Uncertainty-Aware Deep Learning: A Promising Tool for Trustworthy Fault Diagnosis

Recently,intelligent fault diagnosis based on deep learning has been extensively investigated,exhibiting state-of-the-art performance.However,the deep learning model is often not truly trusted by users due to the lack of interpretability of“black box”,which limits its deployment in safety-critical applications.A trusted fault diagnosis system requires that the faults can be accurately diagnosed in most cases,and the human in the deci-sion-making loop can be found to deal with the abnormal situa-tion when the models fail.In this paper,we explore a simplified method for quantifying both aleatoric and epistemic uncertainty in deterministic networks,called SAEU.In SAEU,Multivariate Gaussian distribution is employed in the deep architecture to compensate for the shortcomings of complexity and applicability of Bayesian neural networks.Based on the SAEU,we propose a unified uncertainty-aware deep learning framework(UU-DLF)to realize the grand vision of trustworthy fault diagnosis.Moreover,our UU-DLF effectively embodies the idea of“humans in the loop”,which not only allows for manual intervention in abnor-mal situations of diagnostic models,but also makes correspond-ing improvements on existing models based on traceability analy-sis.Finally,two experiments conducted on the gearbox and aero-engine bevel gears are used to demonstrate the effectiveness of UU-DLF and explore the effective reasons behind.

Analytical Modeling and Mechanism Analysis of Time-Varying Excitation for Surface Defects in Rolling Element Bearings

Surface defects,including dents,spalls,and cracks,for rolling element bearings are the most common faults in rotating machinery.The accurate model for the time-varying excitation is the basis for the vibration mechanism analysis and fault feature extraction.However,in conventional investigations,this issue is not well and fully addressed from the perspective of theoretical analysis and physical derivation.In this study,an improved analytical model for time-varying displacement excitations(TVDEs)caused by surface defects is theoretically formulated.First and foremost,the physical mechanism for the effect of defect sizes on the physical process of rolling element-defect interaction is revealed.According to the physical interaction mechanism between the rolling element and different types of defects,the relationship between time-varying displacement pulse and defect sizes is further analytically derived.With the obtained time-varying displacement pulse,the dynamic model for the deep groove bearings considering the internal excitation caused by the surface defect is established.The nonlinear vibration responses and fault features induced by surface defects are analyzed using the proposed TVDE model.The results suggest that the presence of surface defects may result in the occurrence of the dual-impulse phenomenon,which can serve as indexes for surface-defect fault diagnosis.

Challenges and Opportunities of AI-Enabled Monitoring, Diagnosis & Prognosis: A Review

Prognostics and Health Management(PHM),including monitoring,diagnosis,prognosis,and health management,occupies an increasingly important position in reducing costly breakdowns and avoiding catastrophic accidents in modern industry.With the development of artificial intelligence(AI),especially deep learning(DL)approaches,the application of AI-enabled methods to monitor,diagnose and predict potential equipment malfunctions has gone through tremendous progress with verified success in both academia and industry.However,there is still a gap to cover monitoring,diagnosis,and prognosis based on AI-enabled methods,simultaneously,and the importance of an open source community,including open source datasets and codes,has not been fully emphasized.To fill this gap,this paper provides a systematic overview of the current development,common technologies,open source datasets,codes,and challenges of AI-enabled PHM methods from three aspects of monitoring,diagnosis,and prognosis.

基于Rockfall数值模拟的危岩体运动规律分析

危岩体发育过程具有渐进性,失稳过程具有突变性和强致灾性,危岩崩塌灾害是山区公路工程建设及运营过程中普遍存在的突发性重大安全隐患,前期预测和分析危岩体落石运动特征可为危岩体治理提供参考。本文以兴义环高某特大桥桥台上方的危岩体为例,通过现场调查研究,分析危岩特征,采用Rockfall数值软件模拟危岩落石运动轨迹、落石终点水平位置,分析落石速度及其能量大小。结果表明:危岩体W5发生崩塌后,有大部分落石到达桥面,且在到达桥面时仍有较大动能和转动速度,会对桥面造成严重撞击和破坏,威胁和影响行车安全。建议对危岩体进行防护或加固,以保障公路建设运营安全。

Replacing fishmeal with soybean meal affects survival, antioxidant capacity, intestinal microbiota, and mRNA expression of TOR and S6K1 in Macrobrachium rosenbergii

