Digital Asset Performance Management – Taking Utility Asset Management to the Next Level
- Oct 12, 2020 2:29 am GMTOct 12, 2020 2:36 am GMT
- 893 views
Digital Asset Performance Management – Taking Utility Asset Management to the Next Level
Dan Belmont, Kevin Hade and Kojo Sefah
Asset Management today is more complex and sophisticated than ever before. With the numerous digital enhancements required for grid modernization, the average utility is managing ten times the number of devices it was 20 years ago. Additionally, these devices have electronic components that require more regular care and maintenance. The industry has moved beyond putting a pole in the ground and dressing it e.g., mounting arms and a transformer, and spanning conductors to the customer, and the next pole. Sound asset management strategies provide an array of benefits, including operational efficiencies and cost savings, increased asset life expectancy, improved reliability, and increased safety. More so, in today's digital world, Enterprise Asset Management (EAM) has become known as Asset Performance Management (APM) and continues to evolve toward “Digital Asset Performance Management.” Implementation of a Digital APM solution can reduce utilities Capital and Operations and Maintenance (O&M) spend buy 20 percent or more.
Diligent management of utility assets now requires patches, life cycle refresh, security upgrades, and the like. As the grid evolves to accommodate Distributed Energy Resources (DERs) including Solar Panels, Electric Vehicles charging stations, and battery storage, the complexity and ever-changing landscape of the grid make it even more essential to streamline and digitize the APM process from engineering and construction through lifecycle refresh. Just like Meter Data Management Systems (MDMS) were built to handle smart meter data, utilities now must have more sophisticated asset management systems to care for and manage all the other smart devices properly. According to data from the Federal Energy Regulatory Commission (FERC) Form 1 submittals of large Investor Owned Utilities, after they have deployed all their smart grid assets, their O&M (costs?) have increased. It takes a more measured and frequent effort for utilities to manage digital resources than it did for a pole, transformer, and wires in the past. Utilities are continuing to move toward an asset management-based business for defining long-range plans, budgets, and book-of-work. This article discusses how utilities can take an evolutionary approach to success in monitoring and managing the digital grid in a digital way. Digital APM is the way to increase reliability while driving costs down.
Traditional Utility Asset Management Relies on Preventative Maintenance Schedules
In traditional utility asset management, preventive maintenance is executed to ensure equipment meets specified operating and safety standards. Preventive maintenance refers to planned, regular, and routine maintenance to ensure the continuous operation of electrical assets and equipment. This technique leads to a reduction in unplanned downtime, unanticipated asset failure, and an increase in equipment life expectancy. Digital APM combines data ingestion, analytics, and visualization to improve asset reliability using methods such as condition-based management, risk management, and reliability-centered maintenance.
Preventive maintenance requires some level of consistency by organizations that choose to implement this method of asset management.
Preventive maintenance requires some level of consistency by organizations that choose to implement this method of asset management. The organization is structured to facilitate proper planning, scheduling, and exceptional record keeping. A mature APM program has all assets tracked in a reliable database with installation, inspection, repair, and decommissioning data associated with the assets recorded and tracked.
Asset criticality plays a crucial role in developing a successful preventive maintenance program. Due to the substantial use of utility assets and equipment, asset criticality ensures the right assets are identified and targeted for preventive maintenance programs. Asset criticality involves computing the risk of asset failure and the impact an asset failure will have on utility operations. Failure Modes and Effects Analysis (FMEA) and Root Cause Failure Analysis (RCFA) are reliability principles used across the industry to calculate and determine how critical an asset is to utility operations. FMEA is used to understand how equipment can fail, the probability of the failure, and the risk associated with operations when the failure occurs. RCFA is used to diagnose why a failure occurred and the solutions implemented to resolve the issue. When both principles are applied together, the utility employs proactive and reactive measures in its preventive maintenance program.
Preventive Maintenance Practices
Field equipment such as circuit breakers/reclosers, meters, generators, switches, transformers, etc., deployed by utilities require consistent maintenance to maintain system reliability, improve System Average Interruption Frequency Index (SAIFI) and System Average Interruption Duration Index (SAIDI) metrics, and provide a Return on Investment (ROI). A similar maintenance consistency is necessary for non-field equipment such as control systems (including Energy Management System, Supervisory Control and Data Acquisition (SCADA), Distribution Management System, etc.) and facilities (such as Control Rooms and Data Centers.)
