Methods for handling domain shifts in condition monitoring have proliferated, yet their evaluation often lacks rigorous baseline comparisons and systematic isolation of individual shift factors. This paper proposes a task-focused methodology that structures domain shift studies as a pipeline from data collection through task definition, dataset construction, representation, normalization, model specification, and training. Using bearing fault identification as a case study, we introduce the Relative Performance Drop (RPD) metric and conduct over 600,000 evaluations across the Case Western Reserve University and Paderborn University bearing datasets. Our results reveal that domain shift severity depends strongly on which physical factors vary: rotational speed causes substantial degradation (42.8% RPD), load and force have negligible impact, while shifts in fault type or severity suggest the task itself may require reframing rather than more sophisticated algorithms. Design choices often treated as implementation details can influence cross-domain performance as substantially as model architecture, with optimal choices reversing across collections. A difficulty taxonomy clusters domain pairs by their mean and variance of RPD, distinguishing shifts addressable by simple pipeline design, those requiring dedicated adaptation methods, and those indicating ill-posed task definitions. The accompanying open-source implementation enables application of this methodology to new collections and tasks with full reproducibility.

This study presents the results of fatigue tests on a composite material in which rubber is the matrix, while the reinforcement consists of a single layer of interwoven nylon cords arranged in a 0/90o configuration. The tests were carried out using cyclic, symmetrical three-point bending of composite samples. This method, due to the negligible bending-induced stress, resulted in unidirectional cyclic tension in the sample. The loading was displacement-controlled and performed in two variants: without pre-tension and with pre-tension of approximately 10%. The failure criterion was defined as the point at which the force amplitude dropped to 50% of its initial value. Analysis of the results showed that the fatigue process is well described by the fatigue characteristic in the form of maximum strain in the sample versus the number of cycles to failure. Although the strain-time profiles differed between the non-pre-tensioned and pre-tensioned samples, the observed relationship proved to be universal. The application of pre-tension introduced a wider variety of mean values, enhancing the reliability of the obtained results.

A surrogate neural network model for fatigue assessment and optimisation of an exhaust system is presented in this study. The approach is based on established fatigue analysis tools and employs a parameterised sample geometry. Conventional numerical methods were employed to generate a sufficiently large sample set and to provide stress field data. Fatigue damage calculations were subsequently performed as a post-processing step to prepare input data for the surrogate model. The resulting surrogate model is capable of predicting both the maximum fatigue damage value and its spatial location directly from geometric parameters. Due to the complexity of the prediction task and the wide range of damage values, extensive effort was devoted to model optimisation. Proprietary data pre-processing techniques proved essential for effective neural network training, and the network hyperparameters were tuned to achieve satisfactory predictive performance. To further address the wide range of damage values, A logarithmic transformation with inverse transformation correction was utilised. As sample generation is computationally expensive, the influence of sample size on prediction accuracy was also investigated. The proposed surrogate methodology enables efficient fatigue assessment and is suitable for iterative design and optimisation workflows.

The study reports on the essential level of details in simulations during the development of structural components if reliability-based design is used to ensure their quality and operational safety. A general method, which is initially introduced, is then applied to an indicator spring of a fuse element during assembly and operation stages. First, it is proven that design of simulations based on orthogonal arrays which includes variations of form, material properties and operating conditions within expected scatter limits provides a comparable determination of the scale parameter for the two-parameter Weibull distribution as the experimental observations of the same process. The shape parameter of the distribution tends to be underestimated by the simulations resulting in a higher scatter of the expected properties than experimentally measured. Next, it is shown that the maximum likelihood method to determine representative parameters of the scatter of assembly and operation stages provides a better match with experimental data than the median rank regression. Finally, a high reliability of the indication has been calculated for the fuse element if both the scatter of the assembly and the operation conditions were considered.

This paper presents a generalized elasto-viscoplasticity framework for modelling material behaviour under static, impact, and potentially cyclic loading. A theoretical equivalence between the classical Cowper-Symonds formulation and Chaboche's viscoplasticity law is established, demonstrating that both can be expressed within a unified constitutive structure. Based on this equivalence, a new procedure is proposed for identifying elastoplastic and viscoplastic parameters directly from tensile, compressive, or strain-controlled tests performed at different strain rates. The method eliminates the need for additional creep or relaxation experiments and enables decomposition of the stress-strain response into elastoplastic and viscoplastic contributions. Several variants of the viscoplasticity law are introduced, each satisfying relevant physical constraints, and a regression-based strategy is employed for robust parameter identification. The framework is validated using four materially diverse datasets: X2CrNi18-9 stainless steel, Ti6Al4V titanium alloy, short glass fibre-reinforced polybutylene terephthalate (PBT), and hydroxyl-terminated polybutadiene (HTPB) propellant. In all cases, excellent agreement with experimental data is obtained. The proposed formulation further enables prediction of stress-strain behaviour at the limiting cases of zero and infinitely high strain rates, which cannot be accessed experimentally. The results demonstrate that a single constitutive law can capture rate-dependent responses across metals, polymers, and elastomers, making it suitable for structural analyses under monotonic loading, with potential extension to cyclic, multiaxial, and thermo-mechanical conditions.

