CN118067727B - Rotor welding defect assessment method - Google Patents
Rotor welding defect assessment method Download PDFInfo
- Publication number
- CN118067727B CN118067727B CN202410472963.5A CN202410472963A CN118067727B CN 118067727 B CN118067727 B CN 118067727B CN 202410472963 A CN202410472963 A CN 202410472963A CN 118067727 B CN118067727 B CN 118067727B
- Authority
- CN
- China
- Prior art keywords
- welding
- value
- rotor
- defect
- preset
- 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.)
- Active
Links
- 238000003466 welding Methods 0.000 title claims abstract description 448
- 230000007547 defect Effects 0.000 title claims abstract description 160
- 238000000034 method Methods 0.000 title claims description 55
- 238000001514 detection method Methods 0.000 claims abstract description 90
- 238000011156 evaluation Methods 0.000 claims abstract description 55
- 230000002159 abnormal effect Effects 0.000 claims abstract description 26
- 238000012545 processing Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims description 31
- 238000004458 analytical method Methods 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 21
- 230000007613 environmental effect Effects 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 230000005856 abnormality Effects 0.000 claims description 8
- 238000003709 image segmentation Methods 0.000 claims description 7
- 239000000779 smoke Substances 0.000 claims description 7
- 230000032683 aging Effects 0.000 claims description 6
- 238000013135 deep learning Methods 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 6
- 238000012797 qualification Methods 0.000 claims description 6
- 230000008439 repair process Effects 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 3
- 230000006872 improvement Effects 0.000 abstract description 6
- 238000007781 pre-processing Methods 0.000 abstract description 3
- 238000012549 training Methods 0.000 description 6
- 230000000750 progressive effect Effects 0.000 description 5
- 230000002708 enhancing effect Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000005672 electromagnetic field Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 239000011148 porous material Substances 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- General Factory Administration (AREA)
Abstract
The invention belongs to the technical field of rotor quality detection, in particular to a rotor welding defect evaluation method, which comprises the steps of image acquisition, image processing, defect identification, defect evaluation and evaluation report output; according to the invention, the image acquisition module is used for acquiring the image information of the welding part of the rotor, the image processing module is used for preprocessing the received image, the defect identification module is used for carrying out defect identification on the preprocessed image input model, the defect assessment module is used for carrying out automatic assessment on the defect according to the identified defect information, so that the automatic detection, identification and assessment on the welding defect of the rotor are realized, the welding quality performance of welding equipment in the detection period is comprehensively assessed through the welding quality comprehensive assessment module, potential influencing factors are rapidly judged through judging and analyzing one by abnormal reasons when welding disqualified signals are generated, corresponding improvement measures are carried out by management personnel in a targeted manner, the subsequent welding quality of the rotor is ensured, and the intelligent degree is high.
Description
Technical Field
The invention relates to the technical field of rotor quality detection, in particular to a rotor welding defect assessment method.
Background
The rotor is mainly one of mechanical rotating parts, is a rotating body supported by a bearing, rotates along with the change of acting force such as an electromagnetic field or a hydraulic field, and is an indispensable part in various mechanical equipment such as a motor, a generator, a water turbine, a wind driven generator and the like; the rotor is used as a key component of various rotary machines, and the welding quality of the rotor is directly related to the safe operation and performance stability of mechanical equipment;
The existing assessment of the welding defects of the rotor mainly depends on manual visual inspection, has the problems of strong subjectivity, low efficiency and insufficient detection accuracy, cannot reasonably assess the welding quality condition of the welded rotor in a certain period, is difficult to quickly judge potential influencing factors causing the poor welding quality of the rotor when the welding quality of the rotor is poor in a corresponding period is assessed, is not beneficial to a manager to pertinently make corresponding improvement measures, and has low intelligent degree;
in view of the above technical drawbacks, a solution is now proposed.
Disclosure of Invention
The invention aims to provide a rotor welding defect assessment method, which solves the problems that the assessment of the rotor welding defect in the prior art has strong subjectivity, low efficiency and insufficient detection accuracy, and the welding quality condition of a welded rotor in a certain period cannot be reasonably assessed, and the potential influence factors causing the poor welding quality of the rotor are difficult to quickly judge when the welding quality of the rotor in a corresponding period is assessed, so that the intelligent degree is low.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a method for evaluating welding defects of a rotor, comprising the steps of:
step one, acquiring image information of a rotor welding part through an image acquisition module, and transmitting the acquired image to an image processing module through a server;
The image processing module pre-processes the received image, extracts characteristic information of a rotor welding area by utilizing an image segmentation technology, and sends the pre-processed image to the defect recognition module through the server;
thirdly, a defect recognition module builds a welding defect recognition model based on a deep learning algorithm, a large amount of sample data are trained to enable the model to have the capability of recognizing various welding defects, the preprocessed image is input into the model to conduct defect recognition, and recognized rotor defect information is sent to a defect evaluation module through a server;
step four, the defect evaluation module automatically evaluates the defects according to the identified defect information and combining the welding process requirements and standards, judges whether the defects meet the quality requirements, generates a detailed evaluation report, and sends the evaluation report to the control output module through the server;
And fifthly, controlling the output module to output the assessment report through a display screen or a printer.
