In this episode of the T&D World Live Podcast, we speak with Suresh Kotha, chief information officer for Sacramento Municipal Utility District (SMUD), about how the utility is using digital technology to modernize its electric grid and prepare for a changing energy landscape.

Kotha explains how his role at SMUD encompasses the utility's digital transformation and technology strategy, including advanced systems, artificial intelligence, cybersecurity and tools supporting the utility's 2030 zero-carbon plan. He discusses how SMUD is using technologies such as an Advanced Distribution Management System (ADMS), Distributed Energy Resource Management System (DERMS) and distributed intelligence to improve grid visibility, control and resiliency.

The conversation also examines how SMUD is preparing for large new electric loads, including data centers, hyperscalers and other AI-related developments. Kotha describes the utility's approach to working with customers and developers early in the planning process to identify suitable locations, understand capacity requirements and plan necessary grid upgrades. He also explains how IT supports this work through data forecasting, grid visibility and cybersecurity.

A significant portion of the discussion focuses on SMUD's efforts to support its 2030 zero-carbon goal through grid modernization. Kotha walks through the utility's deployment of ADMS and DERMS and explains how these systems provide greater visibility and control over distributed energy resources such as rooftop solar, utility-scale solar, electric vehicles, batteries, smart thermostats and other behind-the-meter resources. The systems are intended to help SMUD manage congestion, prevent overloads and backfeeds, and maintain grid stability as more distributed resources are connected.

Kotha also discusses SMUD's modernization of its outage management capabilities and deployment of approximately 200,000 next-generation smart meters. By bringing together outage management, distribution management and advanced metering data, the utility is working toward greater real-time visibility into grid conditions and customer restoration.

The interview then turns to wildfire mitigation, an important application of operational technology for a California utility. Kotha describes SMUD's use of LiDAR, pole-mounted sensors, field data and technology-driven vegetation mapping to identify potential hazards and guide vegetation management and infrastructure maintenance. He explains how data and analytics can help the utility focus resources where they are needed to mitigate wildfire risks.

AI and machine learning are another major topic. Kotha distinguishes between traditional AI and machine learning applications, which SMUD has used for areas such as load forecasting and theft detection, and newer generative AI applications. He says SMUD is exploring ways to use AI to improve efficiency, decision-making and customer and employee experiences, while taking a cautious approach to deploying the technology in operational environments because of cybersecurity considerations.

Kotha explains that SMUD has established an AI Governance Council with representatives from across the organization and developed an enterprise-wide policy for the ethical and responsible use of AI. The utility is identifying high-value use cases while maintaining human oversight of AI-assisted processes. One example is “Astro,” a robot dog that uses AI and robotics to assist with inspections of critical infrastructure, particularly in locations that may be difficult or unsafe for field personnel to access.

The discussion also covers SMUD's approach to data privacy and cybersecurity. Kotha describes the utility's private ChatGPT environment, called SmartGPT, which allows employees to use generative AI within SMUD's environment rather than sending organizational data to public AI tools. He also discusses the importance of ongoing monitoring, transparency, responsible deployment and regulatory compliance.

Kotha provides several examples of AI applications currently being explored or deployed, including retrieval-augmented generation applications for employees, contact-center applications, business-process work and SAP-related projects. He also notes that traditional AI and machine learning are being used for load forecasting and theft detection based on granular meter data.

As the interview concludes, Kotha returns to the challenge of rising electricity demand from electrification, population growth, housing, data centers and other large loads. He explains how OT provides real-time visibility into grid conditions, while AMI provides a bottom-up view of customer loads. Together with distribution automation, ADMS, DOMS, data analytics and forecasting, these technologies can help SMUD respond to changing demand, coordinate distributed resources and plan infrastructure investments.

The episode closes with a closer look at SMUD's next-generation smart meters. The utility has deployed 200,000 AMI 2.0 meters capable of providing five-minute interval data, compared with the hourly reads previously used. Kotha explains how the additional data can provide greater visibility into customer loads, behind-the-meter distributed energy resources and transformer health, helping SMUD identify potential equipment needs as electricity demand increases.