HomeRail NewsSmart cable monitoring

Smart cable monitoring

Listen to this article

Signalling power cable failures can be very disruptive to railway traffic. Even short-term interruptions can have a wide impact on performance across the network.

With digital railway technology seeking to increase capacity, the need to detect emerging power distribution cable insulation failures, and the ability to locate and quickly recover from these failures and the associated perturbations, becomes ever more critical.

This has resulted in a radical rethinking of how digital technology and data analytics using artificial intelligence can be integrated into an open architecture using flexible and agile multi-tier platforms for future cable monitoring systems.

Nigel Edwards, Network Rail’s professional head of power distribution HV/LV, said: “We are seeking to build on today’s technology that can be used to support early warning of degradation and combine condition assessment of signalling power cables along with the ability to pinpoint the position of failures.

“This will allow us to revaluate our asset management plans from reactive to predictive regimes that will drive safety, availability and performance improvements. It will support the life extension of existing power cables networks as an alternative to replacement of cables which may be approaching the end of their life, driving cost efficiencies.”

Signalling power systems

An essential sub-system of any railway signalling or traffic management system is the signalling power supply (SPS). This typically includes the following subsystems: source of supply, principal supply point (PSP), signalling power distribution system (SPDS) and signalling equipment. The four subsystems may be discrete entities or contained together within a building, such as a signalbox or rail operating centre (ROC). The majority of distribution systems are autonomous cable networks that are not interconnected although, in new installations, an increasing number are double-end fed systems.

The railway signalling power distribution subsystem is one of the largest low-voltage power distribution networks outside the utility network in the UK. Thousands of kilometres of power cable interconnect the power supply points and signalling equipment housings positioned along the railway via functional supply points.

A signalling power distribution sub-system typically also includes distribution and function supply points transformers, low voltage switchgear, distribution interface transformer assemblies (DITAs), automatic re-configuration systems and feeder protection equipment. A typical configuration of the network can be seen in the diagram.

The reasons for an IT system

Three types of electrical supply earthing systems are defined in BS7671 (typically referred to as the wiring regulations), defined by the two-letter codes TN, TT and IT. The first letter identifies the way that the main power source or transformer is earthed, the second letter how components and subsystems are earthed.

In a TN system, one of the points on the power source or transformer is directly connected with earth (T=terra). All of the equipment on the system is connected back to this single earth connection via the Network, as is common on domestic or simple commercial power systems.

A TT system has every subsystem independently earthed locally, with no ‘earth wire’ connecting the equipment back to the power source. This system is frequently used in telecommunications as it removes any interference from the common earth wire.

The earth connection on an IT system is Isolated from the power source but all of the equipment is independently or collectively earthed (to Terra).

The railway signalling power distribution sub-system commonly uses an IT electrical system, sometimes referred to as an unearthed power supply. The benefit of an IT electrical system over a TN system is that it is tolerant to earth faults.

Earth faults in TN systems cause the protection (fuses) to operate and hence cause the power supply to switch off. This could have significant disruptive safety impact on the railway signalling system.

There are many potential causes of earth faults in a railway environment: rodent damage (rats chewing the cable), cable strikes by tools, crushing due to railway plant, cable joint failure, water ingress, poor cable installation and aging of the cable insulation over time.

Railways share the use of IT electrical systems with other mission-critical systems such as hospital operating theatres, warships, oil and gas platforms, nuclear installations and airport runway lighting.

IT electrical systems continue to operate under an earth fault condition, but such systems are required to be fitted with cable insulation monitoring systems which continuously monitor and detect the presence of earth leakage. When the measured earth fault reaches a threshold, an alarm is sounded by the insulation monitoring system which would normally give maintenance teams advance warning of a developing insulation fault, enabling action before a further fault leads to failure, or more importantly, before a second earth fault can occur on the system.

There is also a safety imperative to repair earth faults quickly before the electrical system presents an electric shock risk. If further earth faults also occur, then the system becomes vulnerable to becoming disconnected by the protection system (fuses).

Insulation monitoring systems form a key part of managing the signalling power distribution network. As they help to keep the signalling systems running, they have a direct impact on train safety and performance.

Asset management challenges

Whilst signalling power systems may not be unique in using IT-configured distribution, their deployment on the railway makes them distinctive when compared to other mission-critical applications. They typically comprise multiple cable feeders or circuits in a range of outdoor environments including surface cable troughs, under-track crossings (UTX) or being directly buried in the ground.

Cables can be subjected to wide temperature variations, humidity changes and water immersion along the length of the cable network. Cables and joints are also subject
to mechanical stresses from the vibration of passing trains and, sometimes, from the movement of cutting or embankment slopes. Typical network cable lengths (multiple feeders) can range from 15 to 70km. Individual signalling power supply feeders can range in length from three to 10 kilometres.