Further, the specific operation process of the fourth step is as follows:
Receiving defect information output by a defect identification module, wherein the defect information comprises the type, the position and the size of the defect, and the corresponding defect information is directly used for judging the severity degree and the influence range of the defect;
Comparing and analyzing the received defect information according to preset welding process requirements and standards; the preset welding process requirements and standards generally comprise various indexes of welding quality, and by comparing the standards with actually detected defect information, whether the corresponding defects exceed the allowable range or not is judged, and whether the corresponding defects meet the quality requirements or not is determined.
Further, in step four, the generated assessment report includes the following:
defect image: the assessment report contains actually detected defect images, and the corresponding defect images intuitively show the appearance of the defects;
Defect type: the type of defect including cracks, pinholes and unfused is explicitly indicated in the assessment report;
Defect location: the specific positions of the defects on the welding pieces are marked in detail in the assessment report;
assessment results: the final evaluation result is recorded in the evaluation report, and whether the defect meets the quality requirement is determined; and for unqualified weldments, corresponding repair suggestions or treatment measures are given in the assessment report.
Further, the server is in communication connection with the welding quality comprehensive evaluation module, the welding quality comprehensive evaluation module obtains welding equipment for rotor welding, the welding quality performance of the welding equipment in a detection period is comprehensively evaluated, welding disqualification signals or welding qualification signals of the corresponding welding equipment are generated through analysis, the welding disqualification signals are sent to the control output module through the server, and the control output module outputs the welding disqualification signals through the display screen.
Further, the specific operation process of the welding quality comprehensive evaluation module comprises the following steps:
Setting a detection period, collecting an evaluation report of the defect quality of a rotor welded by welding equipment in the detection period, marking the number of the rotors with defects meeting the quality requirement as a feasible rotor analysis value, marking the number of the rotors with defects not meeting the quality requirement as a different rotor analysis value, calculating the ratio of the different rotor analysis value to the feasible rotor analysis value to obtain a different rotor detection condition value, comparing the different rotor detection condition value with a preset different rotor detection condition threshold value, and generating a welding failure signal of the corresponding welding equipment if the different rotor detection condition value exceeds the preset different rotor detection condition threshold value; if the abnormal rotor detection condition value does not exceed the preset abnormal rotor detection condition threshold value, generating a welding qualification signal of the corresponding welding equipment.
Further, the server is in communication connection with the abnormal cause judging module, the server sends welding failure signals of corresponding welding equipment to the abnormal cause judging module, when the abnormal cause judging module receives the welding failure signals, the abnormal cause judging module judges and analyzes the welding failure signals one by one through the abnormal cause to generate welding efficiency early warning signals, welding environment early warning signals, welding operation early warning signals or welding equipment early warning signals, and sends the welding efficiency early warning signals, the welding environment early warning signals, the welding operation early warning signals or the welding equipment early warning signals to the control output module through the server, and the control output module outputs the corresponding early warning signals through the display screen.
Further, the specific analysis process of the abnormality cause one by one judgment analysis is as follows:
Acquiring the starting time and the ending time of corresponding welding equipment in rotor welding, marking the interval time between the ending time and the starting time as a rotor welding value, marking the deviation value of the rotor welding value and a preset proper rotor welding standard value as a rotor time offset value, carrying out average calculation on all rotor time offsets of corresponding welding equipment in a detection period to obtain a rotor aging value, and marking the number occupation ratio of the rotor time offsets exceeding a preset rotor time offset threshold in the detection period as a rotor super occupation value;
and carrying out numerical calculation on the rotor aging value and the rotor overexposure value to obtain a welding efficiency early warning value, carrying out numerical comparison on the welding efficiency early warning value and a preset welding efficiency early warning threshold value, and generating a welding efficiency early warning signal if the welding efficiency early warning value exceeds the preset welding efficiency early warning threshold value.
Further, if the welding efficiency early warning value does not exceed a preset welding efficiency early warning threshold value, acquiring environmental temperature data, environmental humidity data and environmental smoke data of the environment where the rotor is positioned in the welding process, marking a deviation value of the environmental temperature data and a preset proper welding temperature value as a welding temperature detection value, acquiring a welding wet detection value in a similar way, and carrying out numerical calculation on the welding temperature detection value, the welding wet detection value and the environmental smoke data to obtain a welding ring condition value;
Performing variance calculation and mean calculation on all welding ring condition values of corresponding environments in a detection period to obtain a welding ring deviation value and a welding ring table value, respectively performing numerical comparison on the welding ring deviation value and the welding ring table value and a preset welding ring deviation threshold value and a preset welding ring table threshold value, and generating a welding environment early warning signal if the welding ring deviation value or the welding ring table value exceeds the corresponding preset threshold value;
If the welding ring deviation value and the welding ring table value do not exceed the corresponding preset threshold values, comparing the welding ring condition value with the preset welding ring condition threshold values, and if the welding ring condition value exceeds the preset welding ring Kuang Yuzhi, judging that the corresponding environment is in a welding damage state; the method comprises the steps of obtaining total duration of a welding damage state of a corresponding environment in a detection period, marking the total duration as a welding damage time condition value, carrying out numerical calculation on the welding damage time condition value and a welding ring table value to obtain a welding ring evaluation value, carrying out numerical comparison on the welding ring evaluation value and a preset welding ring evaluation threshold value, and generating a welding environment early warning signal if the welding ring evaluation value exceeds the preset welding ring evaluation threshold value.
