Composite structures during an operation are subjected to various types of external loading (impact, vibration, cyclic, etc.), which may lead to a decrease in mechanical properties. Previously, many experimental investigations of the mechanical behavior of composites under uniaxial cyclic loading were carried out. Acquisition of new data on the reduction of composite materials’ mechanical characteristics under conditions of multiaxial cyclic loading, as well as verification of existing models for calculation of the residual properties, are relevant. Therefore, this work is devoted to the experimental investigation of the mechanical behavior of fiberglass tubes under proportional cyclic loading. Static and fatigue tests were carried out under tension with torsion conditions. Inhomogeneous strain fields were obtained using a non-contact optical video system VIC-3D. The structural damage accumulation processes were analyzed by an AMSY-6 acoustic emission signals recording system. Surface defects were determined using a DinoLite microscope. Residual dynamic elastic modules were calculated during fatigue tests, and fatigue sensitivity curves were built. Data was approximated using various models, and their high descriptive capability was revealed. Damage accumulation stages were determined. The dependence of the models’ parameters on a stress state were observed. It was concluded that multiaxial cyclic loading leads to a significant decrease in mechanical properties, which should be taken into account in composite structure design.

In this paper, a novel model is presented to describe the composite mechanical properties degradation during cyclic loading. The model is based on cumulative distribution functions using. Weibull probability distribution law and beta distribution are considered. The dependences of the fatigue sensitivity coefficient on the preliminary cyclic exposure are derived. The damage value function derivative using is proposed to define damage accumulation stages boundaries. Model parameters are obtained using experimental data. Determination coefficients are calculated. A high descriptive capability is noted. Rationality and expediency of using cumulative distribution functions as the approximation of experimental data on mechanical characteristics reduction after preliminary cyclic exposure is concluded.

The paper presents an experimental study on regularities of postcritical deformation of carbon-fiber-reinforced plastic (CFRP) under static ±45° tension. The employed test method is based on ASTM D3518. Displacement and strain fields were identified by a digital image correlation method (DIC) using a VIC-3D contactless optical video system. Acoustic emission signals were obtained using an AMSY-6 system. The surface analysis of samples was carried out using a CarlZeiss SteREO Discovery. V12 optical stereomicroscope and a DinoLite microscope. Three experimental test types were considered: active loading, deformation with unloadings, and tension under additional torsion vibrations with various amplitudes. Loading diagrams were constructed; they showed a number of stages in the damage accumulation process. It was analyzed how heterogeneous strain fields develop; a neck development during softening process was observed. It was noted that the loading system rigidity influences the failure moment. The research considered various shear strain calculation methods using a “virtual extensometer” instrument. Composite mechanical properties were obtained. A shear modulus reduction during a plastic strain increase was revealed. The acoustic emission signals were analyzed; three characteristic frequency bands were observed. Most of the contribution to cumulative energy was made by matrix cracking. A reduction of the number of AE signals associated with the violation of adhesion between the fibers and the matrix during postcritical deformation was observed. The research identified basic surface defects. An appearance of the defects corresponds with their identification by the AE system. It was revealed that the presence of additional torsion vibration leads to an increase in the softening stage length. It was concluded that due regard for the postcritical deformation stage and the loading system rigidity is reasonable during the structure strength analysis.

Для обеспечения надежности и безопасности ответственных конструкций, определения их прочностных и деформационных резервов, важным является понимание процессов зарождения и развития макродефектов в твердых телах. При численном моделировании зоны, где произошло разрушение материала, могут быть представлены как области со значительно заниженными жесткостными характеристиками, по аналогии с методом переменных параметров упругости, применяемым при решении краевых задач теории пластичности. Однако формальное использование численных алгоритмов пластичности не всегда приводит к адекватному описанию процессов разрушения, особенно упруго-хрупких тел. В связи с этим в работе рассмотрены такие аспекты численных (базирующихся на методе конечных элементов) алгоритмов процессов разрушения, как пересчет при неизменных граничных условиях напряженно-деформированного состояния после редуцирования жесткости конечных элементов путем организации соответствующей итерационной процедуры, выбор максимального числа «разрушаемых» за итерацию конечных элементов, величины шага нагружения и степени дискретизации расчетной области. Влияние перечисленных аспектов на результаты моделирования процесса разрушения иллюстрируется путем сравнения численных решений задачи деформирования полосы из упруго-хрупкого материала с краевым концентратором напряжений, полученных различными алгоритмами. Построены расчетные диаграммы нагружения, продемонстрирована реализация закритической стадии на макроуровне. Анализируется кинетика процесса разрушения при различной организации итерационной процедуры и разном количестве «разрушаемых» за итерацию элементов. Выявлено, что для более точного описания процесса деформирования и разрушения целесообразнее использовать автоматически подбираемую величину шага нагружения. Получено, что степень дискретизации расчетной области оказывает значительное влияние на результаты моделирования. В связи с этим сделано предположение, что размер конечного элемента должен соответствовать некоторой прочностной константе материала, имеющей размерность длины.

In this paper, a previously developed modification of the Sainz model of multiaxial fatigue is reduced to an invariant form. Model constants were determined for different sets of setup experiments. It was supposed to introduce an additional summand to account for the phase shift between loading modes. The model is used to describe the fatigue behavior of the D16T aluminum alloy. Low-cycle fatigue tests under biaxial loading conditions are presented, with one mode changing cyclically and the other mode remaining constant in magnitude throughout the test. The results of cyclic durability prediction by the modified model provide good convergence.