Method and device for detecting surface cracks of ferromagnetic materials
Through multi-dimensional detection methods, combined with material pre-check, metal magnetic memory, ultrasonic and eddy current detection, the problem of poor accuracy in surface crack detection of ferromagnetic materials is solved, the detection accuracy and reliability are improved, and the safety of ferromagnetic materials is ensured. .
Patent Information
- Application Number
- CN202410594446.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-14
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-05-14
AI Technical Summary
In the prior art, the detection accuracy of surface cracks of ferromagnetic materials is poor, resulting in poor detection effect and safety hazards.
Multi-dimensional detection methods are adopted, including material pre-checking components, metal magnetic memory detection components, ultrasonic detection components and eddy current detection components, and precision detection reports are generated through surface cleanliness verification, crack roughness detection, feature recognition and multi-parameter composite detection.
It improves the accuracy and reliability of surface crack detection of ferromagnetic materials, improves the detection quality, and ensures the safety of ferromagnetic materials.
Smart Images

Figure CN118425288B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of material crack detection, and in particular to a method and device for detecting surface cracks of ferromagnetic materials. Background Art
[0002] Under the influence of external loads such as external force and vibration, ferromagnetic materials are prone to cracks. When cracks occur in ferromagnetic materials, not only the mechanical strength of the ferromagnetic materials is greatly reduced, but also the failure of ferromagnetic material components is easily caused, which eventually brings safety hazards to the equipment containing ferromagnetic materials. Crack detection has an important impact on the application of ferromagnetic materials. In the prior art, there is a technical problem that the accuracy of surface crack detection of ferromagnetic materials is poor, which in turn causes poor detection effect of surface cracks of ferromagnetic materials. Summary of the invention
[0003] The present application provides a method and device for detecting surface cracks of ferromagnetic materials. The method solves the technical problem that the detection accuracy of surface cracks of ferromagnetic materials in the prior art is poor, which in turn causes poor detection effect of surface cracks of ferromagnetic materials. By performing multi-dimensional surface crack detection on ferromagnetic materials, the detection accuracy and reliability of surface cracks of ferromagnetic materials are improved, and the detection quality of surface cracks of ferromagnetic materials is improved.
[0004] In view of the above problems, the present application provides a method and device for detecting surface cracks of ferromagnetic materials.
[0005] In a first aspect, the present application provides a method for detecting surface cracks in ferromagnetic materials, wherein the method is applied to a device for detecting surface cracks in ferromagnetic materials, the device comprising a material pre-verification component, a metal magnetic memory detection component, an ultrasonic detection component and an eddy current detection component, the method comprising: performing a surface cleanliness verification on a target ferromagnetic material according to the material pre-verification component to obtain a surface cleanliness verification result; if the surface cleanliness verification result is passed, activating the metal magnetic memory detection component; performing a surface crack roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain a crack roughness detection result; performing feature recognition according to the crack roughness detection result to obtain a crack existence feature judgment operator; if the crack existence feature judgment operator indicates that a surface crack exists in the target ferromagnetic material, generating a crack precision detection instruction; activating a crack-multi-parameter composite detection submodule based on the crack precision detection instruction, performing a surface crack precision detection on the target ferromagnetic material in combination with the crack roughness detection result, the ultrasonic detection component and the eddy current detection component to obtain a target surface crack detection report.
[0006] In the second aspect, the present application also provides a device for detecting surface cracks of ferromagnetic materials, the device comprising a material pre-verification component, a metal magnetic memory detection component, an ultrasonic detection component and an eddy current detection component, wherein the device further comprises: a surface cleanliness verification module, the surface cleanliness verification module being used to perform surface cleanliness verification on the target ferromagnetic material according to the material pre-verification component to obtain a surface cleanliness verification result; an activation module, the activation module being used to activate the metal magnetic memory detection component if the surface cleanliness verification result is passed; a crack roughness detection module, the crack roughness detection module being used to perform surface crack roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain A crack roughness detection result; a judgment operator acquisition module, the judgment operator acquisition module is used to perform feature recognition according to the crack roughness detection result, and obtain a crack existence feature judgment operator; a precision detection instruction generation module, the precision detection instruction generation module is used to generate a crack precision detection instruction if the crack existence feature judgment operator is that the target ferromagnetic material has a surface crack; a crack precision detection module, the crack precision detection module is used to activate the crack-multi-parameter composite detection submodule based on the crack precision detection instruction, and perform surface crack precision detection on the target ferromagnetic material in combination with the crack roughness detection result, the ultrasonic detection component and the eddy current detection component to obtain a target surface crack detection report.