Further, if the welding criticizing value does not exceed the preset welding criticizing threshold, a welding operation value is obtained through analysis, the welding operation value is compared with the preset welding operation threshold in value, and if the welding operation value exceeds the preset welding operation threshold, a welding operation early warning signal is generated; and if the welding operation value does not exceed the preset welding operation threshold value, generating a welding equipment early warning signal.
Further, the method for analyzing and acquiring the welding operation value comprises the following steps:
Acquiring operators performing rotor welding operation through corresponding welding equipment in a detection period, acquiring the number of times of wrong operation when the operators perform rotor welding operation in the detection period, marking the number of times of wrong operation as a welding wrong table value, acquiring the duration time of each wrong operation, marking the duration time as a welding wrong operation value, summing all the welding wrong operation values in the detection period to obtain the welding wrong value, and marking the number of the welding wrong operation values exceeding a preset welding wrong operation time threshold as a wrong operation high value; and carrying out numerical calculation on the welding error table value, the welding error time value and the error operation high value to obtain a welding operation value.
Compared with the prior art, the invention has the beneficial effects that:
1. In the invention, the image information of the welding part of the rotor is acquired through the image acquisition module, the image processing module preprocesses the received image and extracts the characteristic information of the welding area of the rotor, the defect recognition module inputs the preprocessed image into the model for defect recognition, the defect evaluation module automatically evaluates the defect according to the recognized defect information and generates a detailed evaluation report, and the control output module outputs the evaluation report through the display screen or the printer, so that the automatic detection, recognition and evaluation of the welding defect of the rotor are realized, the detection efficiency and precision are improved, and the influence of human factors is reduced;
2. According to the invention, the welding quality performance of the welding equipment in the detection period is comprehensively evaluated through the welding quality comprehensive evaluation module, the welding disqualification signals or the welding qualification signals of the corresponding welding equipment are generated through analysis, and potential influencing factors causing poor welding quality of the rotor are rapidly judged through judging and analyzing one by abnormal reasons when the welding disqualification signals are generated, so that the management personnel can make corresponding improvement measures in a targeted manner, the subsequent welding quality of the rotor is ensured, and the intelligent degree is high.
Drawings
For the convenience of those skilled in the art, the present invention will be further described with reference to the accompanying drawings;
FIG. 1 is a flow chart of a method according to a first embodiment of the invention;
FIG. 2 is a system block diagram of a first embodiment of the present invention;
Fig. 3 is a system block diagram of the second and third embodiments of the present invention.
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.
Embodiment one: as shown in fig. 1-2, the method for evaluating the welding defect of the rotor provided by the invention comprises the following steps:
The method comprises the steps of firstly, acquiring image information of a welding part of a rotor through an image acquisition module, and sending the acquired image to an image processing module through a server, wherein the image acquisition module comprises a high-resolution camera and a light source device, the high-resolution camera is used for acquiring the image of the rotor, the light source device is used for providing a light source to ensure the definition of an acquired revolving image, and the camera and the light source device are accurately positioned and adjusted through a mechanical arm to ensure the accuracy and consistency of image acquisition.
The image processing module carries out preprocessing on the received image, including denoising (denoising the image through a filter and other algorithms, eliminating or reducing noise interference), enhancing contrast (enhancing the contrast of the image through histogram equalization, contrast stretching and other methods, enabling the characteristics of a welding area to be more prominent, facilitating subsequent image segmentation and characteristic extraction) and the like, so as to improve the image quality; and extracting characteristic information of the rotor welding area by using an image segmentation technology, and sending the preprocessed image to the defect recognition module through the server.
It should be noted that, image segmentation is a process of dividing an image into a plurality of mutually disjoint regions, each region having similar characteristics therein, and there being a significant difference between the different regions; in weld defect detection, it is often necessary to separate the weld area from the background or other non-weld areas in order to further analyze the characteristics of the weld area; in order to realize accurate image segmentation, various algorithms such as threshold segmentation, edge detection, region growth and the like can be adopted, and the algorithms divide the image into different regions according to the characteristics of gray scale, color, texture and the like of the image. In the invention, a proper segmentation algorithm is selected according to the characteristics of the welding area, so that the accurate extraction of the welding area is ensured.
After the welding area is extracted by an image segmentation technology, the characteristic information of the area can be further extracted; such feature information includes shape, size, gray scale distribution, texture, etc. of the welded area, which are of great significance for subsequent defect identification, and the accuracy and integrity of feature extraction will directly affect the effect and accuracy of defect identification.
Thirdly, a defect recognition module builds a welding defect recognition model based on a deep learning algorithm, a large amount of sample data are trained to enable the model to have the capability of recognizing various welding defects, the preprocessed image is input into the model to conduct defect recognition, and recognized rotor defect information is sent to a defect evaluation module through a server.
