Online Fault Diagnosis System Research Articles

An intelligent online fault diagnosis system for gas turbine sensors based on unsupervised learning method LOF and KELM

The performance of gas turbine inevitably grades slowly in service. In order to obtain high-precision state assessment, an intelligent online real-time sensor fault diagnosis algorithm was proposed in this paper. This method can automatically establish an accurate diagnosis model in a short time, realize the rapid diagnosis of sensor faults, and can effectively solve the problem of operation data imbalance. Firstly, wavelet analysis quickly converts the high-dimensional time-series sensor signal into a low-dimensional feature vector. Then, the unsupervised method LOF was used to screen the abnormal sensor signals. Finally, the KELM is used to achieve fast online identify the fault modes of the sensors. This fault diagnosis system achieves a faster diagnosis speed while ensuring the accuracy. The effectiveness of the method proposed in this paper was verified on the operational data of a gas turbine, the diagnosis time is about 0.233 s, which greatly increases the diagnosis speed compared with other methods, at the same time, the proposed method can guarantee more than 95 % of diagnosis accuracy.

Research on online monitoring and fault diagnosis system based on multivariate empirical mode decomposition

Abstract Mechanical equipment is crucial to industrial production, so the monitoring and fault diagnosis of its vibration signals is important to ensure production safety. And this paper mainly researches on an online monitoring and fault diagnosis system. The system makes use of multiple sensors to collect vibration signals from different positions of machines for multivariate signals so that condition of the machines can be monitored and information of local faults can avoid loss. Besides, this paper applies a method for fault feature extraction of mechanical equipment based on multivariate empirical mode decomposition (MEMD), which can accurately extract frequency of fault features to realize fault diagnosis when the features are weak in the early period. And this paper puts forward a design to the overall framework of the system based on technical requirements of the system, then introduces the multi-sensor mechanical fault diagnosis method based on MEMD. On such basis, this paper employs LabVIEW and Python to develop the upper computer software of the system, and experiments are carried out to testify whether this system is viable or not. The research results show that the system developed by this paper has practical application value in condition monitoring and fault diagnosis of mechanical equipment.

Design of AGV Fault Diagnosis Data Acquisition System Based on Cloud Platform

With the development of industrial Internet of Things technology, online fault diagnosis system has become an important part of remote operation and maintenance. Aiming at the problem that the AGV online fault diagnosis system requires higher real-time data, more data volume and faster transmission rate, this paper proposes a cloud platform-based AGV fault diagnosis data acquisition system. The system uses the STM32 single-chip microcomputer to collect the sensor data carried by the AGV, uses the wireless WIFI module to upload the data to the cloud server, and finally saves the data to the cloud database after conversion and verification, which provides a test platform for the Web remote real-time monitoring and online fault diagnosis system. By building an AGV test platform to conduct experiments, the results show that the throughput rate of the system is 90PPS, and the packet loss rate is within 0.1%. The system can meet the data requirements of the online fault diagnosis system, and has good real-time performance and stability.

A Multi-Stage Fault Diagnosis Method for Proton Exchange Membrane Fuel Cell Based on Support Vector Machine with Binary Tree

The reliability and durability of the proton exchange membrane (PEM) fuel cells are vital factors restricting their applications. Therefore, establishing an online fault diagnosis system is of great significance. In this paper, a multi-stage fault diagnosis method for the PEM fuel cell is proposed. First, the tests of electrochemical impedance spectroscopy under various fault conditions are conducted. Specifically, prone recoverable faults, such as flooding, membrane drying, and air starvation, are included, and different fault degrees from minor, moderate to severe, are covered. Based on this, an equivalent circuit model (ECM) is selected to fit impedance spectroscopy by the hybrid genetic particle swarm optimization algorithm, and then fault features are determined by the analysis of each model parameter under different fault conditions. Furthermore, a multi-stage fault diagnosis model is constructed with the support vector machine with the binary tree, in which fault features obtained from the ECM are used as the characteristic inputs to realize the fault classification (including fault type and fault degree) online. The results show that the accuracy of the basic fault test and subdivided fault test can reach 100% and 98.3%, respectively, which indicates that the proposed diagnosis method can effectively identify flooding, drying, and air starvation of PEM fuel cells.