Aluminum alloy AlCu4PbMg (also known as AA2030 or EN AW-2030) is recognized for its high mechanical properties at room temperature, high fatigue performance and excellent machining characteristics but can also be successfully employed as an alloy matrix in cast metal matrix composites. It is, therefore, often the material of choice in automotive or aeronautical industries. As it is important to make accurate predictions of structural behavior either for the manufacturing process or later during the operation of mechanical components, manufactured from the selected material, representative values of material parameters used in simulations are of essential significance. This is especially important in early development stages of products as the number of design-change iterations can be considerably reduced if the predictions of the structural behavior are credible. In this study, it is shown how representative values of material parameters used in elastoplastic and viscoplastic structural analyses are determined. Elastoplastic material properties follow the Ramberg–Osgood relation while viscoplastic material properties are modeled using the Norton law. As the manufacturing or the operation of AlCu4PbMg can take place at variable temperatures, it is discussed how the material parameters for elastoplasticity and viscoplasticity are determined and how they vary across the temperature range. Finally, it is shown how the optimal set of material parameters can be used in both simulations of cooling after casting and operation of a mechanical component.

The Expectation-Maximisation (EM) algorithm is widely used for maximum likelihood estimation in incomplete data problems such as mixture modelling, but it often converges slowly, particularly when mixture components overlap substantially. This study presents a comprehensive empirical evaluation of simple EM acceleration schemes for Gaussian mixture models, comparing linear (STEM), quadratic (SQUAREM), and greedy (line search, golden section) methods across 240 simulated mixture configurations spanning three dimensionalities, four component counts, five overlap levels, and four sample sizes. A key contribution is the first systematic comparison of the three acceleration parameter estimates (α1, α2, α3) in the mixture modelling context: we show that only α3, which is derived as the geometric mean estimate of α1 and α2, provides genuine acceleration, while α1 and α2 consistently increase iteration counts by 50–110% relative to α3, effectively acting as deceleration. With α3, SQUAREM reduces iterations by up to 48% with negligible computational overhead, while greedy methods achieve similar iteration reductions but at 50–110% greater wall-clock time due to repeated log-likelihood evaluations. Crucially, acceleration does not degrade parameter estimation quality under any tested combination of initialisation, overlap, dimensionality, or number of components. We further examine the interaction between acceleration and initialisation, finding that k-means benefits most from acceleration (up to 50% time savings), while the REBMIX (Rough-Enhanced-Bayes MIXture estimation) algorithm benefits least as it already starts near the optimum. Among REBMIX configurations, histogram preprocessing with the outliers mode traversing strategy offers the best trade-off between quality and computational cost. The findings are validated on a real-world Backblaze hard drive failure dataset, confirming the practical utility of EM acceleration. All methods are implemented in the free and open-source R package rebmix, accompanied by full source code.

Sustainable cultivation of urban agricultural areas in the future can be achieved by using automated fleets of processing robots that will be able to adapt their activities to the availability of solar energy. Since this energy source varies depending on the season, it is crucial to consider several factors when designing power systems and algorithms for robot operation. As a part of the study, the impact of available solar energy in Ljubljana on the power supply of a selected Li-ion battery pack has been analysed. The pack contained 7 cells with a total capacity of 20 Ah and a nominal voltage of 24 V, which would be sufficient for approximately 4 hours of operation with 100 W of power. The results show that using the selected size of photovoltaic panels, only one battery pack can be charged on a cloudy January day, whilst the number of possible charges increases up to 20 on a clear January day. The same size of photovoltaic panels allows for the charging of as many as 55 battery packs on a clear June day. Understanding of all influencing factors is therefore key to successful automation, as fluctuations in available solar energy directly affect the performance of a fleet of processing robots.

This paper focuses on optimisation of material parameters to describe the elastoplastic stress–strain relationship in finite element solvers. Two new methods are introduced to minimise the numerical error that occurs in the interspace between the experimental cyclic stress–strain curve and its representation using multilinear interpolation. Specifically, both methods are integrated into a Prandtl operator approach, which can be used to simulate the elastoplastic response of mechanical components subjected to variable thermomechanical loadings. The improvement as compared to standard interpolation is most substantial when the number of yield planes is limited, especially in the case of bilinear stress–strain curves. The innovation of this study is an algorithm that optimises positions of the stress–strain points across the temperature range of interest considering several input temperatures. It is shown that these methods are especially applicable for optimisation of material parameters when the stress–strain curves are available for a range of test temperatures that are needed for simulating thermomechanical fatigue. The improvement in the interpolation using these methods is exhibited for two materials with available experimental results: stainless steel EN 1.4512 and polyamide PA12.

The study analyzed the influence of various factors on the strength and reliable operation of belt conveyors based on rubber composite material. It was shown that the fatigue strength of composites is a very important factor influencing operation. The degradation process results from the variable load of material between the rollers and the belt tension. This load corresponds to three-point bending with sufficient accuracy. The design of an appropriate attachment that will guarantee such a load during tests on a standard testing machine. Conveyor belts tested in this way can then be safely used in conveyors. In addition, the forces and displacements that will occur during the experiment were considered.

Climbing ropes are critical safety equipment, yet there is little data on how various stress factors affect their loadbearing capacity over time. This study investigates the individual and combined effects of wear, cyclic loading and elevated temperature on the quasi-static strength of dynamic 9.8 mm kernmantle ropes. Wear was introduced through a reproducible mechanical abrasion process; dynamic fatigue was simulated with 300 cyclic tensile loads of 15 kN; and temperature effects were investigated by exposing the ropes to 100 degrees C for up to 8 h. The results revealed that: i) severe sheath wear led to an average reduction in rope strength of 16.7 %, with the sheath losing 56.8 % and the core 1.8 % of its strength; ii) an 8-hour heat load at 100 degrees C reduced the load-bearing capacity of the rope by an average of 24.3 %, with the sheath losing 22.3 % and the core 14 % of its strength; iii) surprisingly, cyclic fatigue loading improved the static load-bearing capacity by an average of 5.5 %, possibly due to an internal rearrangement of the rope strands enhancing structural support; iv) the combined effect of wear and temperature was less than additive: maximum wear plus 8 h of heat exposure reduced the strength to 73.9 % of the original value, compared to the predicted combined decrease of 59 % if the effects were additive. These results provide critical insights into rope degradation mechanisms and demonstrate that combined stress factors interact in complex, non-linear ways.

Multiple energy dissipation sources and tuneable mechanical properties make granular materials a promising candidate for the development of superior damping devices. However, in order to utilize all these advantages, a deep understanding of the influence of various granular material parameters on relations between macro and particle-scale responses is necessary. In this study, the discrete element method (DEM) was used to isolate and evaluate the influence of inter-particle friction as one of the key parameters affecting the responses at both scales. The macroscopic response was evaluated based on the volume ratio, damping capacity, stiffness and load transmission, while an in-depth analysis of the inter-particle interactions, coordination number and number of sliding contacts was used to obtain information on particle-scale response. This approach provided valuable insights into the relation between macroscopic and particle-scale responses and showed that intermediate inter-particle friction values create a balanced force network, offering dual benefits in terms of damping capacity and structural rigidity. Our study confirms the promising potential of using granular materials as the main component of high-performance load-bearing damping elements and points to new directions for future research that can lead to significant improvements in both the scientific understanding and industrial applications of granular materials.

This study addresses the challenge of gaining a deeper understanding of charging and discharging mechanisms in lithium-ion batteries to enhance their reliability and safety, necessitating the development of novel modeling techniques. A comprehensive analytical model is introduced, capable of accurately reconstructing the voltage rise during constant-current charging. The novelty of this approach lies in its use of spectral analysis (similar to that employed in linear viscoelasticity) to describe the physical processes occurring during battery charging. The model’s effectiveness was validated using experimental data from a rechargeable lithium-ion battery with a nominal capacity of 25 Ah and a nominal voltage of 3.2 V. The results demonstrate that spectral characterization is a reliable tool for modeling battery response to constant-current charging, with the potential for application in battery lifespan prediction.

As system complexity increases, accurately capturing true system reliability becomes increasingly challenging. Rather than relying on exact analytical solutions, it is often more practical to use approximations based on observed time-to-failure data. Finite mixture models provide a flexible framework for approximating arbitrary probability density functions and are well suited for reliability modelling. A critical factor in achieving accurate approximations is the choice of parameter estimation algorithm. The REBMIX&EM algorithm, implemented in the rebmix R package, generally performs well but struggles when components of the finite mixture model overlap. To address this issue, we revisit key steps of the REBMIX algorithm and propose improvements. With these improvements, we derive parameter estimators for finite mixture models based on three parametric families commonly applied in reliability analysis: lognormal, gamma, and Weibull. We conduct a comprehensive simulation study across four system configurations, using lognormal, gamma, and Weibull distributions with varying parameters as system component time-to-failure distributions. Performance is benchmarked against five widely used R packages for finite mixture modelling. The results confirm that our proposal improves both estimation accuracy and computational efficiency, consistently outperforming existing packages. We also demonstrate that finite mixture models can approximate analytical reliability solutions with fewer components than the actual number of system components. Our proposals are also validated using a practical example from Backblaze hard drive data. All improvements are included in the open-source rebmix R package, with complete source code provided to support the broader adoption of the R programming language in reliability analysis.

