CN104634603A - Early fault diagnosis method for complex equipment - Google Patents
Early fault diagnosis method for complex equipment Download PDFInfo
- Publication number
- CN104634603A CN104634603A CN201510114270.XA CN201510114270A CN104634603A CN 104634603 A CN104634603 A CN 104634603A CN 201510114270 A CN201510114270 A CN 201510114270A CN 104634603 A CN104634603 A CN 104634603A
- Authority
- CN
- China
- Prior art keywords
- fault diagnosis
- neural network
- kernel function
- signal
- svm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses an early fault diagnosis method for complex equipment. The early fault diagnosis method for the complex equipment comprises the specific steps: extracting collection signals through a sensor; carrying out wavelet transform on the extracted collection signals for de-noising; and establishing a fault diagnosis model based on a BP (Back Propagation) neural network and a fault diagnosis model based on an SVM (Support Vector Machine) according to mechanical features and circuit features, and carrying out fault diagnosis. A kernel function in the fault diagnosis model based on the SVM is a polynomial kernel function, a radial basis function (RBF) or a Sigmoid kernel function. According to the early fault diagnosis method, the fault diagnosis of the complex equipment is divided into a fault diagnosis with the mechanical features and a fault diagnosis with the circuit features, and the fault diagnosis models are established respectively according to the mechanical features and the circuit features, so that mechanical tests are easy, many samples can be obtained, quick convergence can be realized by using the BP neural network, and accuracy is higher; furthermore, the circuit sample data is less. With the adoption of the advantages of small samples of the SVM, the fault diagnosis of the complex system, such as the complex equipment, can be realized.
Description
Technical field
The present invention relates to a kind of method for diagnosing faults, specifically a kind of complex apparatus Incipient Fault Diagnosis method.
Background technology
Fault diagnosis mainly refers to monitor equipment state and the judgement of fault, not only need the reason to device fails, position, degree makes correct judgement, and then prevent, reduce breakdown loss, also need the health status of monitoring equipment, especially to the monitoring of initial failure, carry out early warning judgement as early as possible, reduce the generation of catastrophic failure, also the time of origin of future malfunction can be predicted, fully can save Maintenance Resource like this, also be plant maintenance, maintenance provides best maintenance decision foundation, the realization being in particular condition maintenarnce strategy provides possibility.
Fault diagnosis technically, be actually under the guidance of certain breakdown judge strategy, implement automatically to detect to monitoring, diagnosing equipment, namely by analyzing the fault model building acquisition equipment, extract fault signature, then according to predetermined strategy, principle, comprehensive assessment is carried out to the information monitored, finally point out necessary maintenance measures to maintainer.Therefore, fault diagnosis not only needs the subject such as Comtemporary Control Theory, computer science, artificial intelligence, signal transacting, pattern-recognition, statistical mathematics.Common method is mainly divided into two large classes, is the method based on mathematical model and the method based on artificial intelligence respectively, and the method based on mathematical model is a kind of fault detection method of at present most widespread use.This method is based on the signal monitored, and by signal processing analysis, extracts its characteristic information, then contrasts the feature under normal condition, thus discriminatory analysis equipment whether fault, concrete diagnosis principle figure is as shown in Figure 1.
Detection com-parison and analysis for single signal is simple, such as: certain single system output signal is y (t), as shown in Figure 2, under actual nominal situation, output signal y (t) amplitude is at interval [min (y (t)), max (y (t))] in, if dotted line red in Fig. 2 is amplitude interval range, this is under system nominal situation.In theory under nominal situation, output signal y (t), as red solid line in Fig. 2, compares actual measurement output signal and amplitude interval range, and exceeding is then malfunction.Actual measurement output signal is as shown in blue solid lines in Fig. 2.
Its interval range is easily set up in detection for single signal, if complication system, monitor signal is polynary, is difficult to again set up concrete envelope model, is difficult to carry out fault diagnosis by the method.
Summary of the invention
The object of the present invention is to provide a kind of easy to use, complex apparatus Incipient Fault Diagnosis method that accuracy is high, to solve the problem proposed in above-mentioned background technology.
For achieving the above object, the invention provides following technical scheme:
A kind of complex apparatus Incipient Fault Diagnosis method, concrete steps are as follows:
(1) collection signal is extracted by sensor;
(2) Noise Elimination from Wavelet Transform is carried out to extracting the signal gathered;
(3) set up the fault diagnosis model based on BP neural network and the fault diagnosis model based on SVM respectively with regard to mechanical features and circuit feature, carry out fault diagnosis.
