Obtaining real-time operating data on power plant and substation equipment has never been easier. The sensors, transmitters, and systems needed to monitor important parameters are readily available and highly cost-effective. The ability to identify problems early and make repairs prior to catastrophic failure improves system reliability and saves utilities money.
Remote monitoring of electrical components at power plants and utility substations is widespread today and growing. POWER interviewed Tony McGrail, Solutions Director of Asset Management and Monitoring Technology at Doble Engineering, and Ken Elkinson, Online Monitoring Integration Expert at Doble Engineering, to better understand how remote monitoring solutions are being implemented and managed throughout the industry. Their responses provide insight on some of the most helpful technology being used in the field today.
POWER: What are the primary objectives and benefits of implementing remote monitoring at power plants and substations?
McGrail: One of the primary objectives is to gather information about the health of critical assets more frequently than what’s been possible in the past. By doing so, you can extend the lifecycle of assets—a major financial benefit and incentive.
The ability to keep the equipment in service while still gathering data—without scheduled maintenance and downtime—also empowers end-users to make more strategic, proactive, and cost-effective decisions about their assets. Think about it in terms of the tire pressure gauge in a car. Instead of driving until your tires fail, you receive a signal when tire pressure is low so you can plan how to fix it. Remote monitoring provides the same type of benefit and prevents making decisions about asset viability by default. Whether you’re planning for short-term conditions or long-term viability, you can determine the right types of sensors, monitors, data, and plans needed to support that goal.
POWER: What types of sensors are used to collect data and how is data typically transmitted from the equipment to the monitoring system?
Elkinson: There’s a diverse array of sensors available, contingent on the required measurements and assets being monitored. For example, a bushing tap adapter should be used on a transformer bushing to measure leakage current, indicating the bushing’s condition over time. However, for some high-frequency applications such as partial discharge (PD), transducers and antennas should be used to generate appropriate signals.
When it comes to power transformers, oil analysis provides valuable insights into potential problems. Frequent oil sampling can be done cost-effectively, enabling field engineers to monitor conditions on an hourly basis, thereby preventing failures. Other assets like breakers require different sensors to detect issues.
The most effective overall approach to condition monitoring involves identifying potential failure modes and measuring the parameters associated with these modes. Similar to how we might use a human temperature measurement to determine whether someone has a fever—a rise in temperature can signal a potential problem, but an effective diagnosis requires more data.
Data transmission is also situational. The most reliable method involves using wired connections from the sensors to the monitoring equipment. In scenarios where wired connections are not feasible, technologies like Bluetooth and Wi-Fi come into play. The choice of communications employed often depends on the level of comfort the user has with wireless technology, its security, and the robustness of the power supplies to the communications.
POWER: What tools and methods are used for real-time data analysis and interpretation in remote monitoring systems?
McGrail: The tools and methods for remote monitoring depend on what’s expected for a particular asset. When that’s established, a user can place parameters on the typical ranges for load-side voltage and current, cooling status, conductivity, vibration, harmonics, and more. The user will need to decide what constitutes an unusually high or low result to generate a “high-level” alert. Setting up more than one parameter enables a user to move beyond detection to the diagnostic stage by comparing results from multiple tests such as temperature, pressure, or load, and how they interact.
In terms of best practices, place the analysis tools as close to the data source as possible for the most accurate and rapid results. It’s not always practical to send all data back to a central point for analysis and decision-making, especially in critical situations. Back to the tire pressure analogy, if you’re driving through a tunnel without service, and a tire issue arises, you want immediate access to that information. Analysis tools should be readily available at the site of data collection, enabling the rapid detection of issues and the ability to take immediate action.
Additionally, with more powerful computing, utility and power organizations can analyze full fleets of “sister” units, or similarly designed transformers or circuit breakers via remote monitoring. Robust, all-inclusive condition monitoring systems that link testing data and analytics provides teams with a comprehensive view of asset health.
POWER: How is abnormal or faulty equipment behavior detected and addressed?
