CN116935103A - Abnormality identification method for aluminum alloy impeller - Google Patents
Abnormality identification method for aluminum alloy impeller Download PDFInfo
- Publication number
- CN116935103A CN116935103A CN202310798235.9A CN202310798235A CN116935103A CN 116935103 A CN116935103 A CN 116935103A CN 202310798235 A CN202310798235 A CN 202310798235A CN 116935103 A CN116935103 A CN 116935103A
- Authority
- CN
- China
- Prior art keywords
- impeller
- image
- frequency
- abnormal
- aluminum alloy
- 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 35
- 230000005856 abnormality Effects 0.000 title claims abstract description 24
- 229910000838 Al alloy Inorganic materials 0.000 title claims abstract description 19
- 230000002159 abnormal effect Effects 0.000 claims abstract description 36
- 238000007781 pre-processing Methods 0.000 claims abstract description 27
- 238000001914 filtration Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000006243 chemical reaction Methods 0.000 claims abstract description 11
- 238000003384 imaging method Methods 0.000 claims abstract description 10
- 238000001931 thermography Methods 0.000 claims abstract description 10
- 238000001228 spectrum Methods 0.000 claims description 21
- 238000001514 detection method Methods 0.000 claims description 15
- 238000003709 image segmentation Methods 0.000 claims description 14
- 230000010349 pulsation Effects 0.000 claims description 12
- 230000004044 response Effects 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 5
- 239000012530 fluid Substances 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 239000007787 solid Substances 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 241001089723 Metaphycus omega Species 0.000 claims description 2
- 238000010438 heat treatment Methods 0.000 abstract 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 208000037656 Respiratory Sounds Diseases 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000007788 liquid Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011900 installation process Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention discloses an abnormality identification method of an aluminum alloy impeller, and particularly relates to the technical field of data processing, comprising the following steps of S1, acquiring digital images of equipment by using a holder camera, and respectively acquiring impeller rotating speed and impeller angle data under the working condition of normal operation of the impeller; s2, preprocessing impeller images in video streams intercepted in real time by a computer, wherein the preprocessing comprises image conversion, image enhancement and image filtering processing; s3, imaging information data are displayed after the impeller infrared thermal imaging filtering; s4, performing anomaly identification by capturing nonlinear rotation on the spectrogram. The method comprises the steps of firstly obtaining image data of the impeller, preprocessing the image, and then dividing the image. According to the obtained abnormal pixel points of the image gray level, the abnormal heating area of the impeller is identified, and the abnormal data can be effectively identified by analyzing the power curve under different conditions and the probability curve of the historical operation data of the impeller and combining the time sequence characteristics of the data.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an abnormality identification method for an aluminum alloy impeller.
Background
The utility model provides a pump is mechanical equipment of transport liquid, this kind of equipment is mainly used for transporting liquid, there are many different types of water pumps now, there is an impeller formula pump among them, mainly rely on the effect of rotatory impeller to the fluid, transfer energy, increase fluid kinetic energy, the impeller is the key parts of impeller formula pump, the formation of impeller crack trouble of water pump is mainly influenced by the operating condition of impeller, the internal flow condition of auxiliary feed pump impeller is poor under low flow operating condition, form a large amount of vortex in the runner and strike the impeller surface, if the impeller itself has surface defect, can accelerate impeller fatigue and induce impeller to produce the crackle, because long-time operation under water, the frictional force that the rivers produced when the blade rotates can cause the harm to the impeller, along with the increase of water pump life, the problem just needs to take place to the impeller to corrode or the crackle of the impeller pump, whether the scratch trouble of impeller can take place when the impeller formula pump that has been installed when advancing pump operation, in general, only rely on the experience of the operator or carry out the very difficult judgement of the normal scratch of impeller when advancing impeller during operation, the work can influence the work of the impeller, if the efficiency can not be influenced in time, the service life of machine is reduced.
The existing impeller abnormality identification method is characterized in that strain gauges are additionally arranged in an impeller to detect abnormal conditions of the impeller, when the impeller is identified to generate mechanical deformation under the action of external force, the corresponding change of the resistance value of the impeller is detected, the method is relatively direct, but the vibration rate of the working state of a water pump is high, the strain gauges are stuck in the impeller, the sustainable detection time is relatively short, the impeller is generally circular-arc-shaped blades, the strain gauges are relatively difficult in the installation process, and in addition, the requirement of the impeller on vibration prevention is particularly high, so that the method also brings a plurality of challenges.
