Rolling Model Research Articles

A transfer learning model for rolling bearing fault diagnosis based on texture loss strategy and nuclear norm regularization

Abstract In recent years, transfer learning (TL) approaches have seen extensive application in diagnosing bearing faults due to their exceptional performance. However, mechanical noise, equipment aging, and wear lead to notable disparities and differences in the multi-level feature distributions across the source and target domain signals. The issue is addressed by proposing a TL model based on a texture loss strategy and nuclear norm regularization method. First, a feature-enhanced network is designed, which significantly improves the ability to capture local details and long-range dependencies by combining a multi-scale feature extraction module with a dilated residual module. Next, a texture loss strategy is proposed to align multi-scale features across domains by minimizing the Gram matrix of signal features. Finally, a nuclear norm regularization method is proposed to perform low-rank approximation on the signal matrix, facilitating the extraction of more robust feature data and mitigating the risk of overfitting. The experimental results demonstrate that the proposed method achieved an average accuracy of 98.58% on the University of Ottawa bearing fault dataset and 98.11% on the Jiangnan University bearing dataset, surpassing eight other algorithms in bearing fault diagnosis.

Gravitational wave spectrum of chain inflation

Chain inflation is an alternative to slow-roll inflation in which the inflaton tunnels along a large number of consecutive minima in its potential. In this work we perform the first comprehensive calculation of the gravitational wave (GW) spectrum of chain inflation. In contrast to slow-roll inflation the latter does not stem from quantum fluctuations of the gravitational field during inflation, but rather from the bubble collisions during the first-order phase transitions associated with vacuum tunneling. Our calculation is performed within an effective theory of chain inflation which builds on an expansion of the tunneling rate capturing most of the available model space. The effective theory can be seen as chain inflation’s analog of the slow-roll expansion in rolling models of inflation. The near scale-invariance of the scalar power spectrum translates to a quasiperiodic shape of the inflaton potential in chain inflation, with the tunneling rate changing very slowly during the e-folds leading to cosmic microwave background observables. We show that chain inflation produces a very characteristic double-peak GW spectrum: a faint high-frequency peak associated with the gravitational radiation emitted during inflation, and a strong low-frequency peak associated with the graceful exit from chain inflation (marking the transition to the radiation-dominated epoch). There exist very exciting prospects to test the gravitational wave signal from chain inflation at the aLIGO-aVIRGO-KAGRA network, at LISA and /or at pulsar timing array experiments. A particularly intriguing possibility we point out is that chain inflation could be the source of the stochastic gravitational wave background recently detected by NANOGrav, PPTA, EPTA, and CPTA. We also show that the gravitational wave signal of chain inflation is often accompanied by running/ higher running of the scalar spectral index to be tested at future cosmic microwave background experiments. Published by the American Physical Society 2024

Analysis of Gravel Migration Patterns During Vibration Rolling and Their Impact on GCL Performance Based on DEM

In this study, a multilayer composite rolling model consisting of a rolling wheel, a protective layer, a GCL, and a support layer was constructed by the discrete element method (DEM). Soil compaction and gravel migration, and their effects on the GCL, were analyzed from a fine viewpoint, and three key indexes for the safety assessment of the GCL were proposed: local elongation, gravel embedment value, and bentonite allotment number. The results show that the soil porosity and cumulative settlement do not decrease all the time with the number of rolling passes, and there exists an optimal number of rolling passes during the rolling process; the protective layer of gravel soil moves more frequently than the support layer; and the nearly rectangular and nearly elliptical gravels are more likely to rotate. The maximum local elongation of the GCL was 3.79% during the lapping process, and all gravels in contact with the upper boundary of the GCL extruded the GCL to varying degrees during the lapping process. The distribution of bentonite particles is closely related to the contact mode between gravel and GCL.