[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0008] The surface cleanliness of the target ferromagnetic material is checked through the material pre-check component to obtain the surface cleanliness check result; if the surface cleanliness check result is passed, the metal magnetic memory detection component is activated, and the surface crack roughness detection of the target ferromagnetic material is performed according to the metal magnetic memory detection component to obtain the crack roughness detection result; feature recognition is performed according to the crack roughness detection result to obtain the crack existence feature judgment operator; if the crack existence feature judgment operator is that the target ferromagnetic material has a surface crack, a crack precision detection instruction is generated; based on the crack precision detection instruction, the crack-multi-parameter composite detection submodule is activated, and the surface crack precision detection of the target ferromagnetic material is performed in combination with the crack roughness detection result, the ultrasonic detection component and the eddy current detection component to obtain the target surface crack detection report. By performing multi-dimensional surface crack detection on ferromagnetic materials, the accuracy and reliability of surface crack detection of ferromagnetic materials are improved, and the surface crack detection quality of ferromagnetic materials is improved.
[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments of the present invention are briefly introduced below. Obviously, the drawings in the following description only relate to some embodiments of the present invention, but are not intended to limit the present invention.
[0011] Figure 1 The present invention is a schematic flow chart of a method for detecting surface cracks of a ferromagnetic material according to the present invention.
[0012] Figure 2 This is a schematic structural diagram of a device for detecting surface cracks of ferromagnetic materials according to the present application. DETAILED DESCRIPTION
[0013] The present application provides a method and device for detecting surface cracks of ferromagnetic materials. The present application solves the technical problem that the detection accuracy of surface cracks of ferromagnetic materials in the prior art is poor, which in turn causes poor detection effect of surface cracks of ferromagnetic materials. By performing multi-dimensional surface crack detection on ferromagnetic materials, the detection accuracy and reliability of surface cracks of ferromagnetic materials are improved, and the detection quality of surface cracks of ferromagnetic materials is improved.
[0014] Embodiment 1
[0015] Please see attached Figure 1 The present application provides a method for detecting surface cracks of ferromagnetic materials, wherein the method is applied to a device for detecting surface cracks of ferromagnetic materials, the device comprising a material pre-verification component, a metal magnetic memory detection component, an ultrasonic detection component and an eddy current detection component, and the method specifically comprises the following steps:
[0016] Performing surface cleanliness verification on the target ferromagnetic material according to the material pre-verification component to obtain a surface cleanliness verification result;
[0017] Wherein, performing surface cleanliness verification on the target ferromagnetic material according to the material pre-verification component to obtain the surface cleanliness verification result includes:
[0018] According to the material pre-verification component, obtaining material surface state information of the target ferromagnetic material;
[0019] Evaluate the surface cleanliness of the target ferromagnetic material according to the surface state information of the material to obtain the surface cleanliness of the material;
[0020] Determining whether the surface cleanliness of the material is less than a preset cleanliness;
[0021] If the surface cleanliness of the material is less than the preset cleanliness, the surface cleanliness verification result obtained is failure, and a surface cleanliness warning signal is generated.
[0022] Based on the confirmation of those skilled in the art: the degree of cleanliness has a certain influence on the surface crack detection of ferromagnetic materials. If there are impurities such as dirt and grease on the surface of the ferromagnetic material, these impurities may interfere with the propagation of the detection signal, resulting in errors in the detection results, and further affecting the accuracy and reliability of the surface crack detection of the ferromagnetic material. Therefore, before performing surface crack detection on the target ferromagnetic material, the present application first performs a surface cleanliness check on the target ferromagnetic material, thereby preventing the surface impurities of the target ferromagnetic material from affecting the surface crack detection, and improving the accuracy and reliability of the surface crack detection of the target ferromagnetic material.
