The ammunition loading system manipulator is susceptible to gear failure due to high-frequency,heavyload reciprocating motions and the absence of protective gear components.After a fault occurs,the distribution of fau...The ammunition loading system manipulator is susceptible to gear failure due to high-frequency,heavyload reciprocating motions and the absence of protective gear components.After a fault occurs,the distribution of fault characteristics under different loads is markedly inconsistent,and data is hard to label,which makes it difficult for the traditional diagnosis method based on single-condition training to generalize to different conditions.To address these issues,the paper proposes a novel transfer discriminant neural network(TDNN)for gear fault diagnosis.Specifically,an optimized joint distribution adaptive mechanism(OJDA)is designed to solve the distribution alignment problem between two domains.To improve the classification effect within the domain and the feature recognition capability for a few labeled data,metric learning is introduced to distinguish features from different fault categories.In addition,TDNN adopts a new pseudo-label training strategy to achieve label replacement by comparing the maximum probability of the pseudo-label with the test result.The proposed TDNN is verified in the experimental data set of the artillery manipulator device,and the diagnosis can achieve 99.5%,significantly outperforming other traditional adaptation methods.展开更多
Any malfunctions of the actuators of the robots have the potential to destroy the robot’s normal motion,and most of the current actuator fault diagnosis methods are difficult to meet the requirements of simplifying t...Any malfunctions of the actuators of the robots have the potential to destroy the robot’s normal motion,and most of the current actuator fault diagnosis methods are difficult to meet the requirements of simplifying the actuator modeling and solving the difficulty of fault data collection.To solve the problem of real-time diagnosis of actuator faults in the 3-PR(P)S parallel robot,the model of 3-PR(P)S parallel robot and data-driven-based method for the fault diagnosis are presented.Firstly,only the input-output relationship of the actuator is considered for modeling actuator faults,reducing the complexity of fault modeling and reducing the time consumption of parameter identification,thereby meeting the requirements of real-time diagnosis.A Simulink model of the electromechanical actuator(EMA)was constructed to analyze actuator faults.Then the short-term analysis method was employed for collecting the sample data of the slider position on the test platform of the EMA system and feature extraction.Training samples for neural networks are obtained.Furthermore,we optimized the Back Propagation(BP)neural network using the Dung Beetle Optimization Algorithm(DBO),which effectively resolved the weights and thresholds of the BP neural network.Compared to BP and Particle Swarm Optimization(PSO)-BP,the DBO-BP has better convergence,convergence rate,and the best-classifying quality.So,the classification for the different actuator faults is obviously improved.Finally,a fault diagnosis system was designed for the actuator of the 3-PR(P)S parallel robot,and the experimental results demonstrate that this system can detect actuator faults within 0.1 seconds.This work also provides the technical support for the fault-tolerant control of the 3-PR(P)S Parallel robot.展开更多
Based on wavelet packet transformation(WPT), genetic algorithm(GA), back propagation neural network(BPNN)and support vector machine(SVM), a fault diagnosis method of diesel engine valve clearance is presented. With po...Based on wavelet packet transformation(WPT), genetic algorithm(GA), back propagation neural network(BPNN)and support vector machine(SVM), a fault diagnosis method of diesel engine valve clearance is presented. With power spectral density analysis, the characteristic frequency related to the engine running conditions can be extracted from vibration signals. The biggest singular values(BSV)of wavelet coefficients and root mean square(RMS)values of vibration in characteristic frequency sub-bands are extracted at the end of third level decomposition of vibration signals, and they are used as input vectors of BPNN or SVM. To avoid being trapped in local minima, GA is adopted. The normal and fault vibration signals measured in different valve clearance conditions are analyzed. BPNN, GA back propagation neural network(GA-BPNN), SVM and GA-SVM are applied to the training and testing for the extraction of different features, and the classification accuracies and training time are compared to determine the optimum fault classifier and feature selection. Experimental results demonstrate that the proposed features and classification algorithms give classification accuracy of 100%.展开更多