Effect of Rolling Process Parameters on Transverse Cracks on the Surface of Q345 Steel

This article examines the evolutionary behavior of transverse surface cracks in Q345 steel during the hot‐rolling process under various rolling conditions. “V”‐shaped cracks are prefabricated on the billet surface for laboratory hot‐rolling tests, and a corresponding rolling model is developed to validate its accuracy. The slab rolling model is established based on actual onsite rolling process parameters. The aim is to analyze the effects of different rolling conditions on the surface crack morphology and aspect ratio. The results indicate that variations in roll diameter (1400, 1300, and 1200 mm) influence crack propagation, with cracks unfolding earlier when the roll diameter is 1400 mm. At varying initial rolling temperatures (1300, 1200, and 1100 °C), cracks begin to unfold earlier at the highest temperature of 1300 °C. When examining different friction coefficients (0.4, 0.5, and 0.6), cracks unfold earlier at the highest coefficient of 0.6. Similarly, at different rolling speeds (3, 4, and 5 m s−1), cracks unfolded earlier at the highest speed of 5 m s−1. Analysis of changes in crack depth and aspect ratio during rolling shows that the initial few passes have a greater influence on crack depth and unfolding. In the later stages of rolling, cracks continue to unfold, though the depth shows minimal variation, and the aspect ratio tends to approach zero.

Generalisation of the hydrodynamics model method for hot and cold strip rolling application

The present work constitutes a generalization of the hydrodynamic model used to predict the pressures and the rolling speeds during the hot rolling of aluminum and copper strips. The hydrodynamic model with a linear behavior (Newton viscous) of the materials shows good predictions in the literature but its applicability is questionable in non-linear cases, when the materials exhibit viscoplastic or plastic behavior. This work extends the model to accommodate non-linear cases commonly encountered in rolling models (viscoplastic and plastic behaviors). The obtained results are in good agreement with experimental data from the literature. The validated model can, therefore, be considered an enhanced hydrodynamic model for predicting pressures and velocities during both hot and cold rolling of thin strips.

Development and optimization of ultimate performance self-levelling epoxy asphalt concrete (USEA) for prefabricated deck pavement

Developing prefabricated bridge deck pavement systems can fundamentally avoid the drawbacks caused by traditional on-site construction such as poor continuity of the construction process and significant differences in pavement layer performance. Prefabricated bridge deck pavement requires that the bridge pavement is completed with the steel deck together in the factory indoors thus it needs the concrete has better self-leveling characteristics which is rare for existing pavement materials. This article develops the ultimate performance self-levelling epoxy asphalt concrete (USEA) which both have excellent self-leveling properties and comprehensive performance. The study initially established the self-leveling asphalt concrete rolling model based on asphalt film thickness theory and Hertz contact theory and the critical rolling mechanical calculation results states that the aggregate particles and the skeleton amounts are the key elements for fluidity of concrete. Comprehensive the analysis of rolling model and research on the curing reaction process of epoxy resin, the study raises the material configuration of USEA and determine the gradation by the Bailey’s method. To guarantee the better pavement performance of USEA, the study designs the orthogonal test and establish the Entropy-TOPSIS model to optimize the details mortar schemes of concrete. Through the model calculation, the epoxy resin content and the SBS modified asphalt oil-stone ratio are determined as 20 % and 8.0 % respectively. In this scheme, the high temperature stability is 2805 times·mm-1, the low temperature bending strain is 3662με and the fatigue life can reach 1.200 million times. The study provides a material selection for the development of prefabricated bridge pavement system and the technical reference for the development of other self-leveling materials which has constructive practical significance.

Adaptive feature fusion and disturbance correction for accurate remaining useful life prediction of rolling bearings