As the further scheme of the present invention: the described fault diagnosis model based on BP neural network carries out the method for fault diagnosis, and concrete steps are as follows:
(1) described sensor is connected in plant equipment and extracts collection signal;
(2) carry out raw data diagnosis by extracting respectively after the signal gathered carries out Wavelet denoising and sample data is trained, then enter BP neural network and carry out pre-service and feature selecting/extraction respectively;
(3) raw data obtains diagnostic result through BP Neural Network Diagnosis again after pre-service and feature selecting/extraction, and sample data carries out learning training after pre-service and feature selecting/extraction;
(4) diagnosis decision-making is carried out for the diagnostic result obtained, then carry out periodic maintenance, condition maintenarnce and correction maintenance respectively.
As the present invention's further scheme: the described fault diagnosis model Kernel Function based on SVM is Polynomial kernel function, Radial basis kernel function RBF or Sigmoid kernel function.
Compared with prior art, the invention has the beneficial effects as follows:
The fault diagnosis of complex apparatus is divided into the fault diagnosis of mechanical features and the fault diagnosis of circuit feature by the present invention, set up fault diagnosis model respectively with regard to mechanical features and circuit feature, machinery is easily tested, and can obtain sample many, use the convergence of BP neural network fast, accuracy is higher; And circuit sample data is few, uses the small sample advantage of SVM, thus realize the fault diagnosis of the complication systems such as complex apparatus.
Accompanying drawing explanation
Fig. 1 is the concrete diagnosis principle figure of method based on mathematical model.
Fig. 2 is that schematic diagram is analyzed in the detection of single system single signal.
Fig. 3 is method for diagnosing faults schematic diagram in the present invention.
Fig. 4 is the method schematic diagram carrying out fault diagnosis in the present invention based on the fault diagnosis model of BP neural network.
Fig. 5 is the topology diagram of BP neural network in the present invention.
Fig. 6 is the learning training schematic diagram of BP neural network in the present invention.
Fig. 7 is the signal waveforms in the present invention after the extraction of mechanical equipment vibration signal denoising.
Fig. 8 is based on the nonlinear mapping plot of the fault diagnosis model input space of SVM to high-dimensional feature space in the present invention.
Fig. 9 is the method schematic diagram carrying out fault diagnosis in the present invention based on the fault diagnosis model of SVM.
Embodiment
Be described in more detail below in conjunction with the technical scheme of embodiment to this patent.
Refer to Fig. 3, a kind of complex apparatus Incipient Fault Diagnosis method, concrete steps are as follows:
(1) collection signal is extracted by sensor;
(2) Noise Elimination from Wavelet Transform is carried out to extracting the signal gathered;
(3) set up the fault diagnosis model based on BP neural network and the fault diagnosis model based on SVM respectively with regard to mechanical features and circuit feature, carry out fault diagnosis.
Refer to Fig. 4, the described fault diagnosis model based on BP neural network carries out the method for fault diagnosis, and concrete steps are as follows:
(1) described sensor is connected in plant equipment and extracts collection signal;
(2) carry out raw data diagnosis by extracting respectively after the signal gathered carries out Wavelet denoising and sample data is trained, then enter BP neural network and carry out pre-service and feature selecting/extraction respectively;
(3) raw data obtains diagnostic result through BP Neural Network Diagnosis again after pre-service and feature selecting/extraction, and sample data carries out learning training after pre-service and feature selecting/extraction;
(4) diagnosis decision-making is carried out for the diagnostic result obtained, then carry out periodic maintenance, condition maintenarnce and correction maintenance respectively.
Refer to Fig. 5, described BP neural network comprises input layer, middle layer and output layer, learning algorithm in described learning training is supervised learning algorithm, adjusting, allowing BP neural network can make correct reaction according to providing correct input and output to BP neural network.Learning training sample is expressed as: (p
i, d
i), i=1,2 ..., n, wherein, p
ifor sample input data, d
ifor sample exports data, by each neuronic parameter of study adjustment, BP neural network is allowed to produce the result expected, after BP neural network model learning training, then input amendment p
i, the output of BP neural network as far as possible with d
iclose.