Disclosure of Invention
In order to overcome the above-mentioned drawbacks of the prior art, an embodiment of the present invention provides an anomaly identification method for an aluminum alloy impeller, which solves the problems set forth in the above-mentioned background art through image segmentation and fault detection.
In order to achieve the above purpose, the present invention provides the following technical solutions: the method specifically comprises the following steps:
s1, data acquisition: the digital image of the equipment is collected by using a cradle head camera, and impeller rotating speed and impeller angle data under the working condition of normal operation of the impeller are respectively collected;
s2, data preprocessing: preprocessing impeller images in a video stream intercepted by a computer in real time, wherein the preprocessing comprises image conversion, image enhancement and image filtering processing;
s3, image segmentation and fault detection: imaging information data is displayed after the impeller infrared thermal imaging filtering;
s4, identifying abnormal frequency spectrum: and carrying out anomaly identification through nonlinear rotation on the captured spectrogram.
In a preferred embodiment, the data acquisition respectively acquires impeller rotating speed and impeller angle data under the working condition of normal operation of the impeller, the acquisition time interval is 10min, the digital image of the holder camera acquisition equipment is utilized, the acquired image is transmitted into a computer of a monitoring center through a network, and the computer intercepts the impeller image in the video stream in real time.
In a preferred embodiment, the preprocessing of data includes image conversion, image enhancement and image filtering to remove noise, interference and difference contained in the image, image enhancement is implemented by image gray level correction, sharpening and edge enhancement methods, sharpness of abnormal states of the impeller in the image is improved, characteristics including impeller states are highlighted, temperature in the atmosphere influences quality of the infrared thermal image during infrared thermal image acquisition of the impeller, noise is contained in the image, denoising processing is carried out on the infrared thermal image of the impeller, the number of gray level map bits of the acquired infrared thermal image of the impeller is 6, image pixels are 0,255, and the expression of probability density function equation is defined as:
where p (m) represents a probability density function, a represents a black noise point, i.e., a=0, b represents a white noise point, i.e., b=255, p a Representing the probability corresponding to a, P b Representing the probability corresponding to b, using image filtering to order the window range points of the impeller infrared thermal image, the gray value of the image being the center intermediate value of the sequence, and using the equation expression labeled G (i, j) =mean (n (k)), where k represents the number of pixel points, n represents the sequence of gray values, G (i, j) represents the gray value of the image, and L is due to R And L is equal to W Is a component of adaptive median filtering, expressed as:
L R ：R 1 ＝Z med -Z min
R 2 ＝Z med -Z max
wherein Z is min Representing the minimum gray level, Z max Represents the maximum gray level, Z med Represents the median of gray scale, L W The definition is as follows:
L W ：W 1 ＝Z xy -Z min
W 2 ＝Z xy -Z max
wherein Z is xy Representing the gray scale values of coordinates whenR 1 ＞0、R 2 When < 0, it is necessary to switch to L W When R is 1 ＞0、R 2 When < 0, Z is to be added xy Used as output value, whereas the output value is Z med 。
In a preferred embodiment, the image segmentation and fault detection can show imaging information data after the infrared thermal imaging of the impeller is filtered, and according to the obtained infrared thermal image, the gray value of the abnormal area of the impeller device is higher, in order to effectively detect the abnormal area, the image is segmented by adopting a fuzzy mean value method, and is divided into an impeller area and a non-impeller area, and the image segmentation and fault detection comprises the following steps:
s301, inputting subsets with the same gray value into calculation according to the gray value of the image, so that the calculated amount of a data set is reduced, and the segmentation efficiency is improved;
s302, dividing the image histogram by using the same number of pixels as a main rule, and marking the corresponding image gray level as e after dividing, wherein an objective function of the image clustering center is defined as follows:
where T represents the objective function, i, j represents the image pixel gray level, e i 、e j Representing the number of pixels, Q 1 、Q 2 Respectively representing the total number of divided pixels, wherein x represents the maximum value of gray level, alpha and beta represent weights, and when the value of T is the maximum, the original cluster centers of i and j correspond to the gray level;
s303, giving pixel weight to the infrared thermal image of the impeller, weighting the membership degree of the impeller image by using a weighted average method, and marking the membership degree as a mn The formula is described as:wherein m represents the abscissa of the image pixel point, n represents the ordinate, k represents the category, x, y and h represent kernel functions, according to the segmented impeller anomaly image,the gray value of the image is accessed, the abnormal pixel point of the impeller is obtained through mapping, the abnormal area of the impeller is detected, the positioning coordinates of the abnormal area of the impeller are obtained, and the abnormal identification of the impeller is realized.
