Search

• Development of a multi-attribute Bayesian control scheme for fault prediction. • Multivariate hidden Markov model describes the deterioration process. • Computational algorithm to optimize the multi-attribute Bayesian policy. • Development of a real case study and verification using multivariate data. Although Bayesian approaches have been utilized in engineering systems for health prognostics, very little work has been done using Bayesian methods for fault prediction of systems under multiple attributes. To address this issue, in this paper a novel multi-attribute Bayesian control chart is presented for predicting failures of hidden-state systems by jointly considering two performance measures of system operation. The system actual status is represented by a three-state multivariate hidden stochastic process with a normal state, an abnormal state, and a failure state. The working states are unobservable and failure state is observable. Based on the built hidden-state model, a fault prediction scheme integrating both system availability and cost objectives is constructed via a multi-attribute Bayesian control chart to monitor and predict impending risks of the operational systems. The Bayesian control chart alarms when the probability of impending risks reaches a certain control limit, which is optimized and determined by a computational algorithm developed in a semi-Markov decision process framework. The proposed fault prediction scheme provides an appearing feature to jointly consider multiple attributes for hidden-state systems. A real case study of mechanical generators is presented and a comparison with other Bayesian and non-Bayesian methods is also given, which demonstrates the effectiveness and superiority of the proposed approach. [ABSTRACT FROM AUTHOR]

A general methodology for the dynamic modeling and analysis of planar flexible deployable structures consisting of scissor-like elements (SLEs) is presented. This modeling method is based on a comprehensive consideration of the symmetry and array characteristics of deployable structure and on an improved absolute node coordinate formulation (ANCF), which can model the warping of beam section using the locking-free shear deformable beam element. An effective node separation method is proposed to reduce the number of degrees of freedom of the dynamic equation in the ANCF framework, eliminate the constraint equations within and between SLEs and obtain a compact matrix. This reduction method has good adaptability and can be extended to all kinds of scissor deployable structures with array characteristics, whether they are planar or spatial structures. In addition, the modified generalized α method is utilized to solve the motion equations of deployable structure and eliminate the false high-frequency response generated in the calculation process. Finally, the methodology is validated using a cantilever beam case, and the parametric dynamic response of 2 × 2 and 1 × 2 deployable structures is implemented in this paper. The obtained results show that flexibility has an important impact on the dynamic characteristics of large deployable structures, and the deployable structure is unstable near 0 ∘ and 90 ∘ , and its safe working angle is 17 ∘ – 75 ∘ . It is necessary to carry out parametric research on this structures in the design stage because the prediction of these parameters such as initial configuration and component materials can improve the stability and deployment accuracy of deployable structures in orbit. [ABSTRACT FROM AUTHOR]

Generally, the internal states of most engineering systems are inaccessible to sensors under operational conditions, permitting indirect measurements and failures to be observed discretely. For this type of system, a class of proportional hazard (PH) models with multistate process have been proposed for performing prognostics. However, existing studies use a constant degradation mechanism with a very limited number of states, which may fail to model the actual deterioration of the system. Therefore, this article presents a novel PH model that utilizes a nonconstant degradation mechanism with a large number of states to perform health prognostics. A matrix partition-based approximation method is developed to include all the possible state transitions and to estimate important health quantities of the system, such as conditional reliability (CR), mean remaining useful life (MRUL), and remaining useful life distribution (RULD). An online prognostic scheme based on this matrix approach is presented to provide an update with the latest health quantities of the system. The proposed approach provides appealing computational features and avoids difficulties in calculating uncountable transitions for PH model-based prognostics. The effectiveness of the proposed method is demonstrated via a case of helical gearboxes under indirect monitoring and random failure conditions. [ABSTRACT FROM AUTHOR]

