Feature Extraction for Bearing Prognostics Based on Continuous Hidden Markov Model

2014 ◽  
Vol 541-542 ◽  
pp. 1483-1486
Author(s):  
Jian She Kang ◽  
Xing Hui Zhang ◽  
Jin Song Zhao ◽  
Lei Xiao
Keyword(s):  
Fault Diagnosis ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Hidden Markov ◽  
High Reliability ◽  
Degradation Process ◽  
Health Condition ◽  
Diagnosis And Prognosis ◽  
Failure Data ◽  
Fault Diagnosis And Prognosis

Many research papers implemented fault diagnosis and prognosis when there are many history data. However, for some capital and high reliability equipment, it is very difficult to acquire some run-to-failure data. In this case, the fault diagnosis and prognosis become very hard. In order to address this issue, continuous hidden Markov model (CHMM) is used to track the degradation process in this paper. With the degradation, the log-likelihood which is the output of CHMM will decrease gradually. Therefore, this indicator can be used to evaluate the health condition of monitored equipment. Finally, bearing run-to-failure data sets are used to validate the model’s effectiveness

scholarly journals Fault Diagnosis and Prognosis of Bearing Based on Hidden Markov Model with Multi-Features

2020 ◽  
Vol 5 (1) ◽  
pp. 71-84 ◽  
Author(s):  
Weiguo Zhao ◽  
Tiancong Shi ◽  
Liying Wang
Keyword(s):  
Fault Diagnosis ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Hidden Markov ◽  
Wavelet Packet ◽  
Diagnosis And Prognosis ◽  
Pca Method ◽  
The Time Domain ◽  
Low Dimensional ◽  
Fault Diagnosis And Prognosis

AbstractA new approach to achieve fault diagnosis and prognosis of bearing based on hidden Markov model (HMM) with multi-features is proposed. Firstly, the time domain, frequency domain, and wavelet packet decomposition are utilized to extract the condition features of bearing vibration signals, and the PCA method is merged into multi-features to reduce their dimensionality. Then the low-dimensional features are processed to obtain the scalar probabilities of each bearing condition, which are multiplied to generate the observed values of HMM. The results reveal that the established approach can well diagnose fault conditions and achieve the remaining life estimation of bearing.

scholarly journals Symbolic Important Point Perceptually and Hidden Markov Model Based Hydraulic Pump Fault Diagnosis Method

Sensors ◽  
2018 ◽  
Vol 18 (12) ◽  
pp. 4460 ◽  
Author(s):  
Yunzhao Jia ◽  
Minqiang Xu ◽  
Rixin Wang
Keyword(s):  
Fault Diagnosis ◽  
Markov Model ◽  
Hidden Markov Model ◽  
Hydraulic System ◽  
Hidden Markov ◽  
Operating Conditions ◽  
Fault Classification ◽  
Hydraulic Pump ◽  
Diagnosis Method ◽  
Original Time

Hydraulic pump is a driving device of the hydraulic system, always working under harsh operating conditions, its fault diagnosis work is necessary for the smooth running of a hydraulic system. However, it is difficult to collect sufficient status information in practical operating processes. In order to achieve fault diagnosis with poor information, a novel fault diagnosis method that is the based on Symbolic Perceptually Important Point (SPIP) and Hidden Markov Model (HMM) is proposed. Perceptually important point technology is firstly imported into rotating machine fault diagnosis; it is applied to compress the original time-series into PIP series, which can depict the overall movement shape of original time series. The PIP series is transformed into symbolic series that will serve as feature series for HMM, Genetic Algorithm is used to optimize the symbolic space partition scheme. The Hidden Markov Model is then employed for fault classification. An experiment involves four operating conditions is applied to validate the proposed method. The results show that the fault classification accuracy of the proposed method reaches 99.625% when each testing sample only containing 250 points and the signal duration is 0.025 s. The proposed method could achieve good performance under poor information conditions.

Concept Generation Algorithms for Repository-Based Early Design

2006 ◽  
Author(s):  
Jayson Vucovich ◽  
Nikhil Bhardwaj ◽  
Hoi-Hei (Terence) Ho ◽  
Manjeshwar Ramakrishna ◽  
Mayur Thakur ◽  
...  
Keyword(s):  
Markov Model ◽  
Hidden Markov Model ◽  
Design Research ◽  
Hidden Markov ◽  
Functional Model ◽  
Dynamic Programming Algorithm ◽  
Programming Algorithm ◽  
Design Knowledge ◽  
Concept Generation ◽  
Failure Data

Modern product and engineering design research explores methods for formally generating design concepts from stored knowledge. We discuss a design methodology which utilizes archived design knowledge gained from product dissection to aid novice designers in developing new product designs. In this design paradigm, new designs are developed as a model of the product’s intended functionality, rather than a model of actual, physical components. This paper formulates an algorithm to automatically generate a set of components to instantiate such a functional model using archived design knowledge, which maps components to the functions they can satisfy and provides precedents for which components can be connected. In order to avoid generating an exponential number of instantiations, component failure data is leveraged to develop a dynamic programming algorithm. In addition, a method which uses this information to train a Hidden Markov Model is also developed. This Hidden Markov Model is consulted to generate a set of instantiations with low failure rates while avoiding exponential runtime.

Sign in / Sign up

Export Citation Format

Share Document