In a preferred embodiment, the abnormal spectrum identification is carried out by capturing nonlinear rotation on a spectrogram, the monitoring position of the impeller spectrum is selected at a position where the pump body is close to the impeller, when the impeller rotates, the impeller blade top is in contact with the inner wall of the shell, the rigidity of the impeller is increased, the impeller is separated from contact after rebounded by the shell, the rigidity of the impeller is reduced, and transverse free vibration occurs, the rigidity of the impeller changes between contact and non-contact, the changing frequency is the rotation frequency of the impeller, the transverse free vibration of the impeller is overlapped with forced rotation motion, a complex vibration corresponding frequency is generated, the vibration frequency caused by local friction comprises unbalanced rotation speed frequency omega, meanwhile the friction vibration is nonlinear vibration, and therefore 2 omega and 3 omega are contained, some higher harmonics are caused, lower harmonic vibration is also caused, a lower harmonic component omega/n can occur on the impeller, n=2 when the impeller is in heavy friction, n=2, 3 and 4 when the impeller is in light friction, the number of blades is considered, the fault characteristic frequency f of the impeller and the shell are in touch, the fault frequency f=m and the impeller is obviously equal to the frequency of the impeller pump shell, and the frequency of the impeller is obviously different from the vibration, and the frequency of the vibration is obviously different from the vibration frequency of the pump case, and the frequency is compared with the vibration frequency of the pump case, and the frequency of the vibration is obviously different from the vibration frequency of the pump case, and the vibration frequency has a high frequency vibration frequency.
In a preferred embodiment, the device comprises a data acquisition module, a data preprocessing module, an image segmentation and fault detection module and a frequency spectrum abnormality identification module, wherein the digital image of the holder camera acquisition device is utilized, the data preprocessing module preprocesses impeller images in video streams intercepted by a computer in real time, the image preprocessing module comprises image conversion, image enhancement and image filtering processing, the image segmentation and fault detection module shows imaging information data after filtering through impeller infrared thermal imaging, and the frequency spectrum abnormality identification module carries out abnormality identification through nonlinear rotation on a captured spectrogram.
The invention has the technical effects and advantages that:
the method comprises the steps of firstly obtaining image data of the impeller, preprocessing the image, and then dividing the image. According to the method, response frequency spectrums during normal operation of the impeller are collected, the response frequency spectrums during operation are generated in real time through a collecting device which is provided with the impeller frequency spectrums on the impeller, the operation rotating speed of the impeller pump is high, dynamic and static interference effects of fluid and solid overflow parts are obvious, obvious high-frequency impact exists on pressure pulsation of pump shell measuring points of the impeller, main components of the pressure pulsation are high-frequency harmonic waves with frequency conversion and blade passing frequency, and abnormal conditions of the impeller can be identified by comparing the pressure pulsation response frequency spectrums without impeller faults.
Drawings
Fig. 1 is a system flow diagram of a method for identifying anomalies in an aluminum alloy impeller.
Fig. 2 is a system block diagram of an abnormality recognition method of an aluminum alloy impeller.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
The embodiment provides an abnormality identification method for an aluminum alloy impeller as shown in fig. 1, which specifically comprises the following steps:
s1, data acquisition: the digital image of the equipment is collected by using a cradle head camera, and impeller rotating speed and impeller angle data under the working condition of normal operation of the impeller are respectively collected;
s2, data preprocessing: preprocessing impeller images in a video stream intercepted by a computer in real time, wherein the preprocessing comprises image conversion, image enhancement and image filtering processing;
s3, image segmentation and fault detection: imaging information data is displayed after the impeller infrared thermal imaging filtering;
s4, identifying abnormal frequency spectrum: and carrying out anomaly identification through nonlinear rotation on the captured spectrogram.
101. The digital image of the equipment is collected by using a cradle head camera, and impeller rotating speed and impeller angle data under the working condition of normal operation of the impeller are respectively collected;
in this embodiment, specific description is data acquisition, the data acquisition respectively acquires impeller rotation speed and impeller angle data under the working condition when the impeller is in normal operation, the acquisition time interval is 10min, the digital image of the pan-tilt camera acquisition equipment is utilized, the acquired image is transmitted into a computer of a monitoring center through a network, and the computer intercepts the impeller image in the video stream in real time.
