Additionally, the paper presents the main factors to be considered when taking measures to solve voltage quality problems for the whole of Chilectra’s electrical system.
.
Zúñiga Reyes, Cristián Andrés Dept. Power Quality Chilectra S.A. Santiago, Chile [email protected]
Abstract— This paper is presented by Chilectra, a subsidiary of the Endesa Group Spain and the biggest electric utility in Chile. It sets out a methodology for estimating and determining control parameters for the voltage quality to be delivered to its end customers. The methodology is able to estimate the voltage at each end customer location in the low voltage (LV) distribution network, using measurements of voltage, current and power factor in real operating conditions at the feeder substation bus bars in the medium voltage (MV) distribution network. The available information sources, the assumptions used for analysis and the partial results at each stage of the methodology are described. A case study shows the main results of the methodology: the number of end customers with voltage quality problems and the geographical location of these customers. Finally, it shows the improvement obtained as a percentage decrease in end customers subject to bad voltage quality when applying the methodology. Keywords-component; power quality; distribution network; power flow methods.
I.
voltage
quality;
INTRODUCTION
The legal requirements for voltage quality in Chile include restrictions over the voltage levels at the end customer connection points. The requirements consider the nominal voltage at the end customer connection points (high, medium or low voltage). The requirements also consider that the voltage must be within a determined range at least 95% of the time over any seven day measurement period. The legally allowed ranges are presented in Table I.
TABLE I.
LEGALLY ALLOWED VOLTAGE RANGES IN CHILE
Voltag e level
Nominal voltage (kV)
Allowed range as a percentage of the nominal voltage
HV MV LV
110 – 220 12 – 23 0.38 (*)
5% 6% 7.5%
(*) In Chilectra’s electrical system, the nominal voltage in LV network is 400 (V)
Further more, the legal requirements consider the voltage quality evaluation by using two methods: •
Specific distribution network point evaluation. This evaluation determines the individual voltage quality.
•
Statistical evaluation of a group of distribution network points. This evaluation determines the global voltage quality. The group of network points must be selected by statistical methods in order to ensure valid results for all network points.
Currently, the regulating authority evaluates the voltage quality only when it receives an end customer complaint. The regulating authority has yet to begin a systematical evaluation of individual voltage quality. Nevertheless, Chilectra has developed a simulation tool in order to estimate voltage at each end customer connection point. The main objective is to anticipate the voltage quality
problems of end customers. In addition, this simulation tool enables Chilectra to know the current state of voltage quality in Chilectra´s network and to anticipate the results of future regulating authority evaluations. II.
DESCRIPTION OF CHILECTRA’S ELECTRICAL SYSTEM
Chilectra is a utility that delivers the public service of electricity supply to most of Santiago (Chile’s capital). Next Table II presents the main characteristics of Chilectra’s electrical system. III.
DESCRIPTION OF CHILECTRA’S ELECTRICAL SYSTEM
Due to the high number of Chilectra’s end customers and the impossibility of evaluating the voltage quality in all of them, Chilectra has developed a simulation tool to detect voltage quality problems at end customer connection points. This simulation tool estimates voltage at each end customer connection point, by using the operating conditions in the substation bus bars of medium voltage feeders and by using the electrical and topological configuration of the MV and LV distribution networks. The simulation tool uses power flow methods in the MV and LV voltage networks. Figure 1 describes the main stages of the simulation tool. TABLE II.
MAIN CHARACTERISTICS OF THE CHILECTRA’S ELECTRICAL SYSTEM
Characteristic
HV network kilometers HV/MV substations MV network kilometers MV/LV substations LV network kilometers MV feeders Company transformers Private transformers (MV end customers) LV end customers Power flow in the MV distribution network SCADA measurements
Transfer of the MV power flow results to the LV distribution network Voltage, current and power factor in the MV side of distribution transformers
Execution of power flow in the MV network
MV network topology
Transference of MV power flow results to the LV network
No.
