Type Of Failure Research Articles

Cyclic behaviour of welded stainless steel beam-to-column connections: Experimental and numerical study

This paper presents an experimental and numerical programme on six welded stainless steel beam-to-column connections subjected to cyclic loading. The test specimens were designed in accordance with Eurocodes and comprised an I-section beam made from 316 L stainless steel welded to an I-section column made from 304 L stainless steel. Each specimen was subjected to a quasi-static cyclic loading protocol and the load-displacement curves are monitored and reported. The results are analysed in terms of failure characteristics, skeleton curve, strain distribution and energy dissipation. Typical failure modes are combined global flexural and torsional buckling and localised web and flange buckling of the beams. The slope of skeleton curve in the hardening region and the energy dissipation capacity appeared to reduce with the section size. Numerical models capable of successfully replicating the experimentally observed response were developed. The load-displacement behaviour and the failure pattern were compared to the experimental ones, showing a good validation. Overall, it has been demonstrated that good hysteretic performance has been achieved for the welded stainless steel beam-to-column connections.

Fracture Resistance of Endodontically Treated Teeth Restored Using Multifiber Posts Compared with Single Fiber Posts

Purpose. The aim of this in vitro study was to evaluate the fracture resistance and type of failure of two adhesive fiber post systems used to restore endodontically treated teeth. Material and Methods. Twenty-seven extracted premolars were selected and divided into three groups (n=9): a control group restored with direct composite core (group 1), teeth restored with single fiberglass posts (group 2), and teeth were restored with multifiber posts (Biolight Plus System) (group 3). Fracture resistance was measured by applying axial compressive loads parallel to the longitudinal axis of the tooth until failure. Data was analyzed with one-way ANOVA followed by post hoc Turkey tests. Results. The results showed that the mean forces to failure of the control group (0.068 kN) were significantly lower than those restored with either fiberglass post systems (p=0.013). There was no significant difference between the multifiber and the single fiberglass post system in terms of resistance to fracture (p=0.097). Although there are more teeth fractured favorably (above the CEJ) in the Biolight Plus group (77.7%) compared to both the control and fiberglass post groups, it was not statistically significant (p=0.226). Conclusion. Within the limitations of this study, restoring endodontically treated with a multifiber post system is an adequate alternative to single fiberglass post system in terms of resistance to fracture.

How does AI agent (vs. IVR system) service failure impact customer purchase behavior: mediating effect of customer involvement

ABSTRACT In recent years, the use of artificial intelligence (AI) agents in customer service has become a prominent area of study, focusing on their influence over consumer behavior. However, there is a noticeable gap in the current literature regarding AI agent service failures during product information consultations and their effects on customer purchase behavior. The present study, grounded in attribution theory, investigates the effects of various types of self-service failures in both Interactive Voice Response (IVR) systems and AI agents on customer involvement and subsequent purchase behavior. The findings indicate that customers experiencing service failures with AI agents during the consultation stage are more inclined to make purchase behavior after human-led recovery compared to those encountering IVR system failures. Additionally, customer involvement is found to play a pivotal role as a mediator between types of service failures and purchase behavior. The results of this work may offer valuable theoretical and practical insights.

Recycling harmful plastic waste to produce a fiber equivalent to carbon fiber reinforced polymer for reinforcement and rehabilitation of structural members

