CN116481821A - Engine fault early warning method and system based on big data analysis management - Google Patents
Engine fault early warning method and system based on big data analysis management Download PDFInfo
- Publication number
- CN116481821A CN116481821A CN202310408632.0A CN202310408632A CN116481821A CN 116481821 A CN116481821 A CN 116481821A CN 202310408632 A CN202310408632 A CN 202310408632A CN 116481821 A CN116481821 A CN 116481821A
- Authority
- CN
- China
- Prior art keywords
- fault
- identified
- vibration signals
- features
- early warning
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000007405 data analysis Methods 0.000 title claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 49
- 238000004364 calculation method Methods 0.000 claims abstract description 45
- 230000009467 reduction Effects 0.000 claims abstract description 24
- 238000000513 principal component analysis Methods 0.000 claims abstract description 23
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000007726 management method Methods 0.000 claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 12
- 238000012706 support-vector machine Methods 0.000 claims abstract description 11
- 238000000605 extraction Methods 0.000 claims abstract description 9
- 238000012360 testing method Methods 0.000 claims abstract description 6
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000003745 diagnosis Methods 0.000 claims description 13
- 238000013507 mapping Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 238000013523 data management Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000007847 structural defect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
- G01M15/04—Testing internal-combustion engines
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
- G01M15/04—Testing internal-combustion engines
- G01M15/12—Testing internal-combustion engines by monitoring vibrations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses an engine fault early warning method and system based on big data analysis management, which belong to the technical field of engine fault analysis and comprise the steps of acquiring vibration signals of an engine under an unsteady state working condition by using a sensor, acquiring vibration signals to be identified, carrying out calculation and extraction on the vibration signals to be identified, acquiring time domain features, then calculating and extracting frequency domain features in the vibration signals to be identified by using a wavelet packet decomposition algorithm, carrying out dimension reduction processing on the time domain features and the frequency domain features by using a kernel principal component analysis feature algorithm, extracting fault sensitive features, establishing a fault automatic classifier by using a support vector machine algorithm, training and testing a fault subset, carrying out calculation and analysis on the fault sensitive features by using the fault automatic classifier, judging the state of the vibration signals to be identified, and completing identification of the faults of the vibration signals.
Description
Technical Field
The invention belongs to the technical field of engine fault analysis, and particularly relates to an engine fault early warning method and system based on big data analysis and management.
Background
At present, a diesel engine is a common piston engine, is extremely widely applied in industrial production, and is mainly applied to the fields of petroleum, electric power, metallurgy and the like. The camshaft is an important part of the engine, the valve timing of the engine is controlled by virtue of the cam, the camshaft is specially used for controlling the opening and closing time of an intake valve and an exhaust valve of the engine, the rotation speed of the camshaft is only half of that of a crankshaft of the engine, the camshaft can be continuously accelerated and decelerated to rotate in the running process, the idling speed is only hundreds of revolutions per minute, and the speed can reach more than 2000 revolutions per minute when the camshaft is accelerated or operated under heavy load, so that the high rotation speed requires the camshaft of the engine to have better toughness, and serious and more metallurgical defects such as loose structure, internal inclusion, cracks and the like cannot exist in the camshaft. The camshaft is generally composed of three parts, namely a cam, a journal and a shaft bottom, and has a relatively complex shape, and has step change and a large number of hot spots. When the engine works, the pair of the cam on the cam shaft and the valve ejector rod bear extremely high compression stress and high-speed sliding friction force, the lubrication between the cam and the valve ejector rod is poor, and the two parts are extremely easy to scratch and peel, so that the large-area abrasion phenomenon is caused. Meanwhile, due to the defects in the manufacture of the cam shaft, cracks are easily generated at the shaft neck part with concentrated load, and the broken shaft is broken due to long-time crack growth, so that the normal operation of the engine is seriously influenced. For the engine cam shaft, the core part of the cam shaft must have enough toughness, and the cam working surface must have higher hardness, excellent wear resistance and certain corrosion resistance, stable internal structure and stable size and precision, no internal metallurgical and structural defects, high fatigue resistance and long service life, and can well meet the long-time working requirements under different running speeds.
Therefore, the state monitoring is carried out on the engine, the safe, stable and efficient operation of the engine is ensured, the engine vibration signal has very important safe and economic values, and the fault diagnosis method based on the engine vibration signal processing becomes one of the more effective fault diagnosis means used at present because the engine vibration signal has rich state information.
