25.5 C
New York

Will generative AI make the digital twin promise real in the energy and utilities industry?

Published:

A digital twin is the digital representation of a physical asset. It uses real-world data (both real time and historical) combined with engineering, simulation or machine learning (ML) models to enhance operations and support human decision-making.

Overcome hurdles to optimize digital twin benefits

To realize the benefits of a digital twin, you need a data and logic integration layer, as well as role-based presentation. As illustrated in Figure 1, in any asset-intensive industry, such as energy and utilities, you must integrate various data sets, such as:

  • OT (real-time equipment, sensor and IoT data)
  • IT systems such as enterprise asset management (for example, Maximo or SAP)
  • Plant lifecycle management systems
  • ERP and various unstructured data sets, such as P&ID, visual images and acoustic data

Figure 1. Digital twins and integrated data

For the presentation layer, you can leverage various capabilities, such as 3D modeling, augmented reality and various predictive model-based health scores and criticality indices. At IBM, we strongly believe that open technologies are the required foundation of the digital twin.

When leveraging traditional ML and AI modeling technologies, you must carry out focused training for siloed AI models, which requires a lot of human supervised training. This has been a major hurdle in leveraging data—historical, current and predictive—that is generated and maintained in the siloed process and technology.

As illustrated in Figure 2, the use of generative AI increases the power of the digital twin by simulating any number of physically possible and simultaneously reasonable object states and feeding them into the networks of the digital twin.

Figure 2. Traditional AI models versus foundation models

These capabilities can help to continuously determine the state of the physical object. For example, heat maps can show where in the electricity network bottlenecks may occur due to an expected heat wave caused by intensive air conditioning usage (and how these could be addressed by intelligent switching). Along with the open technology foundation, it is important that the models are trusted and targeted to the business domain.

Generative AI and digital twin use cases in asset-intensive industries

Various use cases come into reality when you leverage generative AI for digital twin technologies in an asset-intensive industry such as energy and utilities. Consider some of the examples of use cases from our clients in the industry:

  1. Visual insights. By creating a foundational model of various utility asset classes—such as towers, transformers and lines—and by leveraging large scale visual images and adaptation to the client setup, we can utilize the neural network architectures. We can use this to scale the use of AI in identification of anomalies and damages on utility assets versus manually reviewing the image.
  2. Asset performance management. We create large-scale foundational models based on time series data and its co-relationship with work orders, event prediction, health scores, criticality index, user manuals and other unstructured data for anomaly detection. We use the models to create individual twins of assets which contain all the historical information accessible for current and future operation.
  3. Field services. We leverage retrieval-augmented generation tasks to create a question-answer feature or multi-lingual conversational chatbot (based on a documents or dynamic content from a broad knowledge base) that provides field service assistance in real time. This functionality can dramatically impact field services crew performance and increase the reliability of the energy services by answering asset-specific questions in real time without the need to redirect the end user to documentation, links or a human operator.

Generative AI and large language models (LLMs) introduce new hazards to the field of AI, and we do not claim to have all the answers to the questions that these new solutions introduce. IBM understands that driving trust and transparency in artificial intelligence is not a technological challenge, but a socio-technological challenge.

We a see large percentage of AI projects get stuck in the proof of concept, for reasons ranging from misalignment to business strategy to mistrust in the model’s results. IBM brings together vast transformation experience, industry expertise and proprietary and partner technologies. With this combination of skills and partnerships, IBM Consulting™ is uniquely suited to help businesses build the strategy and capabilities to operationalize and scale trusted AI to achieve their goals.

Currently, IBM is one of few in the market that both provides AI solutions and has a consulting practice dedicated to helping clients with the safe and responsible use of AI. IBM’s Center of Excellence for Generative AI helps clients operationalize the full AI lifecycle and develop ethically responsible generative AI solutions.

The journey of leveraging generative AI should: a) be driven by open technologies; b) ensure AI is responsible and governed to create trust in the model; and c) should empower those who use your platform. We believe that generative AI can make the digital twin promise real for the energy and utilities companies as they modernize their digital infrastructure for the clean energy transition. By engaging with IBM Consulting, you can become an AI value creator, which allows you to train, deploy and govern data and AI models. 

Learn more about IBM’s Center of Excellence for Generative AI

Global Industry Center of Excellence Leader – Energy, Environment & Utilities, IBM Consulting

Work and Asset Management leader, US Communication Sector, IBM Consulting

Related articles

spot_img

Recent articles

spot_img