A fault diagnosis model for proton exchange membrane fuel cell based on impedance identification with differential evolution algorithm

An effective online fault diagnosis system is of great significance to improve the reliability of fuel cell vehicles. In this paper, a fault diagnosis model for proton exchange membrane fuel cells is proposed. Firstly, the tests of electrochemical impedance spectroscopy under different fault types (flooding, drying, air starvation) and fault degrees (minor, moderate, severe) are carried out, and each polarization loss of the fuel cell is denoted by an equivalent circuit model (ECM). Then, the parameters of the ECM are identified by the proposed random mutation differential evolution algorithm. Furthermore, the parameters identified under different fault conditions are used to train and test a probabilistic neural network-based fault diagnosis model. The fault diagnosis model achieves diagnosis accuracies of 100% for the fault type and 96.67% for the fault degree. By setting operating conditions with different fault degrees, the fault diagnosis model proposed in this paper can realize the fault type and fault degree diagnosis, effectively avoiding the misjudgment of fault types, and is effective for improving the reliability of the fuel cell system.

On-Line Open-Phase Fault Detection Method for Switched Reluctance Motors with Bus Current Measurement

This article introduces an on-line fault diagnosis (FD) system to detect and recognize open-phase faults in switched reluctance motors (SRMs). Both tasks, detection and recognition, are based on functions built with the same information but from different sources. Specifically, these functions are constructed from bus current measurement provided by a sensor and from the estimate of such a current provided by an extended Kalman filter (EKF) that performs the estimation from only rotor angular position measurements. In short, the FD system only requires two measurements for employment: bus current and angular position. In order to show its efficacy, results from numerical simulations (performed in a virtual test bench) are presented. Specifically, these simulations involve the dynamics of the SRM, including the magnetic phenomena caused by the analyzed faults. The motor dynamics were obtained with finite element simulations, which guarantee results close to the actual ones.

Research and application of digital virtual maintenance system for steam turbine

After a 430MW machine unit was started with C repair finished, the shafting vibration of it increase rapidly at the phase of on-load. Using the on-line vibration monitoring and fault diagnosis system (TDM system) matched with the unit to check the vibration historical data, and according to the vibration characteristics, oil film vortex fault is diagnosed. In view of this, the vacuum test was carried out on the lubricating oil temperature of the main transformer engine and the condenser of the unit. The results show that the low frequency variable oil temperature and the condenser can be restrained in a large certain load range. The distribution measures of bearing load are adjusted by unit’s adjustment to increase the load of No.3 bearing, so as to eliminate the unit’s low-frequency vibration fault and ensure the normal operation of the unit.

Fault location estimation for series-compensated double-circuit transmission line using EWT and weighted RVFLN

In this paper, empirical Wavelet transform (EWT), Hilbert transform (HT) and weighted random vector functional link network (WRVFLN) are integrated for fault detection, classification, and location estimation in a series capacitor compensated double circuit transmission line (SCCDCTL). The full cycle current signals from the point of fault inception are decomposed using EWT to extract three band-limited modes (BLMs). The four efficacious instantaneous features namely energy, Shannon entropy, the standard deviation of the magnitude, and crest factor are computed from the Hilbert transformed array of the BLMs to construct the feature vector. A diagonal matrix W is computed from the zero sequence current of original six current signals as a weighting factor to categorize the ground fault accurately. Numerous faults are generated with a wide variation of the system conditions such as fault resistance, fault inception angle, fault distance, percentage compensation level, source impedance, line parameters, load angle, and inter-circuit fault in MATLAB/Simulink environments. An efficient WRVFLN computational intelligence technique is proposed to recognize and estimate the location of the faults by taking the extracted suitable feature vector with weight factor as an input. The performances of WRVFLN are compared with the recently developed advanced classifiers such as least-square support vector machine (LSSVM) and extreme learning machine (ELM) in the MATLAB interface. The lesser computational complexity, faster learning speed, superior classification accuracy, accurate fault location estimation, and short event detection time prove that the proposed EWTHT–WRVFLN method can be implemented in the real power system for online fault diagnosis. Finally, the developed system architecture is implemented on the reconfigurable digital field programmable gate array (FPGA) in ISE design suite 14.5 environments to verify the cogency of the proposed method in real-time. The feasibility of the proposed method is tested and validated by using the fast FPGA digital circuitry in a loop.

