The T&D grids of tomorrow will have to be much more complex and interactive than the ones we know today. For example, the rise of renewables and demand-side response in power systems is leading to a new age of real-time controllability. Renewables will make generation more variable while demand-side response will allow energy suppliers and network operators to control consumption in order to balance the system. But to do this they will need to monitor generation as well as consumption patterns and respond in real time. Hence, optimizing grid performance will require analysing so-called ‘big data’ – a vast amount of rapid and reliable data from across the network. So, what does this mean for those who build and operate the components that make up an electricity grid?
When talking about big data and real-time control of power systems, we are, of course, talking about digitalization and microelectronics. Electronics is already widespread in metering, protection and substation automation (MPSA) functionality. However, new demand for control and status monitoring will require electronics to also become a key element of primary equipment in the grid. Transformers, circuit breakers and even cables will have to feature embedded sensors that are in constant communication with the network operations centre. In other words, grids will need to join the celebrated Internet of Things – and in a big way.
A typical substation will feature hundreds of sensors, collecting and sharing information to allow network operators to optimize operations, improve safety, reduce costs and monitor the status of their assets. The critical nature of this information then raises the question: can the data be trusted? If not accurate, the whole function of the grid could be put in jeopardy.
Analysis of results from years of testing at DNV GL’s KEMA Laboratories show that typical first-time failure rates for primary grid components are around 25%. For electronic components, such as MPSA functionality, that figure is about 80%. So, while sensors will be vital for future grids, component manufacturers, network operators and utilities need to be aware of the increased risk of failure and find ways to mitigate it. This is a particular concern given that primary equipment typically has a service life of some 50 years. Although microelectronic components offer extensive self-diagnostic functionality, their operating lifetime is much shorter than for the equipment they are monitoring. This could also have a major impact on long-term system reliability.
Another issue is the ‘electromagnetically polluted’ environment in which these sensors must operate, being surrounded by high voltages and currents. This is a real challenge for components that typically operate at very low voltage levels. Unprotected Internet of Things devices simply will not survive. We will need more robust devices and these will need to be thoroughly tested under realistic conditions.
While there are already standards for electromagnetic compatibility, many in the power industry question how well these apply given the abnormal system conditions often facing the electronics in primary equipment. New standards will be needed that more accurately reflect standard operating and, especially, fault conditions within a substation. Those standards (and the testing based on them) will need to focus on communication between devices. Increased interconnectedness means testing individual components will no longer be sufficient: utilities and grid operators will want to know that the various components work together and ‘talk to’ each other effectively, without interruption. As a result, testing will increasingly need to be carried out at a system level. And, given the complexity of future grid systems, physical testing will be essential for complete assurance. Computer simulations are just not yet up to the job.
Despite these issues, embedding microelectronics into primary components is the only way to give grid operators the tools they need to optimize use of assets in real time. Consequently, at some future point, organizations that design and build power networks will simply stop installing equipment that does not offer this functionality.
Adding sensors to grid equipment is unlikely to significantly increase costs but will bring real added value to TSOs and DSOs. For example, sensors integrated into subsea optical cables will help pinpoint hotspots and monitor their own performance and degradation. This will help operators focus their loading profile and maintenance strategy.
Component manufacturers are already working hard to deliver this added value and functionality. Those that do so successfully will have a real competitive advantage in the marketplace. They will also be laying the foundation on which the interactive and responsive grid of the future can be built.