The defect recognition module is a core part of a rotor welding defect evaluation system, and constructs a welding defect recognition model based on a deep learning algorithm to realize automatic detection and recognition of defects of a welding area; the core of the defect recognition module is the construction of a deep learning model, and a deep learning algorithm can learn characteristic representations from a large amount of data and automatically extract information useful for tasks; in the module, deep learning models such as Convolutional Neural Network (CNN) are adopted to construct welding defect recognition models, and the models gradually extract abstract features of images through multi-layer convolution, pooling and full-connection operation, so that the recognition of welding defects is realized.
In order to train a model with strong recognition capability, a large amount of sample data are prepared for training, the sample data comprise images of normal welding areas and images of welding areas with various defects, and the model can be ensured to have wide applicability and robustness by collecting sample data with different welding processes, different materials and different defect types; and in the training process, the preprocessed image is input into the model, parameters of the model are optimized through a back propagation algorithm, and the model gradually learns the characteristic representation of the welding defect through repeated iterative training, so that the defect can be accurately identified in a new image.
Once the model training is completed, it can be applied to the actual defect identification task; in practical application, firstly, an image processed by an image processing module is input into a defect recognition model, and the model automatically extracts characteristic information in the image and compares the characteristic information with defect characteristics learned in a training process. The model can judge whether the image has defects or not by calculating the similarity or the difference between the features, and output the information such as the type, the position, the size and the like of the defects; the output result of the defect identification module has important significance for subsequent defect evaluation and repair work; the type and the position of the defect are accurately identified, so that the welding quality can be objectively evaluated, corresponding repair measures can be timely taken, and meanwhile, the defect identification module can also provide data support for the optimization of the welding process, and the welding quality and the production efficiency can be improved.
Step four, the defect evaluation module automatically evaluates the defects according to the identified defect information and combining the welding process requirements and standards, judges whether the defects meet the quality requirements, generates a detailed evaluation report, and sends the evaluation report to the control output module through the server; the specific operation process is as follows:
The defect information output by the defect identification module is received, wherein the defect information comprises key parameters such as the type, the position and the size of the defect, and the information is the basis of subsequent assessment work and is directly used for judging the severity degree and the influence range of the defect; comparing and analyzing the received defect information according to preset welding process requirements and standards; among them, the preset welding process requirements and standards generally include various indexes of welding quality, such as uniformity of the weld, no cracks, no pores, etc. By comparing these criteria with the actually detected defect information, it is determined whether the corresponding defect is out of the allowable range and whether it meets the quality requirement.
Further, the generated assessment report includes the following: defect image: the assessment report contains the actually detected defect image, and the corresponding defect image intuitively displays the appearance of the defect, so that the user can understand and analyze the defect image conveniently; defect type: the type of defect is explicitly indicated in the assessment report, including cracks, pinholes, unfused, etc., which helps the user to understand the nature and possible cause of the defect; defect location: the specific positions of the defects on the welding pieces are marked in detail in the assessment report, so that the method has important guiding significance for subsequent repair work; assessment results: the final assessment result is recorded in the assessment report, namely whether the defect meets the quality requirement is determined; and for unqualified weldments, corresponding repair suggestions or treatment measures are given in the assessment report.
And fifthly, the control output module outputs the assessment report through a display screen or a printer, so that the assessment report is convenient for a user to check and analyze, and the control output module can also control and manage the whole system, including functions of starting, stopping, parameter setting and the like.
Embodiment two: as shown in fig. 3, the difference between the present embodiment and embodiment 1 is that the server is communicatively connected with the welding quality comprehensive evaluation module, the welding quality comprehensive evaluation module obtains the welding equipment for performing rotor welding, performs comprehensive evaluation on the welding quality performance of the welding equipment in the detection period, generates a welding failure signal or welding failure signal of the corresponding welding equipment through analysis, and sends the welding failure signal to the control output module through the server, and the control output module outputs the welding failure signal through the display screen, so that the manager can grasp the welding performance condition of the corresponding welding equipment in detail, and make corresponding improvement measures in time as required, thereby helping to ensure the quality of the rotor welded subsequently; the specific operation process of the welding quality comprehensive evaluation module is as follows:
Setting a detection period, preferably twenty-four hours; collecting an evaluation report of the defect quality of a rotor welded by welding equipment in a detection period, marking the number of the rotors with defects meeting quality requirements as a feasible rotor analysis value (namely, the number of the rotors with better welding quality), marking the number of the rotors with defects not meeting quality requirements as a dissimilar rotor analysis value (namely, the number of the rotors with poorer welding quality), and calculating the ratio of the dissimilar rotor analysis value to the feasible rotor analysis value to obtain a dissimilar rotor detection value, wherein the larger the numerical value of the dissimilar rotor detection value is, the worse the rotor quality condition processed by the corresponding welding equipment in the detection period is comprehensively;
Comparing the abnormal rotor detection condition value with a preset abnormal rotor detection condition threshold value, and if the abnormal rotor detection condition value exceeds the preset abnormal rotor detection condition threshold value, indicating that the rotor quality conditions processed by the corresponding welding equipment in the detection period are poor in combination, generating a welding disqualification signal of the corresponding welding equipment; if the abnormal rotor detection condition value does not exceed the preset abnormal rotor detection condition threshold value, indicating that the rotor quality condition processed by the corresponding welding equipment in the detection period is better comprehensively, generating a welding qualification signal of the corresponding welding equipment.