Design of online fault diagnosis and early warning system for coal mine rotating machinery

Design of online fault diagnosis and early warning system for coal mine rotating machinery

Study of on-line monitoring and fault diagnosis system for a naval transformer

Study of on-line monitoring and fault diagnosis system for a naval transformer

A Comprehensive Monitoring System for Online Fault Diagnosis and Aging Detection of Non-Isolated DC–DC Converters’ Components

Online monitoring is an inevitable necessity for dc–dc converters in industrial applications to achieve a fault-tolerant structure, which ensures reliable operation and prevents catastrophic failures. In this paper, a comprehensive monitoring system is proposed for non-isolated dc–dc converters. The proposed approach is mainly focused on monitoring the most prone to failure components of these converters including switching devices and capacitor. Using the proposed monitoring system, switch/diode open circuit and short circuit faults are detected in less than one switching cycle. Furthermore, online monitoring of the capacitor equivalent series resistance and switch on-state voltage as effective precursors of aging is also carried out. Finally, an inter-turn fault detection technique is presented for the converter inductor. The mentioned monitoring tasks are accomplished using only two electrical sensors (for capturing voltages across the converter diode and inductor) and two temperature sensors. The presented monitoring approach is simple, cost-effective, easy to implement, and applicable for all non-isolated dc–dc converters including buck, boost, and buck-boost converters. Performance of the monitoring system is evaluated by considering different scenarios in the experiments carried out on a laboratory prototype. The results confirm the effectiveness of the proposed system for fault diagnosis and aging monitoring in non-isolated dc–dc converters.

Detecting and Learning Unknown Fault States by Automatically Finding the Optimal Number of Clusters for Online Bearing Fault Diagnosis

This paper proposes an online fault diagnosis system for bearings that detect emerging fault modes and then updates the diagnostic system knowledge (DSK) to incorporate information about the newly detected fault modes. New fault modes are detected using k-means clustering along with a new cluster evaluation method, i.e., multivariate probability density function’s cluster distribution factor (MPDFCDF). In this proposed model, a heterogeneous pool of features is constructed from the signal. A hybrid feature selection model is adopted for selecting optimal feature for learning the model with existing fault mode. The proposed online fault diagnosis system detects new fault modes from unknown signals using k-means clustering with the help of proposed MPDFCDF cluster evaluation method. The DSK is updated whenever new fault modes are detected and updated DSK is used to classify faults using the k-nearest neighbor (k-NN) classifier. The proposed model is evaluated using acoustic emission signals acquired from low-speed rolling element bearings with different fault modes and severities under different rotational speeds. Experimental results present that the MPDFCDF cluster evaluation method can detect the optimal number of fault clusters, and the proposed online diagnosis model can detect newly emerged faults and update the DSK effectively, which improves the diagnosis performance in terms of the average classification performance.

A Design of Online Fault Diagnosis System Based on ESR for Electrolytic Capacitors

ESR(Equivalent Series Resistance) is one of the important indicators for evaluating health status of electrolytic capacitors. Due to the structural specificity of the electrolytic capacitors, ESR is difficult to measure directly. In order to monitor the health status of the electrolytic capacitor in real time, an equivalent model for the electrolytic capacitor is established. The ESR frequency characteristics and temperature characteristics are analyzed and tested. The method of online fault diagnosis by evaluating the health status of the electrolytic capacitor with online calculation of ESR is proposed. Comparing the calculated value of ESR real-time with the initial value, when the former increases to a certain threshold value with respect to the latter under the same working condition, the electrolytic capacitor can be considered as failed, and the observer can perform replacement processing according to this mechanism in time. Finally, the effectiveness of the method was verified by using the dSPACE semi-physical experimental platform.

A Portable Fault Site Investigation and Diagnosis Device for Energy Metering Equipment

Currently, the number of on-site installation acquisition terminals and energy meters in the power system is very large. Because of the complexity of the environment, result in more on-site measurement of failure to field operation and maintenance personnel a great deal of pressure, there is an urgent need for fast, automated troubleshooting and fault diagnosis of the metering device. However, at present the domestic market has not specifically for the measurement of field fault detection device. And on-line monitoring fault diagnosis system reported in the literature, can only diagnose the communication of the metering device, which is difficult to meet the actual needs. This paper proposes a portable fault site investigation and diagnosis device for energy metering equipment. The proposed device can be applied to the field of energy metering equipment maintenance, which is proved by our experiments.

A Fault Detection Device for Energy Metering Equipment

Currently, the number of on-site installation acquisition terminals and energy meters in the power system is very large. Because of the complexity of the environment, result in more on-site measurement of failure to field operation and maintenance personnel a great deal of pressure, there is an urgent need for fast, automated troubleshooting and fault diagnosis of the metering device. However, at present the domestic market has not specifically for the measurement of field fault detection device. And on-line monitoring fault diagnosis system reported in the literature, can only diagnose the communication of the metering device, which is difficult to meet the actual needs. This paper proposes a fault detection device for energy metering equipment. The proposed device can be applied to the field of energy metering equipment maintenance, which is proved by our experiments.

