Maximize Reliability, Minimize Risk: Evolving Standards Offer Guidelines for Monitoring Transmission Asset Health
Electric utilities face a growing array of challenges, including an aging infrastructure, limited financial resources and a shrinking workforce.
By Enrique Herrera and G. Matthew Kennedy
Electric utilities face a growing array of challenges, including an aging infrastructure, limited financial resources and a shrinking workforce. It is important for them to develop and implement processes and technologies that will allow them to deliver consistent, reliable power. Condition-based monitoring techniques coupled with effective data collection and analytics are key for utilities to effectively manage asset life, maximize reliability and safety, and drive down costs.
A rapidly evolving set of worldwide industry standards is increasingly affecting how utility owners use both off-line and on-line monitoring to measure and track asset health. Many of these standards have grown out of work by the International Standards Organization (ISO). Others were developed from research performed by the Institute of Electrical and Electronic Engineers (IEEE), the International Electrotechnical Commission (IEC) or the Conseil International des Grand Reseaux Electriques (CIGRE). Together these standards offer power utilities a basic blueprint for building efficient, effective systems to monitor asset health.
The key standard regulating the adoption of a formalized asset management system is ISO 55000. Based on PAS55 (Publically Available Standard) originally developed by the Institute of Asset Management and the British Standards Institution in 2004, ISO 55000 outlines the requirements and offers general guidance in the implementation of efficient asset management systems. Written generically, this series of standards is designed to apply to a wide range of industries. Other standards groups, particularly those influenced by IEC 60812 and ISO13000 series of standards, help asset owners understand how to apply failure mode and effect analysis (FMEA). Today FMEA serves as the key concept utilities owners use to measure asset health.
As utilities have migrated to condition-based monitoring to minimize downtime in utility transmission and generation systems, standards have also emerged to help define best strategies for collecting and analyzing data in real time. Case in point is ISO 18095. Recently sent out to vote, ISO 18095 addresses condition monitoring and diagnostics for power transformers. As part of that process, it defines how asset owners can implement a condition-based monitoring platform and which failure modes an asset owner should target depending on the type of transformer that owner uses.
Once a utility collects data on asset conditions, it must analyze the data to turn it into actionable intelligence. A growing number of transmission and generation utilities are turning to sophisticated analytic tools to create needed intelligence.
Many of these analytic tools also process data from on-line monitoring systems that are often integrated into the assets themselves. Intelligent electronic devices (IEDs), such as load tap changer analyzers, breaker testers, gas testers, bushing monitors and others provide raw data, alerts and alarms that are integrated into the analytic tools. In many cases, these IEDs must comply with multiple IEEE and IEC standards.
The Role of Analytics
Once the data is collected, analytic tools allow utilities to turn that data into useful information and determine if it meets various standards, such as IEEE and IEC, diagnostic methodologies and limits. As an example, the tool might examine a particular transformer to evaluate whether it is in good condition based on test results. Next, the system takes those analytics and aggregates them together to find a failure mode that could cause the transformer to fail. This process must comply with standards such as IEC 60812 and ISO 18095.
The primary goal of this analytical process is to measure each asset’s reliability. Often the output of the FMEA is an asset health index that assigns each asset a score. System can then express the consequences of any failure in terms of its impact on utility finances, availability and reliability, as well as its potential impact on customers, the environment and safety.
By including financial information in asset health consequences, utilities can more easily prioritize their actions. Using this information, they can make better informed business decisions particularly in reference to maintenance strategies where they must often decide whether to repair, replace or refurbish an asset or simply defer work to a later date. Two key standards that impact risk analysis are ISO 14971 and ISO 55000.
Case Study: Austin Energy Monitors its Power Transformers
Austin Energy offers insight into how utilities can take advantage of these tools and follow key asset management standards. Serving Austin, Texas, since 1895, the company is the eighth largest publicly-owned utility in the U.S. The geographically concentrated utility has about 5,400 miles of overhead and 6,000 miles of underground 69kV, 138 kV and 345 kV transmission lines and 72 transmission and distribution substations.
The utility announced plans to augment its off-line testing capabilities with an advanced on-line monitoring system to improve system uptime. The Figure illustrates a new data flow the company created that brings together legacy offline bushing test results and off-line dissolved gas analysis (DGA) lab results with on-line DGA bushing data and operating data through a SCADA system. Initially targeted at power transformers, the system will eventually be extended to circuit breakers and battery banks as well.