Using alternative plant-derived dietary protein to replace fishmeal,combined with practical evaluation indexes,is a recent focus for aquaculture practices.An 8-week feeding experiment with giant freshwater prawn Macrobrachium rosenbergii post-larvae was conducted to determine the eff ects of replacing fi shmeal(FM)with soybean meal in the feed,in terms of growth performance,antioxidant capacity,intestinal microbiota,and mRNA expression of target of rapamycin(TOR)and ribosomal protein S6 kinase B1(S6K1).Four isonitrogenous diets with isocaloric value were prepared to contain 100%,75%,50%,or 25%FM as the protein source(dietary treatments FM100,FM75,FM50,and FM25,respectively).Each diet was fed to post-larval prawns(mean weight 0.045±0.002 g)twice a day in four replicates.No signifi cant diff erence in weight gain was observed among all groups,but the survival rate of prawns fed the FM50 and FM25 diets was signifi cantly lower than that of prawns fed the FM diet.The mRNA expression of both TOR and S6K1 were the lowest in hepatopancreas of prawns fed the FM25 diet.Superoxide dismutase activity of prawns fed the FM25 diet was significantly lower than that of prawns fed FM50.In contrast,the malondialdehyde content was signifi cantly higher in prawns fed FM25 as compared with those fed FM75.The proportion of fishmeal in the diet did not affect the composition of core(phylum-level)intestinal microbiota,but greater fishmeal replacement with soybean meal had a potential risk to increase the relative abundance of opportunistic pathogens in the gut when considered at the genus level.These results suggest that fishmeal replacement with soybean meal should not exceed 50%in a diet for post-larval M.rosenbergii.

Micromechanical Analysis of Interfacial Debonding in Metal Matrix Composites Subjected to off-axis Loading

This study aims to investigate the effects of interfacial debonding and fiber volume fraction on the stressstrain behavior of the fiber reinforced metal matrix composites subjected to off-axis loading.The generalized method of cells(GMC)is used to analyze a representative element whose fiber shape is circular.The constant compliant interface model(CCI)is also adopted to study the response of composites with imperfect interfacial bonding.Results show that for the composites subjected to off-axis loading,the mechanical behaviors are affected appreciably by the interfacial debonding and the fiber volume fraction.

Preparation of Electrospun PVDF Nanofiber Composite Filter Medium and Its Application in Air Filtration

Nanofibrous media with both high particle interception efficiency and robust air permeability has broad technological applications in areas including individual protection, industrial security, and environmental governance. However, producing such filtration media has proven to be extremely challenging. Here we reported an approach to preparing and fabricating a polyvinylidene fluoride (PVDF) nanofiber composite filter medium composed of 2D PVDF nanofiber nets and a stable substrate via onestep electrospinning for effective air filtration. PVDF nanofibers are obtained by adjusting the electrospinning process. With the combined properties of ultrasmall diameter, high porosity, and a bonded scaffold, the resulting PVDF nanofiber composite filter medium exhibits a robust high filtration efficiency of 99.901%(equivalent to an F9 rating) for 0.4 mm particles and a long service life (a large dust holding capacity of 36 g/m^2) for ultrafine airborne particles based on the sieving principle and surface filtration behavior. The successful synthesis of PVDF nanofibers medium would not only make it a promising candidate for air filtration, but also provide new insights into the design and development of composite nanofiber structures for various applications.

Preliminary Study on the Reinforcing Effect of Nanocellulose on Supercapacitor Membranes

In this study,six types of nanocellulose,as paper strengthening agents,were investigated for their reinforcing effects on a supercapacitor membrane by adding them to a slurry.The results indicated that adding 3%nanocellulose bacterial cellulose(BC)-B could effectively increase the tensile strength of the supercapacitor membrane by 36.5%without changing the pore structure of the membrane.Scanning electron microscopy images revealed that adding polyacrylamide to the supercapacitor membrane caused serious adhesion between fibers and affected the pore structure of the supercapacitor membrane,whereas adding BC-B did not produce similar effects.

AI-Enabled Monitoring, Diagnosis & Prognosis

The emerging and development of Artificial Intelligence(AI),especially deep learning,has stimulated its application in various engineering domains.Monitoring,diagnosis and prognosis,as the key elements of intelligence maintenance of manufacturing systems in the era of Industry 4.0,has also benefited from the advancement of AI technology.The main objective of this special issue aims at bringing scholars to show their research findings in the field of monitoring,diagnosis and prognosis driven by AI,and promote its application in intelligent maintenance of manufacturing system in China.Ten papers have been selected in this special issue after rigorous review and they represent the latest research outcomes in this active area.