In order to guarantee quality during mass serial production of motors, a convenient approach on how to detect and diagnose the faults of a permanent-magnetic DC motor based on armature current analysis and BP neural n...In order to guarantee quality during mass serial production of motors, a convenient approach on how to detect and diagnose the faults of a permanent-magnetic DC motor based on armature current analysis and BP neural networks was presented in this paper. The fault feature vector was directly established by analyzing the armature current. Fault features were extracted from the current using various signal processing methods including Fourier analysis, wavelet analysis and statistical methods. Then an advanced BP neural network was used to finish decision-making and separate fault patterns. Finally, the accuracy of the method in this paper was verified by analyzing the mechanism of faults theoretically. The consistency between the experimental results and the theoretical analysis shows that four kinds of representative faults of low power permanent-magnetic DC motors can be diagnosed conveniently by this method. These four faults are brush fray, open circuit of components, open weld of components and short circuit between armature coils. This method needs fewer hardware instruments than the conventional method and whole procedures can be accomplished by several software packages developed in this paper.展开更多
In order to improve the speed and accuracy of analog circuit fault diagnosis,using Back Propagation Neural Network(BPNN),a new method is proposed based on Particle Swarm Optimization(PSO)to adjust weights of BP neural...In order to improve the speed and accuracy of analog circuit fault diagnosis,using Back Propagation Neural Network(BPNN),a new method is proposed based on Particle Swarm Optimization(PSO)to adjust weights of BP neural network.The model can not only overcome the limitations of the slow convergence and the local extreme values by basic BP algorithm,but also improve the learning ability and generalization ability with a higher precision.The response signals of analog circuit is preprocessed by Wavelet Packet Transform(WPT)as the fault feature.The simulation result shows that the proposed method has higher diagnostic accuracy and faster convergence speed,which is effective for fault location.展开更多
Wireless sensor networks are susceptible to failures of nodes and links due to various physical or computational reasons.Some physical reasons include a very high temperature,a heavy load over a node,and heavy rain.Co...Wireless sensor networks are susceptible to failures of nodes and links due to various physical or computational reasons.Some physical reasons include a very high temperature,a heavy load over a node,and heavy rain.Computational reasons could be a third-party intrusive attack,communication conflicts,or congestion.Automated fault diagnosis has been a well-studied problem in the research community.In this paper,we present an automated fault diagnosis model that can diagnose multiple types of faults in the category of hard faults and soft faults.Our proposed model implements a feed-forward neural network trained with a hybrid metaheuristic algorithm that combines the principles of exploration and exploitation of the search space.The proposed methodology consists of different phases,such as a clustering phase,a fault detection and classification phase,and a decision and diagnosis phase.The implemented methodology can diagnose composite faults,such as hard permanent,soft permanent,intermittent,and transient faults for sensor nodes as well as for links.The proposed implementation can also classify different types of faulty behavior for both sensor nodes and links in the network.We present the obtained theoretical results and computational complexity of the implemented model for this particular study on automated fault diagnosis.The performance of the model is evaluated using simulations and experiments conducted using indoor and outdoor testbeds.展开更多
A novel knowledge-based fuzzy neural network （KBFNN） for fault diagnosis is presented. Crude rules were extracted and the corresponding dependent factors and antecedent coverage factors were calculated firstly from ...A novel knowledge-based fuzzy neural network （KBFNN） for fault diagnosis is presented. Crude rules were extracted and the corresponding dependent factors and antecedent coverage factors were calculated firstly from the diagnostic sample based on rough sets theory. Then the number of rules was used to construct partially the structure of a fuzzy neural network and those factors were implemented as initial weights, with fuzzy output parameters being optimized by genetic algorithm. Such fuzzy neural network was called KBFNN. This KBFNN was utilized to identify typical faults of rotating machinery. Diagnostic results show that it has those merits of shorter training time and higher right diagnostic level compared to general fuzzy neural networks.展开更多
Electronic components' reliability has become the key of the complex system mission execution. Analog circuit is an important part of electronic components. Its fault diagnosis is far more challenging than that of...Electronic components' reliability has become the key of the complex system mission execution. Analog circuit is an important part of electronic components. Its fault diagnosis is far more challenging than that of digital circuit. Simulations and applications have shown that the methods based on BP neural network are effective in analog circuit fault diagnosis. Aiming at the tolerance of analog circuit,a combinatorial optimization diagnosis scheme was proposed with back propagation( BP) neural network( BPNN).The main contributions of this scheme included two parts:( 1) the random tolerance samples were added into the nominal training samples to establish new training samples,which were used to train the BP neural network based diagnosis model;( 2) the initial weights of the BP neural network were optimized by genetic algorithm( GA) to avoid local minima,and the BP neural network was tuned with Levenberg-Marquardt algorithm( LMA) in the local solution space to look for the optimum solution or approximate optimal solutions. The experimental results show preliminarily that the scheme substantially improves the whole learning process approximation and generalization ability,and effectively promotes analog circuit fault diagnosis performance based on BPNN.展开更多
