PYPatience Yockey

Patience Yockey

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Patience Yockey has a total of 21 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Artificial Intelligence and Post-Deployment Model Drift in Grid-based Applications

    by Idaho Falls Idaho National Laboratory (Inl), Patience Yockey, Zachary Dalton, Remy Stolworthy, Jeremy Jones, Juliana Ocampo Giraldo, Lauren Ortiz

    Artificial intelligence (AI) and machine learning (ML) are increasingly integrated in grid planning, operations, and asset management as part of the broader digital transformation of the United States’ electric sector. Utilities are adopting AI-enabled analytics to support such functions as anomaly detection, predictive maintenance, forecasting, outage management, and decision support. When responsibly deployed, these applications can provide measurable benefits—improving reliability, reducing unplanned outages, and more efficiently using infrastructure—particularly as grid complexity grows due to electrification, distributed energy assets, and large AI-driven data-center loads. However, while many AI/ML models perform well at deployment, their validity cannot be assumed over time. Changes in load composition, asset behavior, weather patterns, and operating conditions can cause deployed models to diverge from real-world system behavior, a phenomenon known as model drift. When this post-deployment model drift goes unmeasured, operators risk relying on degraded AI/ML models in grid applications where accuracy directly affects economic efficiency and operational reliability

  • Open Access

    Cyber Conservative Operations

    by Patience Yockey, Sam Chanoski, Benjamin Lampe, Virginia Wright

    In an era of increasingly sophisticated and pervasive cyber threats, robust cyber resilience strategies are more critical than ever. This paper introduces the concept of Cyber Conservative Operations, a proactive approach designed to assist critical infrastructure owners and operators in managing risk and maintaining resilience in the face of imminent, yet not occurring, cyber events. By leveraging the principles of Cyber-Informed Engineering (CIE) and the energy sector’s practice of conservative operations, Cyber Conservative Operations offer a framework for planning and executing coordinated active defense and resilience-oriented actions ahead of cyber events. This approach aims to minimize the consequences of digitally-enabled hazards and ensure swift recovery from anticipated cyber threats. Cyber Conservative Operations enhance the ability of owners and operators to execute pre-planned defense and resilience actions, thereby reducing the impact of digitally-enabled hazards. These operations are crucial for addressing impacts that cannot be precisely quantified or forecasted and for preparing organizations for rapid recovery from imminent digital threats. The paper begins with a discussion of conservative operations as applied to the bulk power system (BPS), a current mechanism allowing BPS entities to enact defensive operating plans to mitigate impending grid unreliability. Building on this model, we present a concept for implementing cyber conservative operations at asset owner facilities. The paper concludes with two case studies of cyber conservative operations drawn from high-profile cyber events, illustrating the practical application and benefits of this proactive approach.

  • Open Access

    Adoption of AI in the Utility T&D Sector: Use Cases, Consequence, Assessment and Benefits

    by Emma Stewart, Idaho Falls Idaho National Laboratory (Inl), Patience Christina Yockey, Remy Vanece Stolworthy, Megan Jordan Culler

    Digital transformation and utilization of artificial intelligence (AI) in the electric grid are fundamentally changing the industry’s approach to common problems and enabling a broader paradigm shift in grid planning and operations. The change in approach is circularly both enabling and driving modernization, with load growth and reliable management of data center and AI infrastructure shifting away from planning approaches with relatively predictable behaviors and toward a mix of consumer and industrial choices that surpass human cognitive abilities to process. This movement has potential to condition humans to not understand the system on which the AI depends, while requiring it for development of the necessary infrastructure. Approaches which would address most likely grid conditions and events, such as faults, aging of equipment, and weather, now must also account for large loads which shift not based upon weather or time of day, but the computational load. Quantifying computational load is independent of the traditional grid forecasting variables, where a data center’s aggregate load is determined by user and AI system behavior and decoupled from normal grid planning and operations. AI is both the cause and solution for these challenges, with new grid planning tools integrating massive amounts of decisions into frameworks. The race to dominate the AI field between the United States and China has significant positive and negative implications for both technological advancement and national security. Domination of the field is not an isolated function of AI technology; instead, the U.S. must grow its capacity to handle large AI-driven electronic loads, develop new models, and drive adoption. As America invests heavily in AI research and development, the country’s goal is to achieve breakthroughs that could redefine industries, economies, and military capabilities. However, the intense competition from adversarial nations also heightens security risks, including the potential for cyber espionage, intellectual property theft, and the misuse of AI technologies from the increased digital attack surface. Ensuring the responsible appropriate use and robust security of AI systems is paramount to preventing malicious exploitation and maintaining global stability. The U.S. must navigate this AI race carefully, balancing innovation with international cooperation to address critical security concerns, specifically when it comes to the integration of AI with critical functions such as the electrical power grid. The electric grid is evolving into a complex cyber-physical system where AI is increasingly applied to enhance monitoring, control, and optimization. Grid operators are already using AI to monitor transmission lines and isolate faults, and to predict fluctuations in electricity supply and demand. As they look to grow and thrive in this new reality, utilities must also consider their approach to a secure and responsible use of the load on which the AI is driven along with the AI itself. Many AI applications, will be integrated into these tools, along with new requirements to implement cloud computing into operational technology (OT) operations, which also increases utility use of data centers that house cloud computing platforms. Therefore, risk frameworks must also address these interdependent concerns to enable their responsible and secure use as rapidly as AI adoption. Energy vendors and service providers are racing to deploy AI within their tools, products, and services to demonstrate their acceptance and enriched modernization capabilities. Though acting fast (and even failing fast) are qualities needed to help win the AI race, this comes at a cost to early adopters of these technologies, who will bear the burden of poor implementations, evolving models, and nascent application spaces. When these applications are in critical infrastructure sectors like the grid, the costs of these challenges could be significant (e.g., safety or reliability impacts). This paper focuses on key aspects of AI adoption in the electric energy sector through three modalities: AI model integration inside the organization, purchase of commercial tools with AI integrated, and edge device integration. The authors summarize key risk-informed insights for AI use in electric grid operations and planning, which includes a review of the current use cases for the electric grid and a simple analysis framework to understand AI impact, coupled with application decomposition and consequence-driven analyses of the application’s security and impact and in which case security controls can be applied. By doing this style of analysis, risk-informed controls can be placed around technologies, enabling their use and reducing the risk of the highest consequence repercussions. Like cloud data models, AI introduces new shared risk boundaries among utilities, vendors, regulators, AI developers, maintainers and support. Similar to other d


  • Open Access

    Cyber-Informed Engineering Validation Methods and Guidance

    by Patience Christina Yockey, Daniel Quinones, Reid Robison, Success Oluwole, Benjamin Ruhlig Lampe, Daniel Groves, Jeremy Smith, Andrew Ohrt, Joel Cox

    Validation is an important step in any systems engineering process to ensure the correct system was made to fulfill stakeholders’ needs, goals, and expectations. In the context of Cyber-Informed Engineering (CIE), validation ensures cyber impact is reduced through implemented design choices and CIE requirements. This document details a process in validating CIE-based design choices relative to their effectiveness at mitigating high consequence events. The document includes a case study to illustrate the CIE validation process. The case study explores the implementation of CIE validation within the engineering lifecycle of a chemical mixing plant.