Elkinson: While remote monitoring systems may flag “abnormal” conditions, identifying true abnormalities requires an understanding of the asset’s historical data and general best practices. Adhering to IEEE [Institute of Electrical and Electronics Engineers] or IEC [International Electrotechnical Commission] standards and guidelines establishes a solid framework for defining individual parameter limits—such as temperature and load—for specific assets and applications. Analyzing both conditional and operational data provides context for expected outcomes based on recent usage behavior. This holistic approach is important for accurate assessment and effective response to anomalies within the system.
For example, if a transformer is operating at full load, one could anticipate an increase in its temperature. However, if unusual conditions persist, then an engineer should step in with some form of intervention whether that’s cooling, oil replacement, fault detection, or replacement entirely. This demands a comprehensive understanding of the monitored equipment and the ability to anticipate its potential behavior under varying external conditions.
POWER: What security measures are in place to protect the data and the remote monitoring system from cyber threats?
McGrail: By design, condition monitoring equipment should be put through penetration testing to ensure that the system is hardened to cybercriminal attacks. Basic protective measures include a strong username and password, and automated logouts of the interface after a certain time period. Physically locking the equipment can also be used to prevent unauthorized personnel from accessing as well. In addition, it’s more difficult to hack into a hardwired device, so minimizing wireless communications is another protective measure.
Some organizations are more well-versed in cybersecurity measures than others, but it’s important that each part of a remote monitoring system can operate completely within each firewall. Many power and utility organizations are increasingly using cloud applications and digital twins to better visualize their assets. However, these applications can be susceptible to hacks such as data poisoning, so it’s critical to remain vigilant. Beyond these measures, organizations can also maintain a list of addresses for each device expected to be on a network and apply tools that prevent unknown addresses from joining to further limit outside parties.
POWER: How does remote monitoring help in predictive maintenance of power plant and substation equipment?
Elkinson: Remote monitoring facilitates a shift in asset maintenance strategies from intermittent manual intervention to a continuous monitoring of asset degradation and more strategically planned maintenance. The greater the volume of data available about an asset or a fleet of assets, the higher the probability of preventing excessive deterioration and/or failure. This approach also reduces the resources—time, energy, and funds—required for hands-on testing, which can inadvertently lead to further damage. In short, remote monitoring promotes efficient and targeted maintenance, emphasizing the necessity to maintain what is required, rather than solely following manufacturer guidelines established decades ago.
POWER: Can you provide examples where remote monitoring prevented equipment failure or minimized the consequences of failure?
McGrail: Remote monitoring continuously helps teams identify asset status, future performance, and in some cases, the need for intervention to prevent failure in assets like bushings, transformers, circuit breakers, and cables. “Hard” failures can be classified as explosions or sudden failure, and “soft” failures would be considered instances in which a team predicts failure is approaching and takes it out of service to examine and confirm the need for replacement.
For example, online monitors can detect broad levels of gases in a transformer insulating oil, which can determine deterioration after further sampling. In one case, a dissolved-gas-detection monitor for a 48-MVA 1984-vintage unit transformer provided a low-level alert as indicated by the green background shown in Figure 1.
1. A low-level alert was triggered by a dissolved-gas-detection monitor, prompting operators to investigate the problem further. In the end, the transformer was taken out of service and repaired before catastrophic failure occurred. Courtesy: Doble Engineering
While the change was not significant, the shift did not appear related to the changes in transformer load or operation. The operations team reviewed the data and noted that the unit also had previously high furfural levels and decided to take samples for laboratory analysis. The tests revealed signs of paper aging, and a high winding resistance measurement on the low-voltage (LV) windings, which required further inspection and appropriate draining of the insulating oil.
The result? One of the LV winding leads had significant paper aging, including blackening of the center phase lead and debris from paper disintegration at the base. The team estimated that the lead had only hours before the insulation completely disintegrated, causing potential catastrophic failure. Despite the DGA [dissolved-gas-in-oil analysis] alert displaying a low level, further review for context on the transformer based on historic loading and oil test data revealed a significant issue.