Refer to Fig. 6, the study of BP neural network is divided into two stages: input the learning sample determined, arrange the structure of BP neural network, calculate the weights and threshold that a front iteration obtains, calculate backward from ground floor, solve each neuronic output; Amendment weights and threshold, affects gradient according to total error, from last one deck forward, calculates each weights and threshold of amendment respectively.Two stages replace repeatedly, according to error amendment weights and threshold between layers, until error meets the demands after convergence.
For the engine in diesel-driven generator, the structure of plant equipment rotary system based on the fault diagnosis model of BP neural network is described, refer to Fig. 6, output signal y (t) after extracting the vibration signal denoising of the engine in diesel-driven generator carries out three layers of wavelet decomposition, and the coefficient of dissociation vector of third layer is respectively [y from low to high
30, y
31, y
32, y
33, y
34, y
35, y
36, y
37] then total signal y:
Y=S
30+ S
31+ S
32+ S
33+ S
34+ S
35+ S
36+ S
37formula 1
Wherein, S
30represent node y
30reconstruct restoring signal, S
31represent node y
31reconstruct restoring signal, S
32represent node y
32reconstruct restoring signal, S
33represent node y
33reconstruct restoring signal, S
34represent node y
34reconstruct restoring signal, S
33represent node y
35reconstruct restoring signal, S
36represent node y
36reconstruct restoring signal, S
37represent node y
37reconstruct restoring signal.
Suppose that the low-limit frequency outputed signal in y (t) is 0, highest frequency is 1, is divided into eight frequency ranges like this, as shown in table 1:
The frequency range of signal after table 1 wavelet decomposition
Signal content | Frequency separation | Signal content | Frequency separation |
S 30 | 0~0.125 | S 34 | 0.500~0.625 |
S 31 | 0.125~0.250 | S 35 | 0.625~0.750 |
S 32 | 0.250~0.375 | S 36 | 0.750~0.875 |
S 33 | 0.375~0.500 | S 37 | 0.875~1 |
The determination of input signal: input signal comprises band signal energy and time domain steepness.
Y (t) wavelet decomposition will be outputed signal to eight frequency bands, calculate each band signal energy:
Wherein, E
3jfor frequency band S
3jcorresponding ability, Ts is the analytical cycle of output signal y (t), j=0,1 ..., 7, k=1,2 ..., N, N are sample size, y
jkfor S
3ja signal kth discrete point amplitude.
In order to improve the accuracy rate of BP neural network, each frequency band energy is made normalized:
Thus eight proper vectors of each band signal energy are obtained:
The probability distribution of the vibration signal of described plant equipment rotary system is close to distributing just very much, and time domain steepness is quantitative description measured signal departs from the degree distributed just very much, and time domain steepness computing method are as follows:
Adopt and be worth normalized most:
Wherein,
for the data after normalization, g
ifor the data before normalization, g
i min, g
i maxthe minimum value of data and maximal value before being respectively normalization.
Therefore, the variable of BP nerve network input parameter P is 10:
The determination of output parameter:
For engine diagnosis, described vibration fault comprises tens of type, wherein has several typical fault, and its incidence accounts for more than 95 of sum, and described typical fault comprises rotor unbalance, rotor rubbing, rotor misalignment and bearing's looseness etc.
Therefore using these four kinds of typical faults as BP neural network failure diagnostic model output parameter, be expressed as
D=[d
1, d
2, d
3, d
4], 0 < di≤1 formula 9
Its desirable output state and implication as shown in table 2:
Table 3.2 output state and definition
Fault type | d 1 | d 2 | d 3 | d 4 |
Rotor unbalance | 1 | 0 | 0 | 0 |
Rotor rubbing | 0 | 1 | 0 | 0 |
Rotor misalignment | 0 | 0 | 1 | 0 |
Bearing's looseness | 0 | 0 | 0 | 1 |
Therefore, in BP neural network failure diagnostic model, the input node in input layer is ten, input node comprises eight frequency band energy signals and two time domain steepness, output node in output layer is four, and according to the principle of 2M+1, the hidden node in middle layer is 21.
Described BP neural network model has good non-linear ability, and for actual nonlinear system, its structure is simple, functional.The key of fault diagnosis is that the one being implemented to defective space from fault signature space maps, and reaches the object of fault diagnosis like this.BP neural network has adaptive faculty, network can not only adaptively learn, network size can also be adjusted, in addition, BP neural network has certain fault-tolerance, and what show input pattern information is incomplete, or it is not too responsive to the defect of feature, and obvious for mechanical features in complex apparatus, easily obtain a large amount of fault samples by experiment, ensure that the confidence level of BP neural network.Thus, in BP neural fusion complication system, the fault diagnosis of complex apparatus is effectively feasible.
