Analysis techniques of diagnosis high voltage underground cables

Table of Contents

1.Introduction
1.1 Thesis Objectives

2. LITERATURE REVIEW

3. Differences between Underground Electrical Cables and Overhead Electrical Cables

4. Insulation and Methods of Protection of High Underground Cables
4.1 Phase Comparison Scheme
4.2 Directional Comparison Scheme

5. Economic Aspects of High Underground Cables

6. Technical Aspects of High Voltage Underground Cables

7. Common High Voltage Underground Cables Faults

8. Detection of High Underground Cables Faults

9. Analysis of High Underground Cables Faults pre-locating Methods
9.1 Cable Route tracers
9.2 Selecting a Tracer
9.3 Time Domain Reflectometry (TDR) Tests
9.4 DC Hi-potential Test
9.5 Cable pre-location
9.5.1 Sectionalizing
9.5.2 Electromagnetic Surge Detection
9.5.3 Arc Reflection
9.6 Pinpointing the fault
9.6.1: Acoustic Detection:
9.6.2 Electromagnetic Impulse Detector
9.6.3 Voltage Gradient Test

10. Solution and method to fix faults on High underground Cables faults

11. Discussions of High Underground Cables Recovery Solutions
11.1 Cable smoldering
11.2 Cable chopping
11.3 Cable Stripping
11.4 High-Temperature Incineration

12. Modeling & Simulation of High Voltage Underground Cable
12.1 Model Explanation
12.2 Measurment of Voltage modelling for different fault locations

13. Installing & Testing of High Voltage Underground Cable in Ras Tanura
13.1 Underground High Voltage Cable Constructions & Ratings
13.2 Important Standard & Specifications
13.3 High Voltage Underground Cable Installation
13.3.1 Materials and Equipment
13.3.2 Manpower Organizations
13.3.3. Quality Control
13.3.4. High Voltage Cable Installation Procedure
13.4 69 KV High Voltage Underground Cable Simulation & Testing
13.4.1 Simulation for Fault Points for 69 Kv Underground Cable
13.4.2 Calculation for pre-location faults

14. Conclusion

15. References

Original Content Statement

Hereby I declare that this work consists of an original creation written by me, Hussain Ali Al Mahdi, student number 242023 from the Warsaw University of Technology

From the huge amount of literature covering the topic presented here, several titles have been used as source of knowledge and therefore cited at the end of the document, but the content has not been in any case copied, or plagiarized in any form.

However, if any sentence written in the following chapters appears to be very similar with another in literature, I must say that will be product of coincidence and there is not any sort of intentionality/awareness from me, the author, about this situation. Finally, when external pictures are used, there source will be understated.

. ABSTRACT

This study is focused on installations and faults diagnosis of high voltage underground cables and examined on the various ways through which they can be identified and corrected. It has been found that wear and tear is one of the major issues leading to electric faults like power loss. There has been seen that several causes of failure which could be identified with them and analyze better ways of detecting them. So, it has been looked at how to detect degradation of insulation of the underground cable. After a thorough analysis of fault detection techniques have been explained, the pre-location fault techniques is described; I illustrated many methods which aid to discover the fault location. I also developed a method which contributes in pre-location of the fault based on Matlab simulation and mathematical slope equation. This method was invented after the observation that the voltages measured at the local and remote substations, after filtering and after Fourier analysis, showed a linear behavior that was proportional to the respective distances between the fault location and the mentioned substations or generators. Thus, the developed method exploits such a linearity to estimate the fault location by interpolation, assuming that the parameters of the line are previously known and that dedicated devices are located on site in both substations to provide post-fault measurements of the faulted line. Finally, I clarify many pin-pointing methods to precisely determine the fault location

Thesis Keywords: High Voltage, Underground Cable, Cable Failure, Pre-location Detection, Matlab Simulation, Pin-pointing.

The laying of cables underground is associated with several significant advantages, and this has made it popular for decades. One factor that makes overhead transmission lines undesirable is the fact that they are adversely affected by rough environmental conditions leading to power interruptions. That, however, is not the case with underground lines that pose very minimal threat to the environment The underground cables have their demerits too. These cables are costly and are, therefore, only limited to use within short distances. This other disadvantage of underground cables is the subject of this thesis work. When faults occur in underground cables, total failure is likely to occur because fault diagnosis in the cables is a very cumbersome task. It is very difficult right from fault detection, cable routing, fault location all the way to correction. Besides, the faults experience very poor heat dissipation due insulation. As a result, these cables have low power transferring capacity and are likely to get permanently damaged.