As a key component in the transmission system of high-speed trains, bearings need to withstand certain loads while rotating at high speeds. Once a failure occurs, it can directly affect the safety of train operations. Therefore, it is of great significance to establish a reliable remaining useful life model to ensure train operation safety. Addressing the problems of redundant features in the existing multi-feature fusion process, which affects diagnostic performance, and the spurious fluctuations in the fusion features, which cause inaccurate determination of the start fault time and lead to low prediction accuracy of remaining useful life, we propose a method based on adaptive feature fusion and the autoregressive integrated moving average model for rolling bearing prediction. Firstly, features from the time domain, frequency domain, and entropy are extracted. A feature selection mechanism with minimum redundancy is constructed to screen the optimal sensitive feature set. Secondly, based on adaptive feature fusion, the optimal sensitive feature set is dynamically fused, and the spurious fluctuations of the health index are corrected using linear regression and the 3σ principle. Next, a bottom-up time series segmentation method is employed to divide the health status of the Improved Health Indicators. Finally, a remaining useful life prediction model based on the autoregressive integrated moving average model is established. This study demonstrates that the proposed method effectively identifies features that are most sensitive to degradation trends, accurately determines the initial fault moment of bearings, and achieves effective prediction of the remaining useful life of bearings.

Study on Cavity Filling Defects and Tensile Properties of L-Shaped Profiled Rings

Severe cavity filling defects and poor mechanical properties increase the difficulty in the integrated forming of L-shaped profiled rings due to its asymmetrical section geometry. A novel rolling method of a C-shaped ring was proposed in this study, and two symmetric L-shaped rings were prepared simultaneously. A numerical model of C-shaped ring rolling was established, and the cavity filling defects in different directions and the overall forming defect were defined for a qualitative analysis of the geometry’s accuracy. The effect of the rolling parameters on the forming defects and ring quality was investigated. The forming defects increased with an increase in the groove depth ratio as well as decreases in the groove angle and rolling ratio. The feeding strategy with a constant ring growth velocity led to the best geometric accuracy and strain uniformity of the C-shaped rings. Optimized rolling parameters can be acquired by the Box–Behnken optimization method with multi-objective optimization of the rolling stability and ring quality. An experiment of C-shaped ring rolling was successfully prepared, based on the optimized parameters. The hardness distribution on the cross-section was symmetric and uniform. The C-shaped ring showed obvious anisotropy of the tensile properties of the cast ring’s blank, and heat treatment had little effect on the improvement of the isotropy.

Motion Planning for a Legged Robot with Dynamic Characteristics.

Legged soccer robots present a significant challenge in robotics owing to the need for seamless integration of perception, manipulation, and dynamic movement. While existing models often depend on external perception or static techniques, our study aims to develop a robot with dynamic and untethered capabilities. We have introduced a motion planner that allows the robot to excel in dynamic shooting and dribbling. Initially, it identifies and predicts the position of the ball using a rolling model. The robot then pursues the ball, using a novel optimization-based cycle planner, continuously adjusting its gait cycle. This enables the robot to kick without stopping its forward motion near the ball. Each leg is assigned a specific role (stance, swing, pre-kick, or kick), as determined by a gait scheduler. Different leg controllers were used for tailored tiptoe trajectory planning and control. We validated our approach using real-world penalty shot experiments (5 out of 12 successful), cycle adjustment tests (11 out of 12 successful), and dynamic dribbling assessments. The results demonstrate that legged robots can overcome onboard capability limitations and achieve dynamic mobility and manipulation.

A class of nonlinear ship stability analysis: Stochastic dynamics with time-delayed control in crosswind and wave conditions

The impact of ocean waves and crosswinds on ships has long been a focal point of research for many scholars. This paper proposes a stochastic ship rolling model influenced by crosswinds, ocean waves, and time delays. The safe operating region for ship navigation is presented through the phase space of the system. A quantitative discussion of the system is conducted using the stochastic Melnikov function, and the stochastic P-D bifurcation of the system is discussed using topological data analysis techniques. The research indicates that incorporating time delay feedback can effectively enhance the system’s stability.