Like many utilities, Austin Energy uses OSIsoft’s PI System to collect, store, organize and distribute data. In this new configuration, all SCADA data is fed into Austin Energy’s PI system and then brought into the dobleARMS (Doble’s asset risk management system) asset analysis engine that has a PI Server. Doble takes data from the PI System and combines it with offline data in a holistic asset risk system, running on servers in Doble’s data farm. The data is put through a subject matter expert analysis in Doble’s First Response Analytic Knowledgebase (FRANK). FRANK applies analytics to off-line test data customers remotely store on Doble’s servers and compares that data to the industry at large and standards-based limits. The system aggregates different types of data and performs failure mode analysis along with health and risk indexing. FRANK makes it easy for Austin Energy to easily measure its own test results against other utilities.
Doble uses its secure enterprise gateway, PI Asset Framework (AF), to connect to Austin Energy’s PI Server. All off-line and on-line test data is fed into this system and then sent to the FMEA engine, known as Doble’s Knowledgebase Analytics Engine, where it measures asset health and risk.
One of the early goals of Austin Energy’s project was to improve monitoring of transformer bushings. In 2014, the utility began this phase by installing on-line transformer bushing monitors on three transformers at one of its substations. Bushings are one of the more common causes of transformer failure. Austin Energy wanted to monitor the transformer bushings to ensure they were maintaining integrity over time and avoid any unexpected failure.
The primary method used to track the quality of a bushing is to measure power factor. That test indicates how efficient the bushing’s insulation is operating over time. A switch adapter that downloaded data to an IED was installed. To get an accurate picture of the degradation rate of each bushing, Doble took measurements at various time intervals. By tracking degradation on a daily, weekly and monthly basis, Austin Energy now can implement more effective maintenance strategies, reduce downtime and improve system availability.
At the same time, working with Doble, Austin Energy installed dissolved gas analysis monitors on the three 138 kV transformers. The monitors provided a single value (composite dissolved gas) in ppm based on four gases: hydrogen, carbon monoxide, acetylene and ethylene. These values helped Austin Energy track winding degradation inside the systems, as well as overheating and particle discharge. With these new capabilities, Austin Energy can quantify an asset health index and identify the potential risk of failure for each transformer.
Given the tremendous amount of operating data pouring into a standards-based asset management system, power utility personnel must be able to quickly identify the impact of high asset risks, isolate potential problems, and implement needed repairs and maintenance. To help Austin Energy rapidly perform these tasks, the company uses OSIsoft’s AF to identify business-centric issues on the top of the hierarchy and answer questions such as how much money is being generated on an asset at any given time. In the middle of the AF hierarchy, the management team can locate operating problems and use GIS information to provide a view of the locations. At the bottom of the AF hierarchy, operations and maintenance managers can delve into asset attributes and performance data and then perform the diagnostics needed to meet individual standards.
While maintenance and asset health issues rarely grab the public’s attention, facilities’ managers recognize that both off-line and on-line monitoring are key to improving asset reliability and reducing costs. At the same time, a growing number of worldwide standards offer asset owners guidelines for implementing asset management systems, using diagnostic technology and operating in a regulated environment. By combining state-of-the-art diagnostic capability with a standard data infrastructure, utilities in electric transmission and generation can more easily comply with these new standards and, in the process, capture the expertise of technical experts, optimize operating and capital expenses, and, ultimately, make smarter business decisions. UP
Enrique Herrera is a market principal for OSIsoft, Enrique oversees OSIsoft’s Connected Services business and shapes the company’s strategy to address the Internet of Things. He brings over 25 years of experience from the Microsoft Corp., as an industry specialist and the automotive engineering/manufacturing perspective (Ford Motor Co., Jaguar Cars and Visteon Corp.). Enrique holds a bachelor of science degree in mechanical engineering from the Massachusetts Institute of Technology and a master’s of science degree in advanced automotive engineering from Loughborough University in the UK. He also serves on the board of the Smart Manufacturing Leadership Coalition.
G. Matthew Kennedy is Doble Engineering Co.’s Solutions Director: Enterprise and Data Products, overseeing the complete software and cloud product vision of the company. Kennedy holds a bachelor of science degree in electrical engineering from the University of California, Santa Barbara, where he studied signal and digital signal processing, His post graduate studies continued with the U.S. Navy: Nuclear Power School and Cornell University in product design and development. He is a member of IEEE, IEC, USNC and ISO.