358 41 5,035 27,006 9,586 374 21,626 5,380 1,426,445 Power flow in the LV distribution network Voltage, current and power factor in LV side of distribution transformers
Execution of power flow in the LV network
LV network topology Equivalent model of the distribution transformers
Voltage estimation at each end customer connection point
Figure 1. Main stages of the simulation tool
The simulation tool passes through three main stages to achieve results. Each one of these three stages is described below: A. Execution of power flow in the MV distribution network: The input data for this stage are measurements of voltage, current and power factor on the header of each MV feeder. These measurements are taken every 15 minutes, from Chilectra’s SCADA system and show the real value at that present time. Other input data for this stage is the MV network topology. This information is used to simulate the power flow in the MV network. The available topological information of the MV network includes the geographical location of the MV distribution network and of the protection and switching equipment. The length, cross section and electrical characteristics of the MV distribution network conductors are also included. The information about the MV distribution network is maintained and updated daily in an Informix Database. The power flow in the MV distribution network uses a program developed in Chilectra. The program was developed by using the FORTRAN language and it works in the UNIX operating system. The program calculates voltage, current and power factor on the MV side of all the distribution transformers in the MV distribution network. This power flow method uses an algorithm of aggregated power and it assumes that the power distribution at each transformer is proportional to its nominal power. As an outcome of this stage of the process, we can get an estimation of the voltage, current and power factor on the MV side in each distribution transformer. B. Transfer of MV power flow results to the LV distribution network: This stage considers the transfer of the power flow results obtained on the MV side of each distribution transformer in relation to their corresponding values on their LV side. The input data for this stage are voltage, current and power factor on the MV side at each distribution transformer. These results are obtained from the previous stage. Other input data for this stage is an equivalent model for the distribution transformers. This model considers the nominal voltage, the complex series impedance, the tap changer available positions and the connected tap position for each distribution transformer. The information regarding the complex series impedance (magnitude and angle) is obtained from standard impedance tables of Chilectra´s distribution transformers. The information regarding the connected tap position for each transformer is obtained from Maintenance Unit databases and from measurement programs concerned with the verification of voltage quality problems. A MS-Access database is used to provide the results of this stage: the voltage, the current and the power factor on the LV side at each distribution transformer.
IV.
CASE STUDY
The operating conditions of Chilectra’s electrical system vary throughout the year. The performance of the loads connected to the electrical system of which Chilectra’s system is connected and the performance of the loads connected to Chilectra’s electrical system are both important factors. Consequently, the voltage at the substation bus bars of the MV feeders and the voltage at each end customer connection point vary throughout the year. The main factors that determine the voltage level at end customer connection points are: •
Voltage in the MV bus bars.
•
Voltage drop along the MV feeders.
•
Distribution transformers connected tap positions.
•
Voltage drop along the LV distribution network.
Because the simulation tool developed considers all factors described, it was used to estimate the number of end customers with a voltage level out of the required range. The simulation tool was applied in the period when the variation voltage at the MV bus bars was highest. Figure 2 shows an example with the performance of voltage in the period when its variation is highest, in a MV bus bar of Chilectra’s electrical system. It shows the measurement of voltage on week days (Monday to Friday) for a month. In Figure 2, the performance of an “average voltage day” and the performance of a “minimum voltage day” are also shown. The “average voltage day” is the day showing the average voltage level for each hour of the day. This is calculated by recording the voltage levels every day, at 15 minute intervals for one month, and then taking the average over the month. Similarly, the “minimum voltage day” is the day showing the minimum voltage level for each hour of the day. This is calculated by recording the voltage levels every
11,5
11,0
10,5
Week days
Average voltage day
23:00
22:00
21:00
20:00
19:00
18:00
17:00
16:00
15:00
14:00
13:00
12:00
11:00
9:00
10:00
8:00
7:00
6:00
5:00
4:00
3:00
10,0 2:00
The result of this stage is an estimation of the voltage at each end customer connection point.
12,0
1:00
A program developed in Chilectra is used to calculate power flow results in the LV distribution network. This program uses an aggregated power algorithm to calculate power flow. The program was developed using the C++ language and it has a graphic interface to analyze the LV network.
12,5
0:00
Additionally, the LV network topology is used. Similarly, just as it does for the MV network, the LV network information includes the length and electrical characteristics of the LV network conductors. The geographical location of all end customers is also included. The database used to maintain and to update the information about the LV distribution network is the same as the one used for the MV network information.
day, at 15 minute intervals for one month, and then taking the minimum value over the month. These two simulations were used to obtain a range for the number of end customers with a voltage level out of the allowed range when the simulation tool considers the operating conditions for the voltage in a MV bus bar.
Bus bar voltage (kV)
C. Power flow in the LV distribution network: The input data for this stage are the voltage, the current and the power factor on the LV side at each distribution transformer obtained from the previous stage.
Minimun voltage day
Figure 2. Example with the voltage performance in the period when its variation is highest.
