Lenes-hydrate-preventiont-electrical-heating-2005.pdf

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/242190961

Hydrate Prevention on Long Pipelines by Direct Electrical Heating Article · July 2008

CITATIONS

READS

3

241

5 authors, including: Harald Kulbotten

Arne Nysveen

SINTEF

Norwegian University of Science and Technology

8 PUBLICATIONS   31 CITATIONS   

74 PUBLICATIONS   539 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Optimal Utilization of Smart Generators View project

Design of Variable Speed Generators for Hydropower Applications View project

All content following this page was uploaded by Harald Kulbotten on 02 June 2016. The user has requested enhancement of the downloaded file.

SEE PROFILE

Hydrate Prevention on Long Pipelines by Direct Electrical Heating Atle Lenes1, Jens Kristian Lervik1, Harald Kulbotten1, Arne Nysveen2, Atle Harald Børnes3 1

SINTEF Energy Research, N-7465 Trondheim, Norway 2NTNU, Dept. of El. Power Eng., N-7491 Trondheim, Norway, 3 STATOIL ASA, Forusbeen 50, N-4035 Stavanger, Norway

KEY WORDS ABSTRACT The direct electrical heating system (DEHS) has been developed and qualified for heating of flowlines and is the only system for heating that has been installed on subsea flowlines in the North Sea. Especially for deep-water fields electrical heating of pipelines is attractive for achieving reliable operation of transport flowlines. Experiences on installed systems have proven easy operation and have also shown to imply high reliability for flow assurance. At present DEH has been applied to pipelines of chromium (13Cr) up to 16 km length. Projects are now coming up in the North Sea where application of DEH for pipelines of carbon steel with lengths up to 50 km are considered. A qualification project was initiated in 2003 to expand the technology base for DEH to include these new installation parameters. An important issue has been to obtain the electrical and magnetic characteristics of carbon steel pipes. The difference in magnetic characteristics of individual carbon pipes results in variation of the heated pipe temperature. Carbon Steel materials are more sensitive than 13Cr for thermal and mechanical stresses on the outer surface in regard to DEH properties. For long pipelines a new cable design is essential to avoid high screen voltages caused by capacitive and inductive currents. In order to solve this problem a semi-conductive outer sheath has been developed for continuous transfer of the capacitive current to surrounding ground potential (seawater/seabed). The results from the study is promising and have given basis of rating of DEH to maintain or raise the thermally insulated steel pipe temperature above the critical value for hydrate (typically 15 - 25 oC) or wax formation (typically 20 - 40oC). Furthermore the single-phase power supply to the heating system is normally available from the general topside (platform) power system. Topside equipment for power factor compensation and load symmetry is based on commercial components. Experiences from installed systems prove that the load current can be quite easily adjusted to give the required heating for the different operation modes.

Paper No. 2005-JSC-413

Lenes

Electrical heating, cable design, screen voltage, semi-conductive cable sheath, hydrate and wax prevention.

INTRODUCTION When transporting untreated well stream in ordinary pipelines, the temperature of oil, gas and produced water will drop rapidly due to cooling from the surrounding seawater. The low temperature results in undesired fluid properties. At high pressures hydrates start to precipitate already at temperatures in the range of 20-25oC. Large amounts of hydrate, which is similar to ice crystals, can precipitate on the pipe wall and cause blocking of well stream transport. For some fields wax formation in the flowing crude may also cause operational problems due to increased pressure loss in the pipeline. The viscosity of waxy oil can be of such magnitude, that full “shut in wellhead pressure” will not be sufficient for getting the cold fluid on stream again after long shut downs. The use of chemicals to remove hydrates will in practice mean to use methanol or glycol. These chemicals move the hydrate equilibrium to higher pressure and lower temperature, and thus work in the same way as non-freeze solution in an automobile radiator. The disadvantage with use of chemicals is that large amounts are often needed and implies a risk to the environment if leakage should occur. An obvious way to remove hydrates is to supply heat to the flow. On a platform or on shore this can be done by injecting hot water or steam into the pipe, but subsea this mostly has to be done by using electrical heating. The direct electrical heating system (DEHS) has been developed and qualified for heating of flowlines and is the only system that has been installed on subsea flowlines in the North Sea. Electric heating of pipelines implies reduced investments of depressurizing systems and recovery plants for chemical residual products. Especially for deep-water fields electrical heating of pipelines is attractive for achieving reliable operation of transport flowlines.