Underground cables are single core, and three cores in most cases and are not left naked to be insulated by the air like overhead lines. The cables have sheaths around them made from artificial materials to provide insulation. In the early days, insulation was provided using paper of extruded lead.

The capacitance of underground cables is high and that leads to higher charging current. On the other hand, their inductance is low due to the small space existing between the earth and the conductor.

Faults are not commonly experienced in underground cables. When they occur, however, locating and correcting them is more tedious than in overhead cables. Despite such challenges, faults should be detected and eliminated promptly to ensure that power supply to consumers is not interrupted for long.

A host of techniques exist for classifying and detecting faults in underground electrical cables. The types of faults that occur in underground cables are classified using hybrid, wavelet, and ANFIS methods (Yang, Choi, Lee, Ten, & Lim 2008). Software-based methods, artificial intelligence (AI) methods, traveling wave methods and analytical procedures are used in location of the faults.

Either overhead lines or underground cables can be used to convey power to consumers. The major distinction, as I have already mentioned in the previous paragraphs, between the two cable types is that overhead lines are suspended high in the sky while the underground ones are buried in the ground. Overhead lines are long distance lines and that is why they are preferred in rural areas. The electricity network, therefore, consists majorly of the overhead cabling due to the long distance coverage. In addition, the overhead lines are also popular for low transmission losses. On another front the underground cables, even though the initial cost of laying them is high, last long and are not subject to weather-related interruptions. Because they are not suspended up in the sky, overhead cables are frequently affected by adverse weather conditions and that is a setback in power distribution (Paolucci 2010)

Since underground high voltage cables are buried and, therefore, not exposed to physical damages, the occurrence of faults is rare in them. That, however, does not mean that these cables are entirely immune to faults. The cables occasionally experience faults which are very difficult to diagnose considering the fact that the cables are buried hence difficult to access. A lot of expenses are thus incurred in eliminating faults in underground cables.

The three major types of faults experienced in underground cables are open-circuit fault, earth fault and short-circuit fault. An open-circuit fault results in case the metal conductor breaks up and a megger is used to check it. The wearing out of the insulation cover of a multi-core cable makes its conductors to be directly in contact with each other leading to the second type of fault which is short circuit fault. Finally, the third type of fault is an earth/ground fault that results from conductor-ground contact. Both the short circuit and the earth/ground faults can also be checked using a megger.

The fault types above have been categorized based on what causes them. Alternatively, the classification can also be done based on whether the faults are temporary or permanent. Permanent faults are irreversible result from equipment failure. When there is total insulator failure.

Fault detection in cables can be done in a number of ways. Abnormalities in circuit behavior could be a pointer to the occurrence of a fault. Changes in circuit performance is the cheapest method of diagnosis because tests are not required. The abnormalities could be a smell of a burning plastic due to overheating causing insulation cover to burn up or power outage at circuit parts.

Underground power distribution network is a complex system composed of switchgears, insulators and, conductors. Considering this complexity, manual monitoring of this network may not be effective. As a result, temperature sensors may be included in this system to enable real-time monitoring of the distribution network (Liu & Zhou 2007)

Table of Illustrations

Figure 3.1: Underground cables in trenches. Source......21 Figure 3.2: Overhead lines

Figure 4.1: Cable insulation

Figure 4.2: Single Sheath Bonding,

Figure 4.3 Both end sheath bonded lines

Figure 4.4: Schematic diagram for differential protection scheme

Figure 9.1: cable analyser devise

Figure 9.3: A TDR circuit

Figure 9.4: A Hi-pot test diagram

Figure 9.5.2: Electromagnetic surge detection

Figure 9.5.3: Arc reflection

Figure 9.6.1: Pinpointing faul using Electromagnatic impuls detector

Figure 9.6.3: Voltage Gradient method to detect fault in underground cable

Table 12.1, parameters of the tested high voltage underground cable:

Figure 12.1: System and underground cable model for fault estimation in Simulink

Figure 12.2: Simulink Model Window

Figure 12.3.: Distributed parameter lines in matlab simulation

Figure 12.4: Example of a simulation of the fault behaviour

Figure 12.5: simulation of fault behavior (zoom after 0.25 seconds)

Figure 12.6: fault location plot at point m=

Figure 12.7: Fault location plot at point m=

Figure 12.8: fault location plot at point m=

Figure 12.9: fault location plot at point m=

Figure 12.10: fault location plot at point m=

Figure 12.11: fault location plot at point m=

Figure 12.12: fault location plot at point m=

Figure 12.13: fault location plot at point m=

Figure 12.14: fault location plot at point m=

Figure 12.15: Voltage A and Voltage B vs Fault location

Table 12.2 Estimated fault location calculation

Figure: 13.1 Underground cable routing, 63 Figure 13.2: demolishing of the overhead electrical cables tower