This paper reviews the literature regarding the fatigue strength of rubber composites - with particular emphasis on using rubber as a matrix. Various forms and materials of rubber reinforcement in rubber-based composites were analyzed. The use of 'classic' reinforcements, such as glass, carbon, or aramid fibers, was especially considered. It was noted that carbon reinforcement can take place in various forms - not only in the form of fibers but also, for example, carbon black. Literature data on rubber reinforcement with plastics, silica, or metal elements, e.g., for the production of tires, were analyzed. What is essential from the point of view of ecological aspects is the possibility of strengthening rubber, being a natural material, with natural additives derived from plants. It was observed that in rubber composites, the reinforcement occurs not only in the form of short or granular fibers but also in the form of continuous fibers – especially in the form of cords. The paper focuses on the analysis of the fatigue properties of these composites, identifying the location of the crack and its path, the course of the hysteresis loop, the dependence between and strain under cyclic loading of these materials, and also the computational models used in the fatigue analysis of rubber-based composites.

Coupling the two main approaches for characterizing the fatigue life of rubber materials, crack nucleation and crack growth approach, provides the opportunity to find a more universal approach to fatigue characterization since both approaches characterize the same intrinsic material property. In this work, we have investigated the possibility of coupling the fatigue tests on planar tension, uniaxial and 3D dumbbell specimens. As is common in fatigue testing, the tests were performed at variable frequencies depending on the magnitude of the load and the limits of the testing machine. Fatigue test data for planar tension and 3D dumbbell specimens were taken from our previous works, whereas fatigue tests results on uniaxial specimens are presented herein. First, the fatigue crack growth curve is converted into a crack nucleation fatigue life curve for room temperature and, in the next step, into an isothermal fatigue life curve for the 3D dumbbell specimens to check the applicability of the method. These represent a more complex three-dimensional geometry with a multiaxial stress-strain field that also exhibits considerable internal heat generation and could, in practice, be replaced with a product of arbitrary geometry. With an additional transformation, the room temperature isothermal fatigue curve can be used to estimate the fatigue life for any rubber product regardless of its geometry and loading condition/level. The results presented herein show that the coupling provides a promising basis for the development of a universal method that opens the way to a faster and more versatile fatigue characterization of new rubber compounds.

The paper introduces a user material for Abaqus, detailing the modeling of elasto-viscoplasticity under diverse thermomechanical conditions. Converting constitutive equations into a robust code requires extensive efforts to solve numerous crucial numerical challenges. In addition to deriving the equations, detailing the code is also crucial for an efficient implementation of a rheological model. The algorithm for multiaxial Prandtl operator approach presented here provides both. The subroutines of the numerical code are explained in detail and solutions to ensure numerical stability are demonstrated. The multiaxial Prandtl operator approach allows a simple and effective calculation of fatigue damage, creep damage, e.g., or dissipated energy using available uniaxial methods. To demonstrate practical application, the paper illustrates the usefulness of the code by analyzing perforated plates under tension–compression and shear loading. This contribution enriches the computational modeling of elasto-viscoplasticity for the finite element method.

Hidden corrosion, not detected timely, may significantly affect the durability and integrity of aircraft structures; therefore, improvement of non-destructive testing (NDT) techniques is of high importance for reliable and safe aircraft operation. One of the commonly used NDT techniques for the detection of such a type of corrosion is the D-Sight technique, which gained wide appreciation in the aerospace sector due to its ability to perform low-cost and fast inspections of wide areas. One of the drawbacks of the method is its qualitative character, which makes it difficult to quantify corroded spots and track corrosion growth during operation. The following study aims to present recent advances in the enhancement of this technique by using advanced image processing, numerical simulations, and validation studies using reference methods (classical metrology, and digital image correlation) to make this method suitable for quantification of areas affected by hidden corrosion and monitoring its growth between periodic NDT inspections. The studies are performed on test specimens that simulate hidden corrosion and verified on results of D-Sight inspections of the Polish Air Force military aircraft. The results demonstrated good performance in identification and quantification hidden corrosion spots on D-Sight images. The approach can be used for supporting aircraft inspectors in analysis of inspection results of aircraft under operation that require periodic non-destructive testing.