Network Rail asset manager Graeme Beale explained: “Today’s insulation monitoring systems may help in identifying the presence of an insulation failure but, as many of the railway’s historic signalling power distribution networks are so extensive, it becomes very difficult and challenging to locate the position of the indicated earth faults without undertaking a binary chop.

“Binary chopping is about successively switching off sections of the signalling power distribution network to allow more precise location of faults. Historically, this rudimentary form of fault-finding involved isolating a section of the network and observing whether the circuit protection operated when the supply was restored to the remainder of the network. This effort, and the time it takes, has a hugely detrimental impact on signalling systems, which cannot be operated under such fault- finding scenarios.

“Nowadays, we have more sophisticated and less disruptive techniques available using hand- held, portable earth-fault leakage detectors that work in tandem with the fixed earth-leakage monitor at the supply location. However, this still requires fault-finding teams to be mobilised at short notice, spending many hours searching the railway for faults which could be hidden in UTXs or along the cable route inside cable troughs. The requirement to protect staff from moving trains means this often has to be done at night and in all weathers. It is very much like finding a needle in the proverbial haystack!”

Some signalling power supply networks are up to 60 years old and are increasingly subject to earth faults, due to the aging of cables, which are then difficult to trace.

Such large networks have a high leakage capacitance to earth – a function of length and the age of the cable. Under earth fault conditions, it becomes even more critical to detect, locate and repair earth faults before it presents an electric shock risk.

Cross-industry engagement

As a result, Network Rail has developed a new product specification – Insulation Monitoring and Fault Location Systems for Signalling Power Distribution Systems (R/L2/ SIGELP/27725). This is the company’s new vision for multiple-tier smart cable insulation monitoring to drive performance and asset life extensions.

The new insulation-monitoring standard is the result of an elicitation and stakeholder engagement across the industry and supply chain. A two-year review and remote condition monitoring project, delivered by RSSB in collaboration with the Birmingham Centre for Railway Research and Education, helped to identify constraints and limitations with current technologies and identify methods that could be used for future technologies.

A wide industrial-sector review of cable monitoring systems used for other mission-critical applications was carried out. The results gave insights that could deliver huge benefits if some features of these systems could be transferred to a railway environment. They also shaped the rationale behind some of the requirements for future insulation monitoring systems. A holistic picture of the health of the cable, along with a rich data set, is the key message.

Product and system requirements

The new standard now specifies that insulation-monitoring and fault-location systems shall be suitable for configuration into the three network architectures, which are themselves subdivided into three tiers based on functionality, providing network monitoring (tier 3), sub-network monitoring (tier 2) and sub-network section monitoring (tier 1).

Requirements include the ability to measure, monitor and locate resistance and capacitance for all three tiers. In addition, fault location systems shall have the capability to detect and locate line-to-earth insulation faults on a network. The option to detect and locate other network faults are also specified including line-to-line short-circuits, intermittent short-circuits (arcing) and open circuits.

To determine the degree of cable contamination resulting from moisture ingress and/or insulation damage, requirements for deriving the polarisation index and dielectric absorption ratio have also been specified. In order to complete the exercise and determine the full health of the cable, additional parameters such as voltage, line-to-earth voltage, line current, power factor, temperature and humidity may also be included.

By collecting this data, and highlighting any trends on cable insulation resistance, it will be possible to more accurately predict the health of the cable. This is particularly useful when legacy cables are being life-extended.

Where data trending is used, the ability to perform complex analysis using resistance, capacitance and other system parameters could be incorporated to derive condition status.

Investigation into power system remote monitoring alarm revealed a damaged 650V cable lying in a flooded trough route.

Patterns may emerge that cannot be identified solely by the use of human experience and intuition. Complex analysis of the data, using fuzzy logic and/or learning artificial intelligence neural networks, could be used to predict imminent cable failures and to generate early alarms.

In the longer term, the aspiration is to build capability to rejuvenate damaged cable through cable self- repair properties or through the application of digital signals that excite the structure of the insulation to drive self-healing.

Business realisation

The new platform and system architecture can be used to make the case for reducing or eliminating the need for disruptive cable testing. This would drive significant OPEX savings in maintenance costs and railway disruption. It also could provide alternative cost-effective power supply architecture, in place of other systems, on some installations where the business case is being challenged.

This new specification and the work behind it, forms part of Network Rail’s aim to drive innovation in the supply chain. In fact, these technologies and developments could find outlets in other fields and industries.

Many suppliers and specialists have already been involved in the work behind the new product specification. However, the team is seeking further collaborations and will welcome applications for product approvals. Mark Downes, head of engineering (Infrastructure Projects Central), stated: “We are keen to support the evaluation and introduction of innovative products and processes that drive technology and business change on our infrastructure renewal and enhancement programmes.”

Written by Tahir Ayub

Tahir Ayub is a programme engineering manager (enhancements) at Network Rail Infrastructure Projects (Central) and has been the technical lead for the development of the new standard.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.