CN115774841A - Water turbine cavitation phenomenon intelligent identification method based on combined spectrum feature extraction - Google Patents
Water turbine cavitation phenomenon intelligent identification method based on combined spectrum feature extraction Download PDFInfo
- Publication number
- CN115774841A CN115774841A CN202211507573.4A CN202211507573A CN115774841A CN 115774841 A CN115774841 A CN 115774841A CN 202211507573 A CN202211507573 A CN 202211507573A CN 115774841 A CN115774841 A CN 115774841A
- Authority
- CN
- China
- Prior art keywords
- cavitation
- water turbine
- sound
- feature extraction
- matrix
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/20—Hydro energy
Landscapes
- Hydraulic Turbines (AREA)
Abstract
The invention discloses a water turbine cavitation phenomenon intelligent identification method based on combined spectrum feature extraction, and belongs to the field of signal identification processing. The method comprises the steps of collecting noise data before and after cavitation of a turbine runner model; decomposing firecracker-like sound spectrums and special pressure pulsation sound spectrums in each group of noise data, extracting index parameters representing the characteristics of the two sound spectrums, normalizing the index parameters, mapping the index parameters to different element positions of row vectors and column vectors of a matrix to form a characteristic vector containing main characteristic ID of bubble sound, and forming an instant contour form matrix A representing the main physical sound state of the bubble sound; and inputting the matrix C obtained after the matrix A is corrected into a trained water turbine cavitation identification model, and outputting a cavitation judgment result. According to the method, the mixed sound spectrum of the shot sound spectrum and the special pulsation spectrum in the acquired water turbine primary cavitation data is extracted and input into a training model as a characteristic vector of cavitation recognition, so that intelligent recognition of the primary cavitation phenomenon of the water turbine by a machine can be realized.
Description
Technical Field
The invention relates to a method for identifying a water turbine cavitation phenomenon, in particular to a method for intelligently identifying the water turbine cavitation phenomenon based on combined spectral feature extraction.
Background
The primary cavitation of the water turbine refers to the phenomenon that the cavitation starts to occur when the gas core contained in the liquid rapidly increases after the local pressure in the liquid is reduced to a critical value. It is accompanied by noise and vibration, which leads to a reduction in power generation efficiency, a reduction in output, and an aggravation of hydraulic vibration. Not only influences the service life of the water turbine, but also threatens the safe operation of the hydropower station and the power grid.
Therefore, the identification of the cavitation of the water turbine has important significance on the operation safety of hydropower stations and power grids, and the identification of the cavitation phenomenon of the water turbine is carried out in a manual judgment mode in the actual operation of the industry at present. Namely: the straight taper section of the draft tube made of transparent organic glass is used for manual observation, and the flow state of the vortex belt of the draft tube and the water outlet edge of the runner is observed. In the test process, a stroboscope (a light-dark alternative light source with adjustable frequency) is used for lighting, the frequency of the stroboscope is adjusted to be equal to or close to the rotating speed frequency of a model water turbine, and the runner blade which seems to be static or rotates at a low speed can be clearly seen by naked eyes, so that the cavitation condition of the water outlet edge of the runner blade can be observed. The method has very high requirements on workers, and the workers with at least ten years of working experience can observe and judge whether cavitation exists. The method has strong subjectivity and low accuracy and efficiency.
In the prior art, there is a method for identifying a turbine cavitation acoustic signal by a big data learning method, for example, patent CN113255848A discloses a turbine cavitation acoustic signal identification method based on big data learning. The technical scheme is as follows: the method comprises the steps of obtaining various neural network models based on big data learning, extracting acoustic signal time sequence data of the hydraulic turbine set, utilizing an SOM neural network to perform time sequence clustering based on various operation working conditions under the condition of multiple output of the hydraulic turbine set, and screening characteristic quantities of stable working conditions under the health state of the hydraulic turbine set; and then, a random forest algorithm is introduced to perform feature screening of the multiple measuring points under the stable working condition operation of the hydraulic turbine set, an optimal feature measuring point and an optimal feature subset which have high sensitivity to the prediction model are extracted, finally, a gate control circulation unit is used for establishing a health state prediction model, and the sum of dynamic tolerances of the multiple measuring points is evaluated in a self-adaptive manner to judge whether the equipment has a primary cavitation phenomenon or not and perform early warning reminding.