Abstract The main idea of this research is to manufacture plastic fibers (PFs) by recycling harmful plastic waste after washing it with water and drying it, then cutting it into strips and immersing them in liquid nitrogen, using a rotating iron basin mixer containing iron balls that touch these strips, producing fibers that are woven later to be similar to carbon-fiber-reinforced polymer (CFRP). This process is the first in the world, and the cost of manufacturing these fibers is very low compared to the cost of producing CFRP, as the cost of producing these fibers is 25 times less than the cost of producing CFRP. All the physical, chemical, and mechanical properties of the manufactured PFs were investigated and compared with the properties of CFRP. In this research, it was found that the manufactured PFs have properties similar to the properties of CFRP. The experimental work consists of casting five reinforced concrete beams with a cross-section of 125 mm × 200 mm and a length of 1,200 mm. They were divided into two groups, each group containing two beams, one of which was reinforced by CFRP, and the other was repaired by CFRP. The second group also contained two beams, one of which was reinforced by PF, and the other was repaired by PF, while the fifth beam was a control beam. From the experimental results, a significant increase in the ultimate load of the beams reinforced by PF or CFRP ranged between 45.45 and 51%, respectively. This is due to the role of manufactured PFs and CFRP in restricting the width of cracks in the bending zone of reinforced or repaired concrete beams. Also, there was a change in the type of failure when changing the type of fiber, where the failure was ductile and gradual in the beams reinforced or repaired with manufactured PFs, unlike what was in the beams reinforced or repaired by CFRP, where the failure was brittle and sudden.

Tornado‐induced vibration assessment of construction elevator attached to high‐rise buildings

SummaryAs one of major construction facilities, the safety concern about elevator outer‐attached to slender high‐rise buildings during construction needs to be seriously addressed. To specifically identify the risk of such slender building–elevator systems in extreme tornado wind field, a safety assessment strategy using the critical velocity envelope is systematically developed. To overcome the difficulties in modeling such building–elevator system, a modified generalized flexural‐shear model developed previously was employed to reach an efficient estimate on the response of a case high‐rise building under simulated tornado wind excitations. To account for the effect of the flexibility of high‐rise building on the response of its attached construction elevator, the response of the elevator was obtained by finite element analysis with applied wind forces, as well as the displacement excitations at its supports with the building. To simulate the typical failure modes of elevator in all states, some critical reference velocity‐based envelopes were generated for the safety assessment where the factors of wind velocity, wind direction, and cage level are systematically addressed. The safety envelopes of the building–elevator system were proved to be instinctively efficient for the assessment of its wind resistance under various scenarios.

Mechanism of overburden fracture induced earthquakes in coal seam mining

Mining-induced earthquakes are unnatural seismic events that frequently occur in high-position hard and thick rock strata during coal mining. Considering the frequent occurrence of strong mining-induced earthquakes in the Dongtan mining area, this study analysed the fracture migration characteristics of hard and thick rock strata and the focal mechanism of mining-induced earthquakes based on Volasov’s thick-plate and moment tensor inversion theories. The results showed that the main key strata were difficult to break under single-panel mining conditions because of the thick and high-strength rock strata and breakage of the main key strata is caused by multiple-panel mining. Volasov’s thick-plate theoretical calculation indicated an initial fracture span of the main key strata was 314 m, which is consistent with the actual mining distance of the working face. This verified that strong mining-induced earthquakes were induced by the initial fracture of the main key strata. In coal mining, the pure shear failure type of mining-induced earthquakes indicated the highest percentage, and the shear fracture of rock strata was the primary cause of strong mining-induced earthquakes. The dip angle of the focal fracture surface in mining-induced earthquakes was generally within 15°. Through an analysis of the focal mechanism of mining-induced earthquakes, it has a certain guiding role in explaining the mechanism of mining-induced earthquakes.

Flexural Performance of Reinforced Concrete Beams Strengthened with a Novel High-Strength and High-Toughness Epoxy Mortar Thin Layer