Design and Experiment of 1LFQ-325 Pneumatic Reversible Plough

A pneumatic reversible plough is developed, which complements to the tractor of 25.7-36.8 kW.The plough adopts the cylinder as reversing mechanism between the right and left plough bodies, and the cylinder can substitute the mechanical reversing mechanism. The pneumatic turnover allows the plough to be operated easily and turned over flexibly. Field experiment results show that indicators of plough performance meet the requirements of the relevant national standards.

Hybrid Intelligent Damage Identification of Composite Plate Based on Ensemble Empirical Mode Decomposition and Support Vector Machine

In this paper,a novel method based on strain distribution is presented to determine the presence of damage and its location in composite plate.By building a damage monitoring experimental platform with Fiber Bragg Gratings(FBGs)sensors,impact experiments are made respectively to gain the strain distribution both in heath and damage state.EEMD is used to process the data and IMFs energy feature is evaluated.Then,support vector machine is applied to identify the damage and the testing classification accuracy reaches 92.86%.Finally,by using the influence of the damage position and the propagation path on energy,the damage location is predicted.The experimental results indicate that the proposed method can accurately identify the presence and position of damage.The effectiveness and reliability of the proposed method is verified.

Physics-Informed Deep Neural Network for Bearing Prognosis with Multisensory Signals

Prognosis of bearing is critical to improve the safety,reliability,and availability of machinery systems,which provides the health condition assessment and determines how long the machine would work before failure occurs by predicting the remaining useful life(RUL).In order to overcome the drawback of pure data-driven methods and predict RUL accurately,a novel physics-informed deep neural network,named degradation consistency recurrent neural network,is proposed for RUL prediction by integrating the natural degradation knowledge of mechanical components.The degradation is monotonic over the whole life of bearings,which is characterized by temperature signals.To incorporate the knowledge of monotonic degradation,a positive increment recurrence relationship is introduced to keep the monotonicity.Thus,the proposed model is relatively well understood and capable to keep the learning process consistent with physical degradation.The effectiveness and merit of the RUL prediction using the proposed method are demonstrated through vibration signals collected from a set of run-to-failure tests.

Spatial transformation of general sampling-aliasing frequency region for rotating-blade parameter identification with emphasis on single-probe blade tip-timing measurement

Blade-health monitoring is intensely required for turbomachinery because of the high failure risk of rotating blades.Blade-Tip Timing(BTT)is considered as the most promising technique for operational blade-vibration monitoring,which obtains the parameters that characterize the blade condition from recorded signals.However,its application is hindered by severe undersampling and stringent probe layouts.An inappropriate probe layout can make most of the existing methods invalid or inaccurate.Additionally,a general conflict arises between the allowed and required layouts because of arrangement restrictions.For the sake of economy and safety,parameter identification based on fewer probes has been preferred by users.In this work,a spatial-transformation-based method for parameter identification is proposed based on a single-probe BTT measurement.To present the general Sampling-Aliasing Frequency(SAFE)map definition,the traditional time-frequency analysis methods are extended to a time-sampling frequency.Then,a SAFE map is projected onto a parameter space using spatial transformation to extract the slope and intercept parameters,which can be physically interpreted as an engine order and a natural frequency using coordinate transformation.Finally,the effectiveness and robustness of the proposed method are verified by simulations and experiments under uniformly and nonuniformly variable speed conditions.

2D-to-3D buckling transformability enabled reconfigurable metamaterials for tunable chirality and focusing effect

Recently,multifarious deformation approaches in nature have promoted dynamic manipulation for electromagnetic(EM)waves in metamaterials,and those representative strategies are mainly focused on the modulation of spectral parameters.Several works have also achieved tunable phase-gradient meta-devices.Here,to broaden the modulation freedom of mechanical deformation,we initially propose two reconfigurable metamaterials consisting of mirrored S-shaped meta-atoms selectively bonded on biaxially pre-stretched substrates.Planar meta-atoms with spin-insensitive transmittance are buckled into 3D morphologies to break residual symmetries by releasing the stress and to facilitate spin-dependent transmittance under circularly polarized incidence.Owing to the geometric anisotropy of S-shaped meta-atoms along the x and y axes,3D chiral meta-atoms exhibit discriminate circularly cross-polarized transmittance under opposite spins.The underlying physical mechanism reveals that EM resonance originates from the excitation of electric dipoles and magnetic dipoles,and their cross coupling finally triggers the chiral effects of 3D meta-atoms.By introducing the gradient-phase design that keeps unchanged under various strains,two types of meta-atoms with specified orientations are interleaved to design a double-foci metalens,and its 2D-to-3D morphology transformation shortens the focusing length and facilitates the intensity change of two foci.Our approach in designing reconfigurable EM metamaterials with 2D-to-3D buckling transformability can be further extended toward terahertz even optical wavebands,and it may assist with deriving more applicable multi-functionalities in the aspects of imaging,sensing,and holograms.