Gear transmissions are widely used in industrial drive systems.Fault diagnosis of gear transmissions is important for maintaining the system performance,reducing the maintenance cost,and providing a safe working envir...Gear transmissions are widely used in industrial drive systems.Fault diagnosis of gear transmissions is important for maintaining the system performance,reducing the maintenance cost,and providing a safe working environment.This paper presents a novel fault diagnosis approach for gear transmissions based on convolutional neural networks(CNNs)and decision-level sensor fusion.In the proposed approach,a CNN is first utilized to classify the faults of a gear transmission based on the acquired signals from each of the sensors.Raw sensory data is sent directly into the CNN models without manual feature extraction.Then,classifier level sensor fusion is carried out to achieve improved classification accuracy by fusing the classification results from the CNN models.Experimental study is conducted,which shows the superior performance of the developed method in the classification of different gear transmission conditions in an automated industrial machine.The presented approach also achieves end-to-end learning that ean be applied to the fault elassification of a gear transmission under various operating eonditions and with signals from different types of sensors.展开更多
According to fault type diversity and fault information uncertainty problem of the hydraulic driven rocket launcher servo system（HDRLSS） , the fault diagnosis method based on the evidence theory and neural network e...According to fault type diversity and fault information uncertainty problem of the hydraulic driven rocket launcher servo system（HDRLSS） , the fault diagnosis method based on the evidence theory and neural network ensemble is proposed. In order to overcome the shortcomings of the single neural network, two improved neural network models are set up at the com-mon nodes to simplify the network structure. The initial fault diagnosis is based on the iron spectrum data and the pressure, flow and temperature（PFT） characteristic parameters as the input vectors of the two improved neural network models, and the diagnosis result is taken as the basic probability distribution of the evidence theory. Then the objectivity of assignment is real-ized. The initial diagnosis results of two improved neural networks are fused by D-S evidence theory. The experimental results show that this method can avoid the misdiagnosis of neural network recognition and improve the accuracy of the fault diagnosis of HDRLSS.展开更多
A new neural network model based on multi-layer perceptron for fault diagnosis of hydraulic pumps is presented, and a framework,ranging from fault signal pick and pre-processing to fault diagnosis, is established. Fi...A new neural network model based on multi-layer perceptron for fault diagnosis of hydraulic pumps is presented, and a framework,ranging from fault signal pick and pre-processing to fault diagnosis, is established. Finally a test was done on an axial pist展开更多
>Transformer faults are quite complicated phenomena and can occur due to a variety of reasons.There have been several methods for transformer fault synthetic diagnosis,but each of them has its own limitations in re...>Transformer faults are quite complicated phenomena and can occur due to a variety of reasons.There have been several methods for transformer fault synthetic diagnosis,but each of them has its own limitations in real fault diagnosis applications.In order to overcome those shortcomings in the existing methods,a new transformer fault diagnosis method based on a wavelet neural network optimized by adaptive genetic algorithm(AGA)and an improved D-S evidence theory fusion technique is proposed in this paper.The proposed method combines the oil chromatogram data and the off-line electrical test data of transformers to carry out fault diagnosis.Based on the fusion mechanism of D-S evidence theory,the comprehensive reliability of evidence is constructed by considering the evidence importance,the outputs of the neural network and the expert experience.The new method increases the objectivity of the basic probability assignment(BPA)and reduces the basic probability assigned for uncertain and unimportant information.The case study results of using the proposed method show that it has a good performance of fault diagnosis for transformers.展开更多
The fault diagnosis model for FMS based on multi layer feedforward neural networks was discussed An improved BP algorithm,the tactic of initial value selection based on genetic algorithm and the method of network st...The fault diagnosis model for FMS based on multi layer feedforward neural networks was discussed An improved BP algorithm,the tactic of initial value selection based on genetic algorithm and the method of network structure optimization were presented for training this model ANN(artificial neural network)fault diagnosis model for the robot in FMS was made by the new algorithm The result is superior to the rtaditional algorithm展开更多