1

DIRECT ELECTRICAL HEATING ON LONG FLOWLINES The DEH system (DEHS) (J. K. Lervik, 1998), is supplied from the platform power supply, from which feeder cables provide the electric power to the heating system. One of the two single core feeder cables is connected to the near end of the pipe, and the other to the forward conductor (piggyback cable) is connected to the utmost end of the pipe. The cable connected to the far end is installed in parallel with the pipe as indicated in Fig. 1. For safety and reliability reasons, the direct electrical heating system is electrically connected (“earthed”) to surrounding seawater through several sacrificial anodes for a length of about 50 meters (“current transfer zone”) at both ends where the cables are connected, see Fig. 1.

Power supply (topside) Feeder cables (Dynamic two core cable) Piggyback cable. 20 % generated heat (~0,5 kV/km) Grounded pipeline 70 % generated heat (0 kV/km) Seawater 10 % generated heat (0 kV/km) Transfer zone Fig. 1.

Stationary part

Transfer zone

DEHS schematically shown with the XLPE insulated single core cable strapped to (piggybacked) and connected to the end of the heated flowline. 50 - 70% of the return current will flow in the seawater. The generated heat in the pipeline is approx. 70% of the total heat generation. The voltage of the piggyback cable increases linearly with length.

material (and also depending on other factors, such as cable conductor dimensions, distance between cable and pipe). The generated heat for the steel pipe in the relevant pipeline in is approx. 70%, which gives an estimate for the efficiency of the system. However, for a buried pipeline the generated power in the piggyback cable, which is approx. 20% of the total contribute to heat the pipeline and will increase the efficiency of the system. The generated heat in the seawater is approx. 10%. The current will be transferred between pipe and seawater at the pipe ends (cable connections) in the current transfer zones as indicated in the figure. Due to variation in magnetic permeability for the single pipe joints, and hence the impedance, some current transfer will take place through the sacrificial anodes, which are distributed along the pipeline. To limit this effect the pipeline is divided into sections, each with a limited span of the permeability. An important design basis concerning the current transfer zones is the number, dimension and location of sacrificial anodes.

CABLE DESIGN RELATED TO LONG DEHS In general a high voltage (HV) cable consists of an insulation system (semi-conductive insulation screens and an insulating material, typically cross-linked polyethylene: XLPE), a radial water tight (aluminum tapes within the outer sheath) or diffusion retardant design (swelling tapes), metallic screen wires and an insulating outer sheath. In addition, in order to prevent water flowing longitudinally (e.g. after a cable service failure) in the conductor, a semi-conductive sealing material are filled in between the strands. In order to avoid high screen voltages due to capacitive currents and induced inductive currents it is possible to use a semi-conductive outer sheath with sufficient electrical and mechanical properties as indicated in Fig. 2. However, the HV cable can be subjected to high temperatures in buried seabed sections combined with seawater. In order to reveal if a cable design with a semi-conductive outer sheath can be used in high temperature seawater conditions for several decades, an accelerated ageing program was started (S. Hvidsten, 2005).

Up to now the system has been installed on flowlines of lengths between 5 and 16 km. The operational experiences from these installations have proven the functionality of the DEHS, and hence the system is being evaluated for application on relevant wellstream and transport pipelines.

1. 2. 3.

The offshore market is also developing distributed subsea systems for remote control and power supply. This implies that lengths of transport lines is increasing, which means that DEHS for flow assurance will be more attractive. The feasibility of the heating system for longer pipelines has therefore been studied during the last two years. The priority of this work has been to optimize the DEHS on an approx. 50 km pipeline in the North Sea. A development programme is being carried out with focus on all relevant aspects. This paper presents the results from the work that has been done on the electromagnetic conditions for the heating cable (see Fig. 1) and the thermal model for the heating system related to the selected pipeline.

4. 5.

Fig. 2. The seawater functions as a current return path in parallel to the steel pipe. The return current will be divided between pipe and seawater. The current in the steel pipe is typically between 50 -70% depending of pipe dimensions and magnetic/electrical characteristics of the steel pipe

Paper No. 2005--JSC-413

Lenes

Prototype HV XLPE insulated Cable for a DEH system planned used in the North Sea. 1: Semi-conductive outer sheath. 2: Insulation Screen. 3: XLPE Insulation. 4: Conductor Screen. 5: Conductor with semi-conductive strand sealing material.