Advanced charge transfer technology for highly efficient and long-lived TADF-type organic afterglow with near-infrared light-excitable property

Heavy atom effects and n-π*transitions have been frequently reported to enhance room-temperature organic phosphorescence efficiency but lead to shortage of phosphorescence lifetimes.Unlike these reported studies,we conceive the incorporation of advanced charge transfer(CT)technology to boost room-temperature organic afterglow efficiency and simultaneously maintain afterglow lifetimes.Here we design difluoroboronβ-diketonate(BF2bdk)CT compounds with moderate singlet-triplet splitting energy(ΔEST)of around 0.4 e V,and relatively large spin-orbit coupling matrix elements(SOCME(S_(1)-T_(1)),1–10 cm^(-1))to achieve efficient intersystem crossing(ISC)and moderate rates of reverse intersystem crossing(kRISC,1–10 s^(-1)).The advanced CT technology,which includes multiple electron-donating groups and orthogonal donor-acceptor arrangement,have been found to narrowΔESTand enhance both ISC and RISC.Meanwhile,the organic matrices suppress nonradiative decay of BF2bdk’s T1states by their rigid microenvironment.Consequently,thermally activated delayed fluorescence(TADF)-type organic afterglow materials can be achieved with afterglow efficiency up to 83.0%,long lifetimes of 433 ms,excellent processablility,as well as advanced anti-counterfeiting and information encryption.Furthermore,with the aid of up-conversion materials and through radiative energy transfer,TADF-type afterglow materials with aqueous dispersity and near-infrared light-excitable property have been achieved,which paves the way for biomedical applications.

Full-field dynamic strain reconstruction of an aero-engine blade from limited displacement responses

Blade strain distribution and its change with time are crucial for reliability analysis and residual life evaluation in blade vibration tests.Traditional strain measurements are achieved by strain gauges(SGs)in a contact manner at discrete positions on the blades.This study proposes a method of full-field and real-time strain reconstruction of an aero-engine blade based on limited displacement responses.Limited optical measured displacement responses are utilized to reconstruct the full-field strain.The full-field strain distribution is in-time visualized.A displacement-to-strain transformation matrix is derived on the basis of the blade mode shapes in the modal coordinate.The proposed method is validated on an aero-engine blade in numerical and experimental cases.Three discrete vibrational displacement responses measured by laser triangulation sensors are used to reconstruct the full-field strain over the whole operating time.The reconstructed strain responses are compared with the results measured by SGs and numerical simulation.The high consistency between the reconstructed and measured results demonstrates the accurate strain reconstructed by the method.This paper provides a low-cost,real-time,and visualized measurement of blade full-field dynamic strain using displacement response,where the traditional SGs would fail.

Non-convex sparse optimization-based impact force identification with limited vibration measurements

Impact force identification is important for structure health monitoring especially in applications involving composite structures.Different from the traditional direct measurement method,the impact force identification technique is more cost effective and feasible because it only requires a few sensors to capture the system response and infer the information about the applied forces.This technique enables the acquisition of impact locations and time histories of forces,aiding in the rapid assessment of potentially damaged areas and the extent of the damage.As a typical inverse problem,impact force reconstruction and localization is a challenging task,which has led to the development of numerous methods aimed at obtaining stable solutions.The classicalℓ2 regularization method often struggles to generate sparse solutions.When solving the under-determined problem,ℓ2 regularization often identifies false forces in non-loaded regions,interfering with the accurate identification of the true impact locations.The popularℓ1 sparse regularization,while promoting sparsity,underestimates the amplitude of impact forces,resulting in biased estimations.To alleviate such limitations,a novel non-convex sparse regularization method that uses the non-convexℓ1-2 penalty,which is the difference of theℓ1 andℓ2 norms,as a regularizer,is proposed in this paper.The principle of alternating direction method of multipliers(ADMM)is introduced to tackle the non-convex model by facilitating the decomposition of the complex original problem into easily solvable subproblems.The proposed method namedℓ1-2-ADMM is applied to solve the impact force identification problem with unknown force locations,which can realize simultaneous impact localization and time history reconstruction with an under-determined,sparse sensor configuration.Simulations and experiments are performed on a composite plate to verify the identification accuracy and robustness with respect to the noise of theℓ1-2-ADMM method.Results indicate that compared with other existing regularization methods,theℓ1-2-ADMM method can simultaneously reconstruct and localize impact forces more accurately,facilitating sparser solutions,and yielding more accurate results.