Table III describes the input data (voltage, current and power factor) concerned with the simulations explained previously. In Figure 2, the measurements of voltage are received every 15 minutes, between 17:00 hours and 23:00 hours, and this gives 24 simulation intervals for each case. In summary, for each one of the 24 simulation intervals, the simulation tool uses the input data (voltage, current and power factor in MV bus bars) corresponding to the “average voltage day” and corresponding to the “minimum voltage day” in order to obtain estimated numbers of end customers with a voltage level outside the allowed range. The input data for the MV power flow method, the transference of MV power flow results to the LV side of distribution transformers and the determination of the number of end customers with a voltage level out of the allowed range, was developed by using a MS-Access Database. The execution of the MV power flow was automated by using the Arc-Info Macro Language (AML) in the UNIX operating system. The LV power flow was executed manually in each simulation interval. TABLE III. Input data
Voltage Current and power factor
INPUT DATA FOR SIMULATION CASES Simulation cases
Average voltage day
Minimum voltage day
Monthly average Values used are those measured simultaneously with the voltage reading nearest to average voltage
Monthly minimum Values used are those measured simultaneously with the minimum voltage
Figure 3 also shows that the number of end customers with a voltage level out of the allowed range increases when the MV bus bar decreases. This was expected (in this case, the voltage level outside the allowed range must be lower than 203.5 (V)). For example, if the voltage level on the selected MV bus bar at 19:15 hours is 11.7 (kV), then the number of end customers with voltage level out of the required range is between 430 y 1,200 customers. The simulation tool results enable us to identify those end customers with a voltage level outside the allowed range and to identify the distribution transformers of these customers. In Figure 4, the geographical location of the distribution transformers with end customers with voltage level out of the allowed range is shown. In Figure 4, the distribution transformers are represented by different dots: the grey dots are transformers where there are no voltage quality problems. The other dots represent the severity of the voltage quality problem. The severity of the voltage quality problem in a distribution transformer is measured according to the number of simulation intervals in which a problem occurred. The higher the number of intervals experiencing difficulties the more severe the problem. The maximum severity is 48, which corresponds to a transformer whose end customers have voltage quality problems in each one of the simulation intervals for the “average voltage day” and for the “minimum voltage day”. There were a total of 25 distribution transformers with end customers experiencing voltage quality problems.
5.000
12,5
4.500
12,0
4.000
11,5
3.500
11,0
3.000
10,5
2.500
10,0
2.000
9,5
1.500
9,0
1.000
8,5
500
8,0
-
Number of end customers
13,0
17 :1 17 5 :3 17 0 :4 18 5 :0 18 0 :1 18 5 :3 18 0 :4 19 5 :0 19 0 :1 19 5 :3 19 0 :4 20 5 :0 20 0 :1 20 5 :3 20 0 :4 21 5 :0 21 0 :1 21 5 :3 21 0 :4 22 5 :0 22 0 :1 22 5 :3 22 0 :4 23 5 :0 0
Figure 3 shows the results obtained by using the simulation tool (voltage values related to the two simulation cases are also showed in Figure 2). Additionally, Figure 3 shows the number of end customers with a voltage level outside the required range (number of end customers with voltage lower than 203.5 (V) or voltage higher than 236.5 (V)).
Bus bar voltage (kV )
A case study using a selected MV bus bar of the Chilectra’s electrical system was developed. The selected MV bus bar has a connected load with the characteristics presented in Table IV.
Average voltage Minimum voltage Number of end customers at average voltage Number of end customers at minimum voltage
Figure 3. Results obtained by the simulation in the selected MV bus bar
By identifying the distribution transformers of end customers with voltage quality problems, the corresponding LV distributions networks and their characteristics are also identified. This, added to the verification of the conditions on the ground, allows the utility to take actions in order to improve the voltage quality. These actions could be: change the connected tap position of the distribution transformer to increase the voltage level, or redistribute the loads connected to the distribution transformers in order to decrease the voltage drop along the LV network. These actions are implemented easily and aren’t expensive. Other actions may be considered, if the aforementioned do not prove successful. This case study presents the improvement in the voltage quality of the end customers obtained by changing the connected tap position of the transformers. From Figure 4, a group of 10 distribution transformers whose end customers have voltage quality problems was selected using the following considerations: . !
! ?! ! @ ! ! !!! ! ! !!! ! ! ! ! ! ? ! !! ! (! ! ! ! !! ! ! !! ! > !!! . !! !! ! ?! ! ! ! ?! ! ! ! ! ? ! ! !!! !!! ! ! ! ! !! ! ! ! > ! ! ! ! ! ! ! ! ! > ! !! ! ! ! ! !! ! !!! .! !! ! ! !! ! ! ! ! ! ! > ! ! !! @! !!! ! ! ! ! ! ! ! ! ( !! (! ! ! !! ! !! ! ! !! ! ! ! ! ! ! ! ! ! ! !
! ? ?! !
TABLE IV.
CHARACTERISTICS OF THE LOAD CONNECTED TO SELECTED MV BUS BAR Characteristic
Number of MV feeders Number of company transformers Number of company end customers Number of private transformers (number of private customers)
No.
3 193 17,534 39
#
> !
! !!
! ! ! !! ! ! ! !! ! ! ( ! !! ! @ ! ! ! ! ! ! ! !! ! ! !! ! ! ( !! ! !! ! !! ! ! ! ( ! ! ! ! !! ! ! !!! ! ! !! ! ! ! ! !! ! ! !
! !
!
!
? !
! ! !