The characteristics of cavitation noise of the water turbine need to be known, the characteristics of phenomena related to cavitation of the water turbine need to be known, and information contained in data can be accurately and comprehensively analyzed from sample data, so that the analysis result of test data is combined with the principle of cavitation of the water turbine, different stages of cavitation of the water turbine can be effectively distinguished, and diagnosis and identification of primary model runner cavitation are completed. The characteristic quantity of stable operating mode under the above-mentioned patent is through screening hydraulic turbine unit health status comes the mode of prediction output future short-term stable operating mode information in advance, all is deficient in the degree of accuracy and the discernment efficiency of hydraulic turbine primary cavitation phenomenon discernment.
When the cavitation of the water turbine occurs, micro bubbles are generated at the outer edge of the blade, and meanwhile, multi-state noise is generated, and due to the fact that the multi-state noise has atypical physical sound state characteristics, the test technology based on the current physical sound state configuration cannot accurately identify whether the model runner is cavitated or not through the sound state of the multi-state noise.
Disclosure of Invention
The invention aims to solve the problems of the water turbine cavitation identification technology in the prior art, and provides an intelligent identification method for the water turbine cavitation phenomenon based on combined spectrum feature extraction, which can quickly and accurately realize intelligent identification of the initial cavitation phenomenon of the water turbine.
In order to achieve the above object, the technical solution of the present invention is as follows:
an intelligent identification method for cavitation phenomena of a water turbine based on combined spectral feature extraction is characterized by comprising the following steps:
s1, collecting noise data when cavitation occurs to a turbine runner model;
s2, decomposing mixed signals in the noise data into a plurality of single-wave signals, and extracting a first main frequency signal and a second main frequency signal;
s3, calculating the statistical characteristics of the first main frequency signal and the second main frequency signal respectively to obtain related analysis index parameters and carrying out normalization processing;
s4, mapping the index parameters after normalization processing to different element positions of a row vector and a column vector of the matrix to form a PCSV vector containing the cavitation bubble sound main feature ID;
s5, inputting the PCSV vector as a feature vector into a neural network for training to obtain a cavitation bubble sound recognition model of the water turbine;
and S6, processing immediately acquired bubble sound data of the water turbine, inputting the processed data into a cavitation bubble sound identification model of the water turbine, and outputting a cavitation identification result.
Further, step S5 includes:
s51, giving different weights to index parameters of different main frequency signals in the PCSV vector for weighted average processing, and then arranging and combining the obtained new vectors to find a contour form matrix B which can represent the main physical sound state of the noise data;
and S52, multiplying each index parameter in the profile form matrix B by a correction value respectively to obtain a corrected profile form matrix D containing a plurality of new vectors, wherein the corrected profile form matrix D forms a cavitation bubble sound identification model of the water turbine.
Further, step S6 specifically includes:
s61, collecting instant noise data of a water turbine runner;
s62, decomposing the mixed signal in the instant noise data into a plurality of single-wave signals, respectively calculating the statistical characteristics of the single-wave signals, obtaining related analysis index parameters and carrying out normalization processing;
s63, mapping each index parameter after normalization processing to different element positions of a row vector and a column vector of the matrix to form an instant feature vector containing the bubble sound main feature ID;
s64, giving different weights to index parameters of different single-wave signals in the instant characteristic vector for weighted average processing, rearranging and combining the obtained new vectors, and finding out an instant contour form matrix A which can represent the main physical sound state of the noise data of the section most;
s65, multiplying each index parameter in the instantaneous contour form matrix A by a correction value to obtain a corrected contour form matrix C containing a plurality of new vectors;
and S66, inputting the corrected contour form matrix C into a turbine cavitation bubble sound recognition model trained in advance, and outputting a cavitation judgment result.
Further, if the corrected profile form matrix C is equal to the corrected profile form matrix D, outputting the result that cavitation occurs in the turbine runner, otherwise, outputting the result that cavitation does not occur.
Further, the related analysis index parameters include a time domain index parameter, a power spectral density index parameter and a frequency domain index parameter; the time domain analysis index parameters comprise a peak value vpp, a quartile frequency probability PVpH, a standard deviation St, a kurtosis Ku, a skewness Sk and an information entropy H; the frequency domain indicator parameter includes a center of gravity frequency PsdFc.
Further, the weight assignment range of the index parameter belonging to the first main frequency signal is 0.6 to 0.8.
Further, a 0-mean value standardization method is used for carrying out normalization transformation processing on the related analysis index parameters.