Embodiment III: as shown in fig. 3, the difference between the present embodiment and embodiments 1 and 2 is that the server is in communication connection with the abnormality cause progressive module, and the server sends the welding failure signal of the corresponding welding device to the abnormality cause progressive module, and when the abnormality cause progressive module receives the welding failure signal, the abnormality cause progressive module performs progressive evaluation analysis through the abnormality cause to generate a welding efficiency early warning signal, a welding environment early warning signal, a welding operation early warning signal or a welding device early warning signal;
The welding efficiency early warning signal, the welding environment early warning signal, the welding operation early warning signal or the welding equipment early warning signal are sent to the control output module through the server, the control output module outputs the corresponding early warning signals through the display screen, potential influencing factors causing poor welding quality of the rotor can be rapidly judged, corresponding improvement measures can be made by management staff in a targeted manner, the subsequent welding quality of the rotor is guaranteed, and the intelligent degree is high; the specific analysis process of the abnormality cause one by one judgment and analysis is as follows:
Acquiring the starting time and the ending time of corresponding welding equipment for rotor welding, marking the interval time between the ending time and the starting time as a rotor welding value, and marking the deviation value of the rotor welding value and a preset proper rotor welding standard value as a rotor time deviation value, wherein the larger the value of the rotor time deviation value is, the more abnormal the welding processing of the corresponding rotor is indicated; carrying out average value calculation on all rotor time offset values of corresponding welding equipment in the detection period to obtain rotor aging values, and marking the number occupation ratio of the rotor time offset values exceeding a preset rotor time offset threshold value in the detection period as a rotor overexposure value;
carrying out numerical calculation on a rotor aging value HS and a rotor overestimation value HY through a formula HX=a1 xHS/a2+a2 xHY to obtain a welding efficiency early warning value HX, wherein a1 and a2 are preset proportionality coefficients, and a2 is more than a1 and more than 0; moreover, the larger the value of the welding efficiency early warning value HX is, the more abnormal the welding efficiency in the detection period is shown; and (3) carrying out numerical comparison on the welding efficiency early-warning value HX and a preset welding efficiency early-warning threshold value, if the welding efficiency early-warning value HX exceeds the preset welding efficiency early-warning threshold value, indicating that the welding efficiency in the detection period is abnormal, generating a welding efficiency early-warning signal if the possibility of poor welding quality of the rotor due to the welding efficiency is high, setting the welding efficiency reasonably in time, strictly controlling the welding efficiency, and guaranteeing the welding quality of the rotor.
If the welding efficiency early warning value HX does not exceed the preset welding efficiency early warning threshold value, acquiring environmental temperature data, environmental humidity data and environmental smoke data (data values of the smoke concentration in the corresponding environments) of the environment in which the rotor is positioned in the welding process, marking a deviation value of the environmental temperature data and a preset proper welding temperature value as a welding temperature detection value, and marking a deviation value of the environmental humidity data and the preset proper welding humidity value as a welding wet detection value;
Carrying out numerical calculation on the welding temperature detection value YP, the welding wet detection value YK and the environmental smoke data YF through a formula YX= (rg1+rg2+YK+rg3)/3 to obtain a welding ring condition value YX; wherein, rg1, rg2 and rg3 are preset proportional coefficients, and the values of rg1, rg2 and rg3 are positive numbers; moreover, the larger the numerical value of the welding ring condition value YX is, the worse the welding environment performance at the corresponding moment is, and the welding quality of the rotor is not guaranteed;
carrying out variance calculation and mean calculation on all welding ring condition values of corresponding environments in a detection period to obtain a welding ring deflection value and a welding ring table value, wherein the larger the numerical value of the welding ring deflection value is, the larger the welding environment fluctuation in the welding process in the detection period is, the larger the numerical value of the welding ring table value is, and the worse the welding environment performance in the detection period is; respectively comparing the welding ring offset value and the welding ring table value with a preset welding ring offset threshold value and a preset welding ring table threshold value in numerical value, and if the welding ring offset value or the welding ring table value exceeds the corresponding preset threshold value, indicating that the possibility of poor welding quality of the rotor caused by environmental factors is high, generating a welding environment early warning signal;
If the welding ring deviation value and the welding ring table value do not exceed the corresponding preset threshold values, comparing the welding ring condition value with the preset welding ring condition threshold values, and if the welding ring condition value exceeds the preset welding ring Kuang Yuzhi, judging that the corresponding environment is in a welding damage state; acquiring the total duration of the welding damage state of the corresponding environment in the detection period, marking the total duration as a welding damage time condition value, and determining the total duration as a welding damage time condition value according to a formula Carrying out numerical calculation on the welding loss time condition value WR and the welding ring table value WY to obtain a welding ring evaluation value WF, wherein kp1 and kp2 are preset proportionality coefficients, and the values of kp1 and kp2 are positive numbers;
Moreover, the larger the value of the welding ring evaluation value WF is, the greater the possibility that the rotor welding quality is poor due to environmental factors in the detection period is indicated; and comparing the welding critique value WF with a preset welding critique threshold value, and if the welding critique value WF exceeds the preset welding critique threshold value, indicating that the possibility of poor welding quality of the rotor caused by environmental factors in the detection period is high, generating a welding environment early warning signal so as to strengthen management and control of the welding environment in the subsequent welding process and ensure that the welding process is always in a proper state.