! ! ! ! ! ! ! !
Severity of voltage quality problems !
! (
1-5
! @ ? ! . !
8 - 10
> !
#
12 - 14 16 - 19 32 - 46 MV/LV substation
Figure 4. Geographical location of end customers with voltage level outside the required range
•
Severity of the voltage quality problems
The selected group of 10 distribution transformers and the actions taken to improve the voltage quality problems are shown in Table V. Additionally, Table V also presents actions to complement the change of the connected tap position of the distribution transformers. These complementary actions are to change the distribution transformer and redistribute the loads connected to the transformers. Figure 5 shows the improvement in voltage quality in relation to the execution of all the actions indicated in Table V. From Figure 5, we can obtain the percentage of end customers whose modified voltage is now in the allowed range in each simulation interval. The average improvement for the “average voltage day” and for the “minimum voltage day” is 17%. The improvement obtained by changing the connected tap position of the distribution transformers, is related to the transformer tap changer (number of available positions and voltage variation for each consecutive position) and is related to the voltage drop along the LV distribution network.
TABLE V.
SELECTED GROUP OF 10 DISTRIBUTION TRANSFORMERS AND THE ACTIONS TAKEN TO IMPROVE THE VOLTAGE QUALITY PROBLEMS
Distribu tion transfor mer ID
Actual tap positio n
Modified tap position
55660
12540
11880
53511 57700 5641 10362 50798
12000 12000 12300 12000 12000
11700 11700 12000 11700 11700
54268 7517
12600 11880
12300 11700
8757
12540
12000
6541
11880
11880
Other recommended actions
(*) Change the transformer for another with a tap changer in line with Chilectra’s requirements
Redistribute the loads connected to transformers (important voltage drop along the LV distribution network) (*) Change the transformer for another with a tap changer in line with Chilectra’s requirements (*) Change the transformer for another with a tap changer in line with Chilectra’s requirements Redistribute the loads connected to transformers (important voltage drop along the LV distribution network)
(*)Actually, the Chilectra’s requirement for new distribution transformers indicates that their tap changer must have the following available positions: 12,600-12,300-12,000-11,700-11,400 (V). This tap changer was selected in order to decrease the voltage variation for each consecutive position.
2000
1500
1000
500
0 17 :1 5 17 :3 0 17 :4 5 18 :0 0 18 :1 5 18 :3 0 18 :4 5 19 :0 0 19 :1 5 19 :3 0 19 :4 5 20 :0 0 20 :1 5 20 :3 0 20 :4 5 21 :0 0 21 :1 5 21 :3 0 21 :4 5 22 :0 0 22 :1 5 22 :3 0 22 :4 5 23 :0 0
Number of end customers with voltage quality problems.
Number of end customers
•
Number of end customers at average voltage Number of end customers at minimum voltage Modified number of end customers at average voltage Modified number of end customers at minimum voltage
Figure 5. Improvement obtained by excuting the actions indicated in Table V
Further more, the voltage level at the higher end must be considered in order to ensure that a change to the connected tap position to resolve a low voltage problem doesn’t produce a high voltage problem for customers. If this occurs, action must be taken to redistribute the loads connected to the transformer. This is the case of transformer ID 6541: the next available position for the connected tap position. It increases the voltage by 5% and this increase produces a high voltage problem at end customers connected near to the distribution transformer. In this case, the improvement must be to change the distribution transformer for one with a tap changer with a lower voltage variation for each consecutive position. By considering the factors previously discussed, the methodology for solving voltage quality problems can be applied to all MV bus bars in Chilectra’s electrical system. V.
CONCLUSIONS
This paper presented a simulation tool in order to estimate the voltage at each end customer connection point in Chilectra’s electrical system. The simulation tool uses power flow methods and geographical information about the MV and LV distribution networks. Each one of the three stages of the methodology has been described. The description has included their input data, calculation methods and results. In the case study, the simulation tool has been applied to two simulation cases concerned with voltage variations in MV bus bars. The number of end customer connection points with voltage level outside the required range was obtained by applying the simulation tool using the “average voltage day” and the “minimum voltage day”. This, together with the verification of conditions on the ground related to the simulation, enables Chilectra to take actions in order to solve voltage quality problems. A 17% decrease in the number of end customers with a voltage level outside the required range was obtained by applying the simulation tool in the case study. This improvement was obtained just by changing the connected tap position of 10 distribution transformers whose end customers have voltage quality problems in most of simulation intervals. The quantity of transformers whose end customers have voltage quality problems was 25 .
Additionally, the paper presents the main factors to be considered when taking measures to solve voltage quality problems for the whole of Chilectra’s electrical system.
.
Related Documents
Voltage Quality Improvement In Distribution Networks
November 2019 3
Quality Improvement
June 2020 9
Supplier Quality Improvement
May 2020 17