About reliability characteristics of overhead electrical grids 6-10 kV elements

In the paper, failure outage statistics in 6–10 kV overhead electric grids of one of the largest electric grid companies in the European part of the country were analyzed to clarify the reliability characteristics of their elements. Failure intensity and mean restoration time for covered and uninsulated overhead lines, low-oil and vacuum circuit breaker cells, power transformers and secondary electrical grid elements were obtained. The obtained reliability characteristics of elements should be used in the tasks of technical justification and selection 6–10 kV overhead electrical grid preferred structure. 1 Problem description Problem of power systems reliability performance existed before and continue to be relevant. One of its aspects is traditionally a research of reliability characteristics of electrical grids. At the same time, close attention is paid to the indicators characterizing their failure outages: failure intensity ω and mean restoration time Тr. In the USSR there was centralized collection and processing of statistical data on the failure outages of power system elements. Its latest results are dated to the period 1977-1982 [1], which presents the reliability characteristics of elements of 6–10 kV overhead electric grid: overhead lines (OHL) with uninsulated wires, 6–10/0.4 kV power transformers, and low-oil circuit breaker cells. In plenty of publications for OHL, an extremely wide range of failure intensity is given. So, in paper [1], ω=7.64 1/(year per 100 km), in [3] 2–20 1/(year per 100 km), and in [2] 25 1/(year per 100 km) and even more [4]. In addition, covered overhead lines, (i.e. with wires having a protective insulating outer sheath) apply more and more in the country’s medium voltage grids. Compared with uninsulated wires they have certain advantages in terms of reliability, for example, in [5] it is noted that the covered OHL are protected from overlapping, have high mechanical strength, practically do not form icy-hoarfrost deposits and others. In [6] there was an assumption that the use of covered OHL will reduce the failure intensity of elecrical lines by 2.4 times. The widespread use of covered overhead lines in distributing power systems dates back to beginning 2000-s and now this type of line occupies an increasingly significant share. For power transformers, published failure intensity are also very different. For example, in [1] values of 0.008-0.016 1/year are indicated, and in [2] 0.035 1/year. However, the studies presented were carried out almost 40 years ago. For low-oil circuit breakers, the values of ω=0,009 1/year and Tr=20 h [1], obtained in the 80s of the last century, are similar to the reliability characteristics that are still given in various publications. For example, in [7] values ω=0,01 1/year is presented, including 0,003 1/year fault type of short circuit and 0,007 fault type of break circuit. However, vacuum circuit breakers become more widely used over the last decade in comparison low-oil ones. For vacuum circuit breakers, the expert value ω=0.004 1/year is given in [8]. The same publication contains a reference that, according to leading world manufacturers, the mean operating time to failure of a modern vacuum circuit breaker is 2000 years, whence ω=0.0005 1/year. Note that in the analysis of circuit breaker failure models consider not a only circuit breaker, but a connection failure, i.e. circuit breaker with all related equipment disconnectors, measuring transformers, connecting buses, relay protection and automation devices, insulators and bushings [9]. Thus, the presented information on the reliability characteristic of the elements of 6–10 kV overhead electric grids is outdated or contradictory in a number of positions, and the reliability characteristic of elements of 6–10 kV grids, such as fuses, measuring transformers and etc. are almost not presented in the literature. Therefore, the authors set themselves the task of analyzing failure outage statistics in 6–10 kV overhead electrical grids of one of the largest electric grid companies in the European part of the country and clarifying their reliability characteristic. 1 , 0 Web of Conferences https://doi.org/10.1051/e3sconf/20191390 E3S 139 (2019) 1


