Technical Loss Reduction in Rural Areas-The Case of Saih Al

This article investigates the potential for reducing technical losses in the rural area network of Saih Al Khairat in Thumrait, Oman. Based on the available network data, a power flow model of the system under study is built and the system performance is studied. To reduce losses and improve the voltage profile, different candidate solutions are investigated in coordination with the distribution system operator. An economic evaluation of the different options is conducted.


Introduction
Losses in power systems are classified into two categories: technical losses and non-technical losses Al-Mahroqi, Metwally et al. (2012).Nontechnical losses result from actions external to power systems, for instance, human manipulation or mistakes in meter reading.Electricity theft is one of the main causes of nontechnical losses in some systems Antmann (2009).To reduce losses of this type, some utilities use automated meter reading and advanced metering infrastructure (AMI).In addition, it is possible to use data mining and intelligence-based techniques to detect and reduce non-technical losses Nagi, Mohammad et al. (2008).Even without AMI, non-technical losses can be estimated by comparing the measured energy consumption in a feeder with the energy that the utility bills plus technical losses Neto and Coelho (2013).
In comparison, technical losses are related to the physical characteristics and functions of the electrical network that result in the dissipation of electrical energy as heat.These types of losses occur mainly in low-efficiency equipment and in transmission and distribution lines.Technical losses can be classified as generation losses, transmission losses, and distribution losses.The causes of the technical losses include a low power factor, long lines, unbalanced loading, and overloading.An assessment of technical losses can be made with engineering calculations based on the design of the system.The majority of avoidable technical losses occur where the current is high.
Technical losses represent economic loss for utilities as generating more energy results in higher costs of generation.In addition, as losses result in the generation of more electrical energy to satisfy the generation-load balance requirement, high technical losses contribute to global warming.
According to the Authority of Electricity Regulation (AER), the total losses (technical and non-technical) in Oman's electricity sector were estimated to be 10.2% in 2015, a decrease from 11.6% in 2014.Moreover, losses in the Main Interconnected System (MIS) decreased from 11.6% in 2014 to 10% in 2015, whereas the Rural Areas Electricity Company's (RAEC's) losses increased from 9.2% in 2014 to 10.7% in 2015 (AER 2016).
The main contributions of this article are listed below: 1. Modelling of a practical system in Oman during peak load condition.The system is known to have an under-voltage problem.
2. Simulating the system performance in terms of voltage profile and technical losses using ETAP ® software package.
3. Simulating the system performance considering five solutions that aim for reducing system losses and improving voltage profile.These options were coordinated with the network operator.
4. Conducting a cost-benefit study of potential savings due to loss reduction for the considered options.
These options involve adding reactive compensation elements at selected buses as well as network reconfiguration by adding a new 11 kV feeder.More details of these options are presented in section 5.
Following this introduction, the paper presents a survey of sources of and mitigation techniques for technical losses in distribution systems.Section 3 presents the data of the system under study.As for section 4, it presents and discusses the results of the simulation of the network performance with different options.Finally, section 5 presents a summary of the main conclusions.

Technical Loss Sources and Mitigation Techniques
Technical losses in a power system result naturally from current flow through resistive materials as well as the nonlinear characteristics of some equipment in the grid.The most common example of technical loss is the power dissipated in transformers and transmission lines due to their internal resistance.
For example, the losses in a transmission line can be calculated by determining the difference between the measured energy injected from the source into the transmission line (  ) and the measured energy consumed (  ) by loads located at the end of that transmission line.

Poor Power Factor
In general, losses that occur in conductive materials can be decreased by reducing the current or by reducing the resistance.However, reducing the current is more effective as the loss formula (I 2 R) shows.The magnitude of the current is a function of the apparent power (S), which, in turn, has two components: real power (P) and reactive power (Q).

|𝐼| =|S|/√3|V|
(2) The power factor (PF) is related to the cosine of the angle between the voltage and the current or the ratio of the real power to the apparent power.
The PF decreases as the ratio of reactive power to the real power increases.It is possible to achieve an improvement in PF by using devices, such as capacitors (switched/fixed) that deliver reactive power.A case study involving the reduction of losses using power factor correction was presented in Phetlamphanh, Premrudeepreechacharn et al. (2012).Figure 1 is a simple illustration of the reactive compensation concept.
Distribution utilities require customers to maintain a good (high) power factor to reduce losses.For example, industrial customers in Oman are obliged to maintain a power factor of at least 0.9 (AER 2016).

Unbalanced Loads
Distribution network losses can vary depending on the level of the balancing of the load.In an unbalanced operation mode, voltages and currents are not equally distributed between phases.Different factors can result in unbalanced operation modes.They include unequal phase loading and different line parameters in different phases.
Unbalance commonly occurs in three-phase distribution systems.However, it can be harmful to the operation of power systems.On the generation side, current asymmetry negatively affects efficiency.Unbalance reduces transmission capacity and efficiency Albadi, Hinai et al. (2015).In addition, it reduces the capacity and efficiency of the transformers.Zero sequence current is converted into a circulating current in a delta/wye-grounded transformer, resulting in increased losses.
There are several approaches to reduce the effects of unbalance.It is essential to adopt regulations and standards to ensure that all system components are designed and manufactured to be symmetrical.These components include generators, transformers, transmission lines, three-phase motors, and switching equipment.In addition, imposing standards related to voltage and current unbalance levels is essential.These unbalance levels should be defined for both utilities and customers Albadi, Hinai et al. (2015).Another approach involves revising the connection of single-phase loads on the utility and customer sides.In addition, unbalance can be reduced by using balancing equipment such as single-phase voltage regulators, a dynamic voltage restorer (DVR), surge-protection devices, unified powerflow controllers (UPFC), and energy storage devices Kazibwe, Ringlee et al. (1990), Jouanne and Banerjee (2001).

Transformers Losses
In power distribution networks, the losses in transformers can reach 3% of the total electrical power generated (Ltd 2006).The transformer efficiency can be increased by reducing these losses.The losses in transformers can be classified in two different categories: the core loss or no-load loss category and the load or copper (winding) loss category Al-Badi, Elmoudi et al. (2011).Load losses are not highly sensitive to grid voltage changes, but they are highly sensitive to temperature variations.In the new distribution transformers, the secondary winding takes the form of a cylindrical sheet of aluminium.This is an important consideration in the adjustment of losses for temperature variation.

Network / Feeder Reconfiguration
In some cases, distribution network restructuring to minimize losses is highly costefficient.This option is of interest to efficiencyconscious electric utilities.

System Data
The system under study consists of 12 diesel generators (43 MW and 81 MW) and 103 transformers, as Table 1 indicates.The network has four feeders.One is a 33 kV feeder and three are 11kV feeders.The 33 kV feeder is operated and owned and operated by a large customer and was not modelled due to missing data.Three types of overhead transmission lines are used in the network.They are Panther, Wolf, and Dog ACSR conductors.Moreover, the underground cables in the 11kV feeders come in different sizes (50, 185, 240, and 500 mm 2 ).In the 33 kV feeder, the underground cables have a size of 300 mm 2 .

Power Flow Model
Power flow analysis is widely used in power system operation and planning.The power flow model of the system can be built using relevant network, load, and generation data.The outputs of the power flow model include voltages at different busses and line flows in the network.These outputs are obtained by solving power balance equations: where |  | and |  | are the magnitudes of the voltage at bus  and , respectively;   and   are the associated angles; |  | is the magnitude of the Y-bus element between the two busses; and   is the corresponding angle.
These power balance equations are nonlinear; therefore, iterative techniques such as the Newton-Raphson, the Gauss-Seidel, and the fast-decoupled methods are commonly used Saadat (1999).In this case, ETAP ® software package was used to model and evaluate the case study.
The system losses can be calculated once the power flow problem is iteratively solved.For example, the losses in the branch i-j are the algebraic sum of the power flows.

Load Data
The load data of the power flow model were validated with the recorded data as shown in Fig. 2. The difference between the measured data and that obtained from the model was about 8.7%, 2.35%, and 0.2% for Feeders 1, 2, and 3, respectively.These differences were due to records missing for some of the transformers.
In addition, the recorded peak loads did not occur at the same time.Individual service transformer (11 kV/415 V) load data were included in the model to obtain the feeder load data.

System Technical Losses
Simulation results presented in Table 3 show that technical real power losses represented about 5.18% of the generated power, whereas the reactive power losses were 11.28% of the generated power.The distributions of these losses are given in Figs. 3 and 4.   It was clear that most losses occurred in the overhead line.This result was attributable to the fact that overhead lines dominated the rural area system.Another observation was that three quarters of the losses occurred in Feeder 2. This result was attributed to the long distances connecting the scattered loads in this feeder.

Voltage Profile
The Omani distribution code mandates that the voltage profile be within 6% of the nominal value in distribution networks (33 kV, 11 kV, and 415 V) MJEC, MZEC et al. (2005).The Omani grid code allows for variation from the nominal value of up to 10% in transmission networks (132 kV, 220 kV, and 400 kV) (OETC 2010).
The voltage profile at all busses is given in Fig. 5.The nominal voltage in this system was 11 kV.The voltage drop varied from one bus to another, depending on the loading of each bus and the distance from the power house.In general, the voltage drop was mostly due to the long distances between the service transformers in the network.The 6% limit that the distribution code specified was violated at many busses, especially in Feeder 2.
The main reasons of having this low voltage problem are the growing demand and extension of feeders to connect new customers in this small isolated system.These substantial changes in this area are due to transferring some agricultural activities from Albatinah Governorate to Saih Al Khairat in Al Najd area, where an underground water reservoir was discovered.This transfer is aiming to conserve underground water in Albatinah Governorate and reduce air pollution caused by some agricultural activities.Therefore, to connect new customers, feeders were extended.

Candidate Solutions
Different options for reducing losses and improving the voltage profile were studied in coordination with RAEC.These options are listed below: The voltage profiles of these options in the peak loading condition are shown in Fig. 6.

System Performance with Different Options
Figure 6 presents the improvements in voltage at different locations of the network using the different proposed options.It is clear that Option 2, Option 3, and Option 4 led to significant improvements in voltage.Hence, they will result in better compliance with ±6% voltage limits, which, in turn, will result in increases in equipment lifetime and customer satisfaction.

Economic Evaluation of Different Options
Tables 4 and 5 show the losses and associated energy costs for the aforementioned options.The energy calculation can be performed by determining the loss factor (  ) using the following equation (Gonen 2008).
where   is the Load Factor.Once   is found, the average power loss can be calculated using the peak losses obtained from the power flow simulation.The load factor used in this study is 0.82.Based on this value, the loss factor is 0.7.
Average Loss = F LS * Peak loss (10) The annual energy loss can be calculated using the following formula: Annual Energy Loss = Average Loss * 8760 (11) RAEC estimated the capital cost of the different options.It was estimated that the costs of the switched capacitor banks and the 11 kV feeder were 14,000 OMR/MVAR and 13,000 OMR/km, respectively.
Using these values and the results from the previous table, the annual savings, payback period, and net present value were calculated based on the 10.4% discount rate and 25-year lifetime that RAEC employed.
The net present value (NPV) was calculated for each option.The NPV of a project is the difference between revenues and costs in the current monetary value.In any comparison of investment options, the project with the maximum NPV is the winning one.For a recurring constant annual income, the present value can be found using the following formula: where   is the present value of the recurring annuity, A. In this context, the annuity refers to the annual savings with the implementation of different options.
The payback period (PP) is defined as the length of time required to recover the initial investment in a project.The shorter the length, the more economically attractive to investors the project is.Another benefit of reducing losses is that it brings down fuel costs and subsidies.Additionally, the reduction of losses results in improved conditions for the immediate environment due to the reduction in power generation and CO2 emission.
The results presented in Table 6 show that any of the first four options would recover the costs within 2.5 years.The best option, which had the highest NPV, was Option 4.Although Option 5 had the largest annual savings, it was not economically attractive.This was because it was associated with a high capital cost.

Conclusion
The objective of the article was to investigate different candidate solutions for reducing technical losses and improving the voltage profile of a rural area distribution system.A model of the Saih Al Khairat network, which the Rural Areas Electricity Company owns, was developed, and the load flow solution was obtained using ETAP ® software package.The network data was collected from the Rural Areas Electricity Company and the equipment manufacturers.To improve the voltage profile and reduce losses, five options were studied and ranked according to their economic feasibility.The best option among them was the installation of three pole-mounted capacitor banks: two in Feeder 2 and one in Feeder 1.

Figure 1 .
Figure 1.Illustration of the reactive compensation concept.

Figure 3 .
Figure 3. Distribution of power losses in the system.

Figure 4 .
Figure 4. Distribution of losses in the feeders.

Figure 5 .
Figure 5. Voltage profile at all busses.

Figure 6 .
Figure 6.Voltage profile for different scenarios.
The Rural Areas Electricity Company (RAEC) is an Omani closed joint stock company registered under the Commercial Companies Law of the Sultanate of Oman.The company commenced its operations on the 1 st of May, 2005, following the implementation of a decision that the Ministry of National Economy issued pursuant to the Regulation and Privatization of the Electricity and Related Water Sector law, which was promulgated by Royal Decree 78/2004 (RAEC 2016) (RAEC 2016).RAEC serves customers who are not connected to the Main Interconnected System (MIS) and Dhofar Power System (DPS).Its license and business activities are associated with generation, transmission, and distribution (RAEC 2016).

Table 1 .
Network data summary.There are six 11 kV/415 V transformers, as Table2indicates.Appendices A and B present the details of the loading on these transformers.The single-line diagram is presented in the Appendix C, as is the per unit line data.

Table 4 .
Reduction in losses.

Table 5 .
Cost of losses for different scenarios.

Table 6 .
Economic evaluation for different scenarios.