102. Preprocessing impeller images in a video stream intercepted by a computer in real time, wherein the preprocessing comprises image conversion, image enhancement and image filtering processing;
in this embodiment, specific needs to be described are data preprocessing, where the data preprocessing performs preprocessing on an impeller image in a video stream intercepted in real time by a computer, including image conversion, image enhancement and image filtering processing, so as to remove noise, interference and differences contained in the image, implement image enhancement by an image gray level correction, sharpening and edge enhancement method, improve the sharpness of an abnormal state of the impeller in the image, highlight a feature containing the impeller state, influence the quality of the infrared thermal image due to the temperature in the atmosphere during the acquisition of the infrared thermal image of the impeller, and cause noise in the image, perform denoising processing on the infrared thermal image of the impeller, and define the acquired impeller infrared thermal image gray level map bit as 6 bits, the image pixel point as 0,255, by a probability density function equation expression as follows:
where p (m) represents a probability density function, a represents a black noise point, i.e., a=0, b represents a white noise point, i.e., b=255, p a Representing the probability corresponding to a, P b Representing the probability corresponding to b, using image filtering to order the window range points of the impeller infrared thermal image, the gray value of the image being the center intermediate value of the sequence, and using the equation expression labeled G (i, j) =mean (n (k)), where k represents the number of pixel points, n represents the sequence of gray values, G (i, j) represents the gray value of the image, and L is due to R And L is equal to W Is a component of adaptive median filtering, expressed as:
L R ：R 1 ＝Z med -Z min
R 2 ＝Z med -Z max
wherein Z is min Representing the minimum gray level, Z max Represents the maximum gray level, Z med Represents the median of gray scale, L w The definition is as follows:
L W ：W 1 ＝Z xy -Z min
W 2 ＝Z xy -Z max
wherein Z is xy Representing the gray scale value of the coordinates, when R 1 ＞0、R 2 When < 0, it is necessary to switch to L W When R is 1 ＞0、R 2 When < 0, Z is to be added xy Used as output value, whereas the output value is Z med 。
103. Imaging information data is displayed after the impeller infrared thermal imaging filtering;
in this embodiment, specific description is to perform image segmentation and fault detection, where the image segmentation and fault detection can show imaging information data after the impeller infrared thermal imaging filtering, and according to the obtained infrared thermal image, the gray value of the abnormal area of the impeller device is higher, and in order to effectively detect the abnormal area, the image is segmented by using a fuzzy mean value method, and is divided into an impeller area and a non-impeller area, where the image segmentation and fault detection includes the following steps:
s301, inputting subsets with the same gray value into calculation according to the gray value of the image, so that the calculated amount of a data set is reduced, and the segmentation efficiency is improved;
s302, dividing the image histogram by using the same number of pixels as a main rule, and marking the corresponding image gray level as e after dividing, wherein an objective function of the image clustering center is defined as follows:
where T represents the objective function, i, j represents the image pixel gray level, e i 、e j Representing the number of pixels, Q 1 、Q 2 Respectively representing the total number of divided pixels, wherein x represents the maximum value of gray level, alpha and beta represent weights, and when the value of T is the maximum, the original cluster centers of i and j correspond to the gray level;
s303, giving pixel weight to the infrared thermal image of the impeller, weighting the membership degree of the impeller image by using a weighted average method, and marking the membership degree as a mn The formula is described as:in the formula, m represents the abscissa of the pixel point of the image, n represents the ordinate, k represents the category, x, y and h represent kernel functions, the gray value of the image is accessed according to the segmented abnormal image of the impeller, the abnormal pixel point of the impeller is obtained through mapping, the abnormal area of the impeller is detected, the positioning coordinates of the abnormal area of the impeller are obtained, and the abnormal identification of the impeller is realized.