Further, if the welding ring evaluation value WF does not exceed the preset welding ring evaluation threshold value, acquiring an operator performing rotor welding operation by a corresponding welding device in a detection period, acquiring the number of times of error operation when the operator performs rotor welding operation in the detection period and marking the number of times of error operation as a welding error table value, acquiring the duration of each error operation and marking the duration as a welding error operation value, summing all the welding error operation values in the detection period to obtain the welding error value, comparing the welding error operation value with a preset welding error operation time threshold value, and marking the number of the welding error operation values exceeding the preset welding error operation time threshold value as an error operation high value;
Performing numerical calculation on the welding error table value GY, the welding error value GL and the misoperation high value GW through a formula GK=ew1+ew2+GL/(ew1+ew3) +ew3 to obtain a welding operation value GK, wherein, w1, w2 and w3 are preset proportionality coefficients, and w3 > w1 > w2 > 0; moreover, the larger the value of the welding operation value GK is, the worse the operation performance of operators in the detection period is, and the higher the possibility that the rotor welding quality is poor due to the operation factors of the operators is;
Comparing the welding operation value GK with a preset welding operation threshold value, and if the welding operation value GK exceeds the preset welding operation threshold value, indicating that the possibility of poor welding quality of the rotor caused by operation factors of operators in a detection period is high, generating a welding operation early warning signal, and timely enhancing welding operation training and welding operation supervision of corresponding operators; if the welding operation value GK does not exceed the preset welding operation threshold value, the possibility of poor welding quality of the rotor caused by operation factors of operators in the detection period is indicated to be small, a welding equipment early warning signal is generated, and corresponding welding equipment is overhauled and maintained in time, and equipment supervision is enhanced in the subsequent welding process.
The working principle of the invention is as follows: when the device is used, the image acquisition module is used for acquiring the image information of the welding part of the rotor, the image processing module is used for preprocessing the received image and extracting the characteristic information of the welding area of the rotor, the defect identification module is used for carrying out defect identification on the preprocessed image input model, the defect assessment module is used for carrying out automatic assessment on the defect according to the identified defect information and generating a detailed assessment report, the control output module is used for outputting the assessment report through a display screen or a printer, and the automatic detection, identification and assessment on the welding defect of the rotor are realized through the combination of an image processing technology, a deep learning algorithm and an automation technology, so that the detection efficiency and accuracy are improved, and the influence of human factors is reduced; and the welding quality performance of the welding equipment in the detection period is comprehensively evaluated through the welding quality comprehensive evaluation module, the welding disqualification signals or the welding qualification signals of the corresponding welding equipment are generated through analysis, the welding efficiency early warning signals, the welding environment early warning signals, the welding operation early warning signals or the welding equipment early warning signals are generated through judging and analyzing one by one through abnormal reasons when the welding disqualification signals are generated, potential influence factors causing poor welding quality of the rotor can be rapidly judged, corresponding improvement measures can be made by management staff in a targeted mode, the subsequent welding quality of the rotor is guaranteed, and the intelligent degree is high.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation. The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (9)
1. A method for evaluating welding defects of a rotor, comprising the steps of:
step one, acquiring image information of a rotor welding part through an image acquisition module, and transmitting the acquired image to an image processing module through a server;
The image processing module pre-processes the received image, extracts characteristic information of a rotor welding area by utilizing an image segmentation technology, and sends the pre-processed image to the defect recognition module through the server;
thirdly, a defect recognition module builds a welding defect recognition model based on a deep learning algorithm, a large amount of sample data are trained to enable the model to have the capability of recognizing various welding defects, the preprocessed image is input into the model to conduct defect recognition, and recognized rotor defect information is sent to a defect evaluation module through a server;
step four, the defect evaluation module automatically evaluates the defects according to the identified defect information and combining the welding process requirements and standards, judges whether the defects meet the quality requirements, generates a detailed evaluation report, and sends the evaluation report to the control output module through the server;
fifthly, controlling an output module to output the assessment report through a display screen or a printer;
The server is in communication connection with the abnormal cause judging module, the server sends welding failure signals of corresponding welding equipment to the abnormal cause judging module, when the abnormal cause judging module receives the welding failure signals, the abnormal cause judging module judges and analyzes the welding failure signals one by one through the abnormal cause to generate welding efficiency early warning signals, welding environment early warning signals, welding operation early warning signals or welding equipment early warning signals, and the welding efficiency early warning signals, the welding environment early warning signals, the welding operation early warning signals or the welding equipment early warning signals are sent to the control output module through the server, and the control output module outputs the corresponding early warning signals through the display screen.
