The challenge

Developing a reliable digital twin of a complex system such as an industrial manipulator requires more than an accurate CAD model. While some aspects of system performance depend on the control strategy, others are governed by the mechanical characteristics of the structure, joints, and transmission system.
To accurately predict the robot’s behavior under operating conditions, the mechanical properties of the system must first be identified through experimental testing. This requires a comprehensive FEM–test correlation process, in which the numerical model is calibrated against experimental measurements, allowing the digital twin to faithfully reproduce the real dynamic response of the machine.
SmartCAE’s objective was to develop and validate a finite element model of the manipulator by accurately predicting its modal characteristics and correlating the numerical results with Experimental Modal Analysis measurements.

Our contribution

Starting from the robot’s 3D CAD geometry, SmartCAE developed a detailed finite element model, paying particular attention to the representation of the joints within the kinematic chain, which have a major influence on the dynamic behavior of the system.
An extensive experimental modal testing campaign was then carried out to identify the robot’s natural frequencies and mode shapes. Measurements were performed with the manipulator positioned in multiple configurations to ensure reliable characterization across its operating workspace.
Finally, the numerical model was validated through a systematic FEM–test correlation and model updating process. Using specialized correlation software, joint stiffness values and other key model parameters were calibrated until an excellent agreement was achieved between the simulated and experimentally measured dynamic responses.

Benefits for the client

SmartCAE’s extensive experience in finite element modelling and structural dynamics enabled the customer to rely on a single engineering partner throughout the entire validation process—from numerical modelling and experimental testing to final model calibration.
The validated digital twin provided an accurate representation of the robot’s dynamic behavior, making it possible to identify the joints most susceptible to vibration and positioning errors. These insights supported targeted design improvements, increasing positioning accuracy, structural performance, and confidence in the robot’s behavior before production and deployment.