[0023] The specific process of surface cleanliness verification of the target ferromagnetic material includes: real-time image acquisition of the target ferromagnetic material through the material pre-verification component, obtaining the material surface state information of the target ferromagnetic material, and performing surface cleanliness evaluation on the material surface state information to obtain the material surface cleanliness. Then, by judging whether the surface cleanliness of the material is less than the preset cleanliness, the surface cleanliness verification result is obtained. The surface cleanliness verification result includes pass / fail. If the surface cleanliness of the material is less than the preset cleanliness, the surface cleanliness verification result obtained is fail, and a surface cleanliness warning signal is generated. If the surface cleanliness of the material is greater than / equal to the preset cleanliness, the surface cleanliness verification result obtained is pass.
[0024] Among them, the material pre-verification component includes an image acquisition device in the prior art. The target ferromagnetic material can be any ferromagnetic material (or any ferromagnetic material component) that uses the detection device for surface cracks of a ferromagnetic material to perform intelligent surface crack detection. The material surface state information includes real-time image data corresponding to the target ferromagnetic material. The surface cleanliness of the material is data information used to characterize the surface cleanliness of the target ferromagnetic material. The more impurities such as dirt and grease there are on the surface of the target ferromagnetic material, the lower the surface cleanliness of the target ferromagnetic material, and the smaller the corresponding surface cleanliness of the material. The preset cleanliness is a material surface cleanliness threshold value pre-set and determined by the detection device for surface cracks of a ferromagnetic material. The surface cleanliness warning signal is a warning prompt information used to characterize that the surface cleanliness of the material is less than the preset cleanliness and the target ferromagnetic material needs to be cleaned.
[0025] Preferably, when evaluating the surface cleanliness of the material surface state information, historical data is collected based on the material surface state information to obtain multiple groups of surface cleanliness evaluation data. Each group of surface cleanliness evaluation data includes historical material surface state information and historical material surface cleanliness corresponding to the historical material surface state information. Furthermore, according to the convolutional neural network, the multiple groups of surface cleanliness evaluation data are continuously self-trained and learned to a convergence state, and a surface cleanliness evaluation network can be obtained. The material surface state information is input into the surface cleanliness evaluation network to obtain the surface cleanliness of the material. Convolutional neural network is a type of feedforward neural network that includes convolution calculation and has a deep structure. Convolutional neural network has representation learning ability and can perform translation-invariant classification of input information according to its hierarchical structure. The surface cleanliness evaluation network is a convolutional neural network trained by multiple groups of surface cleanliness evaluation data. The surface cleanliness evaluation network includes an input layer, a hidden layer, and an output layer.
[0026] If the surface cleanliness verification result is passed, activating the metal magnetic memory detection component;
[0027] Performing surface crack and roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain a crack and roughness detection result;
[0028] Wherein, performing surface crack roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain a crack roughness detection result includes:
[0029] Obtaining crack roughness detection accuracy of the target ferromagnetic material;
[0030] Inputting the crack roughness detection accuracy into a probe lift-off feature table to obtain a detection probe lift-off operator;
[0031] Based on the detection probe lift-off operator, the target ferromagnetic material is detected according to the metal magnetic memory detection component to obtain a material magnetic memory signal detection result;
[0032] A surface crack roughness analysis is performed according to the material magnetic memory signal detection result to generate the crack roughness detection result.
[0033] When the surface cleanliness verification result is passed, the metal magnetic memory detection component is activated. The metal magnetic memory detection component includes a metal magnetic memory sensor in the prior art. The metal magnetic memory sensor is an instrument that uses metal magnetic memory detection technology to detect the internal stress concentration part of the material. The metal magnetic memory sensor includes components such as a detection probe. When the metal magnetic memory detection component is used to detect the target ferromagnetic material, the detection probe scans the surface of the target ferromagnetic material, collects the magnetic field signal for analysis, and the probe lift-off value (i.e., the vertical distance between the detection probe and the surface of the target ferromagnetic material) directly affects the amplitude of the magnetic field signal. In order to obtain the best detection effect, the present application inputs the crack roughness detection accuracy into a pre-constructed probe lift-off feature table, matches the probe lift-off value of the crack roughness detection accuracy through the probe lift-off feature table, and obtains the detection probe lift-off operator. The crack roughness detection accuracy includes the metal magnetic memory detection accuracy of the target ferromagnetic material preset by the user end. The probe lift-off feature table includes multiple preset crack roughness detection accuracies, and multiple preset probe lift-off values corresponding to multiple preset crack roughness detection accuracies. The detection probe lift-off operator is a preset probe lift-off value corresponding to the crack roughness detection accuracy.