104. Performing anomaly identification by capturing nonlinear rotation on the spectrogram;
in the embodiment, the specific description is based on the frequency spectrum anomaly identification, the frequency spectrum anomaly identification is carried out by capturing nonlinear rotation on a spectrogram, the monitoring position of the impeller frequency spectrum is selected at the position of the pump body close to the impeller, when the impeller rotates, the local contact moment of the impeller blade top and the inner wall of the shell occurs, the rigidity of the impeller is increased, the impeller is separated from contact after being rebounded by the shell, the rigidity of the impeller is reduced, the transverse free vibration occurs, the rigidity of the impeller is changed between contact and non-contact, the changed frequency is the rotation frequency of the impeller, the transverse free vibration of the impeller is overlapped with the forced rotation movement, the complex vibration response frequency is generated, the rotation speed frequency omega caused by unbalance is contained in the vibration frequency caused by local friction, meanwhile, the friction vibration is nonlinear vibration, therefore, the pump further comprises 2 omega, 3 omega, and the higher harmonics, low harmonic vibration can be caused, low harmonic components omega/n can appear on a spectrogram, n=2 during heavy friction and n=2, 3,4 during light friction, the fault characteristic frequency f of the collision and grinding of the impeller and the shell is m.omega, m=1, 2, … and Z when the vibration of the pump shell is monitored, wherein Z is the number of the blades, the running rotating speed of the impeller type pump is high, the dynamic and static interference effect of fluid and solid overflow parts is obvious, obvious high-frequency impact exists on pump shell measuring point pressure pulsation of the abnormal impeller, the main components of the pressure pulsation are the higher harmonics of the rotating frequency and the blade passing frequency, and the abnormal condition of the impeller can be identified by comparing the pressure pulsation response frequency spectrum with the pressure pulsation without the impeller fault.
Example 2:
the embodiment of fig. 2 provides an abnormality recognition system for an aluminum alloy impeller, which specifically comprises a data acquisition module, a data preprocessing module, an image segmentation and fault detection module and a frequency spectrum abnormality recognition module, wherein digital images of a holder camera acquisition device are utilized to respectively acquire impeller rotating speed and impeller angle data under the working condition of the impeller in normal operation, the data preprocessing module preprocesses impeller images in video streams intercepted by a computer in real time, the image preprocessing module comprises image conversion, image enhancement and image filtering processing, the image segmentation and fault detection module shows imaging information data after filtering through infrared thermal imaging of the impeller, the frequency spectrum abnormality recognition module carries out nonlinear rotation on a capturing spectrogram, and the monitoring position of the impeller frequency spectrum is selected at a position where a pump body is close to the impeller to carry out abnormality recognition through comparison with a pressure pulsation response frequency spectrum without impeller faults.
The formula in the invention is a formula which is obtained by removing dimension and taking the numerical calculation, and is closest to the actual situation by acquiring a large amount of data and performing software simulation, and the preset proportionality coefficient in the formula is set by a person skilled in the art according to the actual situation or is obtained by simulating the large amount of data.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Finally: the foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (10)
1. An abnormality identification method for an aluminum alloy impeller is characterized by comprising the following steps of: the method specifically comprises the following steps:
s1, data acquisition: the digital image of the equipment is collected by using a cradle head camera, and impeller rotating speed and impeller angle data under the working condition of normal operation of the impeller are respectively collected;
s2, data preprocessing: preprocessing impeller images in a video stream intercepted by a computer in real time, wherein the preprocessing comprises image conversion, image enhancement and image filtering processing;
s3, image segmentation and fault detection: imaging information data is displayed after the impeller infrared thermal imaging filtering;
s4, identifying abnormal frequency spectrum: and carrying out anomaly identification through nonlinear rotation on the captured spectrogram.
2. The abnormality identification method of an aluminum alloy impeller according to claim 1, characterized in that: and step S1, data acquisition is carried out to acquire impeller rotating speed and impeller angle data under the working condition of normal operation of the impeller respectively, the acquisition time interval is 10min, the digital image of the cradle head camera acquisition equipment is utilized, the acquired image is transmitted into a computer of a monitoring center through a network, and the computer intercepts impeller images in the video stream in real time.
3. The abnormality identification method of an aluminum alloy impeller according to claim 1, characterized in that: and step S2, preprocessing impeller images in the video stream intercepted in real time by data preprocessing, including image conversion, image enhancement and image filtering processing, so as to remove noise, interference and difference contained in the images, and realizing image enhancement by image gray level correction, sharpening and edge enhancement methods, thereby improving the definition of abnormal states of the impellers in the images and highlighting the characteristics containing the impeller states.