The computer numerical control (CNC) equipment is playing essential role in production and manufacturing, thus, its reliability estimation is crucial for industrial systems. Existing research studies mainly focused on univariate degradation data or multiple correlated data, which might not be accurate in equipment reliability assessment. In real CNC equipment, the degradation data consists of correlated and non-correlated degradation data and the data sample might be scarce, which bring great challenges for maintenance managers to assess risk and prevent failure ahead of fault occurrence. This paper proposes a reliability assessment approach based on multiple non-correlated and correlated degradation data with focusing on small size of data sample. Using the multiple degradation data of CNC equipment, a covariance matrix is built to determine the correlation between the performance signals. If the performance signals are not correlated, the degradation trajectory of the performance signals is curve fitted by least squares support vector machines (LS-SVM). The reliability which indicates the degradation curve not reaching its specified threshold is calculated. If the performance signals are correlated, according to the degradation data, the degradation curve of the performance signal is fitted by multivariate regression using LS-SVM, and the joint probability density function is derived. According to both the reliabilities of non-correlated degradation data and correlated degradation data, the joint reliability of the system is calculated. Finally, using a specific type of CNC machine tool as an example, the performance reliability assessment method is verified. [ABSTRACT FROM AUTHOR]

The study aims to work on multi -level condition-based maintenance (CBM) policy for multi-unit systems under economic dependence. Failure modes of units are classified as hard failure or soft failure and taken into account in this paper. Unlike previous works which only considered setup cost dependence with a single failure mode, two types of economic dependence are considered, which are setup cost-saving, and additional cost and time that can be saved when identical maintenance tasks and expertise are needed. In this study, hard failure provides an opportunity for additional inspection and repair of soft-type units, and soft failure creates an opportunity for replacement of the hard-type unit as its age exceeds opportunistic threshold. They both undergo preventive maintenance (PM) when the level/age exceeds preventive threshold at inspection epochs. The objective of this study is to determine the optimal opportunistic control limit and preventive control limit for each unit such that the system's average long-run maintenance cost per unit time is minimized. The optimization problem is formulated in a semi-Markov decision process (SMDP) framework. The explicit formula for the mean residual life is derived for the proposed model which enables the estimation of the remaining useful life and allows maintenance engineers to plan maintenance early based on the observed information. Finally, the method is illustrated by an industrial case study of a feed system of a machine tool. A comparison with other methods is given, which illustrates the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

Objective: This study aimed to compare the per OPU clinical outcomes for transfer of Day 3 double cleavage-stage embryos (DET) and Day 5 single blastocyst-stage (SBT) in patients with five or fewer good quality embryos on day 3 per occyte pick-up cycle (OPU) in antagonist cycles with consideration of blastocyst formation failure. Methods: This was a retrospective, observational cohort study of 2,116 cases of OPU treated with antagonist protocol in the affiliated Chenggong Hospital of Xiamen University between January 2013 and December 2020. DET was performed in 1,811cycles and SBT was performed in 305 cycles. The DET group was matched to the SBT group by propensity score (PS) matching according to multiple maternal baseline covariates. After PS matching, there were 303 ET cycles in each group. The primary outcomes were the cumulative live birth rate (CLBR), cumulative multiple pregnancy rate(CMPR)per OPU and the number of ET to achieve live birth per OPU. Secondary outcomes were the percentage of clinical pregnancy(CPR), live birth rate(LBR), multiple pregnancy rate(MPR). Results: Following PS mating, the CLBR was slightly higher (48.8% versus 40.3% ; P = 0.041) and the CMPR was significantly higher in the DET group compared to SBT group(44.2% versus 7.9%, P < 0.001). The CPR, LBR and MPR per fresh transfer were higher in DET group compared to SBT group(50.2% versus 28.7%; 41.3% versus 21.5%;29.6% versus 0%, P < 0.001). The number of ET to achieve live birth per OPU in SBT group was obiviously more than in DET group(1.48 ± 0.578 versus 1.22 ± 0.557 ,P < 0.001). Conclusion: With a marginal difference cumulative live birth rate, the lower live birth rate per fresh transfer and higher number of ET per OPU in the SBT group suggested that it might take longer time to achieve a live birth with single blastocyst strategy. A trade-off decision should be made between efficiency and safety. [ABSTRACT FROM AUTHOR]