[0034] Furthermore, the metal magnetic memory detection component is controlled according to the detection probe lift-off operator to detect the target ferromagnetic material, and the material magnetic memory signal detection result is obtained, and the surface crack roughness analysis is performed on the material magnetic memory signal detection result to generate a crack roughness detection result. Among them, the material magnetic memory signal detection result includes a plurality of magnetic field signals obtained by scanning the surface of the target ferromagnetic material by controlling the detection probe of the metal magnetic memory detection component according to the detection probe lift-off operator. The crack roughness detection result includes whether the target ferromagnetic material has surface cracks, and a plurality of surface crack positions corresponding to the target ferromagnetic material. When the target ferromagnetic material does not have surface cracks, the corresponding surface crack position is 0.
[0035] Exemplarily, when performing surface crack roughness analysis on the material magnetic memory signal detection result, historical data is collected according to the material magnetic memory signal detection result to obtain multiple surface crack roughness analysis records. Each surface crack roughness analysis record includes the historical material magnetic memory signal detection result and the historical crack roughness detection result corresponding to the historical material magnetic memory signal detection result. According to the fully connected neural network, multiple surface crack roughness analysis records are cross-supervised trained to obtain a surface crack roughness analysis network. The material magnetic memory signal detection result is input into the surface crack roughness analysis network to obtain a crack roughness detection result. The fully connected neural network is an artificial neural network structure with a relatively simple connection method. The surface crack roughness analysis network includes an input layer, a hidden layer, and an output layer.
[0036] Taking advantage of the unique advantages of metal magnetic memory detection technology in defect location, the surface cracks of the target ferromagnetic material are located through the metal magnetic memory detection component, and the crack roughness detection results are determined, making the surface crack detection of ferromagnetic materials more efficient and accurate.
[0037] Performing feature recognition according to the crack roughness detection result to obtain a crack existence feature judgment operator;
[0038] If the crack existence characteristic judgment operator indicates that the target ferromagnetic material has a surface crack, a crack precision detection instruction is generated;
[0039] The crack roughness detection result is subjected to feature recognition, that is, whether the target ferromagnetic material in the crack roughness detection result has surface cracks is recorded as a crack existence feature judgment operator. The crack existence feature judgment operator is the target ferromagnetic material having surface cracks / the target ferromagnetic material not having surface cracks. When the crack existence feature judgment operator is the target ferromagnetic material having surface cracks, the device for detecting surface cracks of a ferromagnetic material automatically generates a crack precision detection instruction. The crack precision detection instruction is an instruction information used to characterize the presence of surface cracks in the target ferromagnetic material and the need for further crack detection on the target ferromagnetic material.
[0040] Based on the crack precision detection instruction, the crack-multi-parameter composite detection submodule is activated, and the surface crack precision detection of the target ferromagnetic material is performed in combination with the crack rough detection result, the ultrasonic detection component and the eddy current detection component to obtain a target surface crack detection report.
[0041] Constructing the crack-multi-parameter composite detection submodule, wherein the crack-multi-parameter composite detection submodule includes an ultrasonic-crack feature analysis channel, an eddy current-crack feature analysis channel and a crack analysis feature comparison channel;
[0042] The construction of the crack-multi-parameter composite detection submodule includes:
[0043] Obtaining an ultrasonic-crack feature analysis record library, an eddy current-crack feature analysis record library, and a crack analysis feature comparison record library;
[0044] Obtaining a plurality of preset training operators, and randomly selecting the plurality of preset training operators to obtain a first preset training operator, a second preset training operator, and a third preset training operator;
[0045] Performing supervised training on the ultrasonic-crack feature analysis record library based on the first preset training operator to generate the ultrasonic-crack feature analysis channel;
[0046] Performing supervised training on the eddy current-crack feature analysis record library based on the second preset training operator to generate the eddy current-crack feature analysis channel;
[0047] Performing supervised training on the crack analysis feature comparison record library based on the third preset training operator to generate the crack analysis feature comparison channel;
[0048] The ultrasonic-crack feature analysis channel, the eddy current-crack feature analysis channel and the crack analysis feature comparison channel are connected to generate the crack-multi-parameter composite detection submodule.