Further, the mixed signal in the noise data is decomposed using a 0.97 confidence probability assignment function or a mixed sampling function.
Furthermore, the corrected value is a discrimination coefficient or interval, and the corrected value is determined by combining empirical parameters and cavitation test data of the water turbine.
In summary, the invention has the following advantages:
1. the method comprises the steps of extracting feature data of a primary cavitation acoustic signal and a pulse signal of the water turbine, combining to form a feature vector PCSV for judging cavitation of the water turbine, wherein the feature vector PCSV is different from conventional sound elements and is a combination of a brand-new defined sound emission spectrum and a special pulse spectrum, and inputting the feature vector into a neural network for training to obtain a water turbine cavitation bubble sound identification model, so that the identification accuracy and the identification efficiency of the machine on primary cavitation of the water turbine can be improved, and the intelligent judgment accuracy of the cavitation machine by 80% is met;
2. the method adopts a brand-new more accurate physical sound state testing technology to replace the technology based on the traditional physical sound state configuration, avoids the subjectivity and uncertainty of manual testing, realizes the complete replacement of the machine to the manual work, greatly reduces the testing cost, and reduces the cost by more than 50 percent; considerable economic benefits are brought, and digitization and intellectualization of the test technology are realized; and extensibility suitable for other industries.
Drawings
FIG. 1 is a diagram illustrating a set of model runner cavitation time domain noise data collected;
fig. 2 illustrates a cavitation identification determination process according to the present invention.
Detailed Description
In order to more clearly illustrate the present invention, the present invention is further described below with reference to preferred embodiments and the accompanying drawings. It is to be understood by persons skilled in the art that the following detailed description is illustrative and not restrictive, and is not to be taken as limiting the scope of the invention. The terms "first," "second," and the like in the description and claims of the present invention and in the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, or apparatus.
When the cavitation of the water turbine does not occur, the noise signal is mainly low-frequency, the main components of the noise signal are background noises of the water turbine such as water flow sound and electromagnetic noise, the power spectral density spectrum amplitude of the low-frequency component is high, the noise signal has the characteristics of large amplitude and low frequency, and the medium-high frequency band amplitude has the tendency of gradually rising along with the occurrence of the cavitation of the model runner.
As shown in figure 1, the collected noise data of a group of model runner cavitation is analyzed, and the bubble sound state generated on the surface of the runner blade of the water turbine is found to have two characteristics, namely, the bubble sound state is similar to a firecracker sound spectrum, and the bubble sound state is mingled with a special pressure pulsation sound spectrum of the water turbine, namely: the characteristic acoustic state of cavitation is a distinct polymorphic mixed acoustic spectrum comprising an acoustic shot spectrum and a particular pulsation spectrum.
The invention provides a water turbine Cavitation phenomenon intelligent identification method based on combined spectrum feature extraction, which is a Polymorphic Sound intelligent identification method (PSVFR) and helps to realize intelligent judgment of model runner Cavitation phenomenon by a machine by using a newly defined water turbine Cavitation Sound state Vector PCSV (Polymorphic Sound Vector) as a new tool for Cavitation identification.
Specifically, the method comprises the following steps:
the method comprises the following steps of firstly, collecting noise data when a turbine runner model is cavitated;
step two, calculating the statistical characteristics of the noise data:
because the cavitation bubble sound of the water turbine is a mixed sound spectrum of the firecracker-like sound and the special pressure pulsation sound of the water turbine, the bubble sound spectrum in noise data when cavitation of the water turbine is generated is manually (expert) disassembled to obtain related analysis index parameters respectively representing the characteristics of the firecracker-like sound (a first main frequency signal) and the special pressure pulsation sound (a second main frequency signal). The specific operation is as follows:
firstly, decomposing a mixed signal in collected noise data into a plurality of single-wave signals by using a 0.97 confidence probability assignment function or a mixing sampling function, and extracting a first main frequency signal and a second main frequency signal, wherein the two main frequency signals respectively correspond to firecrackers-like sound and special pressure pulsation sound.
Then, the statistical characteristics of the two types of single-wave signals are respectively calculated to obtain related analysis index parameters, wherein the related analysis index parameters comprise a time domain index parameter, a power spectral density index parameter and a frequency domain index parameter.
Wherein the time domain analysis index parameter comprises a peak-to-peak value Vpp; quartile bit frequency probability PVpH; mean value Mean; standard deviation St; kurtosis Ku; skewness Sk; the information entropy H.