The fault diagnosis method can be divided into a model-based method, a signal-based method, a hybrid active method and a data driving method, and most of the traditional data driving method is based on the assumption that the traditional data driving method has a linear structure when extracting vibration signal characteristics, and in the fault diagnosis of mechanical equipment, the more the number of sensors of the accurate monitoring control system is, the more indexes for representing the running state of the equipment are, and the data of states described by a plurality of variables are abstracted to be high-dimensional data. The high-dimensional data provides extremely rich and detailed information about the state of equipment, and the characteristic information is often nonlinear, so that when the number of samples is large, the calculation amount of a program is large, the efficiency of fault diagnosis is seriously reduced, and the real-time requirement of fault diagnosis in actual engineering is difficult to meet.
Disclosure of Invention
Problems to be solved
Aiming at the problems that the existing fault characteristic information is nonlinear, when the number of samples is large, the calculation amount of a program is large, the efficiency of fault diagnosis is seriously reduced, and the real-time requirement of fault diagnosis in actual engineering is difficult to meet, the invention provides an engine fault early warning method and system based on big data analysis and management.
Technical proposal
In order to solve the problems, the invention adopts the following technical scheme.
The engine fault early warning method based on big data analysis and management adopts the following steps:
step 1: the method comprises the steps that a sensor is used for collecting vibration signals of an engine under an unsteady state working condition, and vibration signals to be identified are obtained;
step 2: calculating and extracting a vibration signal to be identified to obtain time domain features, and calculating and extracting frequency domain features in the vibration signal to be identified by using a wavelet packet decomposition algorithm;
step 3: performing dimension reduction processing on the time domain features and the frequency domain features by using a kernel principal component analysis feature algorithm, and extracting fault sensitive features;
step 4: establishing a fault automatic classifier by using a support vector machine algorithm, and training and testing a fault subset;
step 5: and calculating and analyzing the fault sensitive characteristics by using a fault automatic classifier, judging the state of the vibration signal to be identified, and completing the identification of the vibration signal fault.
Preferably, the time domain features include a valid value, a peak-to-peak value, an average value, a square root amplitude, a kurtosis index, a pulse index, and a margin index.
Preferably, the calculation formula of the effective value is as follows:
the calculation formula of the peak value is as follows:
the calculation formula of the peak value is as follows:
x p-p =x max -x min
the calculation formula of the average value is as follows:
the square root amplitude is calculated as follows:
the kurtosis index is calculated as follows:
the pulse index:
the margin index:
wherein x is i I=1, 2, …, N is the number of sample points for the time domain sequence of the signal.
Further, the method comprises the steps of calculating and extracting the vibration signals to be identified by using a wavelet packet decomposition algorithm, namely, firstly, carrying out three-layer wavelet packet decomposition on the vibration signals to be identified, then calculating the energy of signals in each frequency band to obtain signal energy filtering, selecting and extracting the energy ratio of the vibration signals to be identified in different frequency bands, and taking the energy ratio as a frequency domain characteristic parameter.
Further, the method for reducing the dimension in the time domain feature and the frequency domain feature by using the kernel principal component analysis feature algorithm is to firstly combine the time domain feature and the frequency domain feature of the vibration signal to be identified to construct a feature parameter matrix X m×n Then uses nonlinear variation function psi (x i ) And mapping the characteristic parameter matrix into a high-dimensional space F, and then performing characteristic dimension reduction processing by using a principal component analysis characteristic algorithm.
The covariance matrix of the feature space F is:
wherein m is the number of samples, n is the number of features, and T is a proportional parameter.
Furthermore, the feature dimension reduction processing of the principal component analysis feature algorithm is to set a sample matrix as X m×n M is the number of samples, n is the number of features, the sample matrix X is first centralized, and the main direction w with the maximum sample projection variance is determined according to the following formula:
the first k main directions of extraction form a load matrix, w= (W) 1 ,...w k ) The result after extracting the vibration signal characteristics to be identified is X new =XW。
Furthermore, the support vector machine algorithm establishes a fault automatic classifier by mapping the feature vector into a high-dimensional feature space through a kernel function, establishes a classification surface in the high-dimensional feature space, classifies training samples through the classification surface, and enables the distance from the nearest point of sample data to the classification surface to be the largest.