The lithium battery is an essential component of electric cars;prompt and accurate problem detection is vital in guaranteeing electric cars’safe and dependable functioning and addressing the limitations of Back Propa...The lithium battery is an essential component of electric cars;prompt and accurate problem detection is vital in guaranteeing electric cars’safe and dependable functioning and addressing the limitations of Back Propagation(BP)neural networks in terms of vanishing gradients and inability to effectively capture dependencies in time series,and the limitations of Long-Short Term Memory(LSTM)neural network models in terms of risk of overfitting.A method based on LSTM-BP is put forward for power battery fault diagnosis to improve the accuracy of lithium battery fault diagnosis.First,a lithium battery model is constructed based on the second-order RC equivalent circuit and the electro-thermal coupling model,and various lithium battery failures are simulated to examine the fault characteristics.Then,the lithium battery charging and discharging experiments collect,clean,and process the battery data.By constructing a neural network LSTM-BP model,we verified the superiority and accuracy of the LSTM-BP neural network model by comparing the LSTM model and BP model vertically and by comparing the Recurrent Neural Network(RNN)model,the Gated Recurrent Unit(GRU)model,and the Residual Neural Network(ResNet)model of a more advanced architecture horizontally.Finally,the lithium battery fault diagnosis process is summarized through the threshold quantitative criteria,and different faults are diagnosed and analyzed.Theresults show that the LSTM-BP neural network not only overcomes the limitations of the LSTMneural network and BP neural network but also improves the ability to process sequence data and reduces the risk of overfitting.展开更多
To overcome the limitations of the standard feedforward neural network, a novel neural network (i.e., ellipsoidal unit network) is proposed, which is very available for fault diagnosis applications due to its bounded ...To overcome the limitations of the standard feedforward neural network, a novel neural network (i.e., ellipsoidal unit network) is proposed, which is very available for fault diagnosis applications due to its bounded generalization and extrapolation. In this paper, the theory and the structure of such a network are described, and the training algorithm is given based on standard backpropagation algorithm. Then, based on this network, a hierarchical diagnosis network (HDANN) is proposed with respect to multiple faults simultaneous diagnosis for rotating machines. The research results show that HDANN based on ellipsoidal unit networks can obtain more accurate and efficient diagnosis results than single net scheme, and is available for real time condition monitoring and diagnosis of rotating machines.展开更多
A main weak point of back propagation (BP) algorithm is that the search procedure easily falls into the local minimum. In order to solve this problem, a GA ANN algorithm is proposed and applied to fault diagnosis o...A main weak point of back propagation (BP) algorithm is that the search procedure easily falls into the local minimum. In order to solve this problem, a GA ANN algorithm is proposed and applied to fault diagnosis of power transformers. Some examples s展开更多
This paper proposes a Fuzzy Neural Network (FNN) model, which uses a propagation algorithm. A logical operation is defined by a set of weights which are independent of inputs. The realization of the basic And,Or and N...This paper proposes a Fuzzy Neural Network (FNN) model, which uses a propagation algorithm. A logical operation is defined by a set of weights which are independent of inputs. The realization of the basic And,Or and Negation fuzzy logical operations is shown by the fuzzy neuron. A example in fault diagnosis is put forward and the result witnesses some effectiveness of the new FNN model.展开更多
文摘The ammunition loading system manipulator is susceptible to gear failure due to high-frequency,heavyload reciprocating motions and the absence of protective gear components.After a fault occurs,the distribution of fault characteristics under different loads is markedly inconsistent,and data is hard to label,which makes it difficult for the traditional diagnosis method based on single-condition training to generalize to different conditions.To address these issues,the paper proposes a novel transfer discriminant neural network(TDNN)for gear fault diagnosis.Specifically,an optimized joint distribution adaptive mechanism(OJDA)is designed to solve the distribution alignment problem between two domains.To improve the classification effect within the domain and the feature recognition capability for a few labeled data,metric learning is introduced to distinguish features from different fault categories.In addition,TDNN adopts a new pseudo-label training strategy to achieve label replacement by comparing the maximum probability of the pseudo-label with the test result.The proposed TDNN is verified in the experimental data set of the artillery manipulator device,and the diagnosis can achieve 99.5%,significantly outperforming other traditional adaptation methods.