Field equipment such as pole-mounted transformers are widely deployed across a utility grid. Transformers range from approximately 10kVA to 100kVA and are capable of transforming 11kV to 35kV to a low voltage of 400V. Some preventive maintenance programs employed to ensure these transformers do not catastrophically fail include visual inspections, thermal imaging inspections, and aerial inspections.1 Due to their sheer numbers, most utilities use a run-to-fail replacement strategy for their pole-mounted transformers. With a run-to-fail strategy, utilities accept the risk of an equipment failure and perform maintenance after the equipment fails. However, during the life of a distribution pole-mounted transformer, a program of continuous inspection and monitoring is employed to detect early signs of degradation beyond what would generally be expected.1 Other benefits derived from preventive maintenance of the transformers include the detection of leaking oil, detection of hotspots to prevent wildfires, and ensuring they are adequately earthed.
Circuit breakers and reclosers play a critical role in the reliability of the power grid by protecting other electric assets. Low voltage breakers (600V and below), medium voltage breakers (600V – 15kV), and high voltage breakers (15kV and above) are used across the electric grid. They are typically found in an open or closed state and can remain in this state for long periods of time. Based on the geographical location of the breakers/reclosers, they can become dusty, ionized, build up corona, or hampered with foreign materials. Utilities conduct preventive maintenance such as insulation resistance tests, millivolt drop tests, overload tripping tests, connection tests, and mechanical operation tests on their breakers to ensure continuous and reliable operation. These tests are routine and scheduled in 5 to 12-month intervals, making it a considerable O&M expense for utilities. Data collected during these tests are used to make switching, loading, and replacement decisions.
Preventive maintenance is beneficial for equipment deployed by utilities to ensure increased efficiencies and equipment lifecycle maximization
Preventive maintenance is beneficial for equipment deployed by utilities to ensure increased efficiencies and equipment lifecycle maximization; therefore, utilities should move towards a digital paradigm such as predictive maintenance
Moving Towards Predictive and Prescriptive Utility Asset Performance Management (APM)
APM has introduced more data from which utilities can improve their management of assets. This data can be used for predictive maintenance – condition monitoring to determine when an asst is likely to fail - or prescriptive maintenance – identifying why an asset failed to mitigate the risk of future failure. As reported in Utility Dive, "For utilities, APM puts in place these processes on the data and acts as a new age "fortune teller." APM usually focuses on data collection and processing. From there, operators must collaborate on the data to add context. This, in turn, improves asset management and takes the guesswork out of the equation." 2
A prime example of "fortune-telling" is the advanced algorithmic analysis of AMI voltage sag and swell data to predict the failure of transformers and schedule replacement to avoid outages. It is widely understood and accepted in the industry that high- and low-voltage output from a transformer is an indicator of impending failure due to fusion of the windings.3 Without transformer monitoring, utilities with AMI meters are unable to monitor high or low voltage coming into the meter to analyze transformer behavior. However, since there is such a wide variety of voltages, mapping the meters to the associated transformer to create a baseline for a specific transformer is crucial. The relationships between the meter and transformer and corresponding high- or low-voltage spikes can be managed in the asset management system and flag the appropriate transformer for replacement when the measurements reach a particular pattern or threshold. An example of low-voltage issues causing transformer failure shows a profile prior to failure. In the below study (see Figure 1), ComEd has determined that when sag and swell variance increase versus the neighboring average or its own past average that a transformer is destined to fail. Ameren Illinois has also adopted a predictive asset management approach for its transformers through an EAM system, which consolidated asset data from various offline and online sources to create health models.4
Figure 1: Voltage dips due to iron core saturation
By combining asset data with geospatial, environmental, and weather data, machine learning can be used to predict failures in overhead distribution power lines which can inform preventive maintenance scheduling. This can potentially increase reliability and decrease the regulatory and financial risk for the utility. Using data from a Southern California utility, researchers from Southern Methodist University built a machine learning asset failure prediction model to identify overhead distribution line failures.5 They determined that asset age was an important variable, however age data was missing for some assets. They used an algorithm which predicted asset age with 82 percent accuracy to impute the age of assets with missing data. Using the imputed age and other data on the overhead conductors and geographic data (e.g., elevation, slope, miles from coast), their prediction model was able to identify 63 percent of failures of overhead power-lines.5 A model such as this could be used to scope construction projects to mitigate outages. They noted that the prediction capabilities could be improved with more robust data retention on failed assets, and there are opportunities for continuous monitoring by incorporating daily SCADA voltage data and daily weather data into the model.