The frequency domain analysis index parameters comprise the gravity center frequency PsdFc; the frequency standard deviation PsdRvf; the power spectrum band amplitude mean values PsdH (high frequency), psdM (medium frequency), psdL (low frequency).
Because the frequency spectrum characteristic of the cavitation noise signal of the water turbine model is not obvious sometimes, in order to prevent clutter interference, a power spectral density index (PSD) is introduced as a correction quantity.
Under the classification identification application scene, the frequency domain index parameter, the time domain index parameter and the power spectral density index parameter are taken as part of the feature vector together, so that the types of the feature vector are enriched, and the diagnosis accuracy is improved.
Step three, normalization processing
Considering that min-max normalization, that is, the dispersion normalization method cannot eliminate the influence of the dimension on the variance when in use, the 0-mean normalization method is adopted in the present embodiment to perform the normalization transformation on each index parameter of the noise data. The specific operation is as follows: calculating the Mean value Mean and standard deviation St of each index type by using a formula
= (X-Mean)/St calculates the normalized data for each type of indicator.
Step four, building a cavitation bubble sound identification model of the water turbine
And giving different weights to index parameters of different main frequency signals for weighted average processing, and combining to form a mixed bubble sound characteristic parameter capable of representing the cavitation of the water turbine. Wherein the weight assignment range of the index parameters representing the sound spectrum characteristics of the firecracker-like sound spectrum is 0.6 to 0.8.
These refined mixed bubble tone trait parameters are selectively mapped to different row and column element positions of the matrix (position changeable), forming a profile morphology matrix B, e.g., [ Vpp, PVpH, std, ku, sk, IH, FC ]. The outline form matrix B forms a water turbine cavitation bubble sound qualitative model which can be identified by a machine, namely, forms qualitative factors which can be intelligently identified by the machine.
The cavitation of the water turbine is a real-time dynamic process and has complete randomness, so that the cavitation bubble sound qualitative model of the water turbine cannot be completely in one-to-one correspondence with an object model formed by instantly acquired water turbine noise data, and therefore, in combination with artificial (expert) experience parameters, correction coefficients or intervals (such as 0 or 1, [1.1,2] and the like) are respectively preset for each characteristic parameter (Vpp, PVpH, ku, sk and the like) of mixed bubble sound in the profile form matrix B, and are recombined into a correction profile form matrix D after being multiplied by each PCSV vector in the cavitation bubble sound qualitative model of the water turbine, and the correction profile form matrix D forms a usable cavitation bubble sound identification model of the machine water turbine. The cavitation bubble sound identification model can continuously update the form through intelligent learning, and the preset discrimination coefficient or interval can be continuously adjusted along with the increase of the data volume of the cavitation bubble sound.
Step five, identifying cavitation bubble sound of water turbine
Comparing the possible cavitation bubbles collected immediately with the definite cavitation bubbles. The specific process is as follows:
1) Instant noise data of the turbine runner is collected.
2) Decomposing mixed signals in the instant noise data into a plurality of single-wave signals, respectively calculating the statistical characteristics of the single-wave signals, obtaining relevant analysis index parameters and carrying out normalization processing; the classification and normalization method of the correlation analysis index parameters obtained here are the same as in the previous step two and step three.
3) And mapping each index parameter after normalization processing to different element positions of a row vector and a column vector of the matrix to form an instant feature vector containing the bubble sound main feature ID.
4) Giving different weights to index parameters of different single-wave signals in the instant characteristic vector for weighted average processing, rearranging and combining the obtained new vectors, and finding out an instant contour form matrix A which can represent the main physical sound state of the noise data; the instantaneous outline form matrix A forms a water turbine cavitation bubble sound object model which can be identified by a machine, and forms object elements which can be intelligently identified by the machine.
Presetting a discrimination coefficient or interval (such as 0 or 1, [0.97,3.2] and the like) for each characteristic parameter (Vpp, PVpH, ku, sk and the like) of the immediately collected possible cavitation bubble sound respectively, multiplying the discrimination coefficient or interval by each PCSV characteristic vector in the immediately collected water turbine cavitation bubble sound object model, and recombining to form a new modified contour form matrix C, wherein the modified contour form matrix C forms a water turbine cavitation bubble sound discrimination tool model which can be recognized by a machine. The discrimination tool model can continuously update the pattern through intelligent learning, and the preset discrimination coefficient or interval can also be continuously adjusted along with the difference of the time periods of the cavitation bubble sound data collected in real time.
