Utility Week

22 | 13TH - 19TH APRIL 2018 | UTILITY WEEK Operations & Assets Market view N ew sources of renewable energy, such as wind and solar power, and an anticipated growth in electric vehicle use, are driving major change in the util- ity industry. At the same time, fast-growing economies are experiencing an increased demand for power, putting greater pressure on the electric network. In order to forecast supplies more accu- rately, utility companies are turning to ana- lytics to improve grid operations. Just as artificial intelligence (AI) and driverless cars are fuelled by data, the emerging modern electric grid is becoming increas- ingly data driven, with current restrictions and limitations needing to be overcome to ensure utilities have access to the best quality data. Constraints on the network Many utilities use geographic information systems (GIS) to model network behaviour within key operational systems, because they offer the best mechanisms to manage electrical connectivity. However, with emerging technologies on the horizon, limitations are emerging as the GIS network struggles to provide the tools required for efficient operation of the mod- ern grid. Historically, GIS has been used from a design and planning (as-built) perspective rather than an operations perspective. This becomes challenging when utilities require a version of the network that reflects the exist- ing operating state of the model. At the same time, advanced distribution management systems (ADMS) look to make operational decisions without human inter- action, again requiring as-operated content. Field operators also require more timely and precise data than GIS can typically provide. Taking GIS to the next level Optimising GIS to support modern electric grid operations will require a number of modifications, including the following: • improving the speed and timeliness of updates; • more accurate reflections of the physical network; • deployment of machine learning (ML) algorithms to harmonise phase and trans- former connectivity with actual network conditions. A form of AI, ML is an emerging technol- ogy that harmonises data to help the GIS provide the essential accuracy for opera- tions. It enables utilities to analyse and pro- cess different types of data, ensuring the GIS can provide the most accurate information at the right time to bet- ter inform business decisions within the ADMS platform. Field operators can analyse data during a storm, for instance, to assess where and when to distribute different sources of energy during downtime. Utilities must be able to react quickly and effectively in complex and demanding environments. To achieve this, they must harmonise their data with actual operat- ing conditions. Creating harmony between accurate data and as-is conditions requires an intelligent data management solution to align process and system data. By using ML in day-to-day operations, utilities can take advantage of additional intelligence, creat- ing a virtual circle of data quality. Many organisations understand the need to harmonise processes, systems and data in theory. However, legacy organisations and standalone data repositories make consoli- dation and aggregation difficult to achieve in practice. Finding the right model to align this data is the first step to obtaining rich data sets and enhancing modern grid operations. Ensuring data governance success Utilities are made up of different organi- sations and departments, each of which manage various processes, systems and programmes. Typically, these organisations will work separately, oen duplicating data in silos rather than sharing it. However, the modern grid is causing a shi in this paradigm. With data volumes increasingly exponentially, so too are the problems associated with a lack of data governance. Utilities must therefore engage a govern- ance model, assuring alignment between processes, systems and data to meet the demands of the modern grid. Data governance brings together information from multiple sources, and requires blending accountability, agreed service levels and measurement. An intelligent data management solution can provide windows into service levels: for instance, by using dashboards, utilities can enforce the agreed service levels at key points within the utility and manage these restrictions. Adopting a strong governance model will therefore improve their approach to the data lifecycle. Mastering the power of analytics The modern grid requires an environment that thrives on high quality data. Utilities must look at their objectives of creating a safety culture to help inform the models that they need to achieve this. This includes where to start with data quality, building accountable teams, education and knowl- edge sharing, understanding what data quality means and assigning employee ownership. The emergence of renewable power sources, electric cars and the need for a smarter grid are driving major disruption in the electric utility industry. To take advan- tage of these dynamic changes, utilities must empower operations with a level of data quality and homogeneity not typically pre- sent in the network. Key to achieving this is to optimise GIS for modern electric opera- tions, master a data governance model and establish a data quality culture. These com- ponents will enable utilities to overcome cur- rent constraints and restrictions to master the power of analytics and enable essential operations data quality. Dan Beasley, director, utilities, Cyient The importance of rich data If networks want to leverage the smart revolution, it is essential they invest in the right technologies to deliver the rich data sets that are the building blocks of intelligent grids, says Dan Beasley. "Finding the right model is the first step to obtaining rich data sets and enhancing modern grid operations."

• Cover