The flexural performance of RC beams strengthened with a novel high-strength and high-toughness epoxy mortar thin layer was investigated through four-point flexural tests on two contrast beams and two strengthened beams. The effects of this strengthening method on the failure modes, crack distribution, load–deflection curves, and bearing capacity of the RC beams with two reinforcement ratios were studied. The experimental results revealed that the contrast beams exhibited the typical bending failure modes where the failure mode of the reinforced beam is the yielding of the tensile reinforcement of the original beam and then fracture damage of the new epoxy mortar-reinforced thin layer. No debonding phenomenon was observed between the reinforced thin layer and the original concrete, and no visible cracks appeared before the tensile failure occurred in the thin layer. The cracks in the reinforced beams developed slowly, increased in number, and decreased significantly in width and spacing. The stiffness of the strengthened beam increased significantly, while its deformation ductility coefficient noticeably decreased. Compared to the corresponding contrast beams, the cracking load for strengthened beams A1 and B1 increased by 14% and 23%, respectively; the yield load increased by 32% and 40%, respectively; and the peak load increased by 18% and 17%, respectively. Finally, a calculation method for the flexural bearing capacity of RC beams strengthened with the novel epoxy mortar thin layer based on the flat section assumption was proposed. The calculated values showed a good agreement with the experimental values (with errors at −11.73% and 4.14%, respectively), providing a valuable reference for further research and application related to this kind of reinforcement method.

Failures of geometrically asymmetric steel-UHPC composite beams

Geometrically asymmetric steel-ultra-high performance concrete (UHPC) composite structures (SUCSs) suffer several typical failure modes during applications in structural engineering. But the competing mechanism of these failure modes remains a challenge to be urgently overcome. In the present paper, the failure modes of the geometrically asymmetric SUCSs under lateral loads are theoretically predicted and numerically studied. Considering both the geometric asymmetry of the structure and the tensile-compressive asymmetry of UHPC, a theoretical model is proposed to predict the critical failure load for each failure mode (e.g., bending and core shear), and then a failure mechanism map is constructed by comparing each critical failure load. The theoretical predictions are in line with the present numerical results and the previous experimental measurements. The results demonstrate that geometric asymmetry, together with some other geometric parameters and material properties, have remarkable influences on the failure modes of structures. This work provides an effective approach for predicting and designing the failure modes of geometrically asymmetric SUCSs in structural engineering applications, such as undersea tunnels, offshore platforms, and nuclear shielding walls.

Risk of conversion to total knee arthroplasty after surgically treated tibial plateau fractures: an observational cohort study of 439 patients.

We aimed to assess joint failure rate, i.e., subsequent conversion to TKA after surgical treatment of a tibial plateau fracture (TPF). Secondary aims were to explore the association between joint failure and fracture type, and to determine the risk of failure associated with inadequate joint surface reduction. We included all patients ≥ 18 years of age with a surgically treated TPF, treated at Uppsala University Hospital between 2002 and 2015. All fractures were classified according to the Arbeitsgemeinschaft für Osteosynthesefragen/Orthopaedic Traumatology Association (AO/OTA) classification. Postoperative radiographs were evaluated to determine the quality of joint surface inadequate reduction, defined as an articular step-off ≥ 2 mm. The study cohort was linked with the Swedish Arthroplasty Register (SAR) for information on subsequent total knee arthroplasty (TKA). 439 patients (57% women) with a mean age of 55 years (SD 17) were included. According to the AO/OTA classification, the fracture distribution was B1: 4.8%, B2: 10%, B3: 47%, C1: 12%, C2: 6.4%, and C3: 19%. 23 patients (5.2%) were converted to a TKA within 2 years of initial surgery, and 34 patients (7.7%) had been converted by the end of follow-up (16 years). AO/OTA type B3 and C3 had a 6.8 (95% confidence interval [CI] 1.6-29) times greater risk of joint failure compared with B1-2 and C1-C2 at 2 years' follow-up. Inadequate joint surface reduction led to an 8.4 (CI 3.6-20) times greater risk of conversion to TKA at 2 years' follow-up. Overall, 5.2% were converted to a TKA within 2 years. Fracture types AO/OTA B3 and C3 with a comminuted articular surface and inadequate joint surface reduction were strongly associated with joint failure.