4. The abnormality identification method for an aluminum alloy impeller according to claim 3, characterized in that: the temperature in the atmosphere environment can influence the quality of an infrared thermal image, noise is contained in the image, the infrared thermal image of the impeller is subjected to denoising treatment, the number of gray level figures of the collected impeller infrared thermal image is 6, the pixel point of the image is 0,255, and the expression of the probability density function equation is defined as:
where p (m) represents a probability density function, a represents a black noise point, i.e., a=0, b represents a white noise point, i.e., b=255, p a Representing the probability corresponding to a, P b Representing the probability corresponding to b, sorting the range points of the infrared thermal image window of the impeller by using image filtering, wherein the gray value of the image is the central intermediate value of the sequence, and using an equationThe expression is denoted as G (i, j) =medium (n (k)), where k represents the number of pixel points, n represents a sequence of gray values, and G (i, j) represents a gray value of an image.
5. The abnormality identification method for an aluminum alloy impeller according to claim 3, characterized in that: said due to L R And L is equal to W Is a component of adaptive median filtering, expressed as:
L R ：R 1 ＝Z med -Z min
R 2 ＝Z med -Z max
wherein Z is min Representing the minimum gray level, Z max Represents the maximum gray level, Z med Represents the median of gray scale, L W The definition is as follows:
L W ：W 1 ＝Z xy -Z min
W 2 ＝Z xy -Z max
wherein Z is xy Representing the gray scale value of the coordinates, when R 1 >0、R 2 <At 0, it is necessary to switch to L W When R is 1 >0、R 2 <At 0, Z is to be xy Used as output value, whereas the output value is Z med 。
6. The abnormality identification method of an aluminum alloy impeller according to claim 1, characterized in that: in the step S3, imaging information data can be displayed after the image segmentation and fault detection are carried out through impeller infrared thermal imaging filtration, the gray value of an abnormal area of impeller equipment is higher according to the obtained infrared thermal image, in order to effectively detect the abnormal area, the image is segmented by adopting a fuzzy mean value method, the image is divided into an impeller area and a non-impeller area, a subset with the same gray value is input into calculation according to the gray value of the image, and the calculated amount of a data set is reduced.
7. The abnormality identification method for an aluminum alloy impeller according to claim 6, characterized in that: dividing the image histogram by using the same pixel number as a main part, marking the corresponding image gray level as e after dividing, and defining an objective function of the image clustering center as follows:
where T represents the objective function, i, j represents the image pixel gray level, e i 、e j Representing the number of pixels, Q 1 、Q 2 Respectively representing the total number of divided pixels, x represents the maximum value of gray level, and alpha and beta represent weights, when the value of T is the maximum, the original cluster centers of i and j are made to correspond to the gray level.
8. The abnormality identification method for an aluminum alloy impeller according to claim 6, characterized in that: the infrared thermal image of the impeller is endowed with pixel weight, the weighted average method is used for weighting the membership degree of the impeller image, and the membership degree is marked as a mn The formula is described as:in the formula, m represents the abscissa of the pixel point of the image, n represents the ordinate, k represents the category, x, y and h represent kernel functions, the gray value of the image is accessed according to the segmented abnormal image of the impeller, the abnormal pixel point of the impeller is obtained through mapping, the abnormal area of the impeller is detected, the positioning coordinates of the abnormal area of the impeller are obtained, and the abnormal identification of the impeller is realized.
9. The abnormality identification method of an aluminum alloy impeller according to claim 1, characterized in that: in the step S4, the frequency spectrum anomaly identification is carried out by capturing nonlinear rotation on a spectrogram, the monitoring position of the impeller frequency spectrum is selected at a position where the pump body is close to the impeller, when the impeller rotates, the impeller blade tip part is in contact with the inner wall of the shell, the rigidity of the impeller is increased, the impeller is separated from contact after being rebounded by the shell, the rigidity of the impeller is reduced, transverse free vibration occurs, the rigidity of the impeller is changed between contact and non-contact, the changing frequency is the rotation frequency of the impeller, the transverse free vibration of the impeller is overlapped with forced rotation movement, a complex vibration response frequency is generated, the rotation frequency omega caused by unbalance is contained in the vibration frequency caused by local friction, meanwhile, the friction vibration is nonlinear vibration, and therefore 2 omega, 3 omega, higher harmonics are also caused, lower harmonic vibration is generated, lower harmonic component omega/n is generated on the spectrogram, n=2, 3,4 when in heavy friction.