Further, after the time domain features and the frequency domain features are calculated and extracted, scale normalization processing is further performed, and a calculation formula is as follows:
wherein z is the data after scale normalization, x is the original data,delta is the variance of the original data, which is the mean of the original data.
Further, after the detection of the signal sample to be detected is completed, the signal feature subset to be detected is stored into a training sample in the automatic fault classifier according to the fault diagnosis result, and the automatic fault classifier is updated according to the updating of the training sample.
An engine fault early warning system based on big data analysis management, comprising:
the sensing acquisition module is used for acquiring vibration signals in real time under the unsteady state working condition and acquiring vibration signal data to be identified;
the time domain calculation module is used for calculating the vibration signal data to be identified to obtain time domain characteristics;
the frequency domain calculating module is used for calculating and extracting frequency domain characteristics in the vibration signals to be identified by using a wavelet packet decomposition algorithm;
the dimension reduction calculation module is used for performing dimension reduction calculation on the time domain features and the frequency domain features by using a kernel principal component analysis feature algorithm and extracting fault sensitive features;
the fault classification module is used for establishing a fault automatic classifier, carrying out calculation and analysis on fault sensitive characteristics and judging the state of the vibration signal to be identified.
The engine fault early warning method and system based on big data analysis management are characterized in that a sensor is used for collecting vibration signals of an engine under an unsteady state working condition to obtain vibration signals to be recognized, calculation extraction is carried out on the vibration signals to be recognized to obtain time domain features, then a wavelet packet decomposition algorithm is used for calculation extraction of frequency domain features in the vibration signals to be recognized, a kernel principal component analysis feature algorithm is used for carrying out dimension reduction processing on the time domain features and the frequency domain features, fault sensitive features are extracted, a support vector machine algorithm is used for establishing a fault automatic classifier, training and testing fault subsets, the fault automatic classifier is used for carrying out calculation analysis on the fault sensitive features, the state of the vibration signals to be recognized is judged, recognition of vibration signal faults is completed, data error rate is reduced, analysis results are more reliable, diagnosis sensitivity, robustness and accuracy are improved, and the capability of solving complex problems is achieved.
Advantageous effects
Compared with the prior art, the invention has the beneficial effects that:
(1) According to the method, feature calculation is carried out on vibration signal data to be identified through a kernel principal component analysis feature algorithm, dimension reduction processing is carried out through the kernel principal component analysis feature algorithm, fault sensitive features are extracted, a nonlinear change function is used for mapping a feature parameter matrix into a high-dimensional space, and feature dimension reduction processing is carried out through the principal component analysis feature algorithm, so that the purpose of reducing the dimension of an original space is finally achieved, the error rate of the data is reduced, and an analysis result is more reliable;
(2) According to the method, a wavelet packet decomposition algorithm is used for carrying out calculation extraction on the vibration signals to be identified, three-layer wavelet packet decomposition is carried out on the vibration signals to be identified, then energy of signals in each frequency band is calculated, signal energy filtering is obtained, the energy ratio of the vibration signals to be identified in different frequency bands is selected and extracted, the energy ratio is used as a frequency domain characteristic parameter, the sensitivity, the robustness and the accuracy of diagnosis are improved, and the method has the capability of solving complex problems;
(3) According to the invention, a fault automatic classifier is established by using a support vector machine algorithm, the fault sensitive characteristics are calculated and analyzed, the vibration signals which finish the fault identification of the vibration signals are stored into corresponding training samples according to the class of the fault signals, then the fault signals can be used as the training samples to participate in the calculation, training and testing of the next training sample, and as the use times are increased, the training samples in the fault automatic classifier are gradually increased, and the fault classification result of the vibration signal data to be identified is also more and more accurate.
Drawings
In order to more clearly illustrate the technical solutions in embodiments or examples of the present application, the drawings that are required for use in the embodiments or examples description will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application and therefore should not be construed as limiting the scope, and that other drawings may be obtained according to the drawings without inventive effort to those of ordinary skill in the art.
FIG. 1 is a schematic diagram of the method steps of the present invention;
FIG. 2 is a schematic flow chart of the method of the present invention;
fig. 3 is a schematic diagram of a system structure according to the present invention.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments, and that the components of the embodiments of the present application generally described and illustrated in the drawings herein may be arranged and designed in various different configurations.
Thus, the following detailed description of the embodiments of the present application, provided in the accompanying drawings, is not intended to limit the scope of the application as claimed, but is merely representative of selected embodiments of the application, based on which all other embodiments that may be obtained by one of ordinary skill in the art without making inventive efforts are within the scope of this application.