The use of predictive analytics by utilities can reduce the cost of asset replacement by controlling when it will be replaced and improve the reliability of the grid by reducing the likelihood of an outage. However, the use of predictive analytics is not widespread at utilities. According to Utility Analytics Institute's 2018 Analytics Maturity Assessment, 77 percent of utilities surveyed said their predictive analytics capabilities were limited.6 This is because when considering asset management, challenges with data readiness and lack of regulatory guidance can make implementing predictive asset analytics costly or risky, although the data science exists to predict failures.
Until regulatory agencies address proper protocols for responding to insights and the extent to which a utility can be held liable for failing to take remedial action on an idea from a predictive model, utilities will hesitate to implement predictive models
Until regulatory agencies address proper protocols for responding to insights and the extent to which a utility can be held liable for failing to take remedial action on an idea from a predictive model, utilities will hesitate to implement predictive models.5 The utility needs to balance proactive maintenance to avoid negligence, with the cost and logistical challenges of responding to lower confidence insights or potentially false positives. Additionally, challenges with data readiness, including the quality and completeness of data sets, can act as a barrier to implementing predictive analytics. Information coming from inconsistently completed paper forms or missing data from older assets or purged records can reduce the accuracy of models or require processing before it can be used in models.
Technologies Deployed in APM for Data Ingestion and Analysis
The deployment of distribution automation (DA) field devices has generated a plethora of performance data that can be leveraged for identifying asset failures digitally or as an additional dataset to build predictive analytics. The Philadelphia Electric Company (“PECO”) has used radio frequency (RF) noise from its DA devices to identify electrical equipment issues and is looking into the possibility of detecting precursor patterns in the noise to predict future issues.9 Incorporating this RF noise analysis has increased operational efficiency by reducing the time to diagnose problems with substation equipment.
Digital tools are also leveraged in asset management for inspection data. Historically, field changes, or “as-builts” were handwritten in the field and turned in to the drafting department to input changes to the geographical information system (GIS) and SCADA systems and eventually the EAM system for digital ingestion. These data changes could then be used months or years later. This process has been so robust that even today the process is still in use. In modern asset management, however, there are tools to aid in the collection and documentation of new construction, repairs, and changes in the field via GPS enabled handhelds with cameras and other sensors. They can be automatically tied into the EAM system with all of the needed attributes to manage that asset through its useful life. This integration allows almost immediate transference of information into the GIS and SCADA systems.
Another area where digital tools help the APM data ingestion process is in asset inspections. Using tools such as drones for image and data capture tied into the EAM system can create an historical database to analyze for failure modes before assets fail. These tools will even lengthen the useful life of assets by extending them past preventative replacement standards, saving money without sacrificing reliability. One example is utilities use of drones to capture and monitor cracks, or lack thereof, in the tops of poles, thus prescribing replacement when needed or delaying until required.10,11,12
Comparative Benefit Analysis
Implementing EAM and digital APM by utilities leads to asset optimization throughout the entire organization. They allow utilities to draft asset strategies based on real-time data generated and analyzed from in-service devices and equipment. Organizational strategy management around assets becomes enhanced through a uniform decision-making process and a centralized database of asset information and history. This helps increase workflow efficiency and builds a backbone for predictive analytic practices.
According to GE Digital, the installation of Internet of Things (IoT) devices (sensors) at multiple utilities across 5,000 assets in 900 power plants resulted in about 125 TB of data being analyzed daily and delivered over 200 billion tags. This reduced unplanned downtime by up to 5 percent and reduced false alarms by up to 75 percent.7 This benefit translates into monetary value results in substantial cost savings for utilities implementing digital APM strategies as compared to the traditional methods of asset management. Unexpected failures can be reduced by 55 percent due to real-time monitoring, data ingestion, analytics, and visualization.
An EAM system that incorporates financial data and reporting can enable utilities to quickly understand the cost savings associated with rolling out a Digital APM strategy. On the average, a utility that rolls out a Digital APM strategy can save approximately 30 percent in Operational Expenditure (OpEx) and approximately 20 percent in Capital Expenditure (CapEx) compared to traditional asset management practices.8
Maintenance schedules can be better predicted and planned to ensure maximum optimization of labor and resources with a Digital APM strategy. With a holistic, centralized view of asset health, crew work can be optimized by capability and geography when possible to reduce unnecessary trips to the same area for different asset types. Utilities can build an effective and efficient maintenance program by shifting focus to assets with high data opacity.