[0049] Connect the detection device of the surface crack of the ferromagnetic material, and retrieve the ultrasonic-crack feature analysis record library, the eddy current-crack feature analysis record library and the crack analysis feature comparison record library. The ultrasonic-crack feature analysis record library includes multiple ultrasonic-crack feature analysis records. Each ultrasonic-crack feature analysis record includes historical partition ultrasonic detection results (historical echo signals corresponding to historical surface crack partitions), historical ultrasonic-surface crack detection results (historical crack depths, historical crack types, historical crack shapes, etc. corresponding to historical partition ultrasonic detection results). The eddy current-crack feature analysis record library includes multiple eddy current-crack feature analysis records. Each eddy current-crack feature analysis record includes historical partition eddy current detection results (historical eddy current output voltages corresponding to historical surface crack partitions), historical eddy current-surface crack detection results (historical crack depths, historical crack types, historical crack shapes, etc. corresponding to historical partition eddy current detection results). The crack analysis feature comparison record library includes multiple crack analysis feature comparison records. Each crack analysis feature comparison record includes historical ultrasonic-surface crack detection result information, historical eddy current-surface crack detection result information, and historical crack detection comparison degree between historical ultrasonic-surface crack detection result information and historical eddy current-surface crack detection result information.
[0050] Further, the device for detecting surface cracks of a ferromagnetic material is connected, multiple preset training operators are retrieved, and the multiple preset training operators are randomly selected to obtain a first preset training operator, a second preset training operator, and a third preset training operator. Among them, the multiple preset training operators include multiple different model network structures such as BP neural network, fully connected neural network, feedforward neural network, recurrent neural network, long short-term memory neural network, decision tree, support vector machine, etc., which are pre-set and determined by the device for detecting surface cracks of a ferromagnetic material. The first preset training operator, the second preset training operator, and the third preset training operator are all any preset training operators among the multiple preset training operators (the first preset training operator, the second preset training operator, and the third preset training operator can be the same preset training operator or different preset training operators).
[0051] Further, supervised training is performed on the ultrasonic-crack feature analysis record library according to the first preset training operator to generate an ultrasonic-crack feature analysis channel. For example, when the first preset training operator is a BP neural network, 80% of the data information in the ultrasonic-crack feature analysis record library is randomly divided into a training data set. 20% of the data information in the ultrasonic-crack feature analysis record library is randomly divided into a test data set. The training data set is cross-supervised trained according to the BP neural network to obtain an ultrasonic-crack feature analysis channel. The test data set is used as input information and input into the ultrasonic-crack feature analysis channel, and the parameters of the ultrasonic-crack feature analysis channel are updated through the test data set. The BP neural network is a multi-layer feedforward neural network trained according to the error back propagation algorithm. The ultrasonic-crack feature analysis channel is a BP neural network trained by the ultrasonic-crack feature analysis record library. The ultrasonic-crack feature analysis channel includes an input layer, a hidden layer, and an output layer.
[0052] Similarly, supervised training is performed on the eddy current-crack feature analysis record library according to the second preset training operator to generate an eddy current-crack feature analysis channel. Supervised training is performed on the crack analysis feature comparison record library according to the third preset training operator to generate a crack analysis feature comparison channel. Then, the ultrasonic-crack feature analysis channel, the eddy current-crack feature analysis channel and the crack analysis feature comparison channel are connected to generate a crack-multi-parameter composite detection submodule. The construction methods of the eddy current-crack feature analysis channel and the crack analysis feature comparison channel are similar to those of the ultrasonic-crack feature analysis channel, and will not be repeated here. The crack-multi-parameter composite detection submodule includes an ultrasonic-crack feature analysis channel, an eddy current-crack feature analysis channel and a crack analysis feature comparison channel.
[0053] Through multiple preset training operators, a crack-multi-parameter composite detection submodule with strong generalization ability is constructed, thereby improving the accuracy of surface crack detection of ferromagnetic materials.