2. The method for evaluating welding defects of a rotor according to claim 1, wherein the specific operation procedure of the fourth step is as follows:
Receiving defect information output by a defect identification module, wherein the defect information comprises the type, the position and the size of the defect, and the corresponding defect information is directly used for judging the severity degree and the influence range of the defect;
Comparing and analyzing the received defect information according to preset welding process requirements and standards; the preset welding process requirements and standards generally comprise various indexes of welding quality, and by comparing the standards with actually detected defect information, whether the corresponding defects exceed the allowable range or not is judged, and whether the corresponding defects meet the quality requirements or not is determined.
3. A rotor welding defect assessment method according to claim 2, wherein in step four, the assessment report generated comprises the following:
defect image: the assessment report contains actually detected defect images, and the corresponding defect images intuitively show the appearance of the defects;
Defect type: the type of defect including cracks, pinholes and unfused is explicitly indicated in the assessment report;
Defect location: the specific positions of the defects on the welding pieces are marked in detail in the assessment report;
assessment results: the final evaluation result is recorded in the evaluation report, and whether the defect meets the quality requirement is determined; and for unqualified weldments, corresponding repair suggestions or treatment measures are given in the assessment report.
4. The method for evaluating the welding defect of the rotor according to claim 1, wherein the server is in communication connection with the welding quality comprehensive evaluation module, the welding quality comprehensive evaluation module obtains welding equipment for welding the rotor, comprehensively evaluates the welding quality performance of the welding equipment in a detection period, generates welding failure signals or welding failure signals of the corresponding welding equipment through analysis, and sends the welding failure signals to the control output module through the server, and the control output module outputs the welding failure signals through the display screen.
5. The method for evaluating welding defects of a rotor according to claim 4, wherein the specific operation process of the welding quality comprehensive evaluation module comprises:
Setting a detection period, collecting an evaluation report of the defect quality of a rotor welded by welding equipment in the detection period, marking the number of the rotors with defects meeting the quality requirement as a feasible rotor analysis value, marking the number of the rotors with defects not meeting the quality requirement as a different rotor analysis value, calculating the ratio of the different rotor analysis value to the feasible rotor analysis value to obtain a different rotor detection condition value, comparing the different rotor detection condition value with a preset different rotor detection condition threshold value, and generating a welding failure signal of the corresponding welding equipment if the different rotor detection condition value exceeds the preset different rotor detection condition threshold value; if the abnormal rotor detection condition value does not exceed the preset abnormal rotor detection condition threshold value, generating a welding qualification signal of the corresponding welding equipment.
6. The method for evaluating welding defects of a rotor according to claim 1, wherein the specific analysis process of the abnormality cause-by-abnormality judgment analysis is as follows:
Acquiring the starting time and the ending time of corresponding welding equipment in rotor welding, marking the interval time between the ending time and the starting time as a rotor welding value, marking the deviation value of the rotor welding value and a preset proper rotor welding standard value as a rotor time offset value, carrying out average calculation on all rotor time offsets of corresponding welding equipment in a detection period to obtain a rotor aging value, and marking the number occupation ratio of the rotor time offsets exceeding a preset rotor time offset threshold in the detection period as a rotor super occupation value;
and carrying out numerical calculation on the rotor aging value and the rotor overexposure value to obtain a welding efficiency early warning value, carrying out numerical comparison on the welding efficiency early warning value and a preset welding efficiency early warning threshold value, and generating a welding efficiency early warning signal if the welding efficiency early warning value exceeds the preset welding efficiency early warning threshold value.
7. The method for evaluating the welding defect of the rotor according to claim 6, wherein if the welding effect early-warning value does not exceed a preset welding effect early-warning threshold value, acquiring environmental temperature data, environmental humidity data and environmental smoke data of an environment in which the rotor is positioned in a welding process, marking a deviation value of the environmental temperature data and a preset proper welding temperature value as a welding temperature detection value, acquiring a welding wet detection value in a similar way, and performing numerical calculation on the welding temperature detection value, the welding wet detection value and the environmental smoke data to acquire a welding ring condition value;
Performing variance calculation and mean calculation on all welding ring condition values of corresponding environments in a detection period to obtain a welding ring deviation value and a welding ring table value, respectively performing numerical comparison on the welding ring deviation value and the welding ring table value and a preset welding ring deviation threshold value and a preset welding ring table threshold value, and generating a welding environment early warning signal if the welding ring deviation value or the welding ring table value exceeds the corresponding preset threshold value;
If the welding ring deviation value and the welding ring table value do not exceed the corresponding preset threshold values, comparing the welding ring condition value with the preset welding ring condition threshold values, and if the welding ring condition value exceeds the preset welding ring Kuang Yuzhi, judging that the corresponding environment is in a welding damage state; the method comprises the steps of obtaining total duration of a welding damage state of a corresponding environment in a detection period, marking the total duration as a welding damage time condition value, carrying out numerical calculation on the welding damage time condition value and a welding ring table value to obtain a welding ring evaluation value, carrying out numerical comparison on the welding ring evaluation value and a preset welding ring evaluation threshold value, and generating a welding environment early warning signal if the welding ring evaluation value exceeds the preset welding ring evaluation threshold value.