A Hierarchical Fault Diagnosis Model for Planetary Gearbox with Shift-invariant Dictionary and OMPAN

Abstract Planetary gearbox has been widely applied in the mechanical transmission system, and the failure types of planetary gearbox are more and more diversified. The conventional fault diagnosis methods focus on identifying the faults in the fault library, but ignored the faults outside the fault library. However, it is impossible to build a fault library for all failure types. Targeting the problem of identifying the faults outside the fault library, a hierarchical fault diagnosis method for planetary gearbox with shift-invariant dictionary and orthogonal matching pursuit with adaptive noise (OMPAN) is proposed in this paper. By k-means singular value decomposition (K-SVD) dictionary learning method and shift-invariant strategy, a shift-invariant dictionary is constructed so that the normal modulation components of signals can be completed decomposed. OMPAN algorithm is proposed, which uses the white Gaussian noise to improve the solution method of the orthogonal matching pursuit (OMP) algorithm so that it can separate the modulation components in the signal more accurately. The fault feature extraction is developed via shift-invariant dictionary and OMPAN. A hierarchical classifier is proposed with 3 sub-classifiers so that both the faults in the fault library and the faults outside the fault library are identified. The effectiveness of the proposed hierarchical fault diagnosis method is validated by a planetary gearbox experiment. Result show that the proposed shift-invariant dictionary and OMPAN method has achieved a superior performance in highlighting fault features compared with other two sparse decomposition methods. The proposed hierarchical fault diagnosis approach has achieved a good performance both in classification of the faults in the fault library and identification of the faults outside the fault library.

Bond Strength of Resin Cement Following Biomimetic Remineralization: An in vitro Study

Objectives This research assessed the efficacy of self-assembling peptide P11-4 (Curodont protect) in promoting biomimetic remineralization of dentin, thereby stabilizing the hybrid layer and enhancing the long-lasting of the resin-dentin bond. Methods Five premolar teeth were longitudinally sectioned to assess dentin microhardness before and after demineralization, as well as after the application of P11-4 (Curodont protect). For the microtensile test, ten premolar teeth were sectioned perpendicular to their long axes. The teeth were assigned to different groups. Following acid etching and rinsing, samples in control group I received no pretreatment, while the prepared dentin surface of group II was treated with 0.1 mL of P11-4 before the placement of the adhesive restoration. All specimens were stored in distilled water at 37 ± 2 °C for 24 hours and then underwent thermocycling. The microtensile test was measured, and the type of bond failure was evaluated. The obtained information, expressed in megapascals, was analyzed using a one-way analysis of variance (ANOVA), followed by Tukey’s multiple post hoc test. Results There was a statistically significant difference in hardness values after demineralization, followed by a significant increase in hardness after the application of P11-4 (Curodont protect) (p < 0.001, effect size = 0.985). No statistically significant difference was found in the microtensile bond strength between the two groups (p = 0.384, effect size = 0.582). Adhesive-type failure was more commonly observed. Conclusion The use of self-assembling peptide P11-4 (Curodont protect) on dentin resulted in higher microhardness than the control group without adversely affecting resin cement bond strength.

Effect of surface treatment and resin cement type on the bond strength of polyetheretherketone to lithium disilicate ceramic

BackgroundThis study aims to evaluate the effect of surface treatment and resin cement on the shear bond strength (SBS) and mode of failure of polyetheretherketone (PEEK) to lithium disilicate ceramic (LDC). This is suggested to study alternative veneering of PEEK frameworks with a ceramic material.Methodseighty discs were prepared from PEEK blank and from lithium disilicate ceramic. Samples were divided into four groups according to surface treatment: Group (A) air abraded with 110 μm Al2O3, Group (AP) air abrasion and primer application, Group (S) 98% sulfuric acid etching for 60 s, Group (SP) Sulfuric acid and primer. Each group was subdivided into two subgroups based on resin cement type used for bonding LDC:1) subgroup (L) self- adhesive resin cement and 2) subgroup (B) conventional resin cement (n = 10). Thermocycling was done for all samples. The bond strength was assessed using the shear bond strength test (SBS). Failure mode analysis was done at 50X magnification with a stereomicroscope. Samples were chosen from each group for scanning electron microscope (SEM). The three-way nested ANOVA followed by Tukey’s post hoc test were used for statistical analysis of results. Comparisons of effects were done utilizing one way ANOVA and (p < 0.05).ResultsThe highest mean of shear bond strength values was demonstrated in Group of air abrasion with primer application using conventional resin cement (APB) (12.21 ± 2.14 MPa). Sulfuric acid groups showed lower shear bond strength values and the majority failed in thermocycling especially when no primer was applied. The failure mode analysis showed that the predominant failure type was adhesive failure between cement and PEEK, while the remaining was mixed failure between cement and PEEK.ConclusionThe air abrasion followed by primer application and conventional resin cement used for bonding Lithium Disilicate to PEEK achieved the best bond strength. Primer application did not have an effect when self-adhesive resin cement was used in air-abraded groups. Priming step is mandatory whenever sulfuric acid etching surface treatment is utilized for PEEK.