5) And comparing the corrected contour form matrix C with the corrected contour form matrix D, if the corrected contour form matrix C and the corrected contour form matrix D are equal, outputting a judgment result of cavitation bubble sound identification model of the water turbine as 'cavitation occurrence', and if the corrected contour form matrix C and the corrected contour form matrix D are not equal, outputting a result of 'no cavitation occurrence'.
While the present invention has been described in detail with reference to the illustrated embodiments, it should not be construed as limited to the scope of the present patent. Various modifications and changes may be made by those skilled in the art without inventive work within the scope of the appended claims.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications and equivalent variations of the above embodiments according to the technical spirit of the present invention are included in the scope of the present invention.
Claims (9)
1. An intelligent identification method for cavitation phenomena of a water turbine based on combined spectral feature extraction is characterized by comprising the following steps:
s1, collecting noise data when cavitation occurs to a turbine runner model;
s2, decomposing mixed signals in the noise data into a plurality of single-wave signals, and extracting a first main frequency signal and a second main frequency signal;
s3, calculating the statistical characteristics of the first main frequency signal and the second main frequency signal respectively to obtain related analysis index parameters and carrying out normalization processing;
s4, mapping the index parameters after normalization processing to different element positions of a row vector and a column vector of the matrix to form a PCSV vector containing the cavitation bubble sound main feature ID;
s5, inputting the PCSV vector serving as a feature vector into a neural network for training to obtain a cavitation bubble sound recognition model of the water turbine;
and S6, processing immediately acquired bubble sound data of the water turbine, inputting the processed data into a cavitation bubble sound identification model of the water turbine, and outputting a cavitation identification result.
2. The intelligent identification method for the cavitation phenomenon of the water turbine based on the combined spectral feature extraction is characterized in that the step S5 comprises the following steps:
s51, giving different weights to index parameters of different main frequency signals in the PCSV vector for weighted average processing, and then arranging and combining the obtained new vectors to find a contour form matrix B which can represent the main physical sound state of the noise data;
and S52, multiplying each index parameter in the profile form matrix B by a correction value respectively to obtain a corrected profile form matrix D containing a plurality of new vectors, wherein the corrected profile form matrix D forms a cavitation bubble sound identification model of the water turbine.
3. The method for intelligently identifying the cavitation phenomenon of the water turbine based on the combined spectral feature extraction as claimed in claim 2, wherein the step S6 specifically comprises:
s61, acquiring instant noise data of a turbine runner;
s62, decomposing the mixed signal in the instant noise data into a plurality of single-wave signals, respectively calculating the statistical characteristics of the single-wave signals, obtaining related analysis index parameters and carrying out normalization processing;
s63, mapping each index parameter after normalization processing to different element positions of a row vector and a column vector of the matrix to form an instant feature vector containing the bubble sound main feature ID;
s64, giving different weights to index parameters of different single-wave signals in the instant characteristic vector for weighted average processing, rearranging and combining the obtained new vectors, and finding out an instant contour form matrix A which can represent the main physical sound state of the noise data;
s65, multiplying each index parameter in the instantaneous contour form matrix A by a correction value respectively to obtain a corrected contour form matrix C containing a plurality of new vectors;
and S66, inputting the corrected contour form matrix C into a turbine cavitation bubble sound recognition model trained in advance, and outputting a cavitation judgment result.
4. The intelligent identification method for the cavitation phenomenon of the water turbine based on the combined spectral feature extraction is characterized in that if the corrected profile form matrix C is equal to the corrected profile form matrix D, the output result is that the turbine runner generates cavitation, otherwise, the output result is that the turbine runner does not generate cavitation.
5. The intelligent identification method for the cavitation phenomenon of the water turbine based on the combined spectral feature extraction of the claim 1 or 3 is characterized in that the relevant analysis index parameters comprise a time domain index parameter, a power spectral density index parameter and a frequency domain index parameter; the time domain analysis index parameters comprise a peak value vpp, a quartile bit frequency probability PVpH, a standard deviation St, a kurtosis Ku, a skewness Sk and an information entropy H; the frequency domain indicator parameter includes a center of gravity frequency PsdFc.
6. The method for intelligently identifying the cavitation phenomenon of the water turbine based on the combined spectral feature extraction as claimed in claim 2, wherein the weight assignment range of the index parameter belonging to the first main frequency signal is 0.6 to 0.8.
