AI 102 — Reinforcement Learning & Generative AI
Artificial Intelligence
Facilitator:
Adi Punuru, ExxonMobil Corporation
Speakers:
Brent Railey, Chevron Phillips Chemical Company LP
The two most controversial and potentially useful aspects of applied AI reside in Generative AI and Reinforcement Learning. Generative AI has applications in technical training, procedures and guided application development. Reinforcement Learning has applications in nonlinear closed loop optimization. Both areas will be of interest to refinery engineers and managers. This session is an extension of AI 101 which presented a broader view of the areas of application for AI.
Speakers
Session Start End
-
Topic
Digital Transformation: Conditioning Your Workforce for Artificial Intelligence
Artificial Intelligence
Artificial Intelligence (AI) is a transformative technology that is presently influencing, or will eventually influence, nearly every facet of our personal and professional lives. Although much of the information highlights the benefits of AI, it also raises concerns about potential negative impacts and unintended consequences. It's crucial to prepare the workforce for AI to reduce disruptions. Leaders will share their experiences with AI deployments, the methods used to ensure success, and lessons learned for future AI initiatives.
Take Aways:
- Awareness of potential impacts of AI on personnel and culture
- Methods and techniques employed to achieve and sustain positive acceptance of AI
- Insights into how AI is impacting jobs and job roles and defining new valued skill sets
Moderator:
Brent Railey, Chevron Phillips Chemical Company LP
Speakers:
Holly Fitch, Marathon Petroleum Corporation
Tyler Harnos, Big West Oil, LLC
Siva Lakshmanan, DeepHow
Session Start End
-
Topic
Driving Down Coker Giveaway With AI-Based Process Control
Artificial Intelligence
Facilitator:
Atique Malik, AIControl LLC
Speakers:
Chris Harrison, Marathon Petroleum Corporation
This session will discuss the project development and workflow, implementation, and application maintenance of an Imubit DLPC that was used to reduce giveaway on drum cycles on a delayed coker unit. The DLPC was successful in reducing the number of cycles in which the target level was not achieved in the fixed cycle time period. This resulted in an overall reduction in the number of barrels given away in each cycle. The application has achieved great acceptance by Operations.
Participants will:
- Gain an understanding of the Imubit DLPC application to reduce giveway on Coker Drum cycles
- Gain an understanding of the workflow for an Imubit DLPC project
- Gain an understanding of unique challenges and lessons learned from the project
Speakers
Session Start End
-
Topic
Implementing a Refinery Diesel Pool Optimizer Using Neural Networks and Reinforcement Learning
Artificial Intelligence, Advanced Process Controls
Facilitator:
Atique Malik, AIControl LLC
Speaker:
Yangdong Pan, Delek US
Toni Adetayo, Imubit, Inc.
Multi-unit optimization has long been a complex issue in the oil and gas industry. Despite efforts using first-principle models or empirical approaches, challenges persist. However, the emergence of machine learning and AI technologies offers an alternative solution. In particular, AI-based process control technology has shown promise for multi-unit optimization. This session will delve into an example using a distillate system optimizer to understand how these AI models address large-scale optimization challenges and how parent and child models collaborate.
Participants will learn:
- The strength of the technology and its high flexibility of handling core issues
- How the technology deals with the availability of individual units, and cooperates with unit controllers from other advanced control technologies
- How to sustain the technology’s performance and benefits
- A site-wide AI adoption strategy
Speakers
Session Start End
-
AI 101 — Introduction to Artificial Intelligence
Artificial Intelligence
Facilitator:
Rajan Rathinasabapathy, Phillips 66
Speaker:
Brent Railey, Chevron Phillips Chemical Company LP
Demistifying AI by explaining what it is, how it works without a deep dive, no specifics, discussing use cases.
Come prepared with your AI questions by viewing the pre-session AI 101 webinar.
Speakers
Session Start End
-
Topic
Prerequisites for Data Collection and Use in Safety Applications
Artificial Intelligence, Safety
Facilitator:
Atique Malik, AIControl LLC
Speakers:
Brian McClure, Marathon Petroleum Corporation
Emily Stewart, Marathon Petroleum Corporation
Discussion of the basics of good data collection, appropriate methods for filtering data, database requirements and the supporting infrastructure for possible Machine Learning applications.
Come prepared with your questions by viewing the pre-session webinar recording.
Speakers
Session Start End
-
Driving Down Coker Giveaway With AI-Based Process Control
Artificial Intelligence
Facilitator:
Atique Malik, AIControl LLC
Speakers:
Chris Harrison, Marathon Petroleum Corporation
This session will discuss the project development and workflow, implementation, and application maintenance of an Imubit DLPC that was used to reduce giveaway on drum cycles on a delayed coker unit. The DLPC was successful in reducing the number of cycles in which the target level was not achieved in the fixed cycle time period. This resulted in an overall reduction in the number of barrels given away in each cycle. The application has achieved great acceptance by Operations.
Participants will:
- Gain an understanding of the Imubit DLPC application to reduce giveway on Coker Drum cycles
- Gain an understanding of the workflow for an Imubit DLPC project
- Gain an understanding of unique challenges and lessons learned from the project
Speakers
Session Start End
-
Topic
Implementing a Refinery Diesel Pool Optimizer Using Neural Networks and Reinforcement Learning
Artificial Intelligence, Advanced Process Controls
Facilitator:
Atique Malik, AIControl LLC
Speaker:
Yangdong Pan, Delek US
Toni Adetayo, Imubit, Inc.
Multi-unit optimization has long been a complex issue in the oil and gas industry. Despite efforts using first-principle models or empirical approaches, challenges persist. However, the emergence of machine learning and AI technologies offers an alternative solution. In particular, AI-based process control technology has shown promise for multi-unit optimization. This session will delve into an example using a distillate system optimizer to understand how these AI models address large-scale optimization challenges and how parent and child models collaborate.
Participants will learn:
- The strength of the technology and its high flexibility of handling core issues
- How the technology deals with the availability of individual units, and cooperates with unit controllers from other advanced control technologies
- How to sustain the technology’s performance and benefits
- A site-wide AI adoption strategy
Speakers
Session Start End
-