Example 1
As shown in fig. 1 and 2, the engine fault early warning method based on big data analysis and management mainly comprises the following steps:
firstly, a sensor is used for collecting vibration signals of an engine under an unsteady state working condition, vibration signals to be identified are obtained, calculation and extraction are carried out on the vibration signals to be identified, time domain features are obtained, then a wavelet packet decomposition algorithm is used for calculation and extraction of frequency domain features in the vibration signals to be identified, a kernel principal component analysis feature algorithm is used for carrying out dimension reduction processing on the time domain features and the frequency domain features, fault sensitive features are extracted, a support vector machine algorithm is used for establishing a fault automatic classifier, a fault subset is trained and tested, the fault sensitive features are used for calculation and analysis, the state of the vibration signals to be identified is judged, and the identification of the faults of the vibration signals is completed.
And acquiring vibration signals of the engine under an unsteady state working condition by using a sensor, and acquiring the vibration signals to be identified.
And calculating and extracting the vibration signal to be identified to obtain time domain characteristics, wherein the time domain characteristics comprise an effective value, a peak-to-peak value, an average value, a square root amplitude value, a kurtosis index, a pulse index and a margin index.
The calculation formula of the effective value is as follows:
the calculation formula of the peak value is as follows:
the calculation formula of the peak value is as follows:
x p-p =x max -x min
the calculation formula of the average value is as follows:
the square root amplitude is calculated as follows:
the kurtosis index is calculated as follows:
the pulse index:
the margin index:
wherein x is i I=1, 2, …, N is the number of sample points for the time domain sequence of the signal.
And then calculating and extracting frequency domain characteristics in the vibration signals to be identified by using a wavelet packet decomposition algorithm, and calculating and extracting the vibration signals to be identified by using the wavelet packet decomposition algorithm, namely, firstly carrying out three-layer wavelet packet decomposition on the vibration signals to be identified, then calculating the energy of signals in each frequency band to obtain signal energy filtering, and selecting and extracting the energy ratio of the vibration signals to be identified in different frequency bands, wherein the energy ratio is used as a frequency domain characteristic parameter.
After the time domain features and the frequency domain features are calculated and extracted, scale normalization processing is further performed, and a calculation formula is as follows:
wherein z is the data after scale normalization, x is the original data,delta is the variance of the original data, which is the mean of the original data.
Performing dimension reduction processing on the time domain features and the frequency domain features by using a kernel principal component analysis feature algorithm, extracting fault sensitive features, and performing dimension reduction on the time domain features and the frequency domain features by using the kernel principal component analysis feature algorithm by combining the time domain features and the frequency domain features of the vibration signals to be identified to construct a feature parameter matrix X m×n Then uses nonlinear variation function psi (x i ) And mapping the characteristic parameter matrix into a high-dimensional space F, and then performing characteristic dimension reduction processing by using a principal component analysis characteristic algorithm.
The covariance matrix of the feature space F is:
wherein m is the number of samples, n is the number of features, and T is a proportional parameter.
The feature dimension reduction processing of the principal component analysis feature algorithm is to set a sample matrix as X m×n M is the number of samples, n is the number of features, the sample matrix X is first centralized, and the main direction w with the maximum sample projection variance is determined according to the following formula:
||w|| 2 =1
the first k main directions of extraction form a load matrix, w= (W) 1 ,...w k ) The result after extracting the vibration signal characteristics to be identified is X new =XW。
Using a support vector machine algorithm to build an automatic fault classifier, training and testing a subset of faults,
the support vector machine algorithm establishes a fault automatic classifier by mapping a feature vector into a high-dimensional feature space through a kernel function, establishing a classification surface in the high-dimensional feature space, classifying training samples through the classification surface, and enabling the distance from the nearest point of sample data to the classification surface to be maximum.
And calculating and analyzing the fault sensitive characteristics by using the fault automatic classifier, judging the state of the vibration signal to be identified, completing the identification of the vibration signal fault, and after the detection of the signal sample to be detected is completed, storing the signal feature subset to be detected into a training sample in the fault automatic classifier according to the fault diagnosis result, and updating the fault automatic classifier according to the updating of the training sample.