7. The intelligent identification method for the cavitation phenomenon of the water turbine based on the combined spectral feature extraction is characterized in that a 0-mean normalization method is used for carrying out normalization transformation processing on relevant analysis index parameters.
8. The intelligent identification method for the cavitation phenomenon of the water turbine based on the combined spectral feature extraction is characterized in that a 0.97 confidence probability assignment function or a mixing sampling function is used for decomposing mixing signals in noise data.
9. The method for intelligently identifying the water turbine cavitation phenomenon based on the combined spectral feature extraction according to claim 2 or 3, characterized in that the corrected value is a discrimination coefficient or an interval, and is determined by combining empirical parameters and water turbine cavitation test data.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2022114394605 | 2022-11-17 | ||
| CN202211439460 | 2022-11-17 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115774841A true CN115774841A (en) | 2023-03-10 |
Family
ID=85390507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211507573.4A Pending CN115774841A (en) | 2022-11-17 | 2022-11-29 | Water turbine cavitation phenomenon intelligent identification method based on combined spectrum feature extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115774841A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117476039A (en) * | 2023-12-25 | 2024-01-30 | 西安理工大学 | Acoustic signal-based primary cavitation early warning method for water turbine |
| CN117633520A (en) * | 2024-01-26 | 2024-03-01 | 西安理工大学 | Axial flow turbine cavitation primary detection method based on recursive image |
-
2022
- 2022-11-29 CN CN202211507573.4A patent/CN115774841A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117476039A (en) * | 2023-12-25 | 2024-01-30 | 西安理工大学 | Acoustic signal-based primary cavitation early warning method for water turbine |
| CN117476039B (en) * | 2023-12-25 | 2024-03-08 | 西安理工大学 | Acoustic signal-based primary cavitation early warning method for water turbine |
| CN117633520A (en) * | 2024-01-26 | 2024-03-01 | 西安理工大学 | Axial flow turbine cavitation primary detection method based on recursive image |
| CN117633520B (en) * | 2024-01-26 | 2024-04-05 | 西安理工大学 | Axial flow turbine cavitation primary detection method based on recursive image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115774841A (en) | Water turbine cavitation phenomenon intelligent identification method based on combined spectrum feature extraction | |
| CN111037365B (en) | Cutter state monitoring data set enhancing method based on generative countermeasure network | |
| CN104712542B (en) | A kind of reciprocating compressor sensitive features based on Internet of Things are extracted and method for diagnosing faults | |
| EP3516557A1 (en) | Target class feature model | |
| CN113011661B (en) | Aeration control system for river ecological restoration and control method thereof | |
| CN108535635B (en) | EEMD and HMM based analog circuit intermittent fault diagnosis method | |
| CN111853555A (en) | Water supply pipe network blind leakage identification method based on dynamic process | |
| CN110044623A (en) | The rolling bearing fault intelligent identification Method of empirical mode decomposition residual signal feature | |
| CN107831024B (en) | Fan vibration fault state diagnosis method based on multipoint vibration signal characteristic values | |
| CN108846307B (en) | Microseism and blasting event identification method based on waveform image | |
| CN103558519A (en) | GIS partial discharge ultrasonic signal identification method | |
| CN111508517A (en) | Intelligent micro-motor product control method based on noise characteristics | |
| CN105138770A (en) | Spaceflight product reliability simulation evaluating method based on indirect reliability characteristic quantity | |
| CN114782745A (en) | Ocean sound velocity profile classification method and device based on machine learning | |
| CN103852525B (en) | Acoustic emission signal recognition methods based on AR-HMM | |
| CN115730241A (en) | Construction method of cavitation noise identification model of water turbine | |
| CN110020680B (en) | PMU data classification method based on random matrix theory and fuzzy C-means clustering algorithm | |
| CN114298080A (en) | Hydro-turbo generator set monitoring method based on throw data mining | |
| CN111192257B (en) | Method, system and equipment for determining equipment state | |
| CN116484258A (en) | Elevator traction machine bearing fault diagnosis method | |
| CN113822565B (en) | Method for graded and refined analysis of time-frequency characteristics of fan monitoring data | |
| CN106709522B (en) | High-voltage cable construction defect classification method based on improved fuzzy trigonometric number | |
| CN113449966B (en) | Gypsum board equipment inspection method and system | |
| de Oliveira | Univariate extremes; statistical choice | |
| Shaharudin et al. | Performance analysis and validation of modified singular spectrum analysis based on simulation torrential rainfall data |
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 |