EikoSim Multi-Camera DIC: Multi-Camera Digital Image Correlation and Finite Element Model Calibration for Turbine Blades — A Collaborative Project Among Safran, EikoSim, and Ansys
Release time:
2026-05-29 16:40
Source:
This paper presents a multi-camera digital image correlation (DIC) measurement and modeling approach jointly developed by Safran Group, EikoSim, and Ansys, which is used to characterize the mechanical behavior of a turbine blade root specimen under combined tension–torsion loading and to calibrate its finite element model.
01 Introduction
As performance requirements for aerospace components continue to rise, validating simulation models has become a critical step in ensuring design reliability. For parts subjected to complex loading—such as fan module blades—computational accuracy is essential for the robustness of materials and structures.
Safran conducted experimental research on a turbine blade root to characterize its mechanical behavior under combined tension–torsion loading. The study had two objectives: first, to obtain accurate experimental data; and second, to validate the predictive accuracy of computational models. Given the approach adopted by Safran… 3D reinforced composite materials Present Orthotropic Mechanical behavior—these two objectives are particularly important.
In this collaboration, EikoSim leveraged EikoTwin Virtual and EikoTwin DIC The software provides specialized capabilities for test optimization and experimental measurement analysis; Ansys, on the other hand, is responsible for integrating measurement results within the Ansys Workbench environment and calibrating the computational model.
Figure 1 | 3D Braided Composite Fan Blade — Show the types of composite materials used in turbine blades
02 Multi-Camera DIC Test Preparation
Thorough preparation before conducting physical tests helps optimize test setup and improve measurement quality. For this blade specimen study, due to the presence of complex tension–torsion loading and the asymmetry of specimen torsion, comprehensive measurements of displacement and deformation on all specimen surfaces were required; thus, preliminary preparation was especially critical. With the aid of EikoTwin Virtual’s digital test preparation, the research achieved:
- Camera Layout Optimization: Simulate 360° coverage around the test specimen to obtain comprehensive and accurate measurement results.
- Multi-Camera DIC Quality Assessment: Verify the visibility of deformation zones through virtual preparation and estimate the uncertainty of preliminary measurements.
- Minimization of adjustment magnitude: Determine the optimal optical parameter settings during the preparation phase to reduce on-site calibration efforts and save setup time.
This method ensures the feasibility of the experiment while reducing implementation costs and uncertainties, thereby enhancing the reliability of experimental measurements and laying the groundwork for optimal comparisons with numerical simulations.
Figure 2 | Design a measurement layout in EikoTwin Virtual
Figure 3 | Comparison of Virtual Preparation and Actual Test Setup ——Side-by-side presentation of digital readiness and physical test configurations
03 Test setup
The specimen is the tenon root cross-section of a blade, subjected to monotonic tension–torsion combined loading. A multi-camera system arranged around the specimen acquires images every two seconds. Simultaneously, real-time acoustic emission monitoring detects damage initiation; upon detection, the test is immediately halted.
This study focuses on addressing the following key issues:
1 Understand the complex loading conditions of the specimen and verify the assumptions made in the computational model.
2 Detect and locate the origin of damage to define the first failure criterion.
Continuous multi-camera digital image correlation measurements make it possible to conduct a detailed analysis of the strain distribution on all surfaces of the specimen. The complex tension–torsion loading configuration requires the acquisition of reliable and comprehensive data, enabling a precise comparison between experimental displacements and strains and their numerical predictions.
Figure 4 | Images captured by the four main cameras ——Capturing the specimen from different angles
04 Image Processing and Strain Field Calculation
The captured images are analyzed using EikoTwin DIC, which applies the displacement measurement results directly to the simulation mesh. This approach offers multiple advantages:
- Directly associated with simulation: The displacement and strain fields are computed in the same spatial coordinate system (XYZ) as the finite element simulation, facilitating direct comparison and allowing for the consideration of local coordinate systems within composite material elements.
- Transient Analysis: The data are presented as a function of time, allowing examination of the evolution of loading and deformation throughout the entire test.
- Key Moment Selection: The study focuses on moments with well-defined loading conditions prior to test termination, enabling controlled experimental–numerical comparisons.
