PTQ Q2 2025 Issue

specifications, and rapid environmental changes such as rainstorms) needed to ensure product quality and maintain liquids in a distillation column at an appropriate level while making the maximum possible use of waste heat as a heat source. In so doing, it stabilised quality, achieved high yield, and saved energy. Autonomous control AI can greatly con - tribute to the autonomation of production. Autonomous crude switchover A virtual representation, or ‘digital twin’, of a physical asset is an indispensable source of information capable of pro - viding decision support throughout the entire lifecycle of an asset. The digital twin can contain existing information about the asset and can be used for simulation purposes to obtain information about non-measured variables, pre - dict impending alarm states based on current conditions, predict future states, optimise operations, and train opera - tors. It combines real-time data from sensors, IoT devices, and historical records with advanced analytics, AI, and ML to simulate and predict outcomes. This facilitates ongoing optimisation and informed decision-making. The 8D digital twin, an advanced digital representation of a physical asset or system that integrates eight dimensions of data to provide a comprehensive and dynamic model, incorporates the necessary components to help achieve autonomous operations. It enables real-time monitoring and control, predictive maintenance, process optimisation, and safety and risk management in a continuously learning and evolving environment. The eight dimensions of the dig - ital twin are shown in Figure 7 . A process digital twin is an essential component of a digital twin. Utilising ML, the process digital twin auton - omously self-tunes to keep models up to date to meet production plans and schedules. A cloud platform allows multiple models to be managed from one central location. Process digital twins on a cloud platform can combine mul - tiple digital twins and integrate with software applications to provide data intelligence and model insights to support decision-making and advanced levels of automation and autonomy. Autonomous crude switch Refineries and petrochemical plants make numerous crude switches and grade changes. Most often, these are done manually. However, using a combination of APC and AI/ ML, these transitions can be performed autonomously. Accurate and reliable quality predictions have been shown to significantly reduce crude switch time, resulting in less off-spec product and greater production. Industrial Automation To Industrial Autonomy demo IA2IA represents the future of automation in the refining and petrochemical industries. Autonomous operations require data and application integration, AI, and workflow automation across multiple disciplines. A demo was created to show how autonomy can be achieved in a refinery, with particular emphasis on the crude distillation unit. It highlights the difference between semi-autonomy, autonomous orchestration, and

autonomous operations. For the demo, about a dozen dif - ferent applications were integrated on a test bed with a hybrid architecture – most running in the cloud, but some on-premises. The demo was driven by a real-time optimi - sation dynamic model that simulated the plant. Other applications included a 3D digital twin (an advanced virtual representation of a physical object, system, or pro - cess that incorporates three-dimensional spatial data and real-time feedback), an energy management application, APC with AI/ML models, a robotics platform, supervisory control and data acquisition (SCADA), asset management, production accounting, planning and scheduling, opera - tions risk management, and an enterprise resource plan - ning (ERP) system. Besides hosting and integrating many applications, the cloud environment facilitated workflow orchestration. To highlight autonomous operations, several common scenarios were selected: • Autonomous analysis of crude purchasing opportunity. • Production scheduling and updating of schedules due to maintenance issues. • Autonomous crude switch. • Role-based worker enablement at levels 3 and 4 (deci - sion support). • Production accounting. • Energy optimisation. • Decarbonisation. • Robotic inspection and data integration. • Issuing of work permit to perform maintenance. The demo clearly shows what is achievable for down - stream autonomous operations. Benefits There are many benefits to achieving some level of auton - omy in the refining and petrochemical industries. More integration of automation and domain applications will provide higher levels of productivity, flexibility, efficiency, reliability, safety, and profitability. It will reduce or elim - inate human error, provide uninterrupted operations, and remove people from remote or hazardous environments. High levels of industry autonomy require a strong auto - mation base layer. This includes the use of more intelli - gent sensors, remote surveillance, and inspection through traditional approaches, robotics, and drones. Additionally, digital twins, AI, and other analytics will monitor, predict, and mitigate process and equipment failures. Robots and drones will communicate with automation and asset man - agement systems about their mission. They can perform routine operator rounds and inspections, as well as routine maintenance tasks. Tom Fiske is part of Yokogawa’s Global Strategic Technology Center, where he serves as a Principal Technology Strategist. He has more than 30 years of experience in research, product development, project man - agement, and process engineering. He has consulted with end users to address key issues concerning selection, adoption, implementation, use of manufacturing, automation and control, and production and engineering technology. Fiske holds a PhD in chemical engineering from Stevens Institute of Technology and a MSc in management of technol - ogy from the Sloan School at Massachusetts Institute of Technology.

PTQ Q2 2025