The global energy sector is ripe for disruption. This inference draws from a study conducted by the Boston Consulting Group (BCG) in collaboration with MIT. In the study, BCG interviewed 3000 energy industry experts, asking them questions related to the adoption and effects of Artificial Intelligence on the energy sector. On all counts, the energy sector ranked only higher than the public sector, trailing well behind all other industries.
The results demonstrate the sluggish adoption and lack of investment and focus on implementing digital transformation technologies in the energy sector. Such a picture is indicative of an industry that may be about to undergo rapid disruption brought about by digital transformation technologies. In this article, we explore what the future of energy may look like through the lens of disruptive digital transformation.
Internet of Things (IoT)
As a heavily mechanized industry, the energy industry stands to benefit immensely from the implementation of IoT infrastructure. One area that is currently capital intensive and can benefit from this technology is field service maintenance. For energy utilities, monitoring, maintaining and repairing assets is a huge resource sink that can benefit from disruption. IoT can do this. By deploying smart sensors that report on the state of various field assets, energy utilities can deploy “just in time” maintenance interventions that minimize downtime. This is a challenge Samsara, an IoT startup in Silicon Valley is working at solving through Internet-connected smart devices that can be deployed to multiple use-cases, including energy monitoring.
Another area IoT can help transform the energy sector is in load balancing. One challenge energy utilities face when supplying energy is knowing where to apportion energy and at what time. While most utilities have been investing heavily in smart meters and related technologies, the real gains lie in understanding how the entire grid operates in real time. This real-time grid monitoring encompasses the energy generation grid as well as the energy consumption grid, combining both to provide insights that can cut costs and boost efficiency.
Energy utilities currently generate copious amounts of data. However, only 2% of this data is captured, and this without any automation. This means that 98% of all data utilities generate is lost or stored in a raw data format that is currently underutilized. This data represents a tremendous opportunity for utilities to build data-driven asset strategies that focus more on preventive responses to asset deployment and maintenance. This approach will reduce outages and provide real-time feedback on asset performance.
Big data can also impact field workforce enablement. When workers are out in the field, one of the challenges they face is a lack of real-time visibility of assets. This forces them to respond to each issue as it arises, making it hard to predict such issues based on historical data. As such, as more energy utilities embrace big data and move towards automating the collection and processing of data, there could be efficiency gains that impact not only the utility but downstream users as well.
Following closely on big data is predictive analytics. While analytics today has become commoditized and anyone can deploy it to their operations, the real opportunity lies in the type and quality of data being fed into these predictive analytics systems. In the energy sector, the large amounts of raw data being produced by energy generation assets, monitoring assets, customer billing assets and down to even the very assets within homes can provide deep insights into how the entire grid works and how it will work in the future.
Utilities can deploy predictive analytics not only to maintenance and load balancing but also to customer use cases as well. Consider the currently ongoing deployment of smart meters. With predictive analytics, each home can be charged differently based on both historical usage data and predictive models that anticipate usage. This scenario would offer utilities an opportunity to open new marketing models that allow consumers to understand better their billing structures and how to manage their consumption at a granular level.
Safety and asset integrity are at the heart of energy utility operations. For instance, in the event of plant failure, the response is usually first to ensure personnel safety and then next to salvage assets. In such a scenario, downtime, which affects end users, is relegated to a distant third. Enter robotics. With non-human personnel on site, it is possible to reorganize the priority list and focus on minimizing downtime despite the dangers involved in restoring power.
When it comes to robotics, energy utilities lag other industries like oil and gas and manufacturing. While it is understandable that these other industries are not utilities and do not require “on the job” digital transformation, there remains a strong case for utilities to embrace robotics. Better equipped to communicate with IoT devices, process big data and run predictive analytics, robots can transform how utilities run, driving down costs and ramping up facility efficiency.
AI is the thread that links all these other technologies together. With the deployment of AI, a fully autonomous energy plant can be realized. With the previously mentioned technologies, there remains the need for a human element in running the systems. With AI, the human element is eliminated, and a truly smart and autonomous utility realized. With AI, an autonomous plant can manage internal IoT devices, listening to them and monitoring them for performance and efficiency.
The AI would then be able to generate big data from these IoT devices and other systems within the plant. Consequently, through machine learning and predictive analytics, the AI would be able to gain insights and visibility on all processes running within the plant, generating predictive performance models. Finally, by deploying robotic assets, the autonomous plant would then run predictive maintenance and efficiency-boosting optimizations to the plant’s systems. All these activities would operate in a feedback loop that better optimizes the AI for future performance.
One unique challenge energy utilities face is that they cannot rebuild their infrastructure from scratch. The fact that it is impossible to take the grid “offline” for retrofit makes this impossible. However, digital transformation makes it possible to squeeze more juice from existing assets. For instance, PPL Electric managed a 38% service reliability improvement enabled partially by the deployment of advanced analytical capabilities. Forward-thinking utilities will need to start experimenting with similar digital optimization strategies, even as they work towards deploying fully-fledged digital transformation strategies.