Problem description
Problem of power systems reliability performance existed before and continue to be relevant. One of its aspects is traditionally a research of reliability characteristics of electrical grids. At the same time, close attention is paid to the indicators characterizing their failure outages: failure intensity ω and mean restoration time Тr.
In the USSR there was centralized collection and processing of statistical data on the failure outages of power system elements. Its latest results are dated to the period 1977-1982 [1], which presents the reliability characteristics of elements of 6-10 kV overhead electric grid: overhead lines (OHL) with uninsulated wires, 6-10/0.4 kV power transformers, and low-oil circuit breaker cells.
In plenty of publications for OHL, an extremely wide range of failure intensity is given. So, in paper [1], ω=7.64 1/(year per 100 km), in [3] -2-20 1/(year per 100 km), and in [2] -25 1/(year per 100 km) and even more [4]. In addition, covered overhead lines, (i.e. with wires having a protective insulating outer sheath) apply more and more in the country's medium voltage grids. Compared with uninsulated wires they have certain advantages in terms of reliability, for example, in [5] it is noted that the covered OHL are protected from overlapping, have high mechanical strength, practically do not form icy-hoarfrost deposits and others. In [6] there was an assumption that the use of covered OHL will reduce the failure intensity of elecrical lines by 2.4 times. The widespread use of covered overhead lines in distributing power systems dates back to beginning 2000-s and now this type of line occupies an increasingly significant share.
For power transformers, published failure intensity are also very different. For example, in [1] values of 0.008-0.016 1/year are indicated, and in [2] -0.035 1/year. However, the studies presented were carried out almost 40 years ago.
For low-oil circuit breakers, the values of ω=0,009 1/year and Tr=20 h [1], obtained in the 80s of the last century, are similar to the reliability characteristics that are still given in various publications. For example, in [7] values ω=0,01 1/year is presented, including 0,003 1/year -fault type of short circuit and 0,007 -fault type of break circuit.
However, vacuum circuit breakers become more widely used over the last decade in comparison low-oil ones. For vacuum circuit breakers, the expert value ω=0.004 1/year is given in [8]. The same publication contains a reference that, according to leading world manufacturers, the mean operating time to failure of a modern vacuum circuit breaker is 2000 years, whence ω=0.0005 1/year. Note that in the analysis of circuit breaker failure models consider not a only circuit breaker, but a connection failure, i.e. circuit breaker with all related equipment -disconnectors, measuring transformers, connecting buses, relay protection and automation devices, insulators and bushings [9].
Thus, the presented information on the reliability characteristic of the elements of 6-10 kV overhead electric grids is outdated or contradictory in a number of positions, and the reliability characteristic of elements of 6-10 kV grids, such as fuses, measuring transformers and etc. are almost not presented in the literature.
Therefore, the authors set themselves the task of analyzing failure outage statistics in 6-10 kV overhead electrical grids of one of the largest electric grid companies in the European part of the country and clarifying their reliability characteristic. As can be seen from the table 3, the overhead line failure intensity was only 2.3 1/(year per 100 km). In the region, abnormal ice rains preceding the period under review caused numerous failure outages in electric grids of 6-10 kV and long-term restrictions on the power supply of consumers. Further, technical measures were organized, such as accelerated reconstruction of grids, widespread and uncompromising clearing of «no tree» zone OHL, replacing uninsulated OHL with covered OHL, which made it possible to achieve shown in table 3 characteristics.
It should be noted those 3-5 days each year, during which a quarter or more of all overhead line outage occur due to abnormal weather conditions (strong winds, ice rains, etc.).  4 shows that the failure structure of uninsulated OHL and covered OHL proved to be expected. So, the percentage of failures related with the intervention of unauthorized persons and organizations and adverse natural phenomena (tree contact), reaches in both cases 70-80%. In addition, the share of covered OHL failures associated with atmospheric overvoltages is several times higher than that for uninsulated OHL, as was mentioned earlier in [5].

