Objective The discontinuity of the large opening design and the non-uniformity of the high-density cargo distribution in the cargo hold section of large ore carriers pose significant challenges to hull structural safety. In order to improve the ability of the structural health monitoring system to perceive and adapt to the structural safety state of ore carriers.

Method A real-time reconstruction method for the hull stress field under the combined action of global and local loads was proposed. A refined finite element model of the cargo hold section of a large ore carrier was selected as the research object, and the structural characteristics of the hull were retained in a relatively complete manner. The combined action of global-local and time-varying-static loads was comprehensively considered, and complex load conditions that were close to the actual ship navigation environment were constructed. According to the load forms and hull structural response characteristics, a monitoring point layout scheme and a unit load basis with various forms were designed; the monitoring points were arranged at the positions of side shell plates, main decks, bottom girders and longitudinal bulkheads. Based on the multi-source load superposition principle and limited measuring point data, the real-time reconstruction of the structural stress field in the cargo hold section of the large ore carrier was realized. Furthermore, the influence laws of dynamic load amplitude and periodic characteristics on the stress reconstruction accuracy were discussed.

Results The results showed that the method exhibited good prediction accuracy under typical wave conditions, and the prediction results could accurately reflect the spatiotemporal inhomogeneity of the hull structural response. The relative error rate of more than 97% of the reconstructed data points was controlled within 5%, and the maximum relative error rate was approximately 10%. The reconstruction error increased with the increase of load amplitude, yet maintained high reconstruction accuracy within the range of standard design values, with the error value always less than 1.0 MPa. The variation of load period had a more significant impact on the reconstruction results: the reconstruction error increased as the load period shortened. Within the common wave period range, the maximum absolute error of this method did not exceed 6 MPa, which could meet the actual engineering requirements to a certain extent.

Conclusion The proposed methodology demonstrates good engineering applicability for structural stress field reconstruction of large ore carriers, and it can effectively support hull structural health monitoring and residual life assessment.