Comprehensive kinetostatic modeling and morphology characterization of cable-driven continuum robots for in-situ aero-engine maintenance

In-situ maintenance is of great significance for improving the efficiency and ensuring the safety of aero-engines.The cable-driven continuum robot(CDCR)with twin-pivot compliant mechanisms,which is enabled with flexible deformation capability and confined space accessibility,has emerged as a novel tool that aims to promote the development of intelligence and efficiency for in-situ aero-engine maintenance.The high-fidelity model that describes the kinematic and morphology of CDCR lays the foundation for the accurate operation and control for in-situ maintenance.However,this model was not well addressed in previous literature.In this study,a general kinetostatic modeling and morphology characterization methodology that comprehensively contains the effects of cable-hole friction,gravity,and payloads is proposed for the CDCR with twin-pivot compliant mechanisms.First,a novel cable-hole friction model with the variable friction coefficient and adaptive friction direction criterion is proposed through structure optimization and kinematic parameter analysis.Second,the cable-hole friction,all-component gravities,deflection-induced center-of-gravity shift of compliant joints,and payloads are all considered to deduce a comprehensive kinetostatic model enabled with the capacity of accurate morphology characterization for CDCR.Finally,a compact continuum robot system is integrated to experimentally validate the proposed kinetostatic model and the concept of in-situ aero-engine maintenance.Results indicate that the proposed model precisely predicts the morphology of CDCR and outperforms conventional models.The compact continuum robot system could be considered a novel solution to perform in-situ maintenance tasks of aero-engines in an invasive manner.

Improved minimum variance distortionless response spectrum method for efficient and robust non-uniform undersampled frequency identification in blade tip timing

The noncontact blade tip timing(BTT)measurement has been an attractive technology for blade health monitoring(BHM).However,the severe undersampled BTT signal causes a significant challenge for blade vibration parameter identification and fault feature extraction.This study proposes a novel method based on the minimum variance distortionless response(MVDR)of the direction of arrival(DoA)estimation for blade natural frequency estimation from the non-uniformly undersampled BTT signals.First,based on the similarity between the general data acquisition model for BTT and the antenna array model in DoA estimation,the circumferentially arranged probes on the casing can be regarded as a non-uniform linear array.Thus,BTT signal reconstruction is converted into the DoA estimation problem of the non-uniform linear array signal.Second,MVDR is employed to address the severe undersampling issue and recover the BTT undersampled signal.In particular,spatial smoothing is innovatively utilized to enhance the estimation of covariance matrix of the BTT signal to avoid ill-condition or singularity,while improving efficiency and robustness.Lastly,numerical simulation and experimental testing are employed to verify the validity of the proposed method.Monte Carlo simulation results suggest that the proposed method behaves better than conventional methods,especially under a lower signal-to-noise ratio condition.Experimental results indicate that the proposed method can effectively overcome the severe undersampling problem of BTT signal induced by physical limitations,and has a strong potential in the field of BHM.

Basic research on machinery fault diagnostics： Past, present, and future trends

Machinery fault diagnosis has progressed over the past decades with the evolution of machineries in terms of complexity and scale. High-value machineries require condition monitoring and fault diagnosis to guarantee their designed functions and performance throughout their lifetime. Research on machinery Fault diagnostics has grown rapidly in recent years. This paper attempts to summarize and review the recent R＆D trends in the basic research field of machinery fault diagnosis in terms of four main aspects： Fault mechanism, sensor technique and signal acquisition, signal processing, and intelligent diagnostics. The review discusses the special contributions of Chinese scholars to machinery fault diagnostics. On the basis of the review of basic theory of machinery fault diagnosis and its practical applications in engineering, the paper concludes with a brief discussion on the future trends and challenges in machinery fault diagnosis.