Adopting a digital transformation framework How a digital transformation platform can enhance the way operators monitor, manage, and optimise heat exchangers and fired heaters
Edward Ishiyama, Simon Pugh, and James Kennedy Heat Transfer Research, Inc. (HTRI)
A s the energy transition accelerates and digital transformation reshapes operations and decision-making, operators need live performance insight, predictive analytics, and cross-disciplinary collaboration. What is required is a digital transformation platform that combines modern digital twin, AI, and data integration technologies to support near real-time plant decisions ( Heat Transfer Research, Inc., 2025 ). Through direct two-way links to data historian, rigorous modelling, and intuitive dashboards, HTRI’s SmartPM (Smart Performance Monitoring, Predictive and Prescriptive Maintenance) transforms raw plant data into actionable insight (see Figure 1 ), combining: • Physics-based simulation, anchored in HTRI’s validated heat exchanger and fired heater models. • AI-enhanced learning, improving fouling predictions through real-world data. • Dynamic digital twins that evolve with time, operational constraints, and plant dynamics. • Cross-functional collaboration tools for integrated decision-making. Case study 1: Company-wide digital transformation Japan’s ENEOS Group implemented SmartPM, across 11 refineries, integrating 13 thermal network models ( Ishiyama, et al., 2022 ). At the Negishi Refinery, it initially used its conventional cleaning plan to improve the performance of the crude preheat train by increasing the furnace inlet temperature (FIT) by approximately 10ºC compared to a baseline scenario with no cleaning. However, by using SmartPM to simulate multiple cleaning strategies, engineers
Physical assets
Work groups
Data historian
Digital platform
Xist Shell-and-tube
Xace Air coolers and economisers
Xspe Spiral plate
Xphe Plate-to-frame
Xjpe Jacketed pipe
Xhpe Hairpin
Xchanger Suite Modules
developed an optimised schedule that predicted an additional 5°C improvement. The enhanced schedule strategically selected exchangers for cleaning and adjusted the cleaning intervals based on simulation data. Once implemented, the actual FIT measured in the plant aligned closely with SmartPM’s forecast, validating the accuracy of the model. At the Marifu Refinery, the team faced a challenge with the cleaning of the preheat train. Traditionally, they cleaned 20 exchanger shells over three maintenance campaigns, including both chemical and mechanical Figure 1 Schematic workflow connecting physical assets to equivalent digital twin models via HTRI technology
Refining India
33
Powered by FlippingBook