As can be seen from the above description, in this example, vibration signals to be identified are acquired by using a sensor to acquire vibration signals under an unsteady state of an engine, the vibration signals to be identified are calculated and extracted to obtain time domain features, then a wavelet packet decomposition algorithm is used to calculate and extract frequency domain features in the vibration signals to be identified, a kernel principal component analysis feature algorithm is used to perform dimension reduction processing on the time domain features and the frequency domain features, fault sensitive features are extracted, a support vector machine algorithm is used to establish a fault automatic classifier, a fault subset is trained and tested, a fault automatic classifier is used to perform calculation analysis on the fault sensitive features, the state of the vibration signals to be identified is judged, and the identification of the faults of the vibration signals is completed.
Example 2
As shown in fig. 3, the engine fault early warning system based on big data analysis management includes:
the sensing acquisition module is used for acquiring vibration signals in real time under the unsteady state working condition and acquiring vibration signal data to be identified;
the time domain calculation module is used for calculating the vibration signal data to be identified to obtain time domain characteristics;
the frequency domain calculating module is used for calculating and extracting frequency domain characteristics in the vibration signals to be identified by using a wavelet packet decomposition algorithm;
the dimension reduction calculation module is used for performing dimension reduction calculation on the time domain features and the frequency domain features by using a kernel principal component analysis feature algorithm and extracting fault sensitive features;
the fault classification module is used for establishing a fault automatic classifier, carrying out calculation and analysis on fault sensitive characteristics and judging the state of the vibration signal to be identified.
As can be seen from the above description, in this example, vibration signals under unsteady state working conditions are collected in real time by the sensing collection module, vibration signal data to be identified is obtained, the time domain calculation module calculates the vibration signal data to be identified to obtain time domain features, the frequency domain calculation module calculates and extracts frequency domain features in the vibration signals to be identified by using a wavelet packet decomposition algorithm, the dimension reduction calculation module calculates dimension reduction in the time domain features and the frequency domain features by using a kernel principal component analysis feature algorithm, fault sensitive features are extracted, the fault classification module establishes a fault automatic classifier, and performs calculation analysis on the fault sensitive features to determine the state of the vibration signals to be identified.
The foregoing examples have shown only the preferred embodiments of the invention, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that modifications, improvements and substitutions can be made by those skilled in the art without departing from the spirit of the invention, which are all within the scope of the invention.
Claims (10)
1. The engine fault early warning method based on big data analysis and management is characterized by comprising the following steps:
step 1: the method comprises the steps that a sensor is used for collecting vibration signals of an engine under an unsteady state working condition, and vibration signals to be identified are obtained;
step 2: calculating and extracting a vibration signal to be identified to obtain time domain features, and calculating and extracting frequency domain features in the vibration signal to be identified by using a wavelet packet decomposition algorithm;
step 3: performing dimension reduction processing on the time domain features and the frequency domain features by using a kernel principal component analysis feature algorithm, and extracting fault sensitive features;
step 4: establishing a fault automatic classifier by using a support vector machine algorithm, and training and testing a fault subset;
step 5: and calculating and analyzing the fault sensitive characteristics by using a fault automatic classifier, judging the state of the vibration signal to be identified, and completing the identification of the vibration signal fault.
2. The engine fault early warning method based on big data analysis management according to claim 1, wherein: the time domain features comprise an effective value, a peak-to-peak value, an average value, a square root amplitude value, a kurtosis index, a pulse index and a margin index.
3. The engine fault early warning method based on big data analysis management according to claim 2, characterized in that: the calculation formula of the effective value is as follows:
the calculation formula of the peak value is as follows:
the calculation formula of the peak value is as follows:
x p-p =x max -x min
the calculation formula of the average value is as follows:
the square root amplitude is calculated as follows:
the kurtosis index is calculated as follows:
the pulse index:
the margin index:
wherein x is i I=1, 2, …, N is the number of sample points for the time domain sequence of the signal.
4. The engine fault early warning method based on big data analysis management according to claim 3, wherein: the method comprises the steps of calculating and extracting vibration signals to be identified by using a wavelet packet decomposition algorithm, namely, firstly, carrying out three-layer wavelet packet decomposition on the vibration signals to be identified, then calculating the energy of signals in each frequency band to obtain signal energy filtering, selecting and extracting the energy ratio of the vibration signals to be identified in different frequency bands, and taking the energy ratio as a frequency domain characteristic parameter.