2

Numerical calculations of screen currents and system temperatures during production / heating by DEH system was used for determine service conditions.

NUMERICAL CALCULATIONS The first cable design for DEHS is illustrated in Fig. 3a). The DEH cable has applied an insulating outer sheath and a metallic ground screen of stainless steel. The reason for using a relative high resistive screen (grounded at both ends) was to minimize the induced currents and hence not reduce the efficiency of the heating system. A high resistive screen will not influence the thermal loading of the cable. With increasing length of the cable system, the screen voltage will increase. For long cable lengths, two alternative methods for keeping the screen voltage at acceptable levels have been suggested: 1) Increasing the electric conductivity of the outer sheath (semiconductive sheath). In this case a metallic screen may not be required. 2) Intermediate grounding of the screen. The alternative by use of a semi-conductive sheath has been considered as the most attractive for long pipelines. This design, see Fig. 2, eliminates the need for a metallic screen and the charging current is radially conducted directly through the semi-conductive outer sheath to ground (seabed). A circuit equivalent for this design is shown in Fig. 4b). In this case the sheath capacitance can be neglected compared to the sheath admittance. This requires a sheath resistivity of less than 100 ȍm (S. Hvidsten, 2005). However, this assumes that for the case of the pipeline being covered by clay on the seabed, the surrounding clay has a relative low resistivity. This will be considered in Section “Electrical Resistivity of clay”.

Calculations of Screen Voltages In order to reduce the induced voltage of the metallic screen it can be grounded at both ends. A per length circuit equivalent for DEHS is shown in Figure 4a). The impedances were calculated by using numerical software based on the finite element method. The data input for these calculations must be defined. Electrical and magnetic properties for the cable and seawater are well documented, but for the steel pipe material and surrounded soil (seabed) limited data are available and measurements have to be performed. The data for surrounding soil (clay) will be discussed in this article. Experiences have shown that the magnetic properties for pipe steel material may vary significantly even within the same batch and must be provided by measurements for each project, see presentation in the next chapter of this paper. The calculating procedure for each PI-equivalent presented in Figure 4a) are described in (L.M. Wedepohl and D.J. Wilcox, 1973). The calculations of voltages and currents along the cable were performed by subdividing the cable into short sections. Each section was modeled by the exact PI - equivalent. By this method the circuit with terminal conditions was solved using the nodal admittance method. Calculations have shown that for the cable with an insulating outer sheath and a metallic screen of stainless steel the DEHS parameters (voltages and currents) was independent of the seabed resistivity in case of water depths exceeding 50 m at 50/60 Hz. a)

a)

Copper conductor

zc zs

cd

zg

cs

Conductor screen XLPE insulation Insulation screen

b)

zc

High resistance screen

cd

Outer sheath

b)

Fig. 4.

Copper conductor

Conductor screen XLPE insulation

Per length circuit equivalent for 50/60 Hz DEHS for the case with a) Insulating outer sheath and a metallic screen of Stainless Steel. b) Semi-conducting outer sheath and no metallic screens. Zc : Cable impedance, Zs : Sheath impedance, Zg : Ground impedance, Cd: Cable insulation capacitance, Cs: Sheath capacitance.

Insulation screen Semi-conducting outer sheath Fig. 3.

Schematic illustration of two DEHS cable designs: a) Cable equipped with a stainless steel screen and an insulating outer sheath. b) Cable equipped with a semi-conductive outer sheath.

Paper No. 2005--JSC-413

Lenes

As an example, by DEH on a 10” Carbon steel pipeline of 16 km with a U-value of 5 W/m2K, about 1,5 kA is required to keep the pipe content above hydrate formation temperature of 25oC at a seawater temperature of 5oC. Fig. 5 shows the results from calculations of the screen voltages referred to ground and shows that maximum is about 1 kV and appears at about 6 km from the power input. This voltage level is considered to be acceptable considering the material properties of the outer cable sheath. Hence the cable design with metallic screen is feasible for a DEH heating length of 16 km.

3

0,8 0,6 0,4 0,2 0 0

Fig. 5.

5 10 Distance along pipeline [km]

15

Screen to ground voltage for the DEH in Fig. 3a) at 1,5 kA for a 10” Carbon steel pipeline of 16 km length with a U-value of 5 W/m2K to keep the pipe content above hydrate formation temperature of 25oC at a seawater temperature of 5oC. 24 kV XLPE insulated piggyback cable with a conductor cross-section of 1000 mm2.

system. For the 50 km long pipeline the graph for the screen voltage is shown in Fig. 7, indicating the maximum value at approx. 12 km from the near end of the heated length. Max. screen voltage [kV]

10 8 6 4 2 0 0

The relation between the cable length (with a metallic screen) and the maximum screen voltage is shown in Fig. 6. Below 10 km pipeline length the screen voltage is relative small, but from approx. 10 km the screen voltage increases more rapidly and is almost linear with length. For a 50 km pipeline the screen voltage is approx. 9 kV, which is far above acceptable level. As indicated in Fig. 5 the location of the maximum screen voltage occurs not on the geometric midpoint for the pipeline. This location depends on the electrical parameters of the

10

20

30

40

50

Distance along the pipeline [km]

Fig. 7.

The screen voltage vs. distance along the pipeline for the 50 km case.

1,2 1 0,8 0,6 0,4 0,2 0

10

0

8

10

20

30

40

50

Distance along the pipeline [km]

6 4

Fig. 8.

2 0 0

10

20

30

40

50

Pipeline length [km]

Fig. 6.

Max. screen voltage [kV]

1

Max. screen voltage [kV]

Screen voltage to ground [kV]

shall then be obtained by intermediate earthing of the metallic screen conductors/bands along the whole length of the cable. The effect of such earthing is shown in Fig. 8.

Relation between pipe length and screen voltage for the DEH piggyback cable with metallic screen.

The enhanced screen voltages for pipe lengths above 16 km indicate that it is necessary to discharge the screen currents to the surrounding seawater/seabed sediments/soil, to above excessive electric field exposure to the outer (insulating) cable sheath. The alternative design of applying a semi conducting outer sheath as a substitute for the metallic screen and the insulating sheath will hence be attractive. However, this requires a qualification of relevant materials, which shall prove long term properties when exposed to the subsea environment at the actual water pressure, temperature etc. At present the cable design with metallic screen is base case. Discharging of the screen currents

Paper No. 2005--JSC-413

Lenes

Values for the screen voltage along the 50 km pipeline.

The graph in Fig. 8 indicates that with an intermediate earthing system for the cable screen every 5 km along the pipeline, the screen voltage is sufficiently reduced. From an installation point of view it may be required to introduce splices at a similar interval. Earthing of the screen may be performed during the splicing operation of the cable. As seen from the graph the screen voltage decreases along the pipeline, and hence the distance between the earthing points can be increased. At the last section with an earthing distance of 15 km, the screen voltage will be kept below 1 kV.

TEMPERATURES SUBJECTED TO THE CABLE SYSTEM The rating of DEH is strongly dependent on the thermal characteristics of the pipe content in the different phases, i.e. the composition of gas, oil and water. When melting wax plugs or hydrates is required, this will be the dominating part of the heating time, but this case is considered to be a remote event. The operation procedures should be prepared to avoid this situation.

4

d heat development, “worst case When calculating the required situations” from a thermal point of view are considered for both the pipeline and the electric cable. The required heat in the pipe was calculated with the insulated pipe located on the seabed (surrounded by seawater) as shown in the configuration in Fig. 9a). The maximum power (current) needed for avoiding wax and hydrate formation, is then determined by the lowest observed temperature of the seawater. The cable conductor cross section is determined for a buried/rock-dumped pipe with the maximum depth of backfill or taking into account seabed movements, and the maximum seawater temperature of the field. The thermal rating of the cable must also be considered if for any reason heating is required while the pipe is in production mode. Fig. 9b) illustrates the “worst case” in respect to the thermal condition for the cable, which is limited to 90oC.

The maximum cable temperature will then occur about 35-70 hours (dependent of the configuration) after DEH is turned on. The calculations in this paper were performed with a thermal conductivity of 1W/mK for the clay. The maximum temperature occurring in the cable was calculated for different clay/rock configurations as shown in Table 1. These conditions are considered relevant for an installed pipeline / DEH system. The ‘worst case’ condition with respect to the maximum temperature (Case No. 4) is the same as that shown in Figure 9b).

Table 1. Calculation of cable temperatures. Case No.

a)

Seawater

1 Piggyback cable inside protection profile

Thermal insulated pipeline

2 3 4

Seabed

Description Pipe covered to the middle by clay and rock dumped Pipe covered to the top by clay and rock dumped Clay coverage of 0,5 m and rock dumped Clay coverage of 1 m and rock dumped

Steady State [oC]

Transient Case [oC]

62,1

89

70,5

94

80,6

100

88,5

104

The results from the temperature calculations show that at steady state conditions with no production, the temperature is limited to about 89oC. This indicates that the maximum temperature of 90oC rated for the XLPE insulated system was not exceeded for any of the four cases.

Seawater

For the “transient case”, 90oC was exceeded for three of the cases with clay coverage of 0,5 and 1 m, with temperatures ranging from 94 to 104oC. However, when the pipe was covered by clay to the middle and rock dumped the temperature was slightly below 90oC.

Max. rock coverage

Max. clay coverage

ELECTRICAL AND MAGNETIC PROPERTIES OF PIPES Thermal insulated pipeline

Piggyback cable inside protection profile b) Fig. 9. a) Pipeline and DEH cable surrounded by seawater. This case determines the heat requirement. b) Maximum coverage of the pipeline represents the “worst case” in respect to the thermal condition for the cable. An example of thermal rating for DEH for an 18” pipeline with a Uvalue of 4 W/m2K is chosen to give basis for thermal rating of the piggyback cable. The calculations were performed both for the steady state condition with no production, and for the “transient case” with immediately turning on DEH after shutdown of the production with a well stream temperature of 116oC. The required supply current was 1320A to keep the pipe content above 25oC determined by the case with the pipe surrounded by seawater. The calculations were performed with a copper conductor cross section of 1200 mm2.

Paper No. 2005--JSC-413

Lenes

Electrical and magnetic characteristics of steel pipes are important for the heat generation and efficiency of the heating system. These data are normally not available from the steel mill (pipe manufacturer). Measurements in the verification work on the concept indicate that the designation of type (“13Cr”, “Carbon” and “Clad steel”) does not identify the material characteristics properly. Experiences prove that these data should be specified within an uncertainty of ±10%. Until there is obtained a data bank for these material properties, measurements are necessary to provide data for rating of the heating system in detail engineering work. Such measurements have been made on steel pipes for projects in the North Sea. The results from measurements performed on a large number of pipejoints display significant variation indicating that material characteristics are not properly covered by the production quality programme. Table 2 shows typical measured values for different steel pipes. A considerable random variation of effective relative permeability for individual pipes within a pipeline has great influence on the variation in pipe temperatures.

5

Steel Carbon steel Clad steel 13Cr

[10-6 :*m] 0,2 0,2 0,8

Temperature coefficient, D [1/°C] 0,003 0,003 0,0008

Relative permeability 300-1500 300-1500 40-200

Subsequent to delivery from the steel mill, the pipe joints are exposed to heat treatment and grit blasting during the thermal insulation process and later to excessive strain during pipe laying (especially reel vessel). These stresses may influence the electrical and magnetic properties. Measurements indicate that the effect of stresses was different on Carbon steel and 13Cr pipes. In general Carbon steel pipes were more affected by the stresses than 13Cr pipes. Measurements show that the generated heat in the pipe may be reduced by approx. 20 % for Carbon steel pipe after grit blasting. For 13Cr steel pipes the grit blasting has insignificant influence. The reason why the electrical and magnetic properties for Carbon steel materials are more affected by external stresses can explained by the depth of penetration or skin depth, G , given by Eq. 1. The skin depth is approx. 1 mm for Carbon steel materials and 9 mm for 13Cr at 50 Hz, see Table 3. Hence the outer surface of the steel pipe will be active in a DEHS in a depth of twice the skin depth. Typical wall thickness for pipes is from 10 – 20 mm. For a Carbon steel pipe the main current passes through the outer 1 – 2 mm of the pipe cross section, while for a 13Cr pipe almost the whole pipe cross section is an active conductor. It is therefore reasonable that Carbon Steel materials are more sensitive than 13Cr for stresses (thermal and mechanical) on the outer surface in regard to DEH properties.