Refer to Fig. 8-9, described dimension based on the signal N in the fault diagnosis model of SVM after Wavelet Denoising Method inputs, output characteristic sample (x
i, y
i), by selecting nonlinear transformation θ (), realize the mapping F from former space to high-dimensional feature space, wherein x
i∈ R
nas input vector, y
i∈ [-1,1] conduct exports fault category, i=1,2 ..., N, N represent the quantity of training sample.
Build optimum linearity classification function:
F (x)=sign [(w θ (x)+b)] formula 10
Wherein, w is weight vectors; B is deviation or is classification thresholds.
For meeting structural risk minimization, w and b need be found formula 11 set up:
Wherein, w
tw is the complexity of Controlling model, and C, for pre-determining constant, realizes controlling punishment degree to the sample of output error; R
pfor control errors function.
Described control errors function is R
ptwo norms of error ξ, therefore fault diagnosis model can be optimized for:
st.y
i[w
T·θ(x
i)+b]=1-ξ
Because weight vectors may be infinite dimension, formula 12 is converted into following system of equations:
Wherein, 1=[1,1 ..., 1]
t, Y=[y
1, y
2..., y
n]
t, a=[a
1, a
2..., a
n]
t, I is unit matrix, K=θ (x
i)
tθ (x
j) be kernel function.
Described kernel function is Polynomial kernel function, Radial basis kernel function RBF or Sigmoid kernel function.Concrete functional form is as follows: Polynomial kernel function K (x, x')=(xx'+1)
d, wherein, d is the exponent number of kernel function.Radial basis kernel function RBF:K (x, x')=exp (-γ || x-x'||
2), wherein γ > 0, γ is the coefficient of Control Radius.Sigmoid kernel function: K (x, x')=tanh (vx
tx
i+ α).
When kernel function is Radial basis kernel function, adopts least square method to obtain the value of b and a, finally can obtain the failure modes function after Wavelet Denoising Method:
The described fault diagnosis model based on SVM, when training, inputs training sample (x
1, y
1) ..., (x
n, y
n), wherein, x
i∈ X=R
n, Y={1,2 ... m}, i=1,2 ... N, N are number of training, and n is input variable number, and m is fault category number; Definite kernel CWinInetConnection type; Use QUADRATIC PROGRAMMING METHOD FOR to ask target function type optimum solution, and obtain optimum a; Utilize a support vector machine X in Sample Storehouse, substitute into formula 13, obtain deviate b.
The sorting technique of SVM comprise one to one with the method for one-to-many, One-against-one training speed is more faster than one-to-many, and global optimization ability is strong, on training effect, one-against-rest needs the wherein sample of certain class to distinguish from other sample classes to come, it is more complicated that Optimal Separating Hyperplane compares One-against-one, and excessive matching problem more easily occurs.Therefore, One-against-one is more suitable for engineer applied.
The fault diagnosis of complex apparatus is divided into the fault diagnosis of mechanical features and the fault diagnosis of circuit feature by the present invention, set up fault diagnosis model respectively with regard to mechanical features and circuit feature, machinery is easily tested, and can obtain sample many, use the convergence of BP neural network fast, accuracy is higher; And circuit sample data is few, uses the small sample advantage of SVM, thus realize the fault diagnosis of the complication systems such as complex apparatus.
Above the better embodiment of this patent is explained in detail, but this patent is not limited to above-mentioned embodiment, in the ken that one skilled in the relevant art possesses, various change can also be made under the prerequisite not departing from this patent aim.
Claims (3)
1. a complex apparatus Incipient Fault Diagnosis method, is characterized in that, concrete steps are as follows:
(1) collection signal is extracted by sensor;
(2) Noise Elimination from Wavelet Transform is carried out to extracting the signal gathered;
(3) set up the fault diagnosis model based on BP neural network and the fault diagnosis model based on SVM respectively with regard to mechanical features and circuit feature, carry out fault diagnosis.