文摘Any malfunctions of the actuators of the robots have the potential to destroy the robot’s normal motion,and most of the current actuator fault diagnosis methods are difficult to meet the requirements of simplifying the actuator modeling and solving the difficulty of fault data collection.To solve the problem of real-time diagnosis of actuator faults in the 3-PR(P)S parallel robot,the model of 3-PR(P)S parallel robot and data-driven-based method for the fault diagnosis are presented.Firstly,only the input-output relationship of the actuator is considered for modeling actuator faults,reducing the complexity of fault modeling and reducing the time consumption of parameter identification,thereby meeting the requirements of real-time diagnosis.A Simulink model of the electromechanical actuator(EMA)was constructed to analyze actuator faults.Then the short-term analysis method was employed for collecting the sample data of the slider position on the test platform of the EMA system and feature extraction.Training samples for neural networks are obtained.Furthermore,we optimized the Back Propagation(BP)neural network using the Dung Beetle Optimization Algorithm(DBO),which effectively resolved the weights and thresholds of the BP neural network.Compared to BP and Particle Swarm Optimization(PSO)-BP,the DBO-BP has better convergence,convergence rate,and the best-classifying quality.So,the classification for the different actuator faults is obviously improved.Finally,a fault diagnosis system was designed for the actuator of the 3-PR(P)S parallel robot,and the experimental results demonstrate that this system can detect actuator faults within 0.1 seconds.This work also provides the technical support for the fault-tolerant control of the 3-PR(P)S Parallel robot.
基金Supported by the National Science and Technology Support Program of China(No.2015BAF07B04)
文摘Based on wavelet packet transformation(WPT), genetic algorithm(GA), back propagation neural network(BPNN)and support vector machine(SVM), a fault diagnosis method of diesel engine valve clearance is presented. With power spectral density analysis, the characteristic frequency related to the engine running conditions can be extracted from vibration signals. The biggest singular values(BSV)of wavelet coefficients and root mean square(RMS)values of vibration in characteristic frequency sub-bands are extracted at the end of third level decomposition of vibration signals, and they are used as input vectors of BPNN or SVM. To avoid being trapped in local minima, GA is adopted. The normal and fault vibration signals measured in different valve clearance conditions are analyzed. BPNN, GA back propagation neural network(GA-BPNN), SVM and GA-SVM are applied to the training and testing for the extraction of different features, and the classification accuracies and training time are compared to determine the optimum fault classifier and feature selection. Experimental results demonstrate that the proposed features and classification algorithms give classification accuracy of 100%.
文摘In order to guarantee quality during mass serial production of motors, a convenient approach on how to detect and diagnose the faults of a permanent-magnetic DC motor based on armature current analysis and BP neural networks was presented in this paper. The fault feature vector was directly established by analyzing the armature current. Fault features were extracted from the current using various signal processing methods including Fourier analysis, wavelet analysis and statistical methods. Then an advanced BP neural network was used to finish decision-making and separate fault patterns. Finally, the accuracy of the method in this paper was verified by analyzing the mechanism of faults theoretically. The consistency between the experimental results and the theoretical analysis shows that four kinds of representative faults of low power permanent-magnetic DC motors can be diagnosed conveniently by this method. These four faults are brush fray, open circuit of components, open weld of components and short circuit between armature coils. This method needs fewer hardware instruments than the conventional method and whole procedures can be accomplished by several software packages developed in this paper.
基金supported the Science and Technology Research Project of Liaoning Provincial Department of Education
文摘In order to improve the speed and accuracy of analog circuit fault diagnosis,using Back Propagation Neural Network(BPNN),a new method is proposed based on Particle Swarm Optimization(PSO)to adjust weights of BP neural network.The model can not only overcome the limitations of the slow convergence and the local extreme values by basic BP algorithm,but also improve the learning ability and generalization ability with a higher precision.The response signals of analog circuit is preprocessed by Wavelet Packet Transform(WPT)as the fault feature.The simulation result shows that the proposed method has higher diagnostic accuracy and faster convergence speed,which is effective for fault location.
