Creating value from advanced analytics to improve operations

Advanced analytics applications are a key component as power generation, transmission, and distribution companies advance on their digital transformation journeys.

By Giro Iuliano January 31, 2022
Courtesy: Seeq

Digital transformation initiatives within the power generation, transmission and distribution industries continue to evolve with a focus on managing costs, improving grid operations and expanding market opportunities. To date, adoption of these initiatives has been mixed and limited due to a lack of accessible data, which results in an inability to analyze and use data to drive improved operational outcomes.

Power and utility companies, and their customers, expect relevant information on demand, and vendors must supply technologies, such as advanced analytic applications, to meet these needs. Consumer technology has empowered users to get what they want, whenever they want, and most expect an immediate response, along with constant access all the time.

These and other expectations regarding delivery of insights from data can be met with advanced analytics. This can lead to reduced costs, more intelligent control of energy consumption and successful participation in time-of-use and demand-response programs.

At the same time, power and utility organizations can employ end-to-end automated process using advanced analytics to transform customer relationships. This can be done by progressively shifting from reactive to proactive decisions, but issues regarding data access and analysis must first be addressed.

Data access issues, digital transformation, access to analytics

The power and utility industry is under tremendous pressure to modernize systems to meet customer needs, improve cybersecurity, comply with regulations, and achieve better operational performance. Leveraging required technologies, such as software as a service (SaaS), cloud computing, machine learning (ML), the Internet of Things (IoT) and cybersecurity tools, will require significant investment.

Implementing various types of digital transformation initiatives will create huge amounts of data, requiring interoperability and context delivered by advanced analytics applications. Many utility data systems are in silos and are not easily accessible by analytics applications. Digital transformation will require much better data quality, along with advanced analytics to deal with structured and unstructured data sets.

Making this data useful to generate actionable business logic, such as guiding utility customer behavior and achieving grid-supply balance, is not straightforward. Therefore, utilities will need to deploy a range of information technology (IT) solutions to collect, secure and store data — and then analyze the data and create insights.

Spreadsheets have long been the standard method for analyzing data within the power and utility sector, but this type of general-purpose tool is not advanced or agile enough to define relevant context quickly. Data volumes are increasing by significant amounts, and spreadsheets are not suited for tasks such as time-series analysis, effective reporting, performing complex calculations and creating models (Figure 1).

Spreadsheets also require constant upkeep, which introduces errors and data misalignment. Another factor limiting spreadsheets are the nuances of working with time-series data, such as missing data, different time zones, daylight savings time, interpolation types, and logic. With spreadsheets, these issues only can be addressed with complex formulas.

Applying specific business logic and sharing the analysis is difficult, and adjustments for dates and non-uniform intervals of data is very complex, as is making corrections for outliers, and for missing or corrupt data. Spreadsheets also require strict protection against user corruption and for version management, and they offer limited multiuser collaboration.

To deal with these issues, online advanced analytics applications are replacing offline tools like spreadsheets and traditional desktop software.

Figure 1: Spreadsheets cannot keep up with ever increasing amounts of data. Courtesy: Seeq

Figure 1: Spreadsheets cannot keep up with ever increasing amounts of data. Courtesy: Seeq

Advanced analytics provides insights

Digital transformation is an ongoing process that requires investments in skills and new software applications. A combination of integrated information and operations technologies are needed to provide a holistic view of all relevant data in an organization. Engineers, subject matter experts, and data scientists must be able to interact with time-series data to execute data cleansing and contextualization tasks. Implementing these initiatives will lead to improved value creation from the ever-expanding amounts of data.

With advanced analytics applications, the power and utility industry can ease cleansing and contextualization of time-series data. These applications connect to existing data silos and other data sources, on premises and cloud-based, without copying or duplicating data from source systems. Once data sources are connected, the application can be used by engineers and subject matter experts for diagnostic, predictive and descriptive analytics. Advanced analytics applications also provide improved real-time collaboration, enabling greater vision and insights across teams, accelerating the communication of insights across the broader organization.

Data scientists also play a major role in driving digital transformation as they are asked to create new ways to maximize operational efficiencies and predict future consumption. Accurate forecasting is critical for utilities as their portfolios are expanding to include renewables (wind, solar, co-generation energy, and other sources). Data scientists can use data from customers to create demand models that accurately determine when energy will be needed and how external factors, such as weather, will affect forecasts. They can then collaborate within the advanced analytics applications with engineers and subject matter experts to ensure the integrity and relevance of the data being modeled, and to provide optimal outcomes.