5. The engine fault early warning method based on big data analysis management according to claim 4, wherein: the method for reducing the dimension in the time domain features and the frequency domain features by using the kernel principal component analysis feature algorithm comprises the steps of firstly combining the time domain features and the frequency domain features of the vibration signals to be identified to construct a feature parameter matrix X m×n Then uses nonlinear variation function psi (x i ) And mapping the characteristic parameter matrix into a high-dimensional space F, and then performing characteristic dimension reduction processing by using a principal component analysis characteristic algorithm.
The covariance matrix of the feature space F is:
wherein m is the number of samples, n is the number of features, and T is a proportional parameter.
6. The engine fault early warning method based on big data analysis management according to claim 5, wherein the engine fault early warning method based on big data analysis management is characterized in that: the feature dimension reduction processing of the principal component analysis feature algorithm is to set a sample matrix as X m×n M is the number of samples, n is the number of features, the sample matrix X is first centralized, and the main direction w with the maximum sample projection variance is determined according to the following formula:
||w|| 2 =1
the first k main directions of extraction form a load matrix, w= (W) 1 ,...w k ) The result after extracting the vibration signal characteristics to be identified is X new =XW。
7. The engine fault early warning method based on big data analysis management according to claim 6, wherein: the support vector machine algorithm establishes a fault automatic classifier by mapping a feature vector into a high-dimensional feature space through a kernel function, establishing a classification surface in the high-dimensional feature space, classifying training samples through the classification surface, and enabling the distance from the nearest point of sample data to the classification surface to be maximum.
8. The engine fault early warning method based on big data analysis management according to claim 7, wherein: after the time domain features and the frequency domain features are calculated and extracted, scale normalization processing is further performed, and a calculation formula is as follows:
wherein z is the data after scale normalization, x is the original data,delta is the variance of the original data, which is the mean of the original data.
9. The engine fault early warning method based on big data analysis management according to claim 8, wherein: after the detection of the signal sample to be detected is completed, the signal feature subset to be detected is stored into a training sample in the automatic fault classifier according to the fault diagnosis result, and the automatic fault classifier is updated according to the updating of the training sample.
10. Engine fault early warning system based on big data analysis management, characterized by comprising:
the sensing acquisition module is used for acquiring vibration signals in real time under the unsteady state working condition and acquiring vibration signal data to be identified;
the time domain calculation module is used for calculating the vibration signal data to be identified to obtain time domain characteristics;
the frequency domain calculating module is used for calculating and extracting frequency domain characteristics in the vibration signals to be identified by using a wavelet packet decomposition algorithm;
the dimension reduction calculation module is used for performing dimension reduction calculation on the time domain features and the frequency domain features by using a kernel principal component analysis feature algorithm and extracting fault sensitive features;
the fault classification module is used for establishing a fault automatic classifier, carrying out calculation and analysis on fault sensitive characteristics and judging the state of the vibration signal to be identified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310408632.0A CN116481821A (en) | 2023-04-17 | 2023-04-17 | Engine fault early warning method and system based on big data analysis management |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310408632.0A CN116481821A (en) | 2023-04-17 | 2023-04-17 | Engine fault early warning method and system based on big data analysis management |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116481821A true CN116481821A (en) | 2023-07-25 |
Family
ID=87211380
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310408632.0A Pending CN116481821A (en) | 2023-04-17 | 2023-04-17 | Engine fault early warning method and system based on big data analysis management |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116481821A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116736091A (en) * | 2023-08-10 | 2023-09-12 | 湖南遥光科技有限公司 | Electronic system test point expansion method and system, and fault diagnosis method and system |