Table 13.1.1: 69Kv underground cable ratings and parameters

Table13.2.1 Important slandered and specification

Figure: 13.3.4.1: Explanation of the duct bank constructions,

Table 13.3.4 High potential DC test Voltage

Figure 13.4.1: Matlab Simulation for 69 KV Underground High Voltage Cable

Figure 13.4.2: Underground cable 69 kv Fault location at point

Figure 13.4.3: Underground cable 69 kv fault location at point

Figure 13.4.4: Underground cable 69 KV fault location at point

Figure 13.4.5: Underground cable 69 kv location at

Figure 13.4.6: Underground cable 69 kv Fault location at point

Figure 13.4.7: Underground cable 69 KV Fault location at point

Figure 13.4.8: Underground cable 69 KV Fault location at point

Figure 13.4.9: Underground cable 69 KV Fault location at point

Figure 13.4.10: Underground cable 69 KV Fault location at point

Figure 13.4.11: 69 KV underground cable VA & VB VS Fault Locations

Table 13.4 calculation of real “m” fault and estimated fault “m” location

1. Introduction

The demand for electrical power rises every day both in the domestic and industrial sectors of our economy. As a result of that, cheap and efficient power distribution system must be in place at all times. Underground cables have been in use for a long time in power distribution networks due to the advantages associated with underground connections (Keulenaer, 2006). These cables are friendly to the environment, are not interrupted by adverse weather conditions, are less expensive for shorter distances and have got low maintenance costs. The disadvantages of these cables are that they are more expensive than overhead lines, have low power transferring capacity, are likely to get permanently damaged and fault location in underground cables is also difficult.

Faults in underground cables are divided into two general groups that are permanent and incipient faults. Incipient faults develop from aging of insulation material that can be caused by chemical pollution, electrical overstress, severe environmental conditions and mechanical factors. As result of this, incipient faults gradually turn into permanent faults (Barakat, 2014). Fault location is important in ensuring that distribution networks remain reliable as the restoration will be quick, so that power outage time is reduced. The cost of repair will also be low if a fault is located and corrected promptly.

1.1 Thesis Objectives

This thesis is focused majorly how faults can be diagnosed in High voltage underground cables and the possible solution techniques for detecting the faults. Therefore, some objectives are examined to realize this goal. The first objective is to establish the distinction between underground electrical cables and overhead electrical cables. The second one is to investigate insulation and other methods of protection in underground electrical cables and to scrutinize the economic and technical aspects of laying these underground cables. Finally, the last aim is to locate fault when occurred and to recover faulted cables.

2. LITERATURE REVIEW

Underground transmission, high voltage cables are usually used (Keulenaer, 2006). The underground cables may be laid in ducts or directly buried in the ground. For overhead lines, air is the insulation. For underground cables, on the other hand, artificial materials are used for insulation. For that matter, cables are more expensive as compared to overhead lines.

The inductance of cables is low because the spacing between conductor and earth is small, their capacitance, though, is high leading to an even higher charging current. These high voltage cables are usually single cored or three cores and are insulated and protected by sheaths. Initially, cables were insulated with paper of extruded lead.

One of the major challenges experienced in underground cables is a fault. A fault is any abnormal electric current in a cable such as a short circuit, for continuous power supply, faulted cables must be located and isolated.

Many ways have been suggested for classifying and finding faults in underground electrical cables, the location techniques have been grouped as software-based methods, artificial intelligence (AI) methods, travelling wave methods and analytical procedures. Fault classification simply means identifying the type of fault that occurs in a cable. For classification, methods such as hybrid, wavelet, and ANFIS have also been applied (Yang, Choi, Lee, Ten, & Lim 2008).

The wavelet method is based on the multi-resolution analysis (MRA) of voltage and current waveforms. The analysis breaks down signals into resolutions of different levels. The signal is first passed through two discrete wavelet transform filters, high pass filter, and low pass filter. The output samples of high pass filters are called detail coefficient while those of low pass filters are called approximate coefficients. These outputs are referred to us first levels and can further be processed to get other levels, and that is dictated by the degree of resolution needed.

According to Bucher & Franck (2013), the ANFIS method employs a fuzzy logic modeling. As a learning algorithm, this method uses an artificial neural network. With rules defined beforehand, the logic can map inputs to outputs.

Fault classification is usually followed by fault location in fault diagnosis. Fault location refers to finding the exact physical location of a fault. Once the position of the fault has been identified, the faulted cable can then be isolated, and power restored to minimize inconvenience to consumers. As mentioned above, there are several ways of locating faults.