Reliability characteristics of 6-10 kV low-oil and vacuum circuit breaker cells
The failure outage statistics for the cells of low-oil and vacuum circuit breakers 6-10 kV at 611 substations 35-220/.../6-10 kV in the region for the period 2016-2018 were analyzed. The size of the statistical sample, see table 5. As can be seen from the table 5, the share of low-oil circuit breakers is more than half of the installed equipment fleet, despite the fact that for many years they have not been used in new construction or reconstruction.
In the table 6 shows the total number of failure outage and their total duration for the circuit breaker cells from table 5. Based on the data in table 5 and 6, the values of the failure intensity and the mean restoration time of the circuit breakers are calculatedsee table. 7.  Table 7 shows that the failure intensity and the mean restoration time of the vacuum circuit breaker cells are about two times less than the low-oil. One of the indirect cause for this is connected that the number of components and moving parts of a vacuum circuit breaker is half that of a low-oil one. Note that the operational logs that record failure outages usually indicate the time to restore the initial condition of the circuit, but not actual time for repair of the circuit breaker cell. The reason of it is that cells with withdrawable elements are mainly installed at substation. Therefore, reserve withdrawable circuit breaker cell are used to quickly restore initial state of the circuit in case of a serious failure of circuit breaker. Table 8 shows the failure structure of circuit breaker cells at substations. During the period under review, only two cases (2016 and 2018) of loss of airtightness of the vacuum circuit breaker arc chamber were recorded. The low-oil circuit breaker drive shaft failure share of about 20% of all failures. In general, the failure structure of low-oil circuit breaker cells in table 8 and in [8] are comparable. For vacuum circuit breaker cells, the main cause of failures was malfunctions of the control circuits, and for low-oil ones -insulation breakdown.
A comparison of the data from table 7 and the paper [1, 7,8] shows that the obtained reliability characteristics of the circuit breaker cells become more optimistic. It can be assumed that this is because substation cells were analyzed. The operational specificity of substation's circuit breaker cells is the relatively insignificant (several times a year) number of switching and the presence of duty personnel at substation. Perhaps we should expect a deterioration in the reliability characteristics for circuit breaker cells in the auxiliary systems of 6 kV power plants, where technological conditions require much more frequent (once or twice a day [8]) switching. For circuit breakers installed in prefabricated 6-10 kV distribution substation should be expected higher mean restoration time. Firstly, this failure takes extra time to detect it and personnel to arrive at the failure substation. Secondly, the design of the cells does not include withdrawable elements. However, a document, published more than 30 years ago, already described that in the scheme of medium voltage electrical grid, the construction of a 6-10 kV distribution substation should be used only in the present of feasibility studies [11].
Thus, vacuum circuit breaker cells 6-10 kV are approximately two times more reliable than outdated low-oil circuit breaker cells in terms of failure intensity and mean restoration time.  Note that the installed capacity of transformers in city cable grids in this region is three times higher than that in the overhead electric grids of suburban areas, although the number of transformers in both cases is comparable. Moreover, in overhead electric grid installed capacity of transformers 6-10 kV three to four times exceed the maximum of their actual load, possibly due to the crisis that occurred in the country (the period of the 90s of the last century).
In the table 10 shows the total number of failure   outage and their total duration for power transformers  from the table 9.  Based on the data table 9 and 10, the values of  failure intensity and mean restoration time of the  considered power transformers are calculated -see  table 11. From the table 11 it follows that the failure intensity of power transformers in cable electrical grids is about four times higher than in overhead electrical grids. Apparently, this is due to two reasons associated with operation on urban areas: the obviously higher load of power transformers and their installation in enclosed spaces, which worsens cooling conditions. Comparing reliability characteristic from table 11 with previously published data, a significant improvement in the reliability of power transformers is noticeable: the failure intensity decreased from 0.008 (see above) to 0.001 1/year (Table 11), which is a favorable factor.
The failure structure of power transformers 6-10 kV is given in table 12. As can be seen from it, the failure share of transformers associated with internal failures reaches approximately 70%, which must be taken into account during their operation.

Reliability characteristics of 6-10 kV secondary electrical grid elements
In the table 13 shows the reliability characteristics of measuring transformers, fuses (on three phases) and disconnectors. Analysis of the data from the table 3, 7, 11 and 13 leads to the conclusion that the main part of the failures of 6-10 kV overhead electric grids consist of overhead line failures. Failures of other network elements can be neglected in reliability assessment

Conclusion
1. Based on the processing of representative statistical sample, the reliability characteristics of uninsulated and covered OHL in 6-10 kV electrical grids are obtained. It turned out, that the failure intensity of covered OHL has significantly more favorable reliability characteristics compared to uninsulated ones.
2. Vacuum circuit breaker cells 6-10 kV are more reliable than outdated low-oil ones: about two times from the terms of the failure intensity and mean restoration time. At the same time, the actual values of the reliability characteristic obtained at this time interval turned out to be more favorable compared to previously published ones.
3. The reliability characteristics of 6-10 kV power transformers turned out to be much more favorable than previously published values. However, it cannot be ruled out that the equipment fleet and its operating conditions have been updated over the past 40 years since the last reliability studies.
4. The failure intensity of secondary electrical grid elements (measuring transformers, disconnectors, fuses) is significantly lower than overhead lines, circuit breaker cells and power transformers described above. Therefore, in a first approximation, they can be neglected in reliability assessment.
5. The obtained reliability characteristics of the 6-10 kV overhead electric grid elements should be used in the tasks of technical justification and selection their preferred structure.