In another case, an insulating oil monitoring system revealed “high-gas” readings for an oil-insulated cable on an expansion tank located next to a circuit breaker. After several additional online tests, a PD survey revealed a potential issue in the form of a PD source at the line end of the cable. The team decided to de-energize the strip and remove the cable sealing end to locate and repair any damage. As expected, the inspection revealed paper deterioration, and the cable was refurbished and returned to service.
Before returning the cable to service, the team took an oil sample at the line end, with an expectation it would confirm high dissolved gas levels. However, the result showed no indications of high gas levels. Therefore, the team decided to extend the outage to strip down and examine the cable sealing at the breaker end. They ended up discovering burn marks within the insulation—incipient insulation breakdown, which would have led to a potentially catastrophic failure if left in service (Figure 2).
2. Testing results led a team of engineers and technicians to investigate potential issues associated with an oil-insulated cable. They found burn marks, indicating insulation breakdown, and repaired the problem prior to failure. Courtesy: Doble Engineering
Successful remote monitoring, and condition monitoring at its core, requires a solid foundation of communication, thorough planning, and swift action. While it’s convenient to consider a simple path from data collection to diagnostics and intervention, the reality is different. Achieving success in remote monitoring scenarios and preventing failures demands a blend of practical experience and technical proficiency.
POWER: What are some of the common challenges or limitations associated with remote monitoring in power plants and substations, and how can they be mitigated or overcome?
Elkinson: Successfully installing remote monitoring systems involves the challenge of not just collecting and recording data, but also transmitting it to an appropriate location where necessary action can be taken. Whether it’s a simple alert or alarm, or granting full access to the monitor’s interface, ensuring an established pathway for notifying individuals of any occurrences is one of the most crucial aspects.
Avoiding outages is obviously another significant issue. De-energizing assets is often necessary for maintenance or new equipment installation, presenting a challenge, especially with critical components like generators. Strategic planning well in advance can mitigate this challenge.
One of the pivotal challenges in remote monitoring systems is the misconception that the system, once installed, autonomously handles all responsibilities without prior consideration for potential challenges. Consider an instance where an alert triggers due to an individual temperature sensor on a generator. If someone in the control room simply acknowledges the alert without informing others, the issue with the generator remains unaddressed. The challenge lies in creating a robust plan, agreed upon by all pertinent stakeholders, that is open to audit, examination, and immediate action when necessary.
POWER: What are the typical cost considerations when implementing and maintaining such systems?
McGrail: The most important cost to consider when it comes to remote monitoring systems is not the price of the monitor system or installation, but the potential cost associated with business interruptions and reputational risk. Monitoring systems and installation will typically run at only a fraction of the price it takes to replace a failed asset. Not to mention, transporting and installing large critical assets, such as transformers, are much larger and costlier to handle. A new power transformer may cost well over $1,000,000 and take, at the present moment, up to two years to specify, design, and build. Additionally, transportation, installation, commissioning, and related activities can bring the overall costs up to more than $4,000,000.
According to the Ponemon Institute, downtime can cost up to $9,000 a minute at a data center, and in 2018, the Department of Energy estimated outages cost the U.S. economy $150 billion a year—and this has likely only increased since then. Beyond the significant financial exposure, major outages can risk an organization’s reputation, ability to raise capital, and eventually ability to perform. Condition monitoring, appropriately applied and used, is an asset management tool that can reduce this financial exposure significantly.
POWER: How does the integration of remote monitoring impact the workforce and job roles at utilities?
Elkinson: Remote monitoring systems don’t exist to replace staff. Instead, they support staff. Condition monitoring is an important part of the quality control process and will always be a process managed by people, whether it’s integrated into the organization or outsourced to a service. Remote monitoring strategies just offer a more efficient and proactive approach to how utilities maintain and solve asset health issues.
While remote monitoring devices can provide real-time data, the responsibility still rests on the engineer or utility worker to understand, interpret, and decide when to act on those signals. Condition monitoring does not reduce the need for knowledgeable staff, but does allow for these increasingly scarce resources to focus their time most effectively.
—Aaron Larson is POWER’s executive editor.