For travelling wave theory method, voltage and current travelling waves are used in finding the fault (Gilany, Ibrahim & Eldin 2007). Voltage and current waveforms become turbulent at the points of fault occurrence. This turbulence will make them propagate along the power system. The location of a fault will then be calculated from the length of the faulted line, the propagation velocity and time of waveforms. This method is commonly used for high voltage transmission lines. However, it is rarely used for distribution lines because of the many laterals and feeders in the distribution systems. The subsections (laterals and feeders) are likely to create disturbances for the travelling waves in the lines.

In the circuit theory method, faults are located using current and voltage measurements. Impedance values are also used. The type of fault is first identified after which the fault voltage and current are used to work out apparent impedance. The equation of the apparent impedance is then used to calculate the location of the fault. The apparent impedance equation has got two unknowns that are fault distance and, fault resistance. Therefore, when this impedance equation is broken into real and imaginary parts, we can get the location of the fault (Barakat, 2006).

The Artificial Intelligence (AI) method employs three approaches which are Artificial Neural Networks (ANN) method, Fuzzy Logic based method and Expert System method. The location of the fault in ANN method is based on recognizing voltage-current patterns at different fault locations (Ryan,1996).

3. Differences between Underground Electrical Cables and Overhead Electrical Cables

Residential places and business enterprises are connected to the grid through overhead lines and underground cables. These cables could be low- voltage or high-voltage. Overhead lines are supported up in the sky using steel towers or wooden poles (Fig 3.1). They are very common in rural areas because they are long distance lines. Because of the long distances covered, these cables constitute a bigger percentage of the interconnected system. They are also reliable in power delivery (380-kV extra-high voltage) because they experience low transmission losses. The initial cost of laying underground cable infrastructure is high and once laid, they last long. Overhead cables are suspended high in the air and are vulnerable to adverse weather conditions. Overhead cables are, therefore, prone to weather-related interruptions (Paolucci 2010).

Underground cables are buried in the ground (Fig 3.2). They do not cover long distances and are used in medium and low-voltage systems. They are common in built up areas such as towns where demand for electricity is high. However, these cables are also becoming popular in the countryside. These cables have a high cost of installation and maintenance as compared to overhead lines (Baldursdottir, 2010). The failure rate is limited in underground cables. However, the failures are very costly when they occur because these cables are very inaccessible and a lot of time is spent in repair and replacement. The plastic insulations of the underground cables also affect the soil (Megger, 2003).

What is spent in high-voltage underground cable installation, when compared to that of overhead lines, is roughly 10 to 15 times higher. Underground cables and overhead transmission lines have different electrical characteristics. Underground cables have got a much higher shunt capacitance and a much lower series inductance. The lower inductance in underground cables is because of the small spaces between cable conductors (Paolucci 2010).

The shunt capacitance of an underground high voltage cable is larger than that of the overhead counterpart due to several factors. First, the cable conductor is close to the ground potential, and this makes its shunt capacitance high. The cable sheath also contributes to high capacitance just like the insulation whose dielectric constant exceeds that of air.

Calculating the series sequence impedance in underground cable can be difficult comparing with overhead lines. This, however, is not the case in underground cables. This calculation is made difficult in underground cables by the occurrence of magnetic coupling in between the cable phase currents. The coupling is occasionally experienced in cable sheaths and, this is hugely dictated by the bonding type of the sheath. So as to make the series sequence impedance calculations possible, therefore, simultaneous equations are formed. It is by solving these equations that we get the voltage drop in the conductors and sheath of the cables. The series sequence impedances is easier to calculate, though, for single conductor cables except the pipe-type cables (Paolucci 2010).

More dissimilarity between OVL and Underground High Voltage cable is that OVL is not affected by Overload for a certain time whereas Underground High Voltage cable has limited tolerance for overload. In Addition, OVL can be rebuilt in case of faults or changing in design whereas Underground High Voltage cable is not likely to be rebuilt. In case of life expectancy, OVL line can live up to 70 years on other hand UGHV cannot last for more than 40 years. Figure 3.1 and figure 3.2 show the construction of underground cable and overhead lines.

illustration not visible in this excerpt

Figure 3.1: Underground cables in trenches. Source: Anon, (2015). [image] Available at: http://www.windfarmbop.com/category/trenches-and-cabling/ [Accessed 11 Jun. 2015].

illustration not visible in this excerpt

Figure 3.2: Overhead lines. Source: Anon, (2015). [image] Available at: http://electrical-engineering-portal.com/how-hv-transmission-lines [Accessed 8 Apr. 2015].

