Managing a dynamic grid - Automation and intelligence for a data-driven grid
For every utility company, or at least for every DSO, the operational landscape has changed forever. This is thanks to fundamental changes both to the grid, and perhaps even more importantly, to the pursuit of sustainable business performance.
Dramatic changes to the grid have a direct and far-reaching impact on the practice of operational management. We cannot expect the tools and processes we used to manage yesterday’s grids to deliver what we need today and in the future.
So, what’s changed? What are the developments that make today’s power networks so different from their predecessors?
We can break these down into four distinct areas:
- Attitudes and Expectations – sustainability and environmental concern have become universal concerns. The idea that an individual can make a difference affects the decisions made by domestic, commercial, and civic customers. In addition to environmental awareness, all are now increasingly digitally sophisticated. And all utility consumers, of course, remain concerned about the cost of consumption.
- Operational risk – environmental awareness has been raised, in part, because of the increasing extremes in weather – all of which impact both demand and continuity of supply. And with millions of customers, coupled with the responsibility for key infrastructure provision, utilities must be continuously vigilant in the face of cybercrime.
- Renewable revolution – the rise of renewables has had a fundamental impact on grid operations, spanning everything from the arrival of “two-way traffic” through to the need for operational agility across both low-voltage and high-voltage networks.
- Operational innovation – yesterday’s grid management tools and practices seem impossibly crude by today’s standards: smart metering, automation, machine learning and proactive asset management are just a few of the developments impacting the grid and grid management.
How change affects grid management
These manifold changes to the grid have a deep and direct impact on the tools and processes utilities need to manage the grid now and in the future:
- Yesterday’s grid ran a one-way power flow – today we must handle bi-directional flows and more.
- Yesterday’s grid was based around a small number of central generation plants – today’s grids are increasingly distributed and must be correspondingly agile.
- Yesterday’s grid could function with relatively crude forecasting – today’s cannot. Demand and delivery forecasting become dynamic, detailed and massively granular.
Digital impact across the grid
Successful digital transformation in grid management and operations requires a special focus on operational processes. These include engineering, planning and strategy development, predictive maintenance, asset health management, workforce optimization, the management of all aspects of smart metering, and substation optimization.
The development of full-scale digital twins becomes central to this extensive data-driven approach to grid modelling and management.
Naturally, the onboarding of distributed production is of particular concern. AI and machine learning also become integral to effective digital transformation.
Pre-emptive intervention and data analytics increase in importance too. With the management of field service personnel, for example, it is important to understand just how engineers and other field service personnel are deployed and supported to best effect.
There will come a time, when many aspects of grid operation will be executed in an entirely automated manner. In the future, for example, fully self-healing grids will become the norm. – But for now, utilities must ensure that the skilled staff who supervise grid operation and management are supported with precise, timely and actionable information.
Key technology components of grid operations and management
The future of the grid will be shaped by the integration of new technologies, new players and new business models. With this growing complexity, there will be a corresponding extension of the grid ecosystem. It will encompass a wide range both of opensource applications and key technologies. These include cloud, analytics, artificial intelligence, and machine learning, IoT and edge computing, communications, and cybersecurity.
The future will be characterized by data-driven operations. New solutions will provide increased visibility across the grid along with greater insight and intelligence.
Decision support will increase with distributed intelligence and autonomous systems will be supported by comprehensive digital twins. Smart metering and wide area monitoring of the grid already provide additional capabilities.
Synchrophasor measurements and advanced analytics combine with high performance computing and advanced visualization to provide new levels of situational awareness.
The future of the grid will be characterized by data-driven operations. New solutions will provide increased visibility across the grid along with greater insight and intelligence.
The grid operators’ focus has already shifted to the lower voltage levels of the distribution network. Approaches to monitoring and control have moved from centralized to decentralized models, and we anticipate increased local automation and autonomous grid function at the edge.
Advances in IoT and edge computing are central to this deployment of distributed intelligence.
Advanced grid balancing and demand-side management, leveraged by machine learning and AI, will contribute both to grid stability and reliability.
With the increase in data-driven operations, robust data governance frameworks, consolidated data models and standardized data-exchange become ever-more important. These pave the way for the creation of full system digital twins.
There will be a growth in decision-support too. Intelligent alarms and incident management functions lead to advances in decision support. We can expect the same systems to be turned to benefit in training and simulation, advancing the skills and understanding of grid operational staff.
Data-driven operations will also deliver significant benefits both in terms of maintenance and asset optimization.
Excellence in asset management (AM) is critical for grid operators. Integrating asset conditions in the business model reduces both capital and operational costs. Grid operators must optimize the use of grid assets over their lifetime. This requires the best in scheduling and maintenance methodology.
The data-driven utility shifts from time-based and activity-based asset management to condition-based intervention and predictive maintenance for proactive intervention.
Data management and governance are essential components of AM. So too are the risk management methodologies and the mathematical models needed to plot and predict asset degradation and useful life of an asset.