The processed data yield: displacement and strain maps (showing stress distribution), deformed contour plots with magnification factors (aiding in the interpretation of mechanical phenomena), and residual analyses to confirm the absence of visible cracks and verify undamaged regions.
Figure 5 | Visualization of vertical displacements on the four surfaces of the specimen ——Display of displacement distribution on each surface
Figure 6 | Visualization of Specimen Torsion — Displays deformation with a magnification factor
Figure 7 | DIC-related residuals ——Analysis diagram confirming the absence of visible cracks in the undamaged area
05 Collaboration with Ansys and model recalibration
After completing the analysis of experimental results, the data are compared with numerical simulations to recalibrate the computational model and improve its accuracy. During this phase, the experimental results are integrated into Ansys Workbench. A dedicated script ensures a robust comparison between experiments and simulations; its functions include:
- Directly import the measured displacement field into the simulation mesh.
- Visualize and compare experimental measurements with numerical predictions within the same coordinate system.
- Analyze the differences between experimental data and simulations to identify necessary adjustments to the model.
Figure 8 | Comparison of the measured field (left) and the calculated field (right) ——Completed using conversion scripts in Ansys Workbench
06 Definition and Application of Boundary Conditions
By adjusting the model’s boundary conditions, a more accurate fit can be achieved:
- Record the measured displacements at the top and bottom of the specimen.
- Determine the actual applied force and rotation angle
- Apply the boundary conditions derived from the experiment to the simulation.
Initially, a system with Six degrees of freedom The overall torsional model (three translations and three rotations). However, it has been shown that this approach is too rigid to accurately capture local deformation effects.
Figure 9 | Control Section for Defining Boundary Conditions in Ansys Workbench
Figure 10 | Comparison of Measured Displacements with Simulation Results ——"Twist" boundary condition
07 Calibration Optimization Based on Boundary Condition Interpolation
A more flexible approach improves the accuracy of re-registration:
✨ Interpolation of boundary conditions on the test cross-section based on surface measurement results
✨ Better accounting for local deformations, thereby more accurately reproducing mechanical behavior
The Ansys team contributed their expertise, improving script efficiency and optimizing result integration within Workbench. This collaboration reduced the computational complexity of import and calibration workflows, accelerated iterative adjustments between testing and simulation, and ensured robust comparability between numerical predictions and experimental observations. Material parameters were tuned more precisely, enabling better prediction of blade mechanical behavior.
Figure 11 | Interpolation defined within the cross-section based on surface nodes
Figure 12 | Revised results — Final optimized simulation comparison visualization
08 Significance and Prospects for Safran
The primary benefits are: enhancing the predictive accuracy of simulation models through highly detailed experimental data. Incorporating test results directly into the numerical model validation process can increase confidence in the simulations and gradually reduce reliance on physical testing. Calibrating models based on comprehensive measurement data can:
✨ Enhance the accuracy of simulation predictions and deepen the understanding of mechanical phenomena
✨ Enhancing the reliability of numerical models—this is a key enabler for “Analysis-Based Certification” (CbA)
For “Certification Based on Analysis (CbA)”
These advances enable Safran to have greater confidence in test results under complex boundary conditions, making simulation tools more reliable when faced with engineering‑scale loading scenarios, thereby optimizing component designs and reducing time‑consuming and costly experimental testing.
By continuously expanding such research and recalibration efforts, the goal is to increasingly rely on simulation in component development and optimization while reducing the need for physical testing. This reflects a global trend toward advanced digital methods—where experimentation serves to validate and refine simulation models rather than systematically verifying designs one by one.
By leveraging EikoTwin DIC for multi-camera experimental analysis and using Ansys Workbench to calibrate the model, Safran has gained a powerful tool to accelerate innovation and ensure design reliability. This study demonstrates both the significant effectiveness of integrated digital tools within rigorous engineering workflows and the critical importance of collaboration between manufacturers and specialized technical partners in addressing the challenges of the aerospace industry.
Multi-camera DIC,Virtual Construction of DIC Experiments,Simulation vs. Real-World Testing Comparison,FEM-DIC,Strain measurement,Simulation Model Validation