An analytical method for mapping alarm information to model of power grid fault diagnosis

Online fault diagnosis systems have recently been applied in power grids. However, the complex modeling and high‐quality requirements for power grid fault diagnosis have restricted the wide application of online systems. This paper proposes an automatic method for mapping alarm information to a fault diagnosis model. First, we automatically extract the basic logic variables from the alarm information. These can be used in power grid fault diagnoses by employing a key character‐matching algorithm. We then build the associative relationship between electrical devices and circuit breakers based on connection analysis. Finally, an associative matrix of all electrical devices is designed to express the cooperative relationship between different devices in a fault event based on the short‐circuit power mark. Based on these modules, cause–effect events expressing the associated relationship between alarms are established according to the protection configuration principle and protective relay setting principle. Expressing the associated relationship between alarms according to the logical requirements of a fault diagnosis model enables the automatic mapping from alarm information to fault diagnosis model to be realized. The validity of the proposed method for online fault diagnosis is verified using a real fault case that occurred in a power grid.

Bearing fault diagnosis of a permanent magnet synchronous motor via a fast and online order analysis method in an embedded system

Abstract A permanent magnet synchronous motor (PMSM) is a typical electromechanical system widely used in industrial automation. Bearing fault diagnosis is necessary because a bearing is a key and vulnerable component in a PMSM. Order analysis (OA) methods, which include tachometer-based OA and tacholess OA methods, have been proven to be effective tools for diagnosing bearing fault under variable speed conditions. However, tachometer-based OA methods require the installation of an external sensor to obtain rotating speed, whereas tacholess OA methods are usually complicated and require massive computation cost. Traditional OA methods cannot diagnose bearing fault conveniently and timely because of such deficiencies. Thus, a novel fast and online OA (FOOA) method is proposed to realize variable-speed PMSM bearing fault diagnosis in this study. The FOOA method consists of two algorithms. (1) The rotating phase information is extracted from the sinusoidal current of the PMSM, and a series of equal-phase sampling pulses are generated. (2) The bearing signal acquired from a microphone is angular resampled based on the equal-phase sampling pulses. The resampled signal is demodulated, and the envelope order spectrum is calculated for bearing fault identification. The two algorithms are executed sequentially by two micro controller units operating in parallel. Thus, they can be implemented in an embedded system for online fault diagnosis. The effectiveness and flexibility of the proposed FOOA method are validated on both a desktop computer and an embedded system to diagnose different types of defective bearings that are installed on a PMSM test rig.

An Online Fault Diagnosis System for Induction Motors via Instantaneous Power Analysis

ABSTRACTThis article aims to provide a new noninvasive method for the online diagnosis of bearing-localized faults under various loading conditions of the induction motors via instantaneous power analysis. The instantaneous noise variations and sensor offsets are considered to be one of the common factors that yield erroneous fault tracking in an online condition monitoring and fault diagnosis system. An adaptive threshold scheme has been designed to tackle the sensor offsets and instantaneous noise variations for reliable decision making on the existence of fault signatures in an arbitrary environment conditions. The performance of the designed threshold scheme has been evaluated on a motor with various bearing defects operating under various loading conditions. Detailed theoretical and experimental evaluations of several bearing-localized faults are presented. The results indicate the viability and effectiveness of the proposed method.

An Intelligent Fault Diagnosis of Induction Motors in an Arbitrary Noisy Environment

In this paper, an intelligent fault diagnosis system based on instantaneous power spectrum analysis is proposed. The instantaneous noise variations and sensor off-sets are considered to be one of the common factors that yield erroneous fault tracking in an online condition monitoring and fault diagnosis system. The developed system has the capability to detect bearing inner race defects at incipient stages with in an arbitrary noise conditions. An adaptive threshold has been designed to deal with line current noise ambiguities for decision-making on the existence of small fault signatures. The performance of the developed system has been analyzed theoretically and experimentally on a custom designed test rig under various loading conditions of the motor.

Fault detection and diagnosis for gas turbines based on a kernelized information entropy model.

Gas turbines are considered as one kind of the most important devices in power engineering and have been widely used in power generation, airplanes, and naval ships and also in oil drilling platforms. However, they are monitored without man on duty in the most cases. It is highly desirable to develop techniques and systems to remotely monitor their conditions and analyze their faults. In this work, we introduce a remote system for online condition monitoring and fault diagnosis of gas turbine on offshore oil well drilling platforms based on a kernelized information entropy model. Shannon information entropy is generalized for measuring the uniformity of exhaust temperatures, which reflect the overall states of the gas paths of gas turbine. In addition, we also extend the entropy to compute the information quantity of features in kernel spaces, which help to select the informative features for a certain recognition task. Finally, we introduce the information entropy based decision tree algorithm to extract rules from fault samples. The experiments on some real-world data show the effectiveness of the proposed algorithms.