Subgrade soil response to rail loading: Instability mechanisms, causative factors, and preventive measures

The demand for more efficient heavy-haul rail networks over soft subgrades poses significant geotechnical challenges and requires a comprehensive understanding of stress conditions as well as the failure potential of subgrade soil under moving wheel loads and increasing rail speeds. Unfavourable stress conditions in the subgrade can result in various types of failures, three of which are identified in this article: (i) excessive plastic settlement, (ii) progressive shear failure, and (iii) subgrade fluidisation (mud pumping). Through a series of advanced testing schemes using cyclic triaxial, hollow cylinder, and an in-house dynamic filtration apparatus, critical stress conditions and soil characteristics prone to subgrade instability are discussed. The results demonstrate that under adverse combinations of loading frequency (f) and cyclic stress ratio (CSR) the continuous application of cyclic loads can lead to an unstable state of soil where excess pore pressure and axial strain increase rapidly. This study also reveals that low to medium plasticity soils (PI < 22) are more vulnerable to subgrade fluidisation, where the rapid internal migration of pore water transforms the upper soil to a fluid-like state with substantial loss in soil stiffness. The layered response of soil through dynamic filtration tests showed larger hydrodynamic forces induced by differential hydraulic gradients in the top layer during cyclic loading causes moisture to move upwards. Various factors that can influence soil instability such as the degree of compaction degree, clay content, soil fabric and stress rotation are also addressed in this paper. Finally, novel solutions for stabilising subgrade such as a vertical drain-composite system and the use of eco-friendly biopolymers are presented.

Fracture propagation and pore pressure evolution characteristics induced by hydraulic and pneumatic fracturing of coal.

A two-dimensional unsteady seepage model for coal using a finite element program is developed, and the temporal variations of key factors such as water pressure and hydraulic gradient are analyzed in this paper. Additionally, the triaxial rock mechanical experiment and utilized pneumatic fracturing equipment on raw coal samples to investigate both hydraulic and pneumatic fracturing processes are conducted. Through these experiments, the relationship between pressure and crack formation and expansion are examined. The analysis reveals that the pore pressure gradient at the coal inlet reaches its peak during rapid surges in water pressure but diminishes over time. Conversely, the pore pressure gradient at the outlet side exhibits a gradual increase. Hydraulic fracturing is most likely to occur at the water inlet during sudden increases in water pressure. Besides, as the permeability of coal decreases, the duration for seepage stabilization prolongs due to the intensified pore pressure gradient resulting from sudden increases in water pressure. Moreover, an extended period of high hydraulic gradient further increases the risk of hydraulic fracturing. The experimental findings indicate that coal samples initially experience tensile failure influenced by water and air pressure. Subsequently, mode I cracks form under pressure, propagating along the fracture surface and becoming visible. The main types of failure observed in hydraulic and pneumatic fracturing are diametrical tensile failure, and the development of fractures can be categorized into three distinct stages, which contains the initial stage characterized by slight volume changes while water pressure increases, the expansion stage when pressure reaches the failure strength, and the crack closure stage marked by little or even decreasing volume changes during pressure unloading. The acoustic emission signal accurately corresponds to these three stages.