8. The method for evaluating a welding defect of a rotor according to claim 7, wherein if the welding critique value does not exceed a preset welding critique threshold, a welding operation value is obtained by analysis, the welding operation value is compared with a preset welding operation threshold, and if the welding operation value exceeds the preset welding operation threshold, a welding operation early warning signal is generated; and if the welding operation value does not exceed the preset welding operation threshold value, generating a welding equipment early warning signal.
9. The method for evaluating welding defects of a rotor according to claim 8, wherein the method for analyzing and acquiring the welding operation value is as follows:
Acquiring operators performing rotor welding operation through corresponding welding equipment in a detection period, acquiring the number of times of wrong operation when the operators perform rotor welding operation in the detection period, marking the number of times of wrong operation as a welding wrong table value, acquiring the duration time of each wrong operation, marking the duration time as a welding wrong operation value, summing all the welding wrong operation values in the detection period to obtain the welding wrong value, and marking the number of the welding wrong operation values exceeding a preset welding wrong operation time threshold as a wrong operation high value; and carrying out numerical calculation on the welding error table value, the welding error time value and the error operation high value to obtain a welding operation value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410472963.5A CN118067727B (en) | 2024-04-19 | 2024-04-19 | Rotor welding defect assessment method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410472963.5A CN118067727B (en) | 2024-04-19 | 2024-04-19 | Rotor welding defect assessment method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118067727A CN118067727A (en) | 2024-05-24 |
| CN118067727B true CN118067727B (en) | 2024-06-21 |
Family
ID=91095751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410472963.5A Active CN118067727B (en) | 2024-04-19 | 2024-04-19 | Rotor welding defect assessment method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118067727B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105588845A (en) * | 2016-01-04 | 2016-05-18 | 江苏科技大学 | Weld defect characteristic parameter extraction method |
| CN116363089A (en) * | 2023-03-23 | 2023-06-30 | 天地上海采掘装备科技有限公司 | Automatic detection method for welding defects of roller tooth holder of coal mining machine |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102621156B1 (en) * | 2021-07-15 | 2024-01-05 | 주식회사 씨피에스 | Welding area inspection system and method thereof for assessing the risk based on artificial intelligence |
| EP4293621A1 (en) * | 2022-06-13 | 2023-12-20 | Tata Consultancy Services Limited | Weld quality inspection with domain knowledge infused adaptive-network-based fuzzy inference system |
| CN117548911A (en) * | 2023-11-23 | 2024-02-13 | 千思跃智能科技(苏州)股份有限公司 | Connector welding intelligent supervision system based on artificial intelligence |
-
2024
- 2024-04-19 CN CN202410472963.5A patent/CN118067727B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105588845A (en) * | 2016-01-04 | 2016-05-18 | 江苏科技大学 | Weld defect characteristic parameter extraction method |
| CN116363089A (en) * | 2023-03-23 | 2023-06-30 | 天地上海采掘装备科技有限公司 | Automatic detection method for welding defects of roller tooth holder of coal mining machine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118067727A (en) | 2024-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110867196B (en) | Machine equipment state monitoring system based on deep learning and voice recognition | |
| CN107220469B (en) | Method and system for estimating state of fan | |
| CN112669305B (en) | Metal surface rust resistance test bench and rust resistance evaluation method | |
| CN111852792B (en) | Fan blade defect self-diagnosis positioning method based on machine vision | |
| CN111272670A (en) | Dyeing quality detection method and device, storage medium and processor | |
| CN117250208B (en) | Machine vision-based nano-imprint wafer defect accurate detection system and method | |
| CN113406107B (en) | Fan blade defect detection system | |
| CN107895362A (en) | A kind of machine vision method of miniature binding post quality testing | |
| CN110186375A (en) | Intelligent high-speed rail white body assemble welding feature detection device and detection method | |
| CN115576284A (en) | Clothing workshop intelligent management method and system | |
| CN113255654A (en) | Imaging processing method of characteristic data and application thereof | |
| CN116152202A (en) | Equipment appearance detection system based on image recognition technology and infrared thermal imaging technology | |
| CN115060312A (en) | Building material safety monitoring system based on artificial intelligence | |
| CN114170477A (en) | Multimode fusion mechanical defect detection method based on attention-driven graph convolutional network | |
| CN114326741B (en) | Sea water desalination monitoring control system based on quadruped robot | |
| CN111754737A (en) | Online identification and evaluation device and method for installation acceptance of metering device | |
| CN118067727B (en) | Rotor welding defect assessment method | |
| CN108898187A (en) | A kind of method and device of automatic identification power distribution room indicating equipment image | |
| CN113033625A (en) | Industrial air conditioner fault diagnosis method based on neural network and integrated learning fusion | |
| CN111415326A (en) | Method and system for detecting abnormal state of railway contact net bolt | |
| CN111456915A (en) | Fault diagnosis device and method for internal components of fan engine room | |
| CN112508946B (en) | Cable tunnel anomaly detection method based on antagonistic neural network | |
| CN116363397A (en) | Equipment fault checking method, device and inspection system | |
| CN114565883A (en) | Graphic recognition algorithm for operation faults of equipment | |
| CN111626323A (en) | Electric meter state detection and evaluation method based on deep learning |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |