Proactive Condition Monitoring Could Predict and Prevent Electricity Network Failures

Proactive Condition Monitoring Could Predict and Prevent Electricity Network Failures

By Muhammad Shafiq

In today’s world, we all are well aware of the significance of electricity in our lives. Households, hospitals, transportation, education, communication, banking systems, security services, online services, all kinds of sophisticated as well as traditional industrial processes, even the availability of basic needs such as air and water, are highly dependent on the functioning and flawless supply of electrical power. Therefore, reliability of the power supply is a major concern for the electric utilities.

Usually an electric utility consists of a huge infrastructure extended across a wide geographical area (hundreds of kilometers) transporting power from generation sites to hundreds of thousands of individuals as well as bulk users. Transmission lines, substations, and distribution lines are the major components of a utility’s supply network. Considering the example of a typical supply network, an Australian utility* manages 300 substation zones, where each zone typically consists of 6-12 feeders, having 80-150 transformers serving 50,000-100,000 customers. Supply networks operate 24/7 and are always exposed to various environmental and operational stresses which may cause failure to the components.

Insulation Faults

Unplanned power interruptions have a considerable impact in terms of the discontinuance of the various services and the economic loss along with consumer dissatisfaction. As a highlight, in the US electricity sector, power outages and disturbances cost the economy, on average, more than $80 billion annually and sometimes as much as $188 billion in a single year (reported in 2011)**.

Substations and distribution networks are the most interactive part of the supply networks. Transformers, switchgears, protection and measuring equipment, and power lines such as cables and overhead lines are critical components in a distribution network and are highly prone to electrical insulations defects. The percentage of the failures caused by insulation breakdown in cables, cable accessories and transformers has been reported as 89%, 87%, and 84%, respectively. Insulation defects start at a certain stage, progress with time and eventually lead to a failure. Condition monitoring is an efficient tool to predict and prevent an incoming threat to the network components.

Insulation defects initiate the emission of partial discharges (PD) which are high frequency electromagnetic signals resulting in the current pulses of few milliamps amplitude. The current pulses propagate along the lines and can be detected by using suitable sensors. Currently, various techniques have been developed by researchers to detect and locate the PD faults. However, these techniques are mostly available for individual components such as transformers, generators, switchgears, and single line sections. There is a lack of available techniques which can be integrated to monitor the wider part of the network in order to take care of the several components.

Prediction and Prevention

During our work on this topic, we realized that while considering a wider part of the cable network, it has to be taken into account that a cable feeder normally consists of multi-section and multi-branch cable sections. Typically, different types of cables are present along the feeder path and therefore a composite cable infrastructure can be seen there. When the insulation faults occur in such network and PD signals travel along the line, the fault detection and locations tasks can be performed using suitable sensors and methods. Please be noted that fault is a pre-stage of failure which still allows the affected component to work for some more time. The focus of the proactive monitoring is to determine, if there is any fault, where the fault is located, and how far the damage is done. This leads to establish a maintenance strategy to cope with the fault, in order to prevent the incoming failure.

Due to complexity of the network topology, multiple sensors are needed to cover all the components along a cable feeder. The sensor should be installed in a symmetrical manner. The location of the sensors’ installation is of critical importance. The data measured by all sensors is processed. In case, there is faulty component, the pair of the sensors around this component will notify the threatening situation pointing out the victimized component. The notification is sent to a maintenance crew which further executes an onsite diagnostics on the identified component to conclude; replace, repair, or wait. Such approach can prevent the loss of the components and unexpected interruptions.

This article presents the concepts of an integrated monitoring technique for a part of a typical cable feeder. Nowadays, the modern grid is becoming different from the conventional grid. Under the vision of a smart grid, inclusion of renewable energy sources distributed across the network area makes the network topology more meshed and complex having bidirectional flow of power. Along with smart metering and other progressions, there has been an increasing interest for development of improved and automated, monitoring and diagnostic solutions to incorporate the self-healing feature in the smart grid technology.

Further Readings:

Shafiq, et al. “Integration of online proactive diagnostic scheme for partial discharge in distribution networks.” Dielectrics and Electrical Insulation, IEEE Transactions on 22.1 (2015): 436-447.

Shafiq, et al. “Partial discharge diagnostic system for smart distribution networks using directionally calibrated induction sensors.” Electric Power Systems Research 119 (2015): 447-461.

Photo taken in Power system and High Voltage Engineering Laboratory at Aalto University.

Learn more about PreScouter at www.prescouter.com.

Never miss an insight

Get insights delivered right to your inbox

More of Our Insights & Work

Never miss an insight

Get insights delivered right to your inbox

You have successfully subscribed to our newsletter.

Too many subscribe attempts for this email address.

*