Optimal Recloser Setting, Considering Reliability and Power Quality in Distribution Networks

Reclosers and fuses are the commonplace protective devices in distribution networks. A recloser can prevent long-time outages by clearing temporary faults before operation of the fuses in the system. Thus, it decreases the rate of long-term outages and improves system reliability and power quality. Despite positive features of reclosers, each operation of a recloser causes a momentary voltage interruption that exacerbates power quality. Nowadays, power quality issues have become more important because of the increasing use of sensitive equipment to voltage interruptions. According to the mentioned concerns, it seems necessary to set reclosers to strike a balance between power quality and the effectiveness of fuse saving scheme. Thus, we proposed a method to set reclosers. Due to the random nature of faults, the proposed method is stochastic based on the Monte Carlo method. The proposed method determines the optimal number of operations, reclosing intervals, and protection zones. The proposed method efficiency is evaluated according to the simulation results, and the proposed method is capable of establishing an optimal trade-off between power quality and protection efficiency.


Introduction
Nowadays, there is an increasing demand for high-quality and reliable electrical energy and protection systems play a significant role in the improvement of the system reliability [1], [2]. With integrating different energy resources and loads such as wind turbines and electric vehicles in smart grids [3] the system protection is of utmost importance. Since the consequences of low-standard electrical energy impose a considerable economic loss on customers as shown in [4] and [5] that special post-fault tasks are required for minimization of the economic losses. As an example, the cost of power interruptions to U. S. customers is $79 billion annually (divided into sustained outages: 33%, momentary outage: 67%) [6]. Faults are the major source the interruptions in distribution networks [7]. Overcurrent protection is the most common protection system in distribution networks. The statics provided in [8] shows that 70-90% of faults on overhead lines are temporary. Therefore, reclosers can improve the reliability of distribution networks. Previously, interruptions shorter than a few minutes were not considered as a source of worry to the most of the customers [9]. As permanent outages were the main concern of utility operators, protection systems were designed to decrease the system average interruption frequency index (SAIFI), system average interruption duration index (SAIDI), and consequently the energy not supplied (ENS) [8]. In order to achieve these goals, the fuse-saving scheme (also referred to as feeder selective relaying) was employed. In this scheme, reclosers operate faster than the other overcurrent protection (OCP) devices to clear temporary faults [7], [8]. The fuse-saving scheme decreases permanent outages, however, increases the number of momentary voltage interruptions. Nowadays, electronic devices such as microcomputer-based devices and adjustable in Distribution Networks speed drivers are sensitive to momentary voltage interruptions [8]. Therefore, short-time voltage interruptions result in extensive sensitive load shutdowns. It has been reported that momentary outages account for two-third (52.3 billion dollars per year) of the overall power interruption cost in the USA [11], [12]. This indicates the main drawback of the fuse-saving scheme that decreases SAIFI at the expense of an increase in momentary average interruption frequency index (MAIFI). Thus, some utilities employ the fuse-blowing scheme (also known as instantaneous relay blocking) The fuse-blowing scheme decreases MAIFI, however, at the expense of an increase in SAIFI and SAIDI. The effect of reclosers on system outages has been studied in several publications. In [13], a technique proposed to identify the number, type, and location of the protective devices to reduce only SAIFI. Reference [13] is extended in [14] by using a goal programming approach to improve MAIFI. As discussed in [15], a non-linear binary programming solution is utilized to minimize the outage costs in distribution networks. In [16], non-linear binary programming model is used to reduce both SAIFI and SAIDI. In [17], the effects of protective devices on different indices such as MAIFI, SAIFI, SAIDI, and cost of ENS are studied, and it is shown that the cost of ENS is a more precise index than the other indices. Previous authors addressed many aspects of protection system to improve the system reliability. However, the effects of recloser settings, on system reliability and power quality, have not been studied in the literature. Hence, in this paper, we studied the effects of reclosers and proposed an approach based on the Monte Carlo method to set reclosers in distribution networks. The method is also applicable to be implemented in smart micro-grids since the energy resiliency is highly considered in smart grids [18][19][20][21]. The objective of the proposed method is minimization of total outage costs by establishing a trade-off between transient and permanent outages. In this regard, this paper aims at finding: 1) the optimum reach (zone) of a recloser, 2) the number of fast shots, and 3) and the reclosing time intervals. The concerned problem involves many parameters with random nature such as fault location, fault type, fault resistance, and fault nature (i.e., temporary or permanent.) Thus, a stochastic computational method is employed. As the Monte Carlo method has proven to be a beneficial statistical computational technique in attaining approximate answers to the stochastic problems with complex and non-linear parameters [22]. The outcomes of simulations indicate a decrease in outage costs resulted by the proposed method in which the recloser settings are optimally selected without new investment costs. This paper is organized as follows: In Section 2, the recloser and system outage features are described. In Section 3, the proposed method is described. In Section 4, the case study and simulation results are presented and discussed, and in Section 5, conclusion is stated.