Visual study of hydrate particles agglomeration and rolling characteristic in gas-oil-water phase using a flow loop

Natural gas hydrate blockage always leads to safety problems in oil and gas transportation pipelines. It is important to look for the characteristic of hydrate blockage process. The hydrate blockage was visually investigated under gas-oil-water multi-phase flow conditions in a flow loop in this study. The change of pressure drop with water conversion rate variations were detected. The hydrate particle movements were also observed in the visual pipes at different liquid loadings and water cuts. It was found that the hydrate particles may agglomerate and deposit with an obvious rolling process in low water cut conditions. The oil phase will accelerate particles agglomeration in the contact interface of oil, gas and water, and form a large number of flocculent hydrates. A theoretical hydrate particle rolling model was conducted to analysis the experiment results, and the hydrate particle velocity increases first and then decreases with the increase of particle diameter due to the comprehensive influence of van der Waals force and drag force. The results also confirmed the maximum of the pressure drop increased when the liquid loading increased. However, the hydrate formation and blockage speed also slow down due to small component of the gas phase in high liquid loading. In addition, comparing with 20 % and 80 % water cut, 50 % water cut has the fastest hydrate formation speed in different experimental conditions. Furthermore, the maximum water conversion rate increased with have a higher water cut. More hydrate will form in the condition, which will lead to the decrease of flow velocity in the pipeline.

An intelligent hybrid deep learning model for rolling bearing remaining useful life prediction

ABSTRACT Remaining Useful Life (RUL) prediction of rolling bearings is one of the intricate and important issues for equipment intelligent maintenance and health management. Various machine learning models and methods have been applied to rolling bearing RUL prediction. However, a single model cannot effectively extract state information and obtain accurate prediction results, and its generalisation is not stable under the condition of small sample data. Therefore this paper proposes an intelligent hybrid deep learning model for achieving accurate RUL prediction of rolling bearings. Firstly, the one-dimensional vibration signal is transformed into the corresponding two-dimensional time-frequency diagram via Continuous Wavelet Transform (CWT). Secondly, the diagram is input into a Multilayer Perceptron (MLP) consisting of a basic three-layer feed-forward network to obtain a one-dimensional feature vector. And lastly, the obtained feature vector is input into an integrated model based on Deep Autoregressive and Transformer to produce the probability distribution and obtain the prediction results of rolling bearing RUL. Extensive experiments on two rolling bearing datasets show that the proposed model outperforms six other comparative models in extracting bearing fault features and predicting bearing RUL, which demonstrates that the proposed model can effectively extract bearing fault features and accurately predict bearing remaining useful life.

Modeling and investigation of combined processes of casting, rolling, and extrusion to produce electrical wire from alloys Al–Zr system

The results of modeling and research of casting, rolling, and extrusion processes for producing wire from Al–Zr system alloys and determining its physical and mechanical properties have been presented. When implementing combined processes for processing aluminum alloys, the number of operations is significantly reduced, productivity and yield are increased. As a result of the conducted research, the influence of alloy preparation modes and their chemical composition on the physical, mechanical and electrical properties of longish semifinished products from Al–Zr system alloys obtained by combined rolling-extrusion (CRE) and ingotless rolling-extrusion (IRE) methods were investigated. The simulation of the CRE process for one of these alloys of the system was carried out in the DEFORM 3D software package using data on its rheological properties obtained by the hot torsion method. Based on the modeling results, the optimal modes for conducting experimental studies were selected. The features of metal forming were experimentally studied, the temperature-velocity and technological parameters of combined processing were found, as well as the properties of cast billets, including those obtained using an electromagnetic mold (EMM). The results of experimental studies of the CRE and IRE processes suggest that it is possible to obtain longish deformed semifinished products from alloys Al–Zr system with a level of physical, mechanical and electrical properties that meet international standards. At all technological stages, including drawing, the structure of the metal was studied, and data on the physical and mechanical properties of hot-extruded rods and wire in cold-deformed and annealed states was obtained. The heat resistance of wire made from the investigated alloys was studied and it was found that after testing at a temperature of 280 °C and a holding time of 1 h, it satisfies the requirements of the AT3 type standard with a maximum permissible long-term operating temperature of 210 °C. Recommendations are given for industrial implementation; alloys containing 0.15–0.20% zirconium and 0.10–0.15% iron are recommended for the manufacture of AT1 type wire without heat treatment, as well as 0.25–0.30% Zr and 0.2–0.25% Fe for AT3 wire type with heat treatment.