Digital APM development is still in its infancy with considerable room for expansion and innovation in the coming years. Utilities are moving from traditional static-run to failure with break-fix procedures, to an integrated EAM and analytics-based APM program that can improve reliability and operational metrics, while also delivering financial benefits with direct cost savings and societal benefits from reduced outages.
Digital APM is also a 'not an all or nothing' approach. Utilities can take incremental steps, or a phased approach, through the maturity model with corresponding benefits. Conservative tactical investments in people and processes can have just as great or greater effect than similar technology investments. Utilities do not have to invest in a multi-year, extremely expensive, business-wide transformational effort to start benefiting from initiating an APM approach to managing and maintaining its asset base. An asset and operational evaluation is the easiest way to find the best place to start.
Figure 1: Predicting Transformer Failure. Retrieved July 3, 2020, from https://www.tdworld.com/grid-innovations/asset-management-service/article/20971387/predicting-distribution-transformer-failures
1: [PDF] Pole Mounted Transformers - Australian Energy Regulator - Free Download PDF. (n.d.). Retrieved from https://nanopdf.com/download/pole-mounted-transformers-australian-energy-regulator_pdf
2: Friehauf, B. (2020, June 26). Fortune telling, utility style: Asset performance management drives costs down, reliability up. Retrieved July 08, 2020, from https://www.utilitydive.com/news/fortune-telling-utility-style-asset-performance-management-drives-costs-d/580591/
3: Predicting Transformer Failure. Retrieved July 3, 2020, from https://www.tdworld.com/grid-innovations/asset-management-service/article/20971387/predicting-distribution-transformer-failures
4: Ameren Illinois Adapts Predictive Asset Performance Management Approach to Grid Modernization. (n.d.). Retrieved July 5, 2020, from https://electricenergyonline.com/energy/magazine/1223/article/Ameren-Illinois-Adapts-Predictive-Asset-Performance-Management-Approach-to-Grid-Modernization.htm
5: Flamenbaum, R., Pompo, T., Havenstein, C., & Thiemsuwan, J. (n.d.). Machine Learning in Support of Electric Distribution Asset Failure Prediction [Scholarly project]. In Machine Learning in Support of Electric Distribution Asset Failure Prediction. Retrieved from https://scholar.smu.edu/cgi/viewcontent.cgi?article=1096&context=datasciencereview
6: Analytics Maturity Assessment [Pdf]. (n.d.). Utility Analytics Institute.
8: GE: Reducing Power Outages Worldwide with Large-Scale Industrial IoT Implementation - IoT - Internet of Things. (2017, October 08). Retrieved July 2, 2020, from https://iot.do/ge-power-outages-industrial-iot-predix-2017-10
8: How analytics can improve asset management in electric-power networks. (n.d.). Retrieved June 30, 2020, from https://www.mckinsey.com/industries/electric-power-and-natural-gas/our-insights/how-analytics-can-improve-asset-management-in-electric-power-networks
9: Benedict, L. A. (n.d.). From Big Data to Actionable Analytics. Retrieved July 2, 2020, from https://www.tdworld.com/smart-utility/data-analytics/article/21121295/from-big-data-to-actionable-analytics
10: Guntrip, T. SCE Flies Drones for Inspections. (2018, July 19). Retrieved July 14, 2020, from https://www.tdworld.com/electric-utility-operations/article/20971476/sce-flies-drones-for-inspections
11: Ryan, C. Drone vs Helicopters – Utility Inspections. (2020, July 8). Retrieved July 14, 2020, from https://www.exelonclearsight.com/drone-vs-helicopters-utility-inspections/
12: Drones working to ensure public safety. Retrieved July 14, 2020, from https://www.sdge.com/wildfire-safety/aviation-services
Dan Belmont is a managing director in the Energy & Utilities practice at West Monroe based in Chicago. He leads the Midwest team in providing value-rich transformational projects for utilities throughout the US, leveraging experience in telecommunications, customer experience, operational excellence, and engineered IT and OT systems.
Kevin Hade is a manager in the Energy & Utilities practice at West Monroe, based in Chicago, IL. He focuses on building data and analytics capabilities for electric, gas, and water utilities and has managed transformational projects at major US utilities related to analytics strategy and implementation, asset preventive maintenance, Advanced Distribution Management System implementation, and utility telecommunications.
Kojo Sefah is a senior consultant in the Energy & Utilities practice at West Monroe, based in Chicago. He has worked with utilities in their transition to become “Utilities of the Future” through technology conceptualization, research & development and project implementation in the fields of Distributed Energy Resources (DERs), energy & power markets and data analytics.