Fibre optic monitoring takes the heat off cables

20 December 1999



Better information on the thermal behaviour of critical circuits might have avoided the electricity supply failure in Auckland in early 1998. A direct buried 110 kV circuit linking the city's Mt Roskill and Liverpool St substations, under construction at the time, has just been completed. This circuit has fibre optic distributed temperature sensing (DTS), which is used to generate monthly performance reports.


Installation and commissioning of a direct buried 110 kV circuit for Mercury Energy New Zealand (now known as Vector Limited), was completed by Olex Cables in December 1998. The circuit is approximately 9.2 km long and is installed from the Mt Roskill outdoor substation to the Liverpool St Substation in Auckland City.

Due to variation of the soil types along the cable route the trench was backfilled with a selected mix with known thermal properties when fully dried out. Soil types along the cable route varied from gravel to volcanic rock

To detect regions of higher thermal resistivity and protect the system from possible overheating, and also to maximise the use of the circuit, a York Sensors DTS 800-MR distributed temperature sensor was installed at the Liverpool Street substation. This unit has a range of 12 km, a spatial resolution of 2m and is accurate to ± 1°C.

As part of the contract agreement, Olex cables provides a monthly report to Vector on the performance of the 110 kV circuit using data obtained from the DTS and load data provided by Vector. The DTS information and load data is accessed via a modem link over a normal telephone line and is analysed by Olex engineers located in Melbourne (which is some 2000 km from Auckland). This analysis has several objectives:

  • The temperature trace is analysed to accurately locate regions where the temperature is not uniform in relation to the rest of the circuit ie hot spots;
  • The trace is viewed over a 24 hour period to ascertain the consistency of any regions noted above to establish the worst case scenario;
  • The actual mean and maximum current carrying capacity of the circuit is determined. This is done by sorting the data using software associated with the DTS 800. This software has as one of its functions, the capability of being able to convert the DTS data to an ASCII format. Olex Engineers can then use software applications such as Microsoft Excel and MathCad to derive the current rating under the current set of operating conditions.

To maintain a high degree of accuracy each trace represents the average of many traces. Under the current system of monitoring each fibre is sampled about every two hours as the cable has a high thermal time constant.

As was expected, the majority of the trace indicates that the thermal conditions along the cable route are uniform. There are, however, certain regions along the cable route where the cable temperature is higher than average. Initial investigations have indicated that factors beyond the control of both Vector Limited and Olex Cables may be having an adverse effect on the cable behaviour. Vector Limited is now able to take corrective action before the loading on the 110 kV cable becomes such that these regions limit the circuit rating.

These regions are currently under investigation and a programme to determine what factors may be causing the excessive warming is being developed. The main theory being postulated is that some specific electrical services crossing over the cable are producing significant heating of the surroundings and the programme will primarily focus on the services in the affected area.

When DTS systems are used to monitor buried and subsea power cables, the resultant temperature profile enables cable owners to determine the size, location, and potential impact on rating performance of any hot spots caused by localised environmental conditions. For example, a 12 km long power cable can be profiled with sampling points every metre using just one optical fibre. The impact of changes to the cable environment caused by seasonal conditions, climate, and land development, can be continuously tracked to ensure the cable is always correctly rated.

Optical fibre sensing

A single optical fibre has the ability to act as a distributed temperature sensor with potentially many thousands of individual measurement points. Optical fibre sensors for power asset monitoring have significant advantages because they are totally immune to EMC interference and can also be used in hazardous areas.

Temperature measurements are achieved by analysing the Raman back-scatter signals caused when a laser pulse is travelling in optical fibre. The position of each measurement point is determined by its time-of-flight from laser emission.

The laser pulse is transmitted into the fibre via optical directional couplers which allow the returning back-scatter signals to be filtered into a highly sensitive dual-channel receiver where the Raman signals are detected and converted into a digitised format specifying temperature and position.

It is possible to produce a complete 12 km profile to a resolution of 1°C within several minutes, but as the typical time constant of a buried cable is measured in hours or even days then a longer period is usually set. Fast measurements are ideal for obtaining a "snap-shot" for hot spot detection.

Over relatively short measurement distances the sensing fibre is connected to the DTS unit as a loop using the double-ended processing technique. In this method the laser pulses are injected into the fibre from both ends. This method provides automatic calibration for fibre losses, splices, connectors, and microbending effects. Further, in the event of an in-service fibre problem, the sensing fibre can still be measured from both ends up to the loss point providing continuous monitoring.

Typically, multimode fibres are useable for DTS in applications up to 12 km fibre length without significant reduction in measurement precision. This allows up to 6 km of cable to be monitored using the double-ended technique or 12 km using a single-ended connection. In the case of the latter it is necessary to calibrate the remote end of the fibre during installation by independent means.

For temperature monitoring applications above 12 km then it is advisable to consider using single-mode fibres which have a lower attenuation. Current DTS technology based on single-mode fibres enables ranges of up to 30 km to be measured at 2 metre sampling intervals.

A significant advantage for single-mode DTS technology is the ability to provide temperature profiles of power cables where a separate communication fibre optic cable has been installed in the same trench. This allows potential "retrofit" application of temperature monitoring to cables which do not have dedicated sensing fibres.