Rolling bearing fault diagnosis based on correlated channel attention-optimized convolutional neural networks

Abstract In the field of intelligent fault diagnosis, traditional convolutional neural network (CNN)-based models for rolling bearing fault diagnosis are effective in extracting signal features but fall short in identifying subtle fault features in noisy environments. To address this challenge, this paper introduces a correlated channel attention-optimized deep convolutional neural network (CAOCNN) for fault diagnosis. The main innovations of this study include: firstly, the expansion of the convolutional kernel width through dilated convolution and optimized network parameter settings, which broadens the receptive field for feature extraction and effectively suppresses high-frequency noise; secondly, the relevant channel attention mechanism was constructed., which not only considers the channel weights post-global average pooling but also analyzes the correlations between channel features and the global feature center, dynamically adjusting channel weights to enhance model focus on critical features; additionally, the use of the Nesterov momentum optimization algorithm to optimize network parameters, reducing oscillations and increasing efficiency during training. Experimental results demonstrate that the CAOCNN achieved accuracies of 99.71% and 100% on the Case Western Reserve University and Xi’an Jiaotong University rolling bearing datasets, respectively, improving by 2.91% and 7.6% over traditional CNN models. In noisy conditions, T-SNE visual analysis further confirmed the excellent robustness and feature classification capability of the CAOCNN. These achievements validate the effectiveness of the CAOCNN in diagnosing rolling bearing faults in complex noise environments, contributing valuable advancements to the technology of intelligent fault diagnosis.

Investigation of Si, Mn, and C interactions during coiling of hot rolling model alloys

Model alloys were developed to investigate the interaction of Mn, Si, and C during hot rolling. The alloy samples prepared by the method of vacuum-sealed tube heat treatment exhibited unique microstructural features: surface reduced Fe layer, interface reduced Fe layer and internal oxidation layer. The experimental results revealed that scale reduction is mainly controlled by the effect of Si and decarburization behavior, as well as the contribution of Si is more significant. The competing reactions of decarburization substantially reduce the depth of internal oxidation of the alloy. Additionally, combination of Mn and Si greatly facilitated the process of internal oxidation, with Si playing a role in inducing the inward oxidation of Mn. These insights will contribute to a better understanding of hot rolling coiling process, especially the complex interactions of Mn, Si, and C.

Research on dynamic characteristics of rolling bearing slippage under different working conditions

Abstract With the development of aerospace, military and other fields, higher requirements are put forward for the life, reliability and accuracy of rolling bearings. Rolling element slippage, as a phenomenon that cannot be ignored in the work of rolling bearings, affects the accuracy and life of rolling bearings. The present paper discusses the dynamic modelling of rolling bearing slippage, drawing on the research conducted by both domestic and international scholars in this field. The variations in rotational speed and revolution speed, the fluctuations in sliding speed between the inner ring raceway and outer ring raceway, as well as the impact of bearing load and speed on rolling element slippage during the process.

경찰공무원의 퇴직준비역량과 교육참여도가 경제활동의지에 미치는 영향

Today's aging population and extension of life expectancy are increasing the economic activities intention after retirement through partial retirement. This is because the age of physical health has been extended by 20 to 30 years compared to the past, due to advancements in medical technology and self-management. This study analyzes the influence of the retirement competencies and educational experience of police officiers on economic activity intention. Additionally, it analyzes the moderated effect of role models. For this study, an online survey was conducted using a questionnaire among 348 police officer.The statistical analysis software used was SPSS 24.0. It involved factor analysis, reliability testing and multiple regression analysis to verify the hypotheses. The research results were as follows. First, as a result of examining the effect of retire competencies on economic activity intention, it was found that among the retire competencies factors, financial management and digital competencies had a significantly positive effect on economic activity intention. Second, educational experience had a significantly positive effect on economic activity intention. Third, financial management and educational experience had a significantly positive effect on economic activity intention according to roll model.

