A uniform initialization Gaussian mixture model–based guided wave–hidden Markov model with stable damage evaluation performance

Yuan, Shenfang; Zhang, Jinjin; Chen, Jian

doi:10.1177/1475921718783652

Cited by 18 publications

(10 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…27,28 Through constructing GW signal features based on probabilistic statistical models and measuring the changing trend, the uncertainty caused by time-varying conditions can be described and suppressed, so that reliable damage monitoring can be achieved. Different kinds of promising methods have been proposed, including GMM, 29,30 hidden Markov model, 31 Gaussian process. 32,33 Among the above methods, GMM is capable of describing arbitrary probability distribution with a finite number of Gaussian components based on unsupervised learning.…”

Section: Online Ablation Monitoring Of Quartz Ceramic Structurementioning

confidence: 99%

Experimental research on online monitoring of high temperature and high speed airflow ablation of quartz ceramic thermal protection structure using guided wave

Ren

Zheng

Qiu

2023

Structural Health Monitoring

Self Cite

View full text Add to dashboard Cite

Thermal protection structure (TPS) plays a key role in protecting hypersonic flight vehicles from harsh service environment and has been widely used. As a common ablative thermal protection material, quartz ceramic TPS realizes thermal insulation by generating ablation during the reentry phases. The structural state of quartz ceramic TPS therefore is of great importance to the service performance and safety of hypersonic flight vehicles, and shows an urgent need for structural health monitoring (SHM). However, the monitoring of real ablation of quartz ceramic structure is barely considered in the current study, let alone online monitoring under harsh service environment, which brings a great challenge to the implementation of reliable ablation monitoring. Aiming at online ablation monitoring of quartz ceramic structure, this paper performs experimental research by taking advantage of guided wave (GW)-based monitoring method. An ablation experiment and monitoring system is built to generate real ablation under high temperature and high speed airflow and perform GW monitoring. The maximal temperature of the system can reach about 2100 °C. The feasibility of ablation monitoring is verified by analyzing ablation-induced influence on GW signals. Gaussian mixture model (GMM) is adopted to combine with GW method to suppress the high temperature variation caused by uncertainty influence on GW signals during the ablation process, so that reliable monitoring of ablation can be realized. Experimental results show that online monitoring of ablation propagation of quartz ceramic structure under high temperature variation is achieved, which can be further used for ablation quantification evaluation.

show abstract

Section: Online Ablation Monitoring Of Quartz Ceramic Structurementioning

confidence: 99%

Experimental research on online monitoring of high temperature and high speed airflow ablation of quartz ceramic thermal protection structure using guided wave

Ren

Zheng

Qiu

2023

Structural Health Monitoring

Self Cite

View full text Add to dashboard Cite

show abstract

“…These damage detection results are always contaminated with uncertainties, such as measurement noise, environment condition, simplifications, and errors in computational models. [14][15][16][17] Thus, the results can deviate significantly from the real situation. Besides, due to the limited measurement and noise, damage parameter identification is often an ill-posed or ill-conditioned inverse problem.…”

Section: Introductionmentioning

confidence: 97%

IWSHM 2019: Perturbation-based Bayesian damage identification using responses at vibration nodes

Huang

Schröder

2021

Structural Health Monitoring

View full text Add to dashboard Cite

One important topic for structural health monitoring is to achieve accurate damage detection with a small number of noisy sensors and without the requirement of a high-fidelity finite element model. This article adopts the Bayesian probabilistic approach combined with a perturbation model using responses at a few vibration nodes for damage monitoring. First, the node displacement, or the response at vibration node, is adopted in this study for real-time damage assessment with a relatively small number of sensors. Then, the construction method of the node displacement response curves based on the perturbation model is proposed to replace the expensive finite element model. After that, a Bayesian framework integrated the node displacement measurement and the response curves are adopted to acquire the probability distribution of the damage parameter. In this article, the accuracy of the node displacement–based Bayesian framework with the perturbation method is evaluated. The proposed method is applied to a supporting structure of a sailplane under different noise levels.

show abstract

“…Yao et al [26] used GMM to depict two-dimensional observation probability density of each state, and constructs the HMM model by considering the state changes of traffic during the communication. Yuan et al [27] illustrated the problem of stability induced by k-means clustering and applied GMM to the unified initialization of guided wave-HMM, and verified in damage evaluation experiment. Aiming at the problem that part of the discretization range was not achievable in the observation state calculation process, Zhang et al [28] used multi-dimensional GMM to fit motion data to build a GMM-HMM recognition model, which can obtain accurate hidden state numbers according to key frames.…”

Section: Introductionmentioning

confidence: 99%

Diagnosis method based on hidden Markov model and Weibull mixture model for mechanical faults of in-wheel motors

Xue¹,

Liu²,

Ding³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

To effectively monitor the operation state of in-wheel motor used in electric vehicle and ensure the safety of the whole vehicle, a diagnosis method based on hidden Markov model (HMM) and Weibull mixture model (WMM) is proposed for mechanical fault of in-wheel motor, that is simply known as a WMM-HMM diagnosis method. Firstly, vibration signals of in-wheel motor are extracted for sensitive symptom parameters (SPs) which are used to characterize the operation state and establish the observation sequence. Secondly, Weibull mixture model is employed to expand the limited observation sequence under various operating states of in-wheel motor to obtain sufficient observation sequence as the training sample set of HMM, and HMM parameters are determined through combining supervised learning with unsupervised learning algorithm. Then the WMM-HMM diagnosis models are constructed under low and medium speed conditions respectively. Finally, the corresponding fault in-wheel motors are customized and the test bench is built to verify the proposed method. The test results show that the proposed method can accurately identify the mechanical fault state of in-wheel motor under different conditions and has good generalization and applicability in traditional methods comparison.

show abstract

A uniform initialization Gaussian mixture model–based guided wave–hidden Markov model with stable damage evaluation performance

Cited by 18 publications

References 27 publications

Experimental research on online monitoring of high temperature and high speed airflow ablation of quartz ceramic thermal protection structure using guided wave

Experimental research on online monitoring of high temperature and high speed airflow ablation of quartz ceramic thermal protection structure using guided wave

IWSHM 2019: Perturbation-based Bayesian damage identification using responses at vibration nodes

Diagnosis method based on hidden Markov model and Weibull mixture model for mechanical faults of in-wheel motors

Contact Info

Product

Resources

About