The 2025 SIMULIA Energy Day was held in Houston, Texas, on Wednesday, March 19. Our own Ani Rajagopal, Worldwide SIMULIA Fluids Industry Process Consultant, presented MODSIM Story: Machine Learning Based Surrogate Model of Centrifugal Pump Design

Abstract: Centrifugal pumps play a vital role in the energy industry, supporting various applications from extraction to transportation and refining. A key component of these pumps is the diffuser—an assembly of vanes surrounding the rotating impeller—that directs fluid flow. Therefore, a well-designed diffuser is critical to ensure optimal pump performance by balancing efficiency with flow rate and head rise.

The 3DEXPERIENCE Platform’s MODSIM capabilities, powered by SOLIDWORKS, enable seamless integration of data-driven design and simulation while ensuring digital continuity throughout the design cycle. Process automation facilitates the generation of high-fidelity physics simulation data, which serves as ground truth for training machine-learning (ML) models.

Once trained, this ML-based surrogate modeling approach provides fast, accurate insights into key performance indicators, such as total pressure/head rise, and evaluates a large number of design options with a fast turnaround time. Users can create new designs and predict outcomes within seconds through an intuitive, dashboard-like interface within the 3DEXPERIENCE Platform.