Stochastic Roll Dynamics of Smooth and Impacting Vessels in Random Waves

Ship instability is frequently associated with extreme roll motion. In this paper smooth and vibro-impact dynamics of vessels rolling under random wave excitations including nonlinearities is investigated. The vessel is represented by a single degree of freedom roll model. The random excitation due to waves and ice is modeled by a sum of random pulses with Poisson arrival rate. Stochastic P-bifurcation analysis is carried out by examining the variation with parameter of the joint probability density functions (jpdf) of the response computed numerically by the solution of the corresponding Fokker–Planck–Kolmogorov (FPK) equation. The finite element (FE), finite difference (FD), path integral (PI) and radial basis neural (RBN) network formulation methods are used in the solution of the FPK equation. The D-bifurcation analysis is analyzed by computing the largest Lyapunov exponent. The mean upcrossing rate is computed using Rice’s formula. The random wave excitation is also modeled as the response of a second order linear filter to white noise and the random response of the ship is investigated by solution of the corresponding FPK equation of the four dimensional Markov system by the FD method. Non-smooth coordinate transformations like the Zhuravlev and Ivanov transformations are used converting the discontinuous systems to equivalent smooth systems in the impact dynamics analysis. The adaptive time stepping method (ATSM) approach is used to accurately locate the discontinuity point and a Brownian tree approach is used to direct the integration along the correct Brownian path. Different models for nonlinear damping and the restoring moment are considered and their effects on ship stochastic response such as chaotic motion, unbounded rotational motion, impact modulation motion and ship capsizing are investigated.

The Fault Diagnosis of Rolling Bearings Is Conducted by Employing a Dual-Branch Convolutional Capsule Neural Network.

Currently, many fault diagnosis methods for rolling bearings based on deep learning are facing two main challenges. Firstly, the deep learning model exhibits poor diagnostic performance and limited generalization ability in the presence of noise signals and varying loads. Secondly, there is incomplete utilization of fault information and inadequate extraction of fault features, leading to the low diagnostic accuracy of the model. To address these problems, this paper proposes an improved dual-branch convolutional capsule neural network for rolling bearing fault diagnosis. This method converts the collected bearing vibration signals into grayscale images to construct a grayscale image dataset. By fully considering the types of bearing faults and damage diameters, the data are labeled using a dual-label format. A multi-scale convolution module is introduced to extract features from the data and maximize feature information extraction. Additionally, a coordinate attention mechanism is incorporated into this module to better extract useful channel features and enhance feature extraction capability. Based on adaptive fusion between fault type (damage diameter) features and labels, a dual-branch convolutional capsule neural network model for rolling bearing fault diagnosis is established. The model was experimentally validated using both Case Western Reserve University's bearing dataset and self-made datasets. The experimental results demonstrate that the fault type branch of the model achieves an accuracy rate of 99.88%, while the damage diameter branch attains an accuracy rate of 99.72%. Both branches exhibit excellent classification performance and display robustness against noise interference and variable working conditions. In comparison with other algorithm models cited in the reference literature, the diagnostic capability of the model proposed in this study surpasses them. Furthermore, the generalization ability of the model is validated using a self-constructed laboratory dataset, yielding an average accuracy rate of 94.25% for both branches.

Participation as Intimate Act: Audience Reflections on Strategies of Consent in Roll Models

Immersive, interactive, and participatory performances often promise co-creative or otherwise unique roles for spectators, who may be called upon to directly interact with performers, explore performance sites, or help shape plot. The nature of much of this heightened participation is important to note: audience members might be expected to be visible onstage, make choices that affect the show, offer personal information, or engage directly with actors, in an ask for labor that is resultantly accompanied by increased risk, vulnerability, and relationality. This article uses the case study of Roll Models, a longform improv show that enacts a short adventure campaign in the style of Dungeons & Dragons on stage, in order to characterize audience participation as an intimate act. Building on previous scholarship aimed at assessing consent, intimacy, and participation in immersive theatre and live action role-play (Villarreal 2021, Biggin 2017, Machon 2013), we query not only into what participation, consent, and on-boarding processes were present in the performance, but also how audience members themselves viewed such mechanisms, understood and learned their role in the performance, and related to actors and each other.