4. Insulation and Methods of Protection of High Underground Cables

Excessive overheating could damage underground cables and they must, therefore, be protected. The overheating is caused by fault current that flows in the cables due to eddy current losses that generate heat. Overheating will burn and destroy the insulation and even lower the quality of the conductor metal leading to a shorter life for the cable. In Figure 4.1 and example of cable insulation is illusterated.

illustration not visible in this excerpt

Figure 4.1: Cable insulation. Source: ( XLPE insulation cable, EXSYM company Japan 2010)

Faults can cause large-scale and catastrophic damages. As a result, the cable protection mechanisms adopted should have a very short response time to minimize damages. The mechanisms should also have efficient ways of communicating from one cable end to the other (Lopez, Gomez, Cimadevilla & Bolado 2008). Most cable faults are initially associated with the ground and, therefore, the ground fault sensitivity must be given utmost consideration.

High-voltage underground transmission cables do not have electrical properties entirely similar to those of overhead lines. The knowledge of grounding and cable impedances is useful in protecting underground cables. The cable impedance is a factor of magnetic coupling in between the phase currents and sheath current (Tobias N, 2009).

AC carrying conductors experience magnetic field leakage. This magnetic field induces a voltage in all the conductors that lie close to the current carrying conductor (Candelaria & Park, 2011). Thus, cable sheaths must be grounded in the circuit for safety reasons. Sheath bonding arrangement is one of the factors that lead to losses in single-conductor cables. In that regard, bonding and grounding in underground cables should be done in such a manner that reduces the sheath voltages and the losses to the least values possible. Bonding and grounding will also allow fault current to travel to and from, and that protects the circuit from surges that could result from either lightning or switching.

They are usually three types of sheath bonding according to IEE standard which are Single End Bonding or Double End Bonding or Cross Bonding so that very low circulating current is present and consider a good method for prevents heating. A single sheath bonding can be considered the easiest system of bonding which can explained as metallic sheaths are grounded at only one point along their length and crossing all other points and hence, since there is no closed sheath circuit , circulating current will not be presented. Figure 4.2 shows the single sheath bonding

illustration not visible in this excerpt

Figure 4.2: Single Sheath Bonding, lines J.R. Riba Ruiz, Antoni Garcia, X. Alabern Morera (N.D), Circulating sheath currents in flat formation underground power.

However in case of both end sheath bonding , the metallic sheaths are grounded at two end points so that no heating effect in the cable screen. Moreover, it provides a path for fault current. This method is suitable for longer cable section, more economical and has less material required. In figure 4.3 shows the both end sheathing bonding.

illustration not visible in this excerpt

Figure 4.3 Both end sheath bonded Source : J.R. Riba Ruiz, Antoni Garcia, X. Alabern Morera (N.D), Circulating sheath currents in flat formation underground power lines.

Charging current is the major challenge faced in protecting underground cables. This charging current is a fraction of the load current, and its impact is majorly felt in long cables. As a result of charging current, the choice of minimum current fault settings becomes limited. Besides charging currents, high transient currents are caused when circuits are energized and de-energized. The intensity of the transient currents is measured from two parameters, which are its magnitude and frequency. These two parameters are a factor of the circuit breaker specifications and the properties of the circuit being energized. A protection system design must, therefore, take the transient currents into consideration. In that regard, the set current must be larger than the charging current by some threshold amount (the charging current is in steady state). Should the threshold not be met, the protection system will malfunction (Candelaria & Park 2011).

Most of the faults witnessed in cables are permanent, even though, how fast is the protective relay. Reclosing is therefore not allowed as it will worsen the damage. A protective relay system in a circuit does not necessarily need only to respond when faults occur in cables. Relays also respond when other devices connected to the system experience flashovers. In fault inspection, therefore, devices must not be avoided because they can also be sources of faults (Demetrios, 2012).

The protection relay schemes used are the directional comparison, current differential and, phase comparison (Demetries 2012).

The directional comparison scheme is further subdivided into zerosequence-based scheme, the negative sequence directional comparison and, distance protection scheme.

The current differential scheme is usually used in the protection of underground cables because they are less affected by cable characteristics. When the channel for communicating gets out of service distance, the directional overcurrent relays are used as backups.

The current differential relay works by comparing currents. The distinction between local and remote currents will tell where the fault has occurred. When the fault occurs within the protection zone of the relay systems, an action will be taken and breaker will be disconnected. The fault may also occur outside the protection zone.

Figure 4.4: Schematic diagram for differential protection scheme. Source: Protection of High-Voltage AC Cables Demetrios A. Tziouvaras , 2005.

This scheme has got several advantages. It is resistant to system swings and current reversals and is minimally affected by cable characteristics. Besides, the scheme is quick in response, sees the cable wholly and current needed for it to function.

On the other hand, the scheme also has its demerits. When there is an external fault, it is affected by CT saturations that require security logic to restrain. The next demerit is that it can only be used when there are a communication channel and a bandwidth, and it cannot be used as a backup protection.

4.1 Phase Comparison Scheme

In phase comparison, instead of directly comparing local and remote currents like in current differential, the phase angle between them is compared. There is a preset threshold that the phase difference should not exceed. If this limit is exceeded, however, the relay will consider the system to be experiencing an internal fault. Just like the current differential, this scheme is immune to system changes and fluctuations in current. It also monitors the cable from one end to the other and ensures timely protection. It is also less affected by cable characteristics and only requires phase information to operate. In comparison to the current differential scheme, it experiences little CT influence when an external fault occurs. When, in the occurrence of an internal fault, the communication channel fails, both of its terminals trip correctly. It is, however, less sensitive than the current differential relay.

On the demerits, the phase comparison scheme cannot be used as a backup, is dependent on the availability of channel and bandwidth and, is affected by charging currents. Also, when a communication channel fails at the time of occurrence of an external fault, the blocking signal will not be received at the terminal. Such a failure would cause the circuit breaker (CB) to not operate at that end.

4.2 Directional Comparison Scheme

In the directional comparison scheme, the directional zero-sequence is compared to the ground-distance and, the phase-distance, between the local and remote terminals. The directional zero-sequence can also be substituted with the negative-sequence information in the comparison.

In designing this scheme required to be conversant with the properties of the cable. Sheath bonding and grounding knowledge is also necessary for this design. Grounding highly influences zero-sequence. Besides, sheath bonding and cable properties also dictate the zero-sequence mechanism. It is because of the reasons mentioned that it is not regularly used for cable protection (Paolucci 2010).

Distance relays are not popular in protection scheme design because of the inherent factors that make them unsuitable. However, they can be used in blocking and unblocking and also in backup protection. The current differential scheme is mostly used because it is reliable, sensitive, and the settings are easy to calculate. The effects of shunt-reactor and charging currents must be compensated though.

5. Economic Aspects of High Underground Cables

The evaluation of every stage of the life cycle of the cable helps in determining its economic aspect. The life cycle of a cable ranges from procurement, construction, operation and finally to the end of life. The most significant costs are the initial ones. They are costs incurred on procurement and construction. They are the immediate costs and are larger than the later expenses on repair and maintenance. For that matter, it is these costs that have the most significant financial implications in cable installation projects.

The costs incurred in the later stages of underground cable installations are not easy to estimate. It is quite difficult to foretell the magnitude of electrical losses likely to be experienced in the future and the associated costs. The amount of future losses is dependent on the load the line will carry. The cost, however, depends on the fuel cost and on whether there is surplus capacity for the generation.

There are technical options that can be exploited to minimize the cost of placing cables underground. These are options such as being flexible in design rather than strictly holding on to the standard designs. It is such flexibilities that make the cost of installation even more unpredictable.

It is not advisable to use the previous values to estimate the costs of current installations. The price of underground cables is significantly affected by fluctuations in the price of the raw material such as copper. Besides, it is very expensive to manufacture and stock cables, especially the high voltage ones, in large quantities. As a result, the price of cables will be affected by the demandmanufacturing capacity curve.

Comparing the cost of the overhead cables with that of the underground cables would give misleading results. For reliable information, the installation costs for overhead lines and underground cables should be separately worked out for every new installation project. The two costs can then be compared. There are also threats that can be hardly expressed in monetary terms such as destruction of archeological sites, visual intrusion and, damaging habitats.

While examining costs, land-use issues must also be considered. These are issues such as people seeking compensation because trenches pass through their land. On land use still, the underground installation might present problems to future endeavors such as using the land for agriculture or building development.

The cost of laying underground cables in urban areas is higher than in rural areas. In these areas, there are several lines such as telecommunication lines, water lines, sewer lines that restrict mechanized digging and this implies that the trenches are dug by hands at times leading to even higher costs.

Other costs arise from managing traffic flow when the channels cross roads and restrictions on working hours so that residents are not inconvenienced. In some cases, the cables are laid in deep tunnels to overcome the problem of dissipation. It is expensive but necessary at times.

The cost of placing the cables underground is relatively low in rural areas. The spaces are enough, and mechanical diggers can be used in digging the trenches. The conditions are also better than in urban areas where you do not have to support the wall of the trench. The only challenge in rural cable undergrounding is the need to conserve the environment, and this comes with a cost.

When cable designs are improved lighter cables, hence longer drum lengths are obtained. As a result, the cost and time for installation are reduced. Mechanized laying techniques can also be employed to lower the cost of installation especially for low power cables in the countryside.

6. Technical Aspects of High Voltage Underground Cables

There are technical factors that make underground circuits more complex and more expensive as compared to overhead lines. These factors even worsen the situation at higher voltages. Underground cables can be used in areas where overhead lines cannot be used. Such could be areas with insufficient space or with some technical aspects that do not favor the use of overhead lines. For instance, overhead lines cannot be used in vast seas.

Underground cables must be insulated so that they can operate safely at high voltage. Initially, oil impregnated paper was used for insulation. This paper was kept under pressure so that it surrounded the cable completely. In the recent days, however, polyethylene has replaced paper in the cable insulation. The polyethylene insulation is far much superior to the paper insulation and can withstand voltages as high as 500kV.

By placing cables underground, cost and complexity increases because the problem of discharging waste heat arises. For overhead cables, the loss is only due to the conductor resistance. Underground cables have current flowing both in their metal conductors and, sheath. The loss in these cables, therefore, is the total sum of both the conductor and sheath losses. Underground cables loss is also a factor of its insulation. This loss has got a linear relationship with voltage (that is, the square of the cable voltage) and can be noted even when the cable is conducting very little current.

The air surrounding the overhead cables and the earth act as the insulation between it and earth. At the same time, this air acts as the coolant for the cable. For the underground cable, on the other hand, the electrical insulation serves as the thermal insulation and prevents a lot of heat from the conductor. The soil, also, can be a serious barrier for heat dissipation. As a result of this hindrance to heat discharge, the cables are usually surrounded by backfill to enhance heat dissipation. The heat dissipation characteristic makes the underground cable ineffective, thus, losses must be keeping very low. In order to keep the losses low, conductors with larger diameters are used to minimize electrical resistance. Overheating can also be reduced by using copper of low resistance for making cable wires.

Laying cables underground has significant implications on the cost of power distribution. This cost is not fixed but varies depending on whether the cable can be easily accessed or not. It also depends on the size of load to be supplied. The requirement for heat dissipation determines the size of the trenches and the extent of the construction work.

7. Common High Voltage Underground Cables Faults

Cables can be directly buried in the ground or be run along cable ducts in underground distribution systems. The occurrence of faults in the underground cables is rare because these cables are not exposed to physical damages. These faults are taxing to correct should they occur because locating them takes a lot of time. According to Theraja, Patel, Uppal, Panchal, Oza, & Patel (2005), there are three major fault types that occur in underground cables. These are the open-circuit fault, earth fault and short-circuit fault.

An open-circuit fault occurs when the metal conductor breaks up. This kind of fault is checked using a megger. When using a megger, all the three conductors at one end of the cable are joined and then earthed. The megger is then used to measure the resistance between each conductor and ground. The megger reading will be zero resistance in case the circuit is not broken. However, if the circuit is broken, the megger reading will be an infinite resistance.

Conductors of a multi-core cable will directly touch each other if the insulation wears out. Such a direct contact of conductors will result in a short- circuit fault. A megger can also be used to check a short-circuit fault. Here, any two conductors are connected to the terminals of the megger. A zero reading in the megger will not be an indication of short circuit fault between the two conductors. For other conductors, the same step will be repeated taking two conductors at a time.

The third type of fault results from conductor-ground contact. This kind of fault is earth/ground fault. It is also checked using a megger. One terminal of the two terminals of the megger is linked to the conductor suspected to be faulty and the other terminal to earth. A zero output by the megger will be an indication that the conductor is earthed. Similarly, the same process is followed for the other conductors of the cable.

Apart from the fault types described above, there are also permanent and temporary faults. A permanent fault cannot be remedied and usually occurs when equipment fails, when there is total insulator failure or when the conductor breaks. Going by statistics, most faults that occur in underground cables are permanent.

Temporary faults are those that can be corrected after they occur. However, when these kinds of faults are not corrected in good time, their arc can cause catastrophic damages to the elements of the circuit.

There are also flashing faults and faults caused by water. The flashing fault is a type that is not noticeable at low voltages but as the voltage rises to some threshold value, it becomes significant. When water gets into cables with paper insulation, it changes the impedance of the cable, and this may lead to a fault. Plastic cables, however, may operate even when they are wet.

8. Detection of High Underground Cables Faults

There are three major ways of detecting faults in circuits. The first way you can tell that a fault has occurred is when you realize an unusual change in circuit performance or behavior. It is a detection technique that does not require any testing. Behavior such as a smell of burning plastic due to a thermal effect on the insulation or sockets can be an indicator of the existence of a fault. Moreover, a section of the circuit may fail to provide power (Liu & Zhou 2007).

Faults can also be detected by inspecting the circuit. Inspection can be done by walking around checking every element of the circuit. Discolored or burnt insulations, broken wires among other things will be signs of faults. One may even have to disconnect some circuit elements from the network so as to locate the problem.

Finally, faults can also be detected by testing. Testing tools are used to find out if faults have occurred. The Magnetic (AMR) CN Probing is a device used for sensing faults in underground cables. Current flowing in conductors produces magnetic fields. This device measures the currents in the concentric neutrals (CNs) by sensing the associated magnetic fields. Both forward and return currents are measured, and they should have identical phases. Any sign of phase imbalance or lack of CN currents will be an indication of a fault.

Underground power cables are complex systems consisting of protective materials, insulators and, conductors. Temperature sensors may also be integrated into this underground distribution network to allow monitoring of cable conditions every time. An optical fiber is reinforced using a protective metal wire and then incorporated with the copper wires that shield the leakage of electric fields outside the cable (Liu & Zhou 2007).

Optical fibers are very sensitive to temperature changes. Changes to the light under transmission give temperature values along the path of light. With the current sensing techniques, the temperature along the fiber can be measured with a resolution of around one meter. Based on that, any factor causing an increase in the cable temperature can be detected real-time. Such factors could be damaged insulations, changing soil conditions around the cable and human interference.

9. Analysis of High Underground Cables Faults pre-locating Methods

Fault location methods are divided into two categories that are offline and online methods. For the offline methods, the target cable must not be in use when being examined for a fault. This method, again, is usually used by trained persons and requires the use of special machines. The online method, however, uses the values of voltage and current measured in the faulted cable.

The offline methods are further subdivided into another two categories that are tracer and terminal methods. The terminal methods are based on studying the readings taken at each end of the cable suspected to be faulty. For the tracer method, the data used is provided by a professional who walks along the cable route taking measurements. Tracer and terminal methods are accurate even though they need a lot of manpower and take a lot of time. In situations when both offline and online methods fail, tracer methods are employed to locate faults.

Terminal methods are employed only in instances of small fault resistance. In case when the resistance is high, methods that can temporarily convert the fault resistance to a small value are used. One of such methods is burn arc reflection. Alternatively, surge arc reflection and surge pulse reflection can also be used. However, it is the recommendation of IEEE that the methods used in fault location should be those that give results at the lowest voltage possible in the least time. For that matter, most of the offline methods are not used because they are problematic.

Online methods are also further subdivided into two categories; traveling wave-based methods and impedance-based methods. In the impedance-based methods, the impedance of the target cable is measured and then compared with the original cable characteristics. The outcome of the comparison can then be used to locate the fault.

After a cable has been isolated in readiness for locating the fault, there must be an order to follow. Just like in diagnosis, this exercise requires a step- by - step procedure that will guide us to locate the fault the precisely. At the very beginning, it is useful to gather all sorts of relevant information about the cable. The first one is the cable type. The cable could be concentric neutral, tape shield or lead covered. The second information to collect is the insulation type. The insulation used could be paper, XLPE or XPR.

The third thing is to know about the cable conductor’s and its size. if the conductor made of aluminum or copper and if it is stranded or solid. The fourth issue is the cable length. Finally, you should know if there are splices and their locations.

There are devices called cable analyzers which can visually display the conditions in a cable. These devices can act as centers from which faults in a cable can be tested. Those displays may be reflections called signatures or cable traces. The signatures are pulse reflections that result from the variations in the impedance of the cable being measured. These reflections can then be used to calculate the distance of the fault. Figure 9.1 show a sample of this devise.

illustration not visible in this excerpt

Figure 9.1: cable analyser devise. Source: [image] Available at: Protection of High Voltage AC Cables Demetrios A. Tziouvaras , 2005 .

9.1 Cable Route tracers

It is sensible, first of all, to know the route along which the cable runs. Knowing the route is even more important in case the. Tracing the route of an electrical cable is not just an ordinary task.

The task of cable tracing can even be more complex when a lot of plants are installed in the ground. The most popular tracers in use have got two primary modules that are a transmitter and a receiver. The transmitter is an AC generator that supplies signal current into the underground cable. The receiver, on the other hand, senses the electromagnetic field created by the AC in the cable.

For easy trace, they are important factors which must be known to assist for accurate pre-location of fault; they can be listed as follow: A-Type of Cable,

B- Insulation type of cable, C- Conductor metal, D- Length of the cable, E- Splices of the cable.

[...]