U

G

U

P0 Pr f

2

A1 A2

1,6

B1 B2

1,2

0,8

0

10

20

30 40 50 T emperat ure [°C]

60

70

Fig. 10. Results from measurements of resistivity of four different sample of clay taken from two different locations (A and B) in the North Sea.

CONCLUSIONS Numerical calculations of screen voltages for the cable design with insulating sheath show that the voltage can become several kV for long pipelines. Such high voltages can not be accepted. The screen voltages can be reduced by earthing the screen at some locations along the pipeline.

Skin depth [mm] 1 1 9

The new cable design with a semi conducting outer sheath requires no metallic screen. This design requires sufficient electrical conductivity of surrounding seabed.

ELECTRICAL RESISTIVITY OF CLAY The alternative cable design by use of a semi-conductive sheath and no metallic outer screen for long pipelines, requires sufficient conductivity of surrounding seabed to avoid high voltage at the sheath. In general

Paper No. 2005--JSC-413

As see from the results the electrical resistivity of the clay (seabed) was sufficiently low. This indicates that the capacitive currents can easily be transferred to seabed (seawater) avoiding voltage buildup at the sheath.

0

Table 3. Skin depth for different steel for frequency of 50 Hz.

Carbon steel Clad steel 13Cr

Fig. 10 shows results from measurements of the resistivity of the clay taken from two different locations in the North Sea. It was seen that the resistivity decreases with temperature. At about 4oC the resistivity is approximately 1,3 ȍm, decreasing to about 0,3 ȍm at 70oC which is close to that typically measured for seawater (0,2 ȍm). No change in resistivity was observed when the samples were measured again at 20oC after the measurements at elevated temperatures.

0,4

- skin depth - electrical resistivity - permeability of air - relative permeability - frequency

Steel

As the electrical resistivity can be dependent upon water content in the clay samples, also the water content was measured both prior and after the resistivity measurements by using the weight method.

(1)

S P0 Pr f G

Measurements of resistivity have been performed on samples taken from the seabed close to relevant oil fields in the North Sea. The resistivity was measured at different temperatures from about 4oC up to 65oC.

m]

Resistivity

the resistivity of surrounding seabed should not be greater than the max. limit of the cable sheath, i.e. less than 100 ȍm (S. Hvidsten, 2005).

Resistivity [

Table 2. Electrical and magnetic properties for steel pipes.

Lenes

Measurements show that the electrical resistivity of the surrounding seabed of clay was less than required for the semi-conductive outer sheaths of the cable. This indicates that the capacitive currents can

6

easily be transferred to seabed (seawater) avoiding voltage buildup at the sheath. Heating did not cause any drying of the clay in contact with seawater. This indicates that the relative low electrical resistivity of the clay will likely remain during service. The maximum temperature subjected to the DEH cable was strongly dependent upon seawater temperature, clay coverage and operation of the DEHS. Special efforts are required not to exceed the cable temperature limit of 90oC. Carbon Steel materials are more sensitive than 13Cr for thermal and mechanical stresses on the outer surface in regard to DEH properties. Measurements show that the generated heat in the pipe may be reduced by approx. 20 % for Carbon steel pipe after grit blasting. For 13Cr steel pipes the grit blasting has insignificant influence. A reduction of generated heat in the pipe requires an increase of the system current to keep the pipe above the hydrate formation temperature.

Paper No. 2005--JSC-413 View publication stats

Lenes

ACKNOWLEDGEMENT The authors want to thank STATOIL ASA for permitting to publish the research results from the actual project.

REFERENCES J. K. Lervik, H. Kulbotten, A. Lenes, G. Klevjer: “Concept Verification of Direct Heating of Oil & Gas Pipelines”, Phase II,SINTEF Energy Research TR A4588, ISBN 82-594-1126-1, 1998-02-18. S. Hvidsten, A. Bruaset, K. Olafsen, L. Lundegaard, A.H. Børnes: “HV XLPE Cable Design for Direct Electrical Heating of Very Long Flowlines”. ISOPE 2005. L.M. Wedepohl and D.J. Wilcox, “Transient Analysis of Underground Power Transmission System; System-Model and Wave Propagation Characteristics”, Proc. IEE, vol. 120, No. 2, February 1973, pp. 252259. Lervik, J. K.; Kulbotten, H.; Klevjer, G.; Lauvdal, T. ”Direct Electrical Heating of Subsea Pipelines”, ISOPE-93.

7