2. complex apparatus Incipient Fault Diagnosis method according to claim 1, is characterized in that, the described fault diagnosis model based on BP neural network carries out the method for fault diagnosis, and concrete steps are as follows:
(1) described sensor is connected in plant equipment and extracts collection signal;
(2) carry out raw data diagnosis by extracting respectively after the signal gathered carries out Wavelet denoising and sample data is trained, then enter BP neural network and carry out pre-service and feature selecting/extraction respectively;
(3) raw data obtains diagnostic result through BP Neural Network Diagnosis again after pre-service and feature selecting/extraction, and sample data carries out learning training after pre-service and feature selecting/extraction;
(4) diagnosis decision-making is carried out for the diagnostic result obtained, then carry out periodic maintenance, condition maintenarnce and correction maintenance respectively.
3. complex apparatus Incipient Fault Diagnosis method according to claim 1, is characterized in that, the described fault diagnosis model Kernel Function based on SVM is Polynomial kernel function, Radial basis kernel function RBF or Sigmoid kernel function.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510114270.XA CN104634603A (en) | 2015-03-16 | 2015-03-16 | Early fault diagnosis method for complex equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510114270.XA CN104634603A (en) | 2015-03-16 | 2015-03-16 | Early fault diagnosis method for complex equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104634603A true CN104634603A (en) | 2015-05-20 |
Family
ID=53213598
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510114270.XA Pending CN104634603A (en) | 2015-03-16 | 2015-03-16 | Early fault diagnosis method for complex equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104634603A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109406949A (en) * | 2018-12-14 | 2019-03-01 | 国网山东省电力公司电力科学研究院 | Power distribution network incipient fault detection method and device based on support vector machines |
CN109580230A (en) * | 2018-12-11 | 2019-04-05 | 中国航空工业集团公司西安航空计算技术研究所 | A kind of Fault Diagnosis of Engine and device based on BP neural network |
CN111259927A (en) * | 2020-01-08 | 2020-06-09 | 西北工业大学 | Rocket engine fault diagnosis method based on neural network and evidence theory |
CN111413030A (en) * | 2019-01-07 | 2020-07-14 | 哈尔滨工业大学 | Large-scale high-speed rotation equipment measurement and neural network learning regulation and control method and device based on rigidity vector space projection maximization |
CN111539516A (en) * | 2020-04-22 | 2020-08-14 | 谭雄向 | Power grid fault diagnosis system and method based on big data processing |
RU2753151C1 (en) * | 2020-09-23 | 2021-08-12 | Федеральное государственное бюджетное образовательное учреждение высшего образования "Орловский государственный университет имени И.С. Тургенева" (ФГБОУ ВО "ОГУ имени И.С. Тургенева") | Method for vibration diagnostics of rotary systems |
CN117978612A (en) * | 2024-03-28 | 2024-05-03 | 成都格理特电子技术有限公司 | Network fault detection method, storage medium and electronic equipment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5566092A (en) * | 1993-12-30 | 1996-10-15 | Caterpillar Inc. | Machine fault diagnostics system and method |
CN101158873A (en) * | 2007-09-26 | 2008-04-09 | 东北大学 | Non-linearity process failure diagnosis method |
CN101614787A (en) * | 2009-07-07 | 2009-12-30 | 南京航空航天大学 | Analogical Electronics method for diagnosing faults based on M-ary textural classification device |
KR20120027733A (en) * | 2010-09-13 | 2012-03-22 | 한국수력원자력 주식회사 | Rotating machinery fault diagnostic method and system using support vector machines |
CN103512765A (en) * | 2013-09-13 | 2014-01-15 | 中国科学院苏州生物医学工程技术研究所 | Fault detection method for variable learning rate wavelet BP neural network of blood type centrifugal machine |
CN103995237A (en) * | 2014-05-09 | 2014-08-20 | 南京航空航天大学 | Satellite power supply system online fault diagnosis method |
-
2015
- 2015-03-16 CN CN201510114270.XA patent/CN104634603A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5566092A (en) * | 1993-12-30 | 1996-10-15 | Caterpillar Inc. | Machine fault diagnostics system and method |
CN101158873A (en) * | 2007-09-26 | 2008-04-09 | 东北大学 | Non-linearity process failure diagnosis method |
CN101614787A (en) * | 2009-07-07 | 2009-12-30 | 南京航空航天大学 | Analogical Electronics method for diagnosing faults based on M-ary textural classification device |
KR20120027733A (en) * | 2010-09-13 | 2012-03-22 | 한국수력원자력 주식회사 | Rotating machinery fault diagnostic method and system using support vector machines |