Vehicle detection technology is of great significance for realizing automatic monitoring and AI-assisted driving systems. The state-of-the-art object detection method, namely, a class of YOLOv5, has often been used to detect vehicles. However, it suffers some challenges, such as a high computational load and undesirable detection rate. To address these issues, an improved lightweight YOLOv5 method is proposed for vehicle detection in this paper. In the presented method, C3Ghost and Ghost modules are introduced into the YOLOv5 neck network to reduce the floating-point operations (FLOPs) in the feature channel fusion process and enhance the feature expression performance. A convolutional block attention module (CBAM) is introduced to the YOLOv5 backbone network to select the information critical to the vehicle detection task and suppress uncritical information, thus improving the detection accuracy of the algorithm. Furthermore, CIoU_Loss is considered the bounding box regression loss function to accelerate the bounding box regression rate and improve the localization accuracy of the algorithm. To verify the performance of the proposed approach, we tested our model via two case studies, i.e., the PASCAL VOC dataset and MS COCO dataset. The results show that the detection precision of the proposed model increased 3.2%, the FLOPs decreased 15.24%, and the number of model parameters decreased 19.37% compared with those of the existing YOLOv5. Through case studies and comparisons, the effectiveness and superiority of the presented approach are demonstrated. • Novel YOLOv5 with a lightweight design for vehicle detection. • Convolutional block attention module improves model performance to select critical information. • CIoU_Loss function to speed up bounding box regression rate and improve localization accuracy. • Considerably better vehicle detection compared with related methods. • Demonstration of the improved performance using the two real cases. [ABSTRACT FROM AUTHOR]

Accurate identification of the tool wear states in machining of titanium alloys continues to be a thorny problem. Rapid construction of accurate and effective tool wear predictive models with regard to various titanium alloys is indispensable for promoting the development of intelligent manufacturing. This article presents a novel WOA-SVM model that integrates support vector machine (SVM) and whale optimization algorithm (WOA), which is utilized for accurate estimation of the tool wear in end milling of titanium alloy Ti-6Al-4V under variable cutting conditions. The signal features in three domains are extracted from the original cutting forces and constitute the “complete features” that are adopted as the monitoring features. Besides, neighborhood preserving embedding (NPE) is utilized to fuse the monitoring features and realize dimension reduction, which aims at improving the modeling efficiency of the novel WOA-SVM model. The experimental results show that under the premise of guaranteeing prediction accuracy, the utilization of NPE reduces the modeling time-consumption of WOA-SVM by more than 90%. In comparison with the commonly used methods (such as PSO-SVM and GSA-SVM), WOA-SVM takes on comparable prediction accuracy and can reduce the modeling time-consumption by more than 30% in most cases. Moreover, the WOA-SVM model has better prediction accuracy and stability than other classical methods, such as ${k}$ -nearest neighbor ($k$ -NN), feedforward neural network (FFNN), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and classification and regression tree (CART). Therefore, the combination of NPE and WOA-SVM provides a guarantee for rapidly constructing accurate tool wear predictive models in machining of titanium alloys. [ABSTRACT FROM AUTHOR]

To ensure the mission implementation of Autonomous Underwater Vehicles (AUVs), faults occurring on actuators should be detected and located promptly; therefore, reliable control strategies and inputs can be effectively provided. In this paper, faults occurring on the propulsion and attitude control systems of a torpedo-shaped AUV are analyzed and located while fault features may induce confusions for conventional fault localization (FL). Selective features of defined fault parameters are assorted as necessary conditions against different faulty actuators and synthesized in a fault tree subsequently to state the sufficiency towards possible abnormal parts. By matching fault features with those of estimated fault parameters, suspected faulty sections are located. Thereafter, active FL strategies that analyze the related fault parameters after executing purposive actuator control are proposed to provide precise fault location. Moreover, the generality of the proposed methods is analyzed to support extensive implementations. Simulations based on finite element analysis against a torpedo-shaped AUV with actuator faults are carried out to illustrate the effectiveness of the proposed methods. [ABSTRACT FROM AUTHOR]