Fusion Domain-Adaptation CNN Driven by Images and Vibration Signals for Fault Diagnosis of Gearbox Cross-Working Conditions

The vibration signal of gearboxes contains abundant fault information, which can be used for condition monitoring. However, vibration signal is ineffective for some non-structural failures. In order to resolve this dilemma, infrared thermal images are introduced to combine with vibration signals via fusion domain-adaptation convolutional neural network (FDACNN), which can diagnose both structural and non-structural failures under various working conditions. First, the measured raw signals are converted into frequency and squared envelope spectrum to characterize the health states of the gearbox. Second, the sequences of the frequency and squared envelope spectrum are arranged into two-dimensional format, which are combined with infrared thermal images to form fusion data. Finally, the adversarial network is introduced to realize the state recognition of structural and non-structural faults in the unlabeled target domain. An experiment of gearbox test rigs was used for effectiveness validation by measuring both vibration and infrared thermal images. The results suggest that the proposed FDACNN method performs best in cross-domain fault diagnosis of gearboxes via multi-source heterogeneous data compared with the other four methods.

Finding the optimum groin layout for the Konaweha river banks protection via 2D numerical modeling

River bank erosion has become a critical issue, especially for river bends with an alluvial soil type. Mostly, river bank erosions have caused structural failures because many national roads in Indonesia are placed near the river bend areas. Groins can reduce flow velocity, thereby increasing riverbank stability. However, most groins were installed in the river without impacting the river flow due to the improper design. This paper analyzes the use of groins placed at the river bend of the Konaweha River to protect the river bank with a length of 250 m from erosion. The evaluation employs 2D numerical modeling using MIKE21 FM to observe the influence of the groin on the river velocity, flow distribution pattern, and water level. Sixteen model scenarios with four groin configurations were tested for 25-year discharge, 2-year discharge, normal discharge 600 m3/s, and low discharge 197 m3/s to achieve the most effective plan. Based on the simulation results, Scenarios with five groins are recommended to reduce the flow velocity along the outer river bank from 2 to 0.3 m/s, thus minimizing the erosion. In addition, the spacing of groins being twice the groin length is recommended.

Unexpected Directionality of Failures in the 1906 San Francisco Earthquake near Point Reyes Station

This study investigates the directions of structural failures and toppling near Point Reyes Station during the 1906 San Francisco earthquake (Mw 7.9). We examined archives of the Jack Mason Museum of West Marin History and other historical sources for photographs and other evidence relevant to the dynamics of the 1906 rupture in this area. Using historical maps, site investigations, and previously unpublished photographs, we determined the precise locations and orientations of several structures, including a correction to the orientation of the train that was the subject of previous studies. Based on the photographic evidence and written accounts, we estimate the direction of toppling or collapse of each structure. Nearly all objects found were thrown in a direction approximately parallel to the right-lateral San Andreas fault, and in the same direction as the static ground displacement. This suggests that fault-parallel accelerations may have been stronger than fault-normal accelerations, and that the slip on the fault may have begun slowly and stopped more suddenly.

Failure mode engineered high-energy-absorption metamaterials with biomimetic hierarchical microstructures and artificial grain boundaries

For numerous engineering applications, there is a high demand for protective lightweight structures with outstanding energy absorption performance and the ability to prevent catastrophic structural failures. In nature, most species have evolved with hierarchical biological structures that possess novel mechanical properties, including ultrahigh specific energy absorption, progressive laminated failure modes, and ability for crack arrestment, in order to defend themselves from hostile environments. In this study, a novel protective metamaterial having spherical hollow structures (SHSs) was developed with different hierarchical microstructures. An artificial failure mode engineering strategy was proposed by tailoring the microstructures of SHS unit cells. To demonstrate the effectiveness of the proposed method, a composite hierarchical SHS lattice structure was developed using a biomimetic laminated failure mode and through a hardening mechanism, mimicking crystal grain boundaries. The quasi-static compressive results indicated a significant improvement in the specific energy absorption, an enhanced plateau stress magnitude, and an obvious delay in the densification stage for the composite hierarchical SHS lattice owing to the constraining effect of its mesoscale grain boundaries and an increased number of intensively engineered laminated failure levels. This novel type of metamaterial was shown to be immensely beneficial in designing lightweight protective aerospace components such as turbine blade lattice infills.

Stress-strain state analysis and fatigue prediction of D16T alloy in the stress concentration zone under combined tension-torsion load

