Industrial energy users will have the most to gain as IoT and AI is incorporated into systems and processes, writes Ghian Oberholzer.
The internet of things (IoT) has had a big impact on the energy industry. Take, for example, smart water and electricity meters, along with lighting and air-conditioning systems that self-adjust according to occupancy levels and environmental conditions. But there’s another distinct area of the energy industry where IoT is having a massive impact: the supply side.
Industrial IoT (IIoT) is the internet of things applied to monitoring, controlling and optimising industrial processes. Its realisation requires integrating modern IT systems with legacy operational technology systems that predate the internet and digital transformation.
The application of IIoT can produce significant efficiency gains at every stage of energy generation and distribution and is making particular headway in the renewable energy sector.
IIoT in action
If greenhouse gas emissions are to reach net zero by the middle of this century and achieve the Intergovernmental Panel on Climate Change’s target it will be essential to maximise the efficiency of renewable energy generation, storage and distribution.
One basic application of IIoT is the use of sensors to gather information on different machines and enable remote control of energy production and distribution systems. More advanced technologies such as predictive and preventive maintenance, artificial intelligence (AI) and digital twins can bring much greater benefits.
The use of AI to analyse historical data gathered by IIoT systems from machinery can predict failure or breakdown. The technology is widely used in wind turbines, which are hard to access and often located in remote areas.
IIoT-enabled maintenance significantly reduces the risk of this downtime. Operation and maintenance can account for as much as 30% of the cost of power generated over the lifetime of the turbine. About 70% of wind turbine downtime is due to major repairs, according to SKF.
AI with IIoT: a powerful partnership
The outputs of wind and solar systems depend on the weather, which makes it difficult for operators to pre-commit to power deliveries. AI can be used on weather data to add a degree of predictability to power outputs from these systems.
In February 2019, Google’s AI division DeepMind was able to accurately predict the output of wind farms with 700MW capacity 36 hours in advance using weather forecasts.
Because power suppliers can command a higher price when they pre-commit to the amount of power they can supply, use of DeepMind’s forecasts could have boosted the value of the wind farms’ output by roughly 20%.
A digital twin is a digital version of a real-world machine or process, made possible by gathering data on every component of the machine or process. When applied to the energy sector, it has great potential.
The Offshore Renewable Energy Catapult, a UK-based company, created a digital twin of its demonstration offshore wind turbine, which gave its technicians a view of the turbine’s inner workings and equipment from any location. They were able to inspect the turbine, plan operations and deliver training and familiarisation without needing to go offshore to the turbines.
Almost every application of IIoT in the energy sector requires the integration of IT and operational technology networks. However, this integration exposes previously isolated operational technology networks to new kinds of cyber threats that are ever-present in the world of IT.
Bridging the cybersecurity gap between IT and operational technology environments is particularly challenging in the energy sector, where most IIoT assets have been isolated. This isolation meant they were never designed to be secure against the IT-based cyber threats they now face, and they are not compatible with traditional IT cybersecurity tools. Also, they often use protocols that are old, obsolete and proprietary.
Furthermore, thanks to the different technology being used across energy firms’ IIoT environments, attacks can be very difficult to detect. The disparate nature of the network that previously made life difficult for attackers is now working in their favour.
AI can greatly increase the security of IIoT implementations by automatically detecting threats and anomalies. It can also compare historical and predicted behaviour models against online patterns to remove distracting and costly false alarms and prioritise alerts and create detailed profiles of every device on a network so that anything out of the ordinary can be detected.
While integration of IT and operational technology might be bringing new security challenges today, taking proactive steps to protect the new pathways between them will enable energy providers to be more efficient and productive in the future.
Ghian Oberholzer is regional vice-president of technical operations at Claroty.