Mastering the Quality of Electric Utility Data
By Dan Beasley, Cyient
The electric utility industry is undergoing a major transformation being driven by growth of renewable energy and other distributed energy resources (DER), consumer demand, cybersecurity threats, digitalization and an abundance of data. One key initiative in navigating and capitalizing on the transformation is to empower operations with a level of data quality and homogeneity not typically present in the utility. Gathering data to harvest insights and forecast more accurately has enormous potential to optimize utility operations. In the same manner that driverless cars require accurate data and roadway mapping to operate effectively, the emerging modern grid demands accurate data and electric network information.
The Importance of High Quality Data
Quality data enables utilities to understand network and asset behavior, operating conditions and their impact on customer service. Electric networks change routinely, therefore, operations must adjust to dynamic conditions, requiring timely access to quality data. The utility network must accurately measure network behavior to ensure correct observations and provide the quality data needed to optimize operations. Access to accurate system and operations data enables the utility to substantially improve its operations quality and cost efficiency.
In addition, regulators are pressuring utilities to take advantage of rich data sets to improve grid operation and asset management and provide better customer experiences.
Data Limitations in the Modern Grid?
The GIS network has long been the system of record utilities use to model the behaviour of the network within key operations systems. The GIS platform offers the best tools and mechanisms to manage the topology of the connections that represent the electric network. The platform can model the network behavior to meet advancing market conditions, however, a number of constraints limit GIS from meeting the needs of today’s modern grid requirements. These constraints include:
“- GIS is built from a design and planning (as-built) perspective rather than an operations perspective.
“- Electric operations personnel require a version of the network that reflects the existing operating state of the model (as-operated).
“- Electric operations personnel work closely with field personnel who require less, but more precise and timely data than GIS typically can provide.
“- Advanced distribution management systems (ADMS) make operations decisions without human interaction, which requires as-operated content along with extensive correlative asset data.
The work of constructing the network generates a great deal of data and information and the GIS network is the most appropriate place to house that network information. While GIS typically reflects the initial network construction, it rarely incorporates essential operating characteristics.
Quality data enables utilities to understand network and asset behavior, operating conditions and their impact on customer service.
In addition to the GIS network, utilities typically store data in an asset management system. Because many electric utilities operate in a regulated environment they generate a great deal of inspection data that requires organization, storage and access. In addition, they must store customer specific data, including location, service experience, as well as amount and quality of electricity.
The new imperatives of the modern grid require harmonizing data sources into a cohesive story that can help utility personnel make immediate decisions based on customer demand.
Putting Data to Work in the Modern Grid
Utilities generate substantial volumes of data and while the Internet of Things (IoT) proliferates across networks thanks to smart devices, it creates multiple new data points that can put pressure on infrastructure. BI Intelligence, a research service from Business Insider, estimated in its “IoT for Utilities” report in last 2016 that the global installed base of smart meters will increase from 450 million in 2015 to 930 million in 2020. On top of this, distributed energy resources (DER) and legacy IT systems bring fresh challenges to utilities that must manage and interpret greater volumes of information. For example, thousands of mini generation plants can connect across the network, bringing in new data points every minute. A system therefore must gather and maintain multiple sources of data.
Consolidating and aligning the many platforms on which the data resides is essential. ADMS, which lies at the heart of the modern grid, requires utilities to combine data from multiple utility business functions. Utility control center personnel can then use this insight to manage all aspects of the distribution system. A strong ADMS model requires high quality data to support the distribution system’s mathematical analyses.
An ADMS allows utilities to improve grid resiliency and their ability to withstand or recover from a natural disaster quickly, as well as accommodate larger quantities of DER. In addition, it allows reliability, efficiency and survival in a DER world, thus enabling utilities to remain compliant with new regulations. Maintaining data quality and continuing to update the data model are key for ongoing ADMS support.
Three keys enable utilities to make better use of their data and to achieve the modern grid’s mandates. They are:
1. Optimize GIS for modern electric operations
2. Master a data governance model
3. Establish a data quality culture
Optimizing GIS for Electric Operations
GIS is the optimal tool for network management and utilities must make it suitable to support electric operations. This requires utilities to improve the speed and timeliness of network updates; provide operations personnel with an accurate view of the system; and deploy machine learning algorithms to harmonize phase and transformer connectivity with actual network conditions.
Machine learning is an emerging data science that harmonizes vast amounts of data to make the GIS as accurate as possible for operations. Machine learning can leverage several types of data. One of the most useful is voltage data, which is available in modern metering systems. Machine learning helps the GIS provide the most accurate information at the right time, which contributes to good decision making within the ADMS platform. Current and accurate data trusted by field operations personnel is especially useful during times of crisis, such as during a natural disaster when utility operators must determine where to distribute the various sources of energy to reduce downtime.
The modern grid enables utilities to react quickly and effectively in a complex and demanding environment. To enable the intelligence offered by machine learning, utilities must harmonize data with actual operating conditions. Creating this harmony between data and “as-is” or “as-switched” conditions requires an intelligent data management solution (IDMS) to align utility process and system data. By harnessing machine learning, asset behaviour can add additional intelligence into the network creating a virtual circle of data quality.
While many utilities understand the need to harmonize processes, systems and data, legacy organizations and stand-alone data repositories make consolidation and aggregation difficult. Finding the right model and system to align this data is the first step to obtaining high quality, actionable data and improving modern grid services quality.
Mastering a Data Governance Model
Electric utilities are large organizations made up of many discrete organizations, each of which manages various programs, processes and systems. Typically, these organizations work separately, often duplicating, not sharing data. As a result, harmonising data systems and processes is a radically new concept. Simply put, data harmony has not been essential to a top utility operator. The modern grid, however, is changing this paradigm. The increasing volume of data is exacerbating the problems associated with a lack of data governance. Today’s mandate, therefore, is to engage a governance model assuring process, system and data alignment to meet modern grid demands.
Data governance enables the utility to aggregate data across multiple processes and systems, and requires blending accountability, agreed service levels and measurement. An IDMS must provide windows into service levels. A good example is a dashboard that can help management enforce the agreed service levels at key points within the utility and manage constraints. A strong governance model is important to integrating disparate utility systems.
Establishing a Data Quality Culture
The modern grid requires a culture that thrives on the generation and assimilation of high quality data. Achieving this standard aligns with a quality culture. Utilities can look at objectives of creating a safety culture to help them achieve this culture in other areas. These ideas include defining where data quality begins, building accountable teams, educating to create knowledgeable management, understanding what data quality means and creating ownership at the employee level.
Utilities that optimize GIS for modern electric operations, master a data governance model and establish a data quality culture are on their way to overcoming current constraints and limitations to enable essential operations data quality.