文摘Wireless sensor networks are susceptible to failures of nodes and links due to various physical or computational reasons.Some physical reasons include a very high temperature,a heavy load over a node,and heavy rain.Computational reasons could be a third-party intrusive attack,communication conflicts,or congestion.Automated fault diagnosis has been a well-studied problem in the research community.In this paper,we present an automated fault diagnosis model that can diagnose multiple types of faults in the category of hard faults and soft faults.Our proposed model implements a feed-forward neural network trained with a hybrid metaheuristic algorithm that combines the principles of exploration and exploitation of the search space.The proposed methodology consists of different phases,such as a clustering phase,a fault detection and classification phase,and a decision and diagnosis phase.The implemented methodology can diagnose composite faults,such as hard permanent,soft permanent,intermittent,and transient faults for sensor nodes as well as for links.The proposed implementation can also classify different types of faulty behavior for both sensor nodes and links in the network.We present the obtained theoretical results and computational complexity of the implemented model for this particular study on automated fault diagnosis.The performance of the model is evaluated using simulations and experiments conducted using indoor and outdoor testbeds.
基金Project supported by the National Major Science and Technology Foundation of China during the 10th Five-Year Plan Period(No.2001BA204B05-KHK Z0009)
文摘A novel knowledge-based fuzzy neural network （KBFNN） for fault diagnosis is presented. Crude rules were extracted and the corresponding dependent factors and antecedent coverage factors were calculated firstly from the diagnostic sample based on rough sets theory. Then the number of rules was used to construct partially the structure of a fuzzy neural network and those factors were implemented as initial weights, with fuzzy output parameters being optimized by genetic algorithm. Such fuzzy neural network was called KBFNN. This KBFNN was utilized to identify typical faults of rotating machinery. Diagnostic results show that it has those merits of shorter training time and higher right diagnostic level compared to general fuzzy neural networks.
基金National Natural Science Foundation of China(No.61371024)Aviation Science Fund of China(No.2013ZD53051)+2 种基金Aerospace Technology Support Fund of Chinathe Industry-Academy-Research Project of AVIC,China(No.cxy2013XGD14)the Open Research Project of Guangdong Key Laboratory of Popular High Performance Computers/Shenzhen Key Laboratory of Service Computing and Applications,China
文摘Electronic components' reliability has become the key of the complex system mission execution. Analog circuit is an important part of electronic components. Its fault diagnosis is far more challenging than that of digital circuit. Simulations and applications have shown that the methods based on BP neural network are effective in analog circuit fault diagnosis. Aiming at the tolerance of analog circuit,a combinatorial optimization diagnosis scheme was proposed with back propagation( BP) neural network( BPNN).The main contributions of this scheme included two parts:( 1) the random tolerance samples were added into the nominal training samples to establish new training samples,which were used to train the BP neural network based diagnosis model;( 2) the initial weights of the BP neural network were optimized by genetic algorithm( GA) to avoid local minima,and the BP neural network was tuned with Levenberg-Marquardt algorithm( LMA) in the local solution space to look for the optimum solution or approximate optimal solutions. The experimental results show preliminarily that the scheme substantially improves the whole learning process approximation and generalization ability,and effectively promotes analog circuit fault diagnosis performance based on BPNN.
基金supported byan ENGAGE Grant from the Natural Sciences and Engineering Research Council of Canada(NSERC),[funding reference number 11R01296].
文摘Gear transmissions are widely used in industrial drive systems.Fault diagnosis of gear transmissions is important for maintaining the system performance,reducing the maintenance cost,and providing a safe working environment.This paper presents a novel fault diagnosis approach for gear transmissions based on convolutional neural networks(CNNs)and decision-level sensor fusion.In the proposed approach,a CNN is first utilized to classify the faults of a gear transmission based on the acquired signals from each of the sensors.Raw sensory data is sent directly into the CNN models without manual feature extraction.Then,classifier level sensor fusion is carried out to achieve improved classification accuracy by fusing the classification results from the CNN models.Experimental study is conducted,which shows the superior performance of the developed method in the classification of different gear transmission conditions in an automated industrial machine.The presented approach also achieves end-to-end learning that ean be applied to the fault elassification of a gear transmission under various operating eonditions and with signals from different types of sensors.
文摘According to fault type diversity and fault information uncertainty problem of the hydraulic driven rocket launcher servo system（HDRLSS） , the fault diagnosis method based on the evidence theory and neural network ensemble is proposed. In order to overcome the shortcomings of the single neural network, two improved neural network models are set up at the com-mon nodes to simplify the network structure. The initial fault diagnosis is based on the iron spectrum data and the pressure, flow and temperature（PFT） characteristic parameters as the input vectors of the two improved neural network models, and the diagnosis result is taken as the basic probability distribution of the evidence theory. Then the objectivity of assignment is real-ized. The initial diagnosis results of two improved neural networks are fused by D-S evidence theory. The experimental results show that this method can avoid the misdiagnosis of neural network recognition and improve the accuracy of the fault diagnosis of HDRLSS.
