MicrosoftProductionEvidence: Medium55/100

Duke Energy cuts methane leaks and boosts grid reliability

Duke Energy, a major US energy company, set a goal to reach net-zero methane emissions from its gas distribution business by 2030, aiming to surpass regulatory requirements. To achieve this, Duke partnered with Accenture and Avanade to co-innovate an AI-powered, end-to-end methane monitoring and predictive maintenance platform based on Microsoft Azure. The system harnesses satellite data, AI, and analytics to monitor and quantify emissions from gas pipelines and assets, presenting prioritized findings in visual dashboards usable at multiple organizational levels. This new approach enables near real-time leak detection, allowing workers to find and respond to leaks in minutes instead of days with traditional inspections. Advanced geolocation and AI-powered predictions help identify vulnerabilities and stop leaks before they become critical, vastly improving operational resilience. The platform sets a new industry standard for how methane emissions are tracked and managed, with plans to scale across all asset types and regions. Leadership at Duke Energy already benefits from more accurate, holistic insights for decision-making. Major reductions in methane emissions are already being achieved, directly aiding in meeting environmental commitments and safeguarding communities. The system is expected to accelerate Duke Energy’s journey toward net-zero goals and provide a model for other utilities worldwide.

Organization
Duke Energy
Published
April 2025

Planned next steps

  • The source says this outcome is planned: Set a new industry standard.
  • Supports Duke Energy in hitting net-zero methane goals for 2030.
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
Customer
Duke Energy
Provider
Microsoft
Maturity
Production
Linked source
accenture.com

Advanced geolocation and AI-powered predictions help identify vulnerabilities and stop leaks before they become critical, vastly improving operational resilience

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 2 of 2

  • 1Real-time Methane Leak Detection Using AI and Satellite Data
  • 2Predictive Maintenance for Gas Distribution Networks
  • Partnered with Accenture and Avanade to create a first-of-its-kind, Azure-based, AI-powered monitoring platform.
  • Utilized satellite data, AI, and analytics for emissions detection, quantification, and visualization.
  • Provided near real-time dashboard insights to leadership and teams at all levels.
  • Enabled predictive maintenance through advanced analytics and geolocation.
  • Streamlined leak detection and enabled immediate response.
  • Workers can identify methane leaks in minutes, down from days previously.
  • Operational system resilience improved, helping avoid major incident risks.
  • Improved insight and decision-making at all business levels.
  • Establishes a path for utilities industry standards on emissions monitoring.
  • Direct emissions reduction and lower environmental impact.
Architecture

The solution uses satellite data to continuously monitor methane emissions from gas distribution infrastructure, feeding information into the Azure cloud. AI analytics prioritize and visualize emissions data on dashboards. Geolocation tools enable rapid leak identification and predictive maintenance. Immediate alerts are generated to mobilize repair teams, with information flowing across the organization in real time.

Sources & evidence1
Evidence: Medium55/100Evidence strength
  • Customer explicitly identified
  • Deployment status explicitly supported
  • Primary source available
  • Technical implementation details available
Type: Case StudyPublished: Apr 14, 2025Publisher: accenture.comEvidence: PrimaryConfidence: High

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Was this useful?

Community

Comments

No published comments yet.