10. The abnormality identification method of an aluminum alloy impeller according to claim 1, characterized in that: when the pump shell is monitored to vibrate, the fault characteristic frequency f of the impeller and the shell in collision and grinding is m.omega, m=1, 2, … and Z, wherein Z is the number of blades, the running rotating speed of the impeller pump is high, the dynamic and static interference effect of fluid and solid flow-through parts is obvious, the pressure pulsation of the measuring point of the pump shell with abnormal impeller has obvious high-frequency impact, the main components of the pressure pulsation are high-frequency harmonic waves with the frequency of rotation and the passing frequency of the blades, and the abnormal condition of the impeller can be identified by comparing the frequency of the response of the pressure pulsation with the pressure pulsation without impeller faults.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310798235.9A CN116935103A (en) | 2023-07-03 | 2023-07-03 | Abnormality identification method for aluminum alloy impeller |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310798235.9A CN116935103A (en) | 2023-07-03 | 2023-07-03 | Abnormality identification method for aluminum alloy impeller |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116935103A true CN116935103A (en) | 2023-10-24 |
Family
ID=88379850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310798235.9A Pending CN116935103A (en) | 2023-07-03 | 2023-07-03 | Abnormality identification method for aluminum alloy impeller |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116935103A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107564008A (en) * | 2017-08-02 | 2018-01-09 | 西安电子科技大学 | Rapid SAR image segmentation method based on crucial pixel fuzzy clustering |
| CN109685786A (en) * | 2018-12-20 | 2019-04-26 | 龙口味美思环保科技有限公司 | A kind of non-destructive testing of birds, beasts and eggs face crack and automatic grading method |
| JP2020042519A (en) * | 2018-09-10 | 2020-03-19 | 沖電気工業株式会社 | Abnormality detection device, abnormality detection method, and abnormality detection program |
| CN111044277A (en) * | 2019-12-31 | 2020-04-21 | 苏州欣皓信息技术有限公司 | Fault diagnosis system and method for pump station unit |
| CN111539935A (en) * | 2020-04-24 | 2020-08-14 | 江苏大学 | Online cable surface defect detection method based on machine vision |
| CN112446006A (en) * | 2020-09-27 | 2021-03-05 | 国网山西省电力公司电力科学研究院 | Scale parameter adjustable morphological filtering method for non-linear rotation signal of gyroscope of unmanned aerial vehicle |
| CN113482945A (en) * | 2021-06-29 | 2021-10-08 | 中电华创电力技术研究有限公司 | Fan vibration fault diagnosis method and device based on vibration characteristic value |
| CN114119503A (en) * | 2021-11-05 | 2022-03-01 | 华南师范大学 | Retina blood vessel segmentation method and device based on IFCM clustering |
| CN114542402A (en) * | 2022-03-17 | 2022-05-27 | 西安热工研究院有限公司 | Wind power blade fault type online diagnosis method and system based on multi-parameter analysis |
| CN114576152A (en) * | 2020-12-01 | 2022-06-03 | 格兰富控股联合股份公司 | Water pump state monitoring system, monitoring method, device, electronic equipment and medium |
| CN115263644A (en) * | 2022-06-15 | 2022-11-01 | 哈尔滨电机厂有限责任公司 | Intelligent early warning method for faults of top cover of water turbine |
| CN115656319A (en) * | 2022-10-09 | 2023-01-31 | 辽宁红沿河核电有限公司 | Method and system for monitoring cracks of water pump impeller based on vibration signals |
-
2023
- 2023-07-03 CN CN202310798235.9A patent/CN116935103A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107564008A (en) * | 2017-08-02 | 2018-01-09 | 西安电子科技大学 | Rapid SAR image segmentation method based on crucial pixel fuzzy clustering |
| JP2020042519A (en) * | 2018-09-10 | 2020-03-19 | 沖電気工業株式会社 | Abnormality detection device, abnormality detection method, and abnormality detection program |
| CN109685786A (en) * | 2018-12-20 | 2019-04-26 | 龙口味美思环保科技有限公司 | A kind of non-destructive testing of birds, beasts and eggs face crack and automatic grading method |
| CN111044277A (en) * | 2019-12-31 | 2020-04-21 | 苏州欣皓信息技术有限公司 | Fault diagnosis system and method for pump station unit |