These models can also be applied to asset data in the grid to analyze and identify equipment headed for failure. This predictive approach empowers teams to plan and budget for repairs and replacements. Advanced analytics applications play a vital role in the monitoring and diagnostics analysis used to determine root causes of equipment failures by focusing on the necessary data sets.

These applications also play an important role in the predictive analytics and pattern recognition required to identify when maintenance is necessary on assets, reducing reactive maintenance and associated downtime. These processes often become repeatable analyses to monitor all performance levels against models or conditions. This can be done at scale and be securely leveraged by the entire organization.

Use cases for analytics in power generation, transmission, distribution

As power generation, transmission, and distribution companies continue their digital transformation journeys, many have quickly discovered that developing strategies built on a foundation of diagnostic, predictive, and descriptive analytics has transformed their operations, saving significant time and money.

Demand management

Effectively managing real-time power prices often results in sharp load reductions when prices increase because commercial energy consumers deploy on-site generation or shut down equipment for several minutes or more to avoid paying top dollar. The independent system operator (ISO) needs to manage the power supply for these load reduction events to prevent outages when load returns to normal.

ISO engineers needed to develop a load forecast model to identify load-change events, and to quantify the magnitude of each event based on price changes. They determined the amount of load reduced in response to high-price spikes by identifying the load reduction times and performing analytics. A report was created summarizing results, enabling operators to prepare for how much load would return when the power price came down to normal levels.

Creating a model to better understand and prepare for load-reduction events helps prevent service disruptions while maintaining customer satisfaction among power producers and end consumers of energy. The analytics solution helps the power utility and the ISO to quickly identify load responses from consumers to high-price events (Figure 2).

Operators now can anticipate power grid requirements following a high-price event.

Figure 2: This report shows the amount of power reduction during high power price events, and it displays how much power the independent system operator can expect when the price returns to normal. Courtesy: Seeq

Figure 2: This report shows the amount of power reduction during high power price events, and it displays how much power the independent system operator can expect when the price returns to normal. Courtesy: Seeq

Data analytics enable transformer preventive maintenance

Knowing when to perform maintenance on power transformers requires deep understanding of the available data. This is best done by transitioning from calendar- to condition-based maintenance using data from a variety of sources, including nameplate information, insulating fluid test results, diagnostic tests (such as dissolved gas analysis and electrical tests), maintenance schedules, real-time data (such as cooling performance) and more. Developing and refining transformer health analytics to take advantage of this large volume of data is difficult, especially when applied to numerous assets supplied by a variety of vendors.

An example of one such analytic is dissolved gas analysis (DGA), the study of dissolved gases in the oil used to insulate the transformer’s electrical components. When this oil breaks down and becomes less effective, it releases gases within the oil. The distribution of these gases is related to the type of electrical fault, and the rate of gas generation indicates the severity of the fault. Many faults like arcing, overheating, and partial discharge only can be detected by analyzing gases.

Many electrical utilities have a DGA program. A DGA program often consists of manually sampling the oil and sending the sample to a laboratory for analysis every one to four years. Advanced analytics applications can be used to aggregate the data, evaluate the required formulas, and scale the analytics across numerous assets. Methods include:

  • IEEE C57-104 Total Dissolved Combustible Gases.
  • IEC 60599.
  • Roger’s ratio.
  • Dornenburg’s state estimation.
  • Duval’s triangle.
Figure 3: Trend with a health score for a transformer, including thresholds and deviations. Courtesy: Seeq

Figure 3: Trend with a health score for a transformer, including thresholds and deviations. Courtesy: Seeq

Advanced analytics help generate value from time-series data

From updating IT systems with new digital technologies to capturing value from data insights to better use of machine learning and artificial intelligence, power generation, transmission, and distribution face similar digital transformation challenges. With advanced analytics applications, these organizations can address these and other issues by generating value from time-series data. Advanced analytics provide the rapid iteration and refinement of analytics required to identify more accurate load recovery, improve transformer maintenance, and more, delivering the improved experience these companies and their customers expect.

Giro Iuliano, industrial principal at Seeq Corp. Edited by Chris Vavra, web content manager, Control Engineering, CFE Media and Technology, cvavra@cfemedia.com.

Original content can be found at Control Engineering.


Author Bio: Giro Iuliano is an industry principal at Seeq Corporation. He brings nearly 20 years of sales and business development experience to Seeq, following a decade working with power and utilities customers at OSIsoft, and five plus years at Oracle and Itron. Iuliano holds a bachelor’s degree in electrical engineering from Northeastern University.