Fracture Characterization and Damage Accumulation Modelling of DP1180 Steel under Proportional and Non-Proportional Loading

Abstract

Lightweighting remains a primary objective in the automotive industry, driven by the need to reduce fuel consumption and greenhouse gas emissions while meeting stringent crashworthiness standards. Advanced High Strength Steels (AHSS), such as Dual Phase 1180 (DP1180), have gained prominence due to their excellent strength-to-weight ratio. However, their complex fracture behavior under multiaxial and non-proportional loading conditions presents challenges for accurate failure prediction in structural simulations. This thesis aims to address these challenges through a systematic experimental investigation and modelling framework tailored to the fracture and damage response of DP1180 steel. The first phase of this work investigates the influence of gauge length on fracture strain in shear-dominated specimens. Conventional Digital Image Correlation (DIC) techniques were refined to enhance local strain measurement accuracy, focusing on strain localization in the shear zone. A series of tests were performed using butterfly shear specimens with varying gauge lengths to assess the lengthscale sensitivity of fracture strain. The results confirmed a strong dependence of measured fracture strain on the gauge geometry, reinforcing the need for standardized specimen design and DIC post-processing protocols. An optimized experimental configuration and robust DIC-based post-processing strategy were established to ensure consistent strain measurements for subsequent studies. The second component focuses on fracture under proportional loading conditions using uniaxial tension tests. Multiple specimen geometries were employed, including standard dogbone and notched samples, as well as conical hole expansion tests, to evaluate the fracture behavior of DP1180 under various constraints. Since fracture initiation under uniaxial tension is complicated by post-necking deformation, post-mortem surface strain analysis was performed to estimate local fracture strains. The study provided a reliable set of fracture strains for proportional loading conditions, allowing for direct comparison between different geometries and stress states. These results form the baseline for calibration and validation of fracture models under simple loading histories. The third phase of the work extends the investigation to combined loading paths involving simple shear and uniaxial tension. This approach enabled the evaluation of fracture behavior under intermediate stress states between pure shear and uniaxial tension. The resulting force-displacement responses and post-mortem strain measurements were used to validate the predictive capability of an existing phenomenological fracture model without necessitating re-calibration. The observed agreement between simulation and experiment under these combined stress states provides a robust validation of the model and highlights the versatility of the butterfly test methodology. To further extend the applicability of the framework, a novel experimental approach was developed to characterize fracture under non-proportional (bi-linear) loading paths. In this methodology, specimens were subjected to controlled proportional loading, after which miniature fracture specimens were extracted along different orientations and stress states. These samples were subsequently tested to failure, capturing the influence of pre-straining on fracture response. The collected data enabled an assessment of existing damage accumulation models under realistic forming conditions. Comparison with model predictions revealed that strain path changes significantly affect fracture strain evolution, especially for loading sequences that cross between tension- and shear-dominated states. These findings demonstrate the limitations of path-independent fracture criteria and underscore the importance of incorporating load history effects into damage modelling strategies. Overall, this thesis presents a comprehensive experimental framework for fracture characterization of AHSS under a wide range of loading conditions. The key contributions include: (1) development of a reliable shear fracture testing methodology that quantifies gauge-length sensitivity in DIC-based strain measurements, demonstrating variations in measured fracture strain depending on the selected length scale, (2) resolution of fracture strain identification under uniaxial tension through the combined use of multiple specimen geometries and post-mortem surface strain analysis, enabling the construction of a consistent proportional fracture dataset across a range of stress triaxialities, (3) validation of a phenomenological fracture model under combined shear–tension loading paths without re-calibration, showing good agreement between experimental observations and numerical predictions across intermediate stress states; and (4) development and application of a two-stage experimental methodology for evaluating fracture under non-proportional loading histories, providing a systematic assessment of path-dependent damage accumulation. Experimental results demonstrated that non-proportional loading generally leads to reduced fracture strains compared to monotonic proportional loading, with pronounced deviations governed by strain-path sequence and material anisotropy. Evaluation of the Generalized Incremental Stress State–Dependent Damage Model (GISSMO) showed that a damage exponent of 𝑛 = 2 provided the most consistent agreement with experimentally measured fracture strains across the investigated non-proportional loading conditions. Based on experimental repeatability and strain-field reliability, a hierarchy of confidence in the non-proportional fracture data was established, with v-bending tests exhibiting the highest confidence, followed by mini-biaxial, hole expansion, and shear tests. Collectively, these findings advance the understanding of path-dependent fracture and damage accumulation in DP1180 steel and provide experimentally validated guidance for improving the fidelity of forming and crashworthiness simulations involving advanced high-strength steels.