Reclosers and Systems Outages
Reclosers exert considerable influence on both transient and permanent outages. Although there are some typical operating sequences for reclosers, it is more appropriate that operating sequence (shuts) of a recloser are selected based on the network specifications. Reclosers affect the system reliability and power quality in three ways as follows, 1) The effect of the shot number: each fast-shot gives temporary faults a chance to be cleared without causing a permanent outage. Therefore, if the number of fast-shots is increased, permanent outages, and consequently, SAIFI will decrease. On the other hand, as the number of fast-shots increases, a less number of loads can withstand the repetitive momentary voltage interruptions [23]. It leads to an increase in momentary outages and MAIFI. 2) Reclosing interval: reclosing interval is defined as the open-circuit time between an automatic opening and the succeeding automatic reclosure [1]. According to (1) and Figure 1, as reclosing interval increases, Risk of Arc Re-ignition (RAR) will decrease [24], [25]. Where is the switching surge frequency distribution, , ∆ is the transient flashover probability at different auto-reclosing dead times, and are the minimum and maximum switching over voltages. It is concluded from (1) and Figure 1 that longer reclosing intervals lead to less permanent outages caused by temporary faults. However, each device withstands an specific duration of voltage interruptions. Therefore, by increasing reclosing intervals, more equipment will be dropped out. 3) Reach (Zone) of reclosers: according to [26], and referring to Figure 2(a), the reach (or zone) of a recloser is defined as a section of a power network that the recloser operates for faults inside it before the other protective devices, while devices outside operate before the recloser. In other words, the fuse-saving scheme is used inside the reach while the fuse-blowing scheme is employed for the faults outside the reach. The reach (zone) of a recloser, contrary to differential or distance protection, is dependently settable. That is, the wideness of a recloser reach is related not only to its own curve but also to the time.
Current characteristics (TCC) curves of downstream protective devices. Referring to Figure 2(a) and (b), we can change the reach of the recloser by moving its fast-shot curve in selectivity diagram. For example, referring to Figure 2, if we decide that the recloser saves all the fuses in Reach 1, its fast-shot curve must be placed below the minimum melting time (MMT) curve of F1. In order to extend the recloser zone to reach 2, then the fast-shot must be placed below the MMTs of F1, F2, and F3. However, it is not always possible to place a recloser fast-shot curve below all the fuse MMTs, considering errors, delays, and grading time (also called as discrimination time or time interval) [27], [13].
The faults outside a reach cause a permanent outage for the loads downstream since a recloser clears temporary faults inside its zone. However, an extension to the reach of a recloser results in a decrease in permanent outages. On the other hand, fast operations of a recloser cause momentary voltage interruptions for downstream loads and it is highly probable that these interruptions cause momentary outages, especially for sensitive loads. Therefore, as the reach of a recloser extends the number of the momentary outages in the system increases.

The Proposed Method
The main goal of the proposed approach is to minimize the total outage cost resulted from a recloser and also to strike a balance between transient and permanent outages. Optimal setting of a recloser involves parameters with random natures such as fault location, fault type, fault resistance, and fault nature (temporary or permanent). These inputs are obtained based on historical data. Therefore, deterministic analyses of such data are inefficient as the input uncertainties may be ignored, while a stochastic method handles those uncertainties. In order to incorporate uncertainties the Monte Carlo method, a computational algorithm relying on repeated random sampling to obtain numerical results is used in the proposed method. The optimum settings for a recloser are achieved through the following steps and the flowchart is shown in Where " # , $ # and # are, respectively, power consumption, restoration time and energy cost for load ( and ) is the number of loads that suffer momentary outages. Cost of the permanent outage (ENS cost) is also calculated using (3) as follows Where " + is total power consumption of the downstream loads, $ , and are restoration time of the network and energy cost, respectively. 1. By repeating the steps 2 and 3, faults are generated and the costs are calculated. This procedure is repeated until the Mont Carlo method converges for each possible recloser setting. 2. Finally, the recloser settings that result in the lowest total cost among all possible settings is selected as the optimum settings. The single-line diagram of the system under study is depicted in Figure 4. In this network, fuses have been used to protect lateral lines and a recloser has been installed in the substation. Table 1 and Table 2 show the line parameters and load data, respectively. The type of transformers is DYn and its ratio is 20/0.4 kV. Four possible settings for the recloser are considered as follows: Case 1: two fast shots with a 1.5-second reclosing interval. Case 2: one fast shot with a 1.5-second reclosing interval. Case 3: two fast shots with a 0.5-second reclosing interval. Case 4: one fast shot with a 0.5-second reclosing interval. The total number of faults in this network is considered as 15 faults per year and the probabilities of different fault types are as follows: LG = 75%, 2LG = 17%, 3LG = 3%, 2L = 3%, and 3L = 2%. In addition, 80% of faults in this network are temporary. Table 3 presents the percentage of temporary faults that are cleared in each recloser shot according to the reclosing interval. The probability of fault occurrence in each line is proportionate to the line length.

Case Study
The fault resistance has a normal distribution with a mean value of 5 Ω and a standard deviation of 1 Ω [7]. The loads based on their sensitivity to power quality problems (i. e., voltage sag and momentary voltage interruption) are classified into three categories as follows: (A) Highly sensitive, (B) medium sensitive, and (C) low sensitive as shown in Table 4. The energy cost and the average duration of supply restoration for this network are 150 $/(MW·Hour) and 2.5 Hour, respectively. The average cost of shots is considered as $10 per shot [17]. By setting a number of shots and reclosing interval based on the abovementioned cases and repeating the proposed Monte Carlo method temporary and permanent outage costs, and case 3 is selected as the optimum setting for the recloser in the test system.

Conclusion
Reclosers are the essential part of the power system and can prevent long-term outages by detecting and interrupting temporary faults. Due to the quick operation of reclosers, some power quality issues may happen in the system. In this paper, a Monte Carlo based method has been proposed for setting reclosers. In this method the optimal number of operations, the reclosing intervals, and the protection zones is determined. The simulation results validate the efficiency of the proposed method.