文摘A new neural network model based on multi-layer perceptron for fault diagnosis of hydraulic pumps is presented, and a framework,ranging from fault signal pick and pre-processing to fault diagnosis, is established. Finally a test was done on an axial pist
基金Project Supported by National Natural Science Foundation of China （ 50777069 ）.
文摘>Transformer faults are quite complicated phenomena and can occur due to a variety of reasons.There have been several methods for transformer fault synthetic diagnosis,but each of them has its own limitations in real fault diagnosis applications.In order to overcome those shortcomings in the existing methods,a new transformer fault diagnosis method based on a wavelet neural network optimized by adaptive genetic algorithm(AGA)and an improved D-S evidence theory fusion technique is proposed in this paper.The proposed method combines the oil chromatogram data and the off-line electrical test data of transformers to carry out fault diagnosis.Based on the fusion mechanism of D-S evidence theory,the comprehensive reliability of evidence is constructed by considering the evidence importance,the outputs of the neural network and the expert experience.The new method increases the objectivity of the basic probability assignment(BPA)and reduces the basic probability assigned for uncertain and unimportant information.The case study results of using the proposed method show that it has a good performance of fault diagnosis for transformers.
文摘The fault diagnosis model for FMS based on multi layer feedforward neural networks was discussed An improved BP algorithm,the tactic of initial value selection based on genetic algorithm and the method of network structure optimization were presented for training this model ANN(artificial neural network)fault diagnosis model for the robot in FMS was made by the new algorithm The result is superior to the rtaditional algorithm
文摘The lithium battery is an essential component of electric cars;prompt and accurate problem detection is vital in guaranteeing electric cars’safe and dependable functioning and addressing the limitations of Back Propagation(BP)neural networks in terms of vanishing gradients and inability to effectively capture dependencies in time series,and the limitations of Long-Short Term Memory(LSTM)neural network models in terms of risk of overfitting.A method based on LSTM-BP is put forward for power battery fault diagnosis to improve the accuracy of lithium battery fault diagnosis.First,a lithium battery model is constructed based on the second-order RC equivalent circuit and the electro-thermal coupling model,and various lithium battery failures are simulated to examine the fault characteristics.Then,the lithium battery charging and discharging experiments collect,clean,and process the battery data.By constructing a neural network LSTM-BP model,we verified the superiority and accuracy of the LSTM-BP neural network model by comparing the LSTM model and BP model vertically and by comparing the Recurrent Neural Network(RNN)model,the Gated Recurrent Unit(GRU)model,and the Residual Neural Network(ResNet)model of a more advanced architecture horizontally.Finally,the lithium battery fault diagnosis process is summarized through the threshold quantitative criteria,and different faults are diagnosed and analyzed.Theresults show that the LSTM-BP neural network not only overcomes the limitations of the LSTMneural network and BP neural network but also improves the ability to process sequence data and reduces the risk of overfitting.
文摘To overcome the limitations of the standard feedforward neural network, a novel neural network (i.e., ellipsoidal unit network) is proposed, which is very available for fault diagnosis applications due to its bounded generalization and extrapolation. In this paper, the theory and the structure of such a network are described, and the training algorithm is given based on standard backpropagation algorithm. Then, based on this network, a hierarchical diagnosis network (HDANN) is proposed with respect to multiple faults simultaneous diagnosis for rotating machines. The research results show that HDANN based on ellipsoidal unit networks can obtain more accurate and efficient diagnosis results than single net scheme, and is available for real time condition monitoring and diagnosis of rotating machines.
文摘A main weak point of back propagation (BP) algorithm is that the search procedure easily falls into the local minimum. In order to solve this problem, a GA ANN algorithm is proposed and applied to fault diagnosis of power transformers. Some examples s
文摘This paper proposes a Fuzzy Neural Network (FNN) model, which uses a propagation algorithm. A logical operation is defined by a set of weights which are independent of inputs. The realization of the basic And,Or and Negation fuzzy logical operations is shown by the fuzzy neuron. A example in fault diagnosis is put forward and the result witnesses some effectiveness of the new FNN model.
