PSI - Issue 84

Silvia Manarin et al. / Procedia Structural Integrity 84 (2026) 231–239

238

Figure 4 (a) Boxplot of ϑ values for 7-wire strand configuration’s measurements without outliers (b) Regression plots for 7-wire strand configuration's measurements without outliers.

measurements more accurately reflect the real stress state, particularly when surface preparation is adequate and measurement artefacts are excluded. These improvements highlight the importance of data cleaning in ensuring the validity of XRD-based stress evaluations, especially in complex geometries such as in-situ prestressing strands. Based on the data collected, it is not possible to define a universal rule for outlier removal. However, it is important to note that this process may involve the exclusion of one or more points, depending on local measurement conditions. Since this procedure reduces the number of valid data points available to the algorithm, it becomes crucial to collect the highest possible number of initial measurements. This approach ensures sufficient redundancy in the dataset to allow for robust statistical filtering without compromising the reliability of the final results. 6. Conclusion The experimental activities described in this work confirmed the potential of XRD technology as a reliable non destructive method for estimating current stresses in prestressing steel elements. Tests conducted on both single wires and 7-wire strands demonstrated that, with appropriate surface preparation and calibration procedures, the XRD system can provide accurate and repeatable stress measurements. Even without considering the presence of residual stresses at the surface, at the typical ranges of applied load in the in-service pretensioned steel members, i.e. upper 800 MPa, sufficiently accurate predictions were obtained. Surface cleaning proved to be essential for reducing measurement variability. In parallel, the optimization of acquisition parameters—such as incidence angle, acquisition time, and number of points—allowed for a balanced compromise between accuracy and operational feasibility. The transition from laboratory-controlled single wire tests to more complex strand tests revealed the additional challenges posed by the geometry and internal stress distributions typical of real prestressed elements. Nonetheless, the XRD system maintained an acceptable level of repeatability, supporting its applicability to in-situ investigations. However, this experimental campaign represents only a first, preliminary study that needs to be extended to exceed the limits of laboratory conditions. More accurate studies are needed to investigate the effects of the residual stresses’ presence, and how they are impacting the measure at different loading levels.