CN103512765A (en) * | 2013-09-13 | 2014-01-15 | 中国科学院苏州生物医学工程技术研究所 | Fault detection method for variable learning rate wavelet BP neural network of blood type centrifugal machine |
CN103995237A (en) * | 2014-05-09 | 2014-08-20 | 南京航空航天大学 | Satellite power supply system online fault diagnosis method |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109580230A (en) * | 2018-12-11 | 2019-04-05 | 中国航空工业集团公司西安航空计算技术研究所 | A kind of Fault Diagnosis of Engine and device based on BP neural network |
CN109406949A (en) * | 2018-12-14 | 2019-03-01 | 国网山东省电力公司电力科学研究院 | Power distribution network incipient fault detection method and device based on support vector machines |
CN111413030A (en) * | 2019-01-07 | 2020-07-14 | 哈尔滨工业大学 | Large-scale high-speed rotation equipment measurement and neural network learning regulation and control method and device based on rigidity vector space projection maximization |
CN111259927A (en) * | 2020-01-08 | 2020-06-09 | 西北工业大学 | Rocket engine fault diagnosis method based on neural network and evidence theory |
CN111259927B (en) * | 2020-01-08 | 2022-08-05 | 西北工业大学 | Rocket engine fault diagnosis method based on neural network and evidence theory |
CN111539516A (en) * | 2020-04-22 | 2020-08-14 | 谭雄向 | Power grid fault diagnosis system and method based on big data processing |
RU2753151C1 (en) * | 2020-09-23 | 2021-08-12 | Федеральное государственное бюджетное образовательное учреждение высшего образования "Орловский государственный университет имени И.С. Тургенева" (ФГБОУ ВО "ОГУ имени И.С. Тургенева") | Method for vibration diagnostics of rotary systems |
CN117978612A (en) * | 2024-03-28 | 2024-05-03 | 成都格理特电子技术有限公司 | Network fault detection method, storage medium and electronic equipment |
CN117978612B (en) * | 2024-03-28 | 2024-06-04 | 成都格理特电子技术有限公司 | Network fault detection method, storage medium and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104634603A (en) | Early fault diagnosis method for complex equipment | |
Gou et al. | Aeroengine control system sensor fault diagnosis based on CWT and CNN | |
CN102855638B (en) | Detection method for abnormal behavior of vehicle based on spectrum clustering | |
CN102705303B (en) | Fault location method based on residual and double-stage Elman neural network for hydraulic servo system | |
CN102122133B (en) | Self-adaption wavelet neural network abnormity detection and fault diagnosis classification system and method | |
CN109444656A (en) | A kind of inline diagnosis method of deformation of transformer winding position | |
CN103914735B (en) | A kind of fault recognition method and system based on Neural Network Self-learning | |
CN104390776B (en) | A kind of redundance aileron actuator fault detect, diagnosis and performance estimating method | |
CN103868690B (en) | Rolling bearing state automatic early warning method based on extraction and selection of multiple characteristics | |
CN102707256B (en) | Fault diagnosis method based on BP-Ada Boost nerve network for electric energy meter | |
CN107505133A (en) | The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM | |
CN104809722A (en) | Electrical device fault diagnosis method based on infrared thermography | |
CN106596116A (en) | Vibration fault diagnosis method of wind generating set | |
CN109063308A (en) | A kind of health evaluating method based on depth quantum learning | |
CN104616033A (en) | Fault diagnosis method for rolling bearing based on deep learning and SVM (Support Vector Machine) | |
CN104408302A (en) | Bearing variable-condition fault diagnosis method based on LMD-SVD (Local Mean Decomposition-Singular Value Decomposition) and extreme learning machine | |
CN109213121A (en) | A kind of fan braking system card cylinder method for diagnosing faults | |
CN102589884A (en) | Method for diagnosing failure of airplane generator bearing based on GentleBoost | |
CN110829417A (en) | Electric power system transient stability prediction method based on LSTM double-structure model | |
CN105606914A (en) | IWO-ELM-based Aviation power converter fault diagnosis method | |
CN103558519A (en) | GIS partial discharge ultrasonic signal identification method | |
CN110110809A (en) | The construction method of fuzzy automata based on Machine Fault Diagnosis | |
CN202119467U (en) | Self-adaptive wavelet neural network categorizing system of anomaly detection and fault diagnosis | |
WO2021114320A1 (en) | Wastewater treatment process fault monitoring method using oica-rnn fusion model | |
CN105241665A (en) | Rolling bearing fault diagnosis method based on IRBFNN-AdaBoost classifier |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150520 |