CN117131425A (en) * | 2023-10-25 | 2023-11-28 | 广东德力宏展智能装备有限公司 | Numerical control machine tool processing state monitoring method and system based on feedback data |
CN117370878A (en) * | 2023-12-07 | 2024-01-09 | 山东第一医科大学第一附属医院(山东省千佛山医院) | Epidermis extraction and positioning method and system based on spine joint vibration information |
CN117540243A (en) * | 2023-10-23 | 2024-02-09 | 巨能摩托车科技有限公司 | Intelligent motorcycle engine fault detection system based on vibration monitoring |
CN117690506A (en) * | 2024-02-01 | 2024-03-12 | 华农恒青科技股份有限公司 | Prediction method of process boundary and evaluation method of feed production process |
CN118585801A (en) * | 2024-07-31 | 2024-09-03 | 青岛鑫硕包装材料有限公司 | Carton production line fault prediction management method and system based on data analysis |
-
2023
- 2023-04-17 CN CN202310408632.0A patent/CN116481821A/en active Pending
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116736091A (en) * | 2023-08-10 | 2023-09-12 | 湖南遥光科技有限公司 | Electronic system test point expansion method and system, and fault diagnosis method and system |
CN117540243A (en) * | 2023-10-23 | 2024-02-09 | 巨能摩托车科技有限公司 | Intelligent motorcycle engine fault detection system based on vibration monitoring |
CN117540243B (en) * | 2023-10-23 | 2024-05-24 | 巨能摩托车科技有限公司 | Intelligent motorcycle engine fault detection system based on vibration monitoring |
CN117131425A (en) * | 2023-10-25 | 2023-11-28 | 广东德力宏展智能装备有限公司 | Numerical control machine tool processing state monitoring method and system based on feedback data |
CN117131425B (en) * | 2023-10-25 | 2024-02-27 | 广东德力宏展智能装备有限公司 | Numerical control machine tool processing state monitoring method and system based on feedback data |
CN117370878A (en) * | 2023-12-07 | 2024-01-09 | 山东第一医科大学第一附属医院(山东省千佛山医院) | Epidermis extraction and positioning method and system based on spine joint vibration information |
CN117370878B (en) * | 2023-12-07 | 2024-03-05 | 山东第一医科大学第一附属医院(山东省千佛山医院) | Epidermis extraction and positioning method and system based on spine joint vibration information |
CN117690506A (en) * | 2024-02-01 | 2024-03-12 | 华农恒青科技股份有限公司 | Prediction method of process boundary and evaluation method of feed production process |
CN117690506B (en) * | 2024-02-01 | 2024-04-26 | 华农恒青科技股份有限公司 | Prediction method of process boundary and evaluation method of feed production process |
CN118585801A (en) * | 2024-07-31 | 2024-09-03 | 青岛鑫硕包装材料有限公司 | Carton production line fault prediction management method and system based on data analysis |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116481821A (en) | Engine fault early warning method and system based on big data analysis management | |
CN111412978B (en) | Reciprocating machinery abnormity detection method based on fault-free vibration signal | |
CN111046945B (en) | Fault type and damage degree diagnosis method based on combined convolutional neural network | |
CN112257530B (en) | Rolling bearing fault diagnosis method based on blind signal separation and support vector machine | |
CN111562108A (en) | Rolling bearing intelligent fault diagnosis method based on CNN and FCMC | |
CN111539152B (en) | Rolling bearing fault self-learning method based on two-stage twin convolutional neural network | |
CN109990891B (en) | Diesel engine vibration impact source identification method based on multi-measuring-point information fusion | |
CN115563553B (en) | Aviation bearing fault diagnosis method of deep migration map convolution network under fluctuation working condition | |
CN110333077B (en) | Rolling bearing fault diagnosis method based on approximate support vector | |
CN113188794B (en) | Gearbox fault diagnosis method and device based on improved PSO-BP neural network | |
CN108709744B (en) | Motor bearing fault diagnosis method under variable load working condition | |
CN112487709B (en) | Marine diesel engine fault tracing method based on sensitivity analysis | |
CN101661029A (en) | Die casting quality on-line detection method | |
CN108956142A (en) | A kind of bearing fault recognition methods | |
CN110595778A (en) | Wind turbine generator bearing fault diagnosis method based on MMF and IGRA | |
CN111076934A (en) | Method for diagnosing potential fault of bearing based on S transformation | |
Jafarian et al. | Vibration analysis for fault detection of automobile engine using PCA technique | |
CN114462480A (en) | Multi-source sensor rolling mill fault diagnosis method based on non-equilibrium data set | |
CN115587290A (en) | Aero-engine fault diagnosis method based on variational self-coding generation countermeasure network | |
CN113314142B (en) | Engine fault detection method, system, storage medium and equipment | |
CN106404443A (en) | Hybrid hoisting device fault forecasting platform and data sample acquisition method thereof | |
CN117807538A (en) | Fault detection method based on dynamic threshold and trend prediction | |
CN117191396A (en) | Gear box fault diagnosis method based on two-stage migration | |
CN108536961B (en) | Marine diesel engine air valve quantitative detection device | |
Wu et al. | A fault diagnosis method of rotating machinery based on multi-scale fuzzy entropy |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |