A significant part of Asset Management's mission is protecting Western Area Power Administration's (WAPA's) most important physical equipment. Power transformers, which convert voltage to different levels, rank high on that list.
Ranging from US$1.5 million to US$4 million each, transformers are among WAPA's most expensive pieces of transmission equipment. They also have one of the longest lead times to procure. A method for calculating a health index score for these assets is needed for many maintenance and investment decisions.
One revealing factor in a transformer's health index score is the megavolt ampere, or MVA, that passes through the transformer. The more heavily loaded transformers tend to see a shorter lifespan than lightly loaded ones.
Scattered across WAPA territory
Although the connection between MVA load and asset health is relatively straightforward, getting this data from all of the transformers is anything but. The sheer quantity of data alone makes collection, storage, and analysis a daunting task. WAPA's supervisory control and data acquisition, or SCADA, system receives transformer data every 2 to 4 secs. It's not just a mountain but a mountain range of information.
To complicate things further, WAPA's different regions have different SCADA systems and methods of storing the data. Asset Management does not have access to these systems and uses the Maximo record-keeping system as their data repository for holding health-index information. Despite the importance of these assets to system reliability, WAPA had no standard practice, routine, or tools for collecting and updating transformer MVA data.
The Asset Management Program Initiation Project set out to collect that data in 2015 to calculate the first transformer health index scores, and the experience was eye opening. Maintenance had to request three years' worth of data from SCADA and that data had to be sifted and scrubbed to make sure the information correlated to the right equipment. This process had to be done one asset at a time because there was no established method for verification.
"It was a huge undertaking that took months," recalled Information Technology Specialist Matthew Bailey. Only then could the data be entered — manually — into Maximo. "We had to 'shrink' the data. Just transferring it all at once into Maximo would overload the system."
This lack of integration meant that the time- and labor-intensive process was only conducted every three years, which left planners and analysts making critical calculations with old data. As Bailey learned more about how the data was used, it became clear that the data-management practice was ripe for an overhaul.
New tools change game
Fortunately, the evolving technology landscape presented the opportunity. Over the next few years, WAPA licensed the enterprise version of the PI Historian — PI referring to "plant interface" — which allowed WAPA to expand its use for data storage throughout the organization.
This move toward a standardized platform brought access to new tools, including one called Asset Framework. "This was the missing link in connecting SCADA data with other systems," said Bailey.
Asset Framework allowed data stored in PI Historian to be organized in more meaningful ways, such as adding asset identification data with devices used in the field. It also enabled PI Historian to store and calculate the near-constant data input in smaller chunks that could be transferred to Maximo more frequently. A pilot project was the next step to test whether or not these capabilities could be applied to transformer loading data.
Testing through teamwork
The Rocky Mountain and Desert Southwest regions were the first to build and deploy PI Asset Framework and send that data to WAPA's central PI system. This made the regions' transformer-loading data available for transferring from PI Historian to Maximo.
The effort to create an automated process for linking data in PI Historian to assets in Maximo was launched in spring 2020. Using RM and DSW transformers as the sample data, the proposed framework assigned identification numbers to assets that could be used across platforms. The pilot also established an integration between TIBCO and PI Asset Framework that ultimately enables PI data to be sent to systems such as Maximo.
An interdisciplinary group came together in the spirit of collaboration to achieve that goal. Team members from IT, Asset Management specialists from Headquarters and the regions and subject matter experts on Maximo, PI Historian, and enterprise architecture all contributed to the project.
There were a lot of challenges and obstacles involved, Bailey acknowledged. "But the team has done a great job of overcoming them," he said. "And we did it all remotely during the pandemic!"
Part of the bigger picture
The framework successfully retrieved accurate, weekly MVA data, calculated transformer load, and transferred that data from PI Historian to Maximo. Trended over time, this data will refine health index analytics and paint a more detailed picture of the condition and performance of critical assets in WAPA's fleet. Another significant benefit of the methodology is that it reduces the time commitment and human error that are part of a manual process.
"Data integrity is a key tenet of Asset Management's strategy to provide consistent and clean asset data to our end-use customers," said Chris Lyles, vice president of Asset Management. "Developing this framework minimizes the human element of manual load calculation and helps us achieve a much better end-state."
The success of the project suggests that the process is scalable and could yield the same benefits to other regions as their data is brought into PI Historian in the next couple of years.
"We proved that we can get data out to our consumers much more quickly with this new framework," Bailey noted. "It can be applied to other assets besides transformers eventually."
The standardized PI system that provided the foundation for the pilot is a central piece of WAPA's Data as a Strategic Asset effort. Implementing central storage for data makes it more accessible and consistent in presentation.
As it becomes easier across WAPA to locate and analyze information for business decisions, both short- and long-term users will gain greater appreciation for the value of real-time data.