Practical application

A circuit rating is primarily governed by the thermal resistivity of the medium surrounding the cables. For direct buried cables the surrounding medium in the trench in which the cables are buried will normally be a backfill that has a guaranteed maximum thermal resistivity. The thermal resistivity of soil outside the backfill area will depend on its structure and to a very large degree its moisture content.

DTS collects and measures cable temperature data and provides a means by which a cable rating can be compared to its operating temperature. Analysis of the collected data, in conjunction with the system load data, using IEC 287 through software packages such as Excel, MathCad and/or a finite element analysis packages such as Sirolex, enables the user to establish and then predict the cable behaviour along the entire route.

To establish a relationship between the distance along the fibre route compared to the cable route, a Non Temperature Sensitive (NTS) trace can be downloaded from the DTS. A NTS trace is basically an OTDR trace and the location of splices in the optical fibre can be established in relation to the cable route. Then, in conjunction with a route drawing which shows and lists the services that cross over the system, hot spots can be very accurately located. As an added measure, a zone can also be programmed into the DTS that will trigger an alarm message if the cable temperature in that zone exceeds a predetermined temperature limit.

Fitting the fibre

The cable used for the Auckland project was originally designed for use installed in air in a tunnel. The design of the cable was such that graded index 50/125 µm optical fibres in a loose tube were provided with the helical copper wire screen of the cable. Incorporating the optical fibre in the copper wire screen enables the conductor temperature to be more accurately calculated and the loose tube is protected from mechanical damage that may occur during handling of the cable.

The cable was installed in a horizontal flat formation and the fibre spliced together in each joint bay. Splicing the fibres together in such a fashion enabled each phase of the circuit to be monitored for its entire length thus ensuring that any problem regions are not only accurately located but can be verified using the data from the other phases.

A single length of fibre was also strapped to the outside of the phases for two sections of the cable route and then laid in a loop in the ground. The purpose of this additional fibre is to measure the temperature along the outside of the cable and also to provide a reference for the ground temperature.

Both methods of fibre installation for the purpose of temperature measurement have some advantages and disadvantages. Installation of the fibre within the power cable enables a far quicker response to changes in load and hence follows the cyclic behaviour of the cable temperature better than a separate fibre cable that is strapped to the outside of the power cable.

The fibre that is incorporated into the cable (internal fibre), displays behaviour that is consistent with the cable loading. On the other hand, the fibre that is strapped to the outside of the power cable (external fibre) has a smoothing effect, indicating little change in temperature due to cyclic loading of the cable. This fibre, therefore, is more representative of the trend in cable temperature, which as expected, follows the changes in the temperature of the installation environment.

Numerical analysis

The international standard IEC 287 is used as a basis for all current rating calculations. The permissible current rating is given by:

The thermal resistivities of the cable components are known (T1, T2 and T3), thus the only variables that do significantly vary are the AC resistance of the conductor (R), and the external thermal resistance (T4), which is directly related to the thermal properties of the medium surrounding the cable.

Given that the backfill placed in the trench has a known thermal resistivity, it is assumed for the purpose of the rating calculation that only the thermal resistivity of soil surrounding the trench can vary.

Having the optical fibre integrated into the copper wire screen of the cable enables the conductor temperature to be calculated accurately. The ambient ground temperature is obtained from a buried loop of fibre, These parameters, as well as the load, are then inserted into the IEC equation. The soil thermal resistivity is then varied until the calculated current rating matches the measured current. Once the soil thermal resistivity has been calculated the conductor temperature is reset to the maximum operating temperature and the circuit rating determined.

Sampling each phase on a regular basis is enabling the cyclic (daily, monthly, and eventually yearly) behaviour of the Auckland cable to be established thus providing a basis from which the cable's thermal behaviour can be analysed. The chart indicates the cyclic nature of the cable loading and also clearly indicates that as the Auckland soil temperature cools as winter approaches, the cable operating temperature is also decreasing. As more data are gathered a seasonal profile of cable behaviour that may be detrimental can be predicted.

Systems are currently being developed to directly read circuit load data as well as data generated by the DTS via a data acquisition unit. This real time data can then be analysed and manipulated to predict current operating condition and capacity, as well as capacity for future loading for a number of different continuous and short term loading scenarios. This RTTR (real time thermal rating) system could automatically generate alarms to indicate possible dangers and thermal instabilities and could give prior warning of such situations.

Conclusions

The experience of Vector Limited in New Zealand has highlighted the need to be able to monitor the thermal behaviour of critical circuits. As the demands for assets to be fully utilised increases, it is becoming imperative to have a monitoring system such as DTS in place to enable the prediction and detection of possible problems such as thermal runaway.

Such a system also allows trends to be monitored, thus giving time for appropriate actions to be taken prior to emergency situations arising. The data obtained from the DTS enable the distributor to maximise the use of a circuit during periods of peak loading, eg mid-summer, when the soil around the cables may begin to dry out.

The real-time nature of the data allows the cable behaviour to be carefully and closely monitored, thus allowing optimum use of the circuit.



Linkedin Linkedin   
Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.