[0054] Divide the target ferromagnetic material into regions according to the crack roughness detection result to obtain a plurality of surface crack partitions;
[0055] Based on the multiple surface crack partitions, the target ferromagnetic material is subjected to precise surface crack detection according to the ultrasonic detection component and the ultrasonic-crack feature analysis channel to obtain ultrasonic-surface crack detection results;
[0056] Performing ultrasonic detection fitness analysis on the target ferromagnetic material and generating ultrasonic fitness instructions;
[0057] The ultrasonic detection fitness analysis of the target ferromagnetic material includes:
[0058] Obtaining basic material information of the target ferromagnetic material;
[0059] Obtaining a material ultrasonic attenuation record library, and performing data fusion on the material ultrasonic attenuation record library to generate an ultrasonic attenuation identification domain;
[0060] Inputting the material basic information into the ultrasonic attenuation identification domain to obtain a matching ultrasonic attenuation coefficient;
[0061] Determining whether the matched ultrasonic attenuation coefficient is less than a preset ultrasonic attenuation coefficient;
[0062] If the matched ultrasonic attenuation coefficient is less than the preset ultrasonic attenuation coefficient, obtaining the ultrasonic adaptation instruction;
[0063] If the matched ultrasonic attenuation coefficient is greater than / equal to the preset ultrasonic attenuation coefficient, an ultrasonic taboo instruction is generated, and ultrasonic detection of the target ferromagnetic material is tabooed according to the ultrasonic taboo instruction.
[0064] The surface area of the target ferromagnetic material is divided according to the crack roughness detection result to obtain a plurality of surface crack partitions. Each surface crack partition includes a surface area corresponding to each surface crack position in the crack roughness detection result on the surface of the target ferromagnetic material.
[0065] Based on the confirmation of those skilled in the art: when ultrasound propagates in a material, the ultrasound attenuation rates corresponding to materials of different materials or structures are different. When the ultrasound attenuation is severe (the ultrasound attenuation rate is high), the material is not suitable for ultrasonic detection. Therefore, before performing ultrasonic detection on the target ferromagnetic material, this application first performs an ultrasonic detection suitability analysis on the target ferromagnetic material.
[0066] The specific process of ultrasonic detection fitness analysis of the target ferromagnetic material includes: connecting a detection device for surface cracks of the ferromagnetic material, reading the basic material information of the target ferromagnetic material, and retrieving the material ultrasonic attenuation record library. The basic material information includes the material composition information and size structure information corresponding to the target ferromagnetic material. The material ultrasonic attenuation record library includes multiple sample material basic information. Moreover, each sample material basic information has a correspondingly identified sample ultrasonic attenuation rate.
[0067] Furthermore, the material ultrasonic attenuation record library is data fused, that is, each sample material basic information in the material ultrasonic attenuation record library is recorded as the material ultrasonic attenuation input feature, and the sample ultrasonic attenuation rate corresponding to each sample material basic information is recorded as the material ultrasonic attenuation output feature, to obtain the ultrasonic attenuation recognition domain. The ultrasonic attenuation recognition domain includes multiple material ultrasonic attenuation input features and multiple material ultrasonic attenuation output features.
[0068] Further, the basic information of the material is input into the ultrasonic attenuation identification domain to obtain the matching ultrasonic attenuation coefficient, and it is determined whether the matching ultrasonic attenuation coefficient is less than the preset ultrasonic attenuation coefficient. If the matching ultrasonic attenuation coefficient is less than the preset ultrasonic attenuation coefficient, an ultrasonic adaptation instruction is generated. On the contrary, if the matching ultrasonic attenuation coefficient is greater than / equal to the preset ultrasonic attenuation coefficient, an ultrasonic taboo instruction is generated, and ultrasonic detection of the target ferromagnetic material is prohibited according to the ultrasonic taboo instruction, that is, according to the ultrasonic taboo instruction, ultrasonic detection is not performed on the target ferromagnetic material, and the partition ultrasonic detection results and ultrasonic-surface crack detection results are both recorded as 0.
[0069] Among them, the matching ultrasonic attenuation coefficient includes the ultrasonic attenuation rate corresponding to the basic information of the material. The preset ultrasonic attenuation coefficient includes the ultrasonic attenuation rate threshold predetermined by the detection device for surface cracks of the ferromagnetic material. The ultrasonic adaptation instruction is an instruction information used to characterize that the matching ultrasonic attenuation coefficient is less than the preset ultrasonic attenuation coefficient, and the target ferromagnetic material can be ultrasonically detected. The ultrasonic taboo instruction is an instruction information used to characterize that the matching ultrasonic attenuation coefficient is greater than / equal to the preset ultrasonic attenuation coefficient, the ultrasonic attenuation of the target ferromagnetic material is serious, and the target ferromagnetic material is not suitable for ultrasonic detection.
