News Article

The Autonomous Grid IN THE AGE OF THE Artificial Intelligence of Things

Deploying advanced analytics and Artificial Intelligence (AI) to unlock the power of IoT devices is already transforming the utility industry. Previously, Zpryme explored how the utility industry was using machine learning in coordination with the network- connected devices that comprise the Internet of Things (IoT). since then, the AI has become more powerful, and its ability to interpret the data captured from IoT devices has increased. This report explores how AI and IoT work together to deliver everything from improved threat detection to better customer engagement for utilities.

As the pace of digital technology adoption accelerates, AI and IoT will begin to fuse into an artificial intelligence of things (AIoT), where the devices deployed by utilities are able to simultaneously gather data and learn from it in real-time to improve enterprise decision making and operational efficiency. This survey examines the future state for utilities using AI and IoT, and the deployment challenges they currently face, and provides guidance on how to accelerate business value through AIoT.

Key Findings

• 69% of utilities agree IoT is critical to the company’s success, and 57% are already using IoT technology.
• 52% of utilities agree AI is critical to the company’s success. 27% report that they are already using AI while only 16% have a specific and comprehensive AI strategy
• Utilities recognize the business value of using AI and IoT in conjunction but have not yet expansively deployed either technology
• 55% of utility respondents say that using AI and IoT in coordination will be crucial for the long-term viability, success, and growth of the industry.

Digital Transformation Underway

According to IoT Analytics2, there are now more than 7 billion IoT devices and an additional 10 billion connected devices like smartphones, tablets, and laptops. These numbers are projected to grow by an additional 3 billion by the end of 2020. Prominent utility IoT devices include smart meters, line sensors, and intelligent switches. With all these connected devices gathering an ever-increasing amount of data, making this data useful is becoming an imperative.

Machine learning was first deployed by utilities to analyze large volumes of diverse IoT data. In earlier research, utilities reported that the top three benefits they were experiencing from using machine learning techniques were improved cybersecurity, better data-driven decision making, and better customer service. These business cases for machine learning were not surprising, as the advanced algorithms that drive machine learning can optimize the monitoring and control of the power grid.

In the 2016 report, Zpryme found that utilities certainly saw the nascent benefits of IoT and machine learning. There is now a broader recognition of the power AI and IoT have to impact the industry; however, there Js still a long way to go for these technologies to achieve their maturity.

The components of AI and status of utility adoption

• Machine learning automates analytical model building. It uses methods from neural networks, statistics, operations research, and physics to find hidden insights in data without explicitly being programmed for where to look or what to conclude
• A neural network is a type of machine learning that is made up of interconnected units (like neurons) that processes information by responding to external inputs, relaying information between each unit. The process requires multiple passes at the data to find connections and derive meaning from undefined data
• Deep learning uses huge neural networks with many layers of processing units, taking advantage of advances in computing power and improved training techniques to learn complex patterns in large amounts of data. Common applications include image and speech recognition
• Cognitive computing is a subfield of AI that strives for a natural, human-like interaction with machines. Using AI and cognitive computing, the goal is for a machine to simulate human processes through the ability to interpret images and speech – and then speak coherently in response
• Computer vision relies on pattern recognition and deep learning to recognize what’s in a picture or video. When machines can process, analyze and understand images, they can capture images or videos in real time and interpret their surroundings
• Natural language processing (NLP) is the ability of computers to analyze, understand and generate human language, including speech. The next stage of NLP is natural language interaction, which allows humans to communicate with computers using normal, everyday language to perform tasks

Nearly 70% of utilities that participated in this survey said that IoT is critical to their company’s future success.

How Utilities Are Using AI and IOT

With the hype of IoT finally settling into reality, you might think that devices spanning the energy value chain would be internet connected and highly sensored. Think again. Two years after our initial report, the industry has made slow and steady progress in deploying IoT in support of major programs, but we are still on the cusp of expansive proliferation. Metering is still the top use case, followed closely by other grid-side applications.

Today’s investments are more focused on optimizing the grid, but the future expansion plans are focused on the customer. Figures 4 and 5 show the progress that utilities are making in deploying IoT devices and applying artificial intelligence to optimize programs and gain real-time insights. At the end of 2018, more utilities have deployed IoT technology than AI. However, there is commonality in the areas of focus. Metering/meter data management (50% and 35%), outage management (46% and 31%), and cybersecurity (43% and 24%) are currently the programs where utilities have made the most progress.

While most utilities are not currently using AI and IoT in tandem, there are some exciting examples from companies that are leading the way. “Our focus has been automation and analytics,” said one survey respondent. “We are using our AMI program to create an asset failure prediction model that is helping us make better decisions. However, we still have a long way to go as an organization to ensure that our people have been trained to make data available for everyone who needs it. The hardest thing is managing the data when there’s petabytes available. The key is to get in a system where people can get the data from disparate sources.”

As AIoT-driven modernization occurs, utilities expect to see a reduction in restoration times and the ability to make better decisions throughout the enterprise. Using IoT devices in the home and throughout the grid will allow AI to strategically analyze the myriad of different systems and find connections between them to improve outcomes. This should invariably lead to better engagement as energy efficiency and renewable programs spark a new connection between the utilities and their customers.

Utilities have made significant strides in the past two years going from pilots to fully implementing IoT and AI programs. However, as the volume and diversity of data grows at an almost exponential rate, the business case for turning that data into strategic operational decision making will drive utilities towards machine learning, advanced algorithms, and other components of AI.