Analytical solution for estimating the minimum required strength of sill mat by considering failure due to rock wall closure

Underhand cut-and-fill mining has been widely used in underground mining operations, especially when the rock mass or orebody is of poor quality or prone to rockburst due to high stress. In such cases, mining workers should carry out all production activities under the cemented backfill roof or sill mat instead of a highly fractured and unstable rock roof or a strong rock roof with a high potential of rockburst. Therefore, the stability and required strength of the sill mat are critical issues for mining engineers. In 1991, Mitchell considered that sill mat could fail by caving, sliding, rotation, and flexure. Mitchell also proposed an analytical solution to determine the minimum required strength of the sill mat for each type of failure based on two stiff or immobile rock walls. However, recent publications using numerical modeling and field measurements indicate that the compressive stresses in the sill mat induced by rock wall closure due to a stope excavation beneath the sill mat can be significant. It is thus highly necessary to investigate the required strength of the sill mat by considering rock wall closure. In this study, the crushing failure of sill mat due to rock wall closure generated by underground excavation and a new failure mode called "crushing and caving" is revealed by numerical modeling. An analytical solution corresponding to each failure mode is then developed to estimate the minimum required cohesion () of the sill mat. A criterion is also proposed to determine if the sill mat fails by crushing or crushing-and-caving failure. The proposed analytical solution does not involve any correction coefficients. The validity of the proposed analytical solution is demonstrated by numerical modeling. The proposed analytical solution can thus be employed to predict the of sill mat subjected to wall closure generated by underlying stope excavation.

Dynamic human error risk assessment of group decision-making in extreme cooperative scenario

For groups in extreme environments, their reliability requirements in complex operational systems are high due to isolation from society, physical constraints for extended periods of time, and exposure to significant hazards. This study attempts to solve the problem of dynamic quantitative risk analysis of decision-making process for extreme group during complex cooperative scenarios. Based on an extended IDA team cognitive model, typical failure modes and performance quantification mechanisms of team communication were analyzed. The performance curve was utilized to distinctively model the cognitive behaviors of operators within the team, driving the cognitive transmission paths of group decision-making in a hybrid event tree. Error calibration of human error due to time uncertainty was conducted. The steps and techniques for dynamic risk analysis of decision-making of extreme groups under complex, dynamic, and ambiguous scenarios were summarized. The proposed method was validated through simulation in a typical powered diving task of manned submarine.

Influence of adhesive systems containing different functional monomers on the longevity of bond strength at different dentin depths

This study evaluated the influence of adhesive systems containing different functional monomers on the longevity of bond strength (BS) to superficial (SD) and deep dentin (DD). Thirty molars were sectioned longitudinally, to obtain regions of SD (closest to the dentin-enamel junction) and DD (closest to the pulp chamber). Adhesive systems were applied (n = 10): Clearfil SE Bond (containing 10-MDP), FL-Bond II (containing 4-MET) and OptiBond Universal (containing GPDM). Microcylinders were made of resin composite (Filtek Z250) in SD and DD. After 24 h or 180 days of storage in Simulated Body Fluid (SBF) solution, BS tests were carried out. The fracture mode was evaluated by scores. ANOVA showed that there was no statistically significant difference between the adhesives in terms of BS (p = 0.5156). There was no significant difference between storage times for the three adhesive systems (p = 0.4703). There was a significant interaction between storage time and dentin depth (p = 0.0175), and in the time interval of 24 h, irrespective of the adhesive system, BS was significantly higher in SD than in DD (p < 0.05). In the time interval of 24 h, there was a significant association between the adhesive and the type of failure in SD (p = 0.0085) (Fisher's Exact test). Under these conditions, only OptiBond Universal showed cohesive failures in dentin (10 %) and mixed failures (15 %). After 180 days of storage, the percentages of early failures in deep dentin were 25.0 % (OptiBond Universal), 16.7 % (Clearfil SE Bond) and 0.0 % (FL-Bond II), with a significant difference between adhesives regarding early failures (p = 0.0410). The adhesive systems showed similar BS at different dentin depths over time, irrespective of the composition of functional monomers. FL Bond II was the only adhesive system that did not show early failures mode in DD within 180 days.