[0070] By performing ultrasonic detection fitness analysis on the target ferromagnetic material, ultrasonic adaptation instructions / ultrasonic taboo instructions are adaptively generated, thereby improving the fitness and reliability of surface crack detection of ferromagnetic materials.
[0071] According to the ultrasonic adaptation instruction, activating the ultrasonic detection component and the ultrasonic-crack feature analysis channel;
[0072] Performing ultrasonic testing on the plurality of surface crack partitions according to the ultrasonic testing component to obtain partition ultrasonic testing results;
[0073] The partitioned ultrasonic detection results are input into the ultrasonic-crack feature analysis channel to generate the ultrasonic-surface crack detection results.
[0074] Based on the multiple surface crack partitions, the target ferromagnetic material is subjected to precise surface crack detection according to the eddy current detection component and the eddy current-crack feature analysis channel to obtain an eddy current-surface crack detection result;
[0075] Inputting the ultrasonic-surface crack detection result and the eddy current-surface crack detection result into the crack analysis feature comparison channel to obtain a crack detection comparison result;
[0076] The target surface crack detection report is generated according to the ultrasonic-surface crack detection result, the eddy current-surface crack detection result and the crack detection comparison result.
[0077] According to the ultrasonic adaptation instruction, the ultrasonic detection component and the ultrasonic-crack feature analysis channel are activated. According to the ultrasonic detection component, ultrasonic detection is performed on multiple surface crack partitions to obtain partition ultrasonic detection results. The partition ultrasonic detection results are input into the ultrasonic-crack feature analysis channel to obtain ultrasonic-surface crack detection results. Among them, the ultrasonic detection component includes an ultrasonic sensor in the prior art. Ultrasonic detection refers to transmitting ultrasonic waves to multiple surface crack partitions through the ultrasonic detection component, and receiving echo signals reflected by multiple surface crack partitions, and obtaining corresponding crack feature information (crack depth, crack type, crack shape, etc.) by analyzing and processing the echo signals. The partition ultrasonic detection results include the echo signals corresponding to each surface crack partition. The ultrasonic-surface crack detection results include crack feature information such as crack depth, crack type, crack shape, etc. corresponding to the partition ultrasonic detection results.
[0078] Further, eddy current detection is performed on multiple surface crack partitions according to the eddy current detection component to obtain partition eddy current detection results. The partition eddy current detection results are input into the eddy current-crack feature analysis channel to obtain eddy current-surface crack detection results. Among them, the eddy current detection component includes an eddy current sensor in the prior art. Eddy current detection refers to inputting an alternating current into the eddy current probe of the eddy current detection component, and the alternating current will induce an alternating magnetic field (i.e., the eddy current probe magnetic field). Under the action of the eddy current probe magnetic field, the target ferromagnetic material will also generate current and magnetic fields with different distribution densities. When there are cracks in the target ferromagnetic material, the distribution of the current and magnetic field signals of the target ferromagnetic material will change, and these changes will be received by the eddy current probe through induction, thereby affecting the voltage change in the coil. The crack characteristics of the target ferromagnetic material can be judged by analyzing the specific changes in the voltage signal (i.e., the eddy current output voltage). The partition eddy current detection results include the eddy current output voltage corresponding to each surface crack partition. The eddy current-surface crack detection results include crack parameters such as crack depth, crack type, and crack shape corresponding to the partition eddy current detection results.
[0079] Further, the ultrasonic-surface crack detection results and the eddy current-surface crack detection results are input into the crack analysis feature comparison channel to obtain the crack detection comparison results, and the target surface crack detection report is obtained by combining the ultrasonic-surface crack detection results and the eddy current-surface crack detection results. Among them, the crack detection comparison results include the crack detection comparison degrees corresponding to the ultrasonic-surface crack detection results and the eddy current-surface crack detection results. The crack detection comparison degree is data information used to characterize the consistency between the ultrasonic-surface crack detection results and the eddy current-surface crack detection results. The higher the consistency between the ultrasonic-surface crack detection results and the eddy current-surface crack detection results, the smaller the corresponding crack detection comparison degree. The target surface crack detection report includes the ultrasonic-surface crack detection results, the eddy current-surface crack detection results and the crack detection comparison results. By performing multi-dimensional surface crack detection on the target ferromagnetic material, a comprehensive target surface crack detection report is obtained, which improves the surface crack detection effect of the ferromagnetic material.