| CN111539935A (en) * | 2020-04-24 | 2020-08-14 | 江苏大学 | Online cable surface defect detection method based on machine vision |
| CN112446006A (en) * | 2020-09-27 | 2021-03-05 | 国网山西省电力公司电力科学研究院 | Scale parameter adjustable morphological filtering method for non-linear rotation signal of gyroscope of unmanned aerial vehicle |
| CN114576152A (en) * | 2020-12-01 | 2022-06-03 | 格兰富控股联合股份公司 | Water pump state monitoring system, monitoring method, device, electronic equipment and medium |
| CN113482945A (en) * | 2021-06-29 | 2021-10-08 | 中电华创电力技术研究有限公司 | Fan vibration fault diagnosis method and device based on vibration characteristic value |
| CN114119503A (en) * | 2021-11-05 | 2022-03-01 | 华南师范大学 | Retina blood vessel segmentation method and device based on IFCM clustering |
| CN114542402A (en) * | 2022-03-17 | 2022-05-27 | 西安热工研究院有限公司 | Wind power blade fault type online diagnosis method and system based on multi-parameter analysis |
| CN115263644A (en) * | 2022-06-15 | 2022-11-01 | 哈尔滨电机厂有限责任公司 | Intelligent early warning method for faults of top cover of water turbine |
| CN115656319A (en) * | 2022-10-09 | 2023-01-31 | 辽宁红沿河核电有限公司 | Method and system for monitoring cracks of water pump impeller based on vibration signals |
Non-Patent Citations (3)
| Title |
|---|
| 张瑞强等: "基于红外热成像检测技术的变电设备异常发热故障检测", 《制造业自动化》, vol. 44, no. 09, 25 September 2022 (2022-09-25), pages 0 - 3 * |
| 曲佳;兴成宏;郭淑萍;赵黎辉;: "精确诊断烟机机组动静件摩擦故障", 风机技术, no. 05, 26 October 2009 (2009-10-26), pages 77 - 79 * |
| 郑祥豪: "可逆式水泵水轮机运行状态监测与智能故障诊断研究", 《中国博士学位论文全文数据库_工程科技Ⅱ辑》, 15 March 2023 (2023-03-15), pages 2 - 2 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110148130B (en) | Method and device for detecting part defects | |
| Peng et al. | Study of tool wear monitoring using machine vision | |
| Saufi et al. | Low-speed bearing fault diagnosis based on ArSSAE model using acoustic emission and vibration signals | |
| WO2019119659A1 (en) | Method and equipment for monitoring vortex-induced vibration for wind turbine generator set | |
| CN112487890B (en) | Bearing acoustic signal fault diagnosis method based on parallel sparse filtering | |
| CN108919116B (en) | Ocean current generator unbalanced stator current fault diagnosis method based on MCCKAF-FFT-Softmax | |
| Zhang et al. | Mask-MRNet: A deep neural network for wind turbine blade fault detection | |
| Ye et al. | Multiscale weighted morphological network based feature learning of vibration signals for machinery fault diagnosis | |
| Sun et al. | Curvature enhanced bearing fault diagnosis method using 2D vibration signal | |
| CN116257747A (en) | Early abnormality detection method for rotor system of fusion kurtosis and convolution self-encoder | |
| CN116524205A (en) | Sewage aeration automatic detection and identification method | |
| CN109612726A (en) | A kind of multiple superstage analysis method extracted for vibration signal characteristics | |
| CN110674991A (en) | OCSVM (online charging management system VM) -based method for detecting abnormality of primary fan of thermal power plant | |
| CN116823725A (en) | Aeroengine blade surface defect detection method based on support vector machine | |
| CN116935103A (en) | Abnormality identification method for aluminum alloy impeller | |
| Zhang et al. | An image processing approach to machine fault diagnosis based on visual words representation | |
| CN117093938A (en) | Fan bearing fault detection method and system based on deep learning | |
| CN113920547A (en) | Glove detection method and system based on neural network | |
| CN112291193B (en) | LDoS attack detection method based on NCS-SVM | |
| CN110766707B (en) | Cavitation bubble image processing method based on multi-operator fusion edge detection technology | |
| CN111145109B (en) | Wind power generation power curve abnormal data identification and cleaning method based on image | |
| CN110533644B (en) | Insulator detection method based on image recognition | |
| CN114022418A (en) | Lithium battery pole piece defect detection method and system | |
| CN115526867B (en) | Method and device for extracting crack defects of X-ray image of carbon fiber composite gas cylinder | |
| CN110309966B (en) | Dredging energy consumption prediction method based on partial least square method |
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 |