Comparison of De-Torque and Failure Load Evaluation of Selective-Laser-Sintered CoCr, CAD-CAM ZrO, and Machined Implant Abutment/Restoration

Aim: This study aimed to compare the torque loss, fracture load, compressive strength, and failure types of selective-laser-sintered cobalt chromium (SLM-Co-Cr), computer-aided design and computer-aided manufacturing zirconium oxide (CAD-CAM-ZrO), and machined titanium (Ti) implant abutments. Methods: Thirty endosseous dental implants were vertically embedded with machined Ti (control group), CAD-CAM-ZrO, and SLM-Co-Cr abutments. Abutment fabrication involved CAD-CAM milling and SLM technology. The de-torque assessment included preload reverse torque values (RTVs), cyclic loading, and post-RTVs using a customized protocol. Fracture load assessment employed ISO-14801 standards, and statistical analysis was conducted using ANOVA and Tukey Post hoc tests (p &lt; 0.05). Results: In pre-load RTVs, SLM-Co-Cr showed the lowest mean torque loss (24.30 ± 2.13), followed by machined Ti (27.33 ± 2.74) and CAD-CAM-ZrO (22.07 ± 2.20). Post-load RTVs decreased for all groups. Fracture load and compressive strength were highest for SLM-Co-Cr, with significant differences among groups (p &lt; 0.001). Fracture types included abutment failures in SLM-Co-Cr and machined Ti, while CAD-CAM-ZrO exhibited crown separation with deformation. Conclusion: SLM-Co-Cr-fabricated implant abutments exhibited superior stability and resistance to rotational forces, higher fracture loads, and greater compressive strength compared to CAD-CAM-ZrO and machined Ti.

Corrugated Box Damage Classification Using Artificial Neural Network Image Training

ABSTRACTThis paper proposes a novel packaging evaluation method using corrugated box images and an artificial neural network (ANN). An ANN works in a way similar to that of neurons in a human brain: by making connections between a trained dataset and the new images provided after training. The ANN has been implemented in the industry in various ways but limited in packaging evaluation. This paper is focused on the corrugated box damage prediction using ANN with the Orange Data Mining platform. By capturing the damaged corrugated box images with an ANN, damaged products can be identified allowing a decision to be made as to what type of package failure occurred. One of the benefits to using an ANN to evaluate corrugated box images is that it allows for the evaluation of package protection in a real distribution environment as compared to a controlled lab setting. In turn, this reduces the cost of testing, as the package failure will have been identified with the assistance of the ANN, rather than full retesting to identify where damage occurred. This process would also reduce costs associated with the usage of materials for testing, due to the lower number of test samples required.

Mesh stiffness calculation of defective gear system under lubrication with automated assessment of surface defects using convolutional neural networks

As typical failure mode of gear system, the tooth surface pitting, spalling can be detected in most long-running gear system, especially under heavy-load and high-speed condition, or under lubricant-starvation condition, the tooth pitting is characterized by irregular contour and random distribution. Most previous study on the defective gear system mainly based on manually detection of defective region, or just rely on geometric simplification of defects, leading to inaccurate results with low-efficient method, therefore, the machine-vision-based defect inspection method is proposed in the study of defective gear system. First, the pitting defects on gear tooth surface is detected and segmented based on involutional neural network U-net, then the tooth surface with segmented defective region is mapped to the elastohydrodynamic lubrication model of spur gear system, finally, the tribological behavior in addition to the mesh stiffness under lubrication condition of defective spur gear system are investigated and discussed. The results reveal that the machine-vision-based defects inspection could improve the accuracy and efficiency of the failure study for gear system.