Engineering machines and components are proneto structural failures during their service time due to certaintechnical reasons and also due to some unforeseencircumstances. The technical breakdowns sometime lead tohigh economic imbalance and can also be fatal to life andproperty. Predicting the failure and evaluating the breakagecharacteristics of engineering components are crucial indetermining the life of the component and also increasetheir maintenance and safety in daily life. This research study deals with the modelling andnumerical simulations of an aluminium alloy specimen in3D stress-state and thereby predicting the fatigue failure ofthe material subjected to external cyclic loadings. Topredict the failure of a component, a specimen with aninduced crack can be evaluated through cyclic loadingprocess. It is based on the fact that the presence of a crackstends to modify the stresses present locally on thecomponent that the elastic deformation and the stressesattributed with them are totally insufficient for the designagainst fracture. It is based on the assumption that thespecimen undergoes complete fracture when the crackreaches its critical size even though the stress at the criticalcrack tip is much lower than the yield stress of thecomponent. The critical size of the crack is based on theapplication of the load and the number of load cycles itundergoes.The main aim of this research is to present andvalidate the numerical method for the study of theinfluence of cracks present in the engineering components.Finite element method was applied for numericalsimulation. In this study the tension, torsion, combined tension-torsion and fatigue loads was applied. Theexperimental testing data of mechanical properties wasused in numerical simulation as input data. This researchstudy investigates the three-dimensional stress-strain stateand fatigue prediction of D16T aluminium alloy which ispredominantly used in the aerospace and automobileindustries for their high strength-to-weight ratio and muchbetter physical properties. The different specimen modelsare then analysed and the most efficient one was selectedfor the preliminary experimental tests.

Automated system for diagnosing the structural failures of the unmanned vessel

The studies show that individual elements of a vessel component that are subjected to variable-repeated loads may be destroyed even at lower voltages than any other calculated component (unit). This problem is particularly important for the dynamically developing direction of unmanned navigation, which operation may be carried out with a reduced crew, in its absence – both remotely and autonomously. Conditions of cyclic variable loads in corrosive media also play an important role with a total number of operation cycles, which may reach many millions over the entire operation period. It should be noted that the statistics show that about 80% of accidents in maritime transport are caused by fatigue effects. This paper is devoted to the development of software that is able to predict fatigue failure of units and structures. The paper also descriunmanned vessel the application within an unmanned vessel, as well as the use of described mathematical algorithm in a device with the possibility of projection of augmented reality on vessel units for further prediction. The described method of predicting the strength and simplifying the service is relevant when remote security controls are required.

The throat of Brazilian Agricole production: The breakdown of vertical silos

In the current world economical conjecture, the accentuated growth in Brazilian production and exportation in the Agro-Food industry sets the country as a central role in the field. Implementing storage units in Agricole properties and industrial sites constitutes a necessary premise to keep the country as a competitive member in the global scenery. The vertical silos present themselves as alternative solutions. However, a great sum of the existent silos does not match the ideal conditions of operation due to the insufficient knowledge of the pressure variations that happen in time and inner space, of the flow, and the variables that affect the behavior of the stored products. Such situations have contributed to accidents and breakdowns in silos. With this increasing demand for the storage capacity in silos, the full understanding of the structural functioning, failure prevention, and optimization of the structural components became of great importance. Thus, this work aimed to present some events that happened in the country in the last years and bring the main causes of structural failures, highlighting the failures that are directly related to the type of flow in the interior of the silo; wacky charging and/or discharging; change in the properties of the stored products (specified weight, actual inner friction angle, friction angle against the wall) and instability in the support columns. It follows that a great number of accidents in these structures could have been avoided or reduced with previous knowledge, from the silo designer and those responsible for the operation, about the most important occurrences in silos.

Structural failures diagnosis using a hybrid artificial intelligence method / Diagnóstico de falhas estruturais utilizando um método híbrido de inteligência artificial

This paper presents a Wavelet-artificial immune system algorithm to diagnose failures in aeronautical structures. Basically, after obtaining the vibration signals in the structure, is used the wavelet module for transformed the signals into the wavelet domain. Afterward, a negative selection artificial immune system realizes the diagnosis, identifying and classifying the failures. The main application of this methodology is the auxiliary structures inspection process in order to identify and characterize the flaws, as well as perform the decisions aiming at avoiding accidents or disasters. In order to evaluate this methodology, we carried out the modeling and simulation of signals from a numerical model of an aluminum beam, representing an aircraft structure such as a wing. The results demonstrate the robustness and accuracy methodology.

Studying the Normal Operation of Grain Harvesters within the Warranty Period

Introduction. Grain harvesters are used for no more than two months within a year. They must have maximum operating reliability, since even short downtime during the harvesting period result in large crop losses. The purpose of the study is to identify the causes of combine harvester failures within the warranty period. Materials and Methods. Identifying consequences of failures and ensuring the reliability of grain harvesters are based on an improved classification of failures. In the process of studying, there have been proposed the ways to solve the problem of combine harvester downtime based on the analysis of the time for grain harvester troubleshooting. The category of severity of failure consequences was taken into account. Results. Through monitoring in the period from 2018 to 2020, there were found failures of units and systems of grain harvesters with low reliability indexes within the warranty period. Most of the failures (59.2%) were found in Russian-manufactured combines, of which operational failures are 55.9%, structural failures – 26.7%, and production failures – 17.4%. The general patterns of changes in the average time for combine troubleshooting have been determined. A geometric model of a detail for the trouble-free operation of combines (header auger shaft) was created. The finite element analysis (ANSYS) was used to identify parts, which are subject to maximum workload. There have been identified units and parts, which fail to function within the warranty period, because of design and technological defects. Discussion and Conclusion. In order to reduce the time to find the consequences of failures, it is necessary to create a more extensive network of enterprises providing a wide range of services, improve the organization of technical service and expand direct links with the manufacturers of equipment in order to respond quickly and make the necessary design and technological decisions.