[0080] In summary, the method for detecting surface cracks of ferromagnetic materials provided in this application has the following technical effects:
[0081] 1. Perform surface cleanliness verification on the target ferromagnetic material through the material pre-verification component to obtain the surface cleanliness verification result; if the surface cleanliness verification result is passed, activate the metal magnetic memory detection component, and perform surface crack roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain the crack roughness detection result; perform feature recognition according to the crack roughness detection result to obtain the crack existence feature judgment operator; if the crack existence feature judgment operator is that the target ferromagnetic material has surface cracks, generate crack precision detection instructions; based on the crack precision detection instruction, activate the crack-multi-parameter composite detection submodule, combine the crack roughness detection result, ultrasonic detection component and eddy current detection component to perform surface crack precision detection on the target ferromagnetic material, and obtain the target surface crack detection report. By performing multi-dimensional surface crack detection on ferromagnetic materials, the accuracy and reliability of surface crack detection of ferromagnetic materials are improved, and the surface crack detection quality of ferromagnetic materials is improved.
[0082] 2. Taking advantage of the unique advantages of metal magnetic memory detection technology in defect location, the surface cracks of the target ferromagnetic material are located through the metal magnetic memory detection component, and the crack roughness detection results are determined, making the surface crack detection of ferromagnetic materials more efficient and accurate.
[0083] 3. By performing ultrasonic detection fitness analysis on the target ferromagnetic material, ultrasonic adaptation instructions / ultrasonic taboo instructions are adaptively generated, thereby improving the fitness and reliability of surface crack detection of ferromagnetic materials.
[0084] Embodiment 2
[0085] Based on the same inventive concept as the method for detecting surface cracks of a ferromagnetic material in the aforementioned embodiment, the present invention also provides a device for detecting surface cracks of a ferromagnetic material, the device comprising a material pre-verification component, a metal magnetic memory detection component, an ultrasonic detection component and an eddy current detection component, please refer to the attached Figure 2 , the device further comprises:
[0086] A surface cleanliness verification module, the surface cleanliness verification module is used to perform surface cleanliness verification on the target ferromagnetic material according to the material pre-verification component to obtain a surface cleanliness verification result;
[0087] An activation module, the activation module being used to activate the metal magnetic memory detection component if the surface cleanliness verification result is passed;
[0088] A crack roughness detection module, the crack roughness detection module is used to perform surface crack roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain a crack roughness detection result;
[0089] A judgment operator acquisition module, the judgment operator acquisition module is used to perform feature recognition according to the crack roughness detection result to obtain a crack existence feature judgment operator;
[0090] A precision detection instruction generation module, the precision detection instruction generation module is used to generate a crack precision detection instruction if the crack existence feature judgment operator is that the target ferromagnetic material has a surface crack;
[0091] A crack precision detection module, the crack precision detection module is used to activate the crack-multi-parameter composite detection submodule based on the crack precision detection instruction, combine the crack roughness detection result, the ultrasonic detection component and the eddy current detection component to perform surface crack precision detection on the target ferromagnetic material, and obtain a target surface crack detection report.
[0092] Furthermore, the device further comprises a verification result generating module to perform the following operation steps:
[0093] According to the material pre-verification component, obtaining material surface state information of the target ferromagnetic material;
[0094] Evaluate the surface cleanliness of the target ferromagnetic material according to the surface state information of the material to obtain the surface cleanliness of the material;
[0095] Determining whether the surface cleanliness of the material is less than a preset cleanliness;
[0096] If the surface cleanliness of the material is less than the preset cleanliness, the surface cleanliness verification result obtained is failure, and a surface cleanliness warning signal is generated.
[0097] Furthermore, the device further comprises a roughness detection result generating module to perform the following operation steps:
[0098] Obtaining crack roughness detection accuracy of the target ferromagnetic material;
[0099] Inputting the crack roughness detection accuracy into a probe lift-off feature table to obtain a detection probe lift-off operator;
[0100] Based on the detection probe lift-off operator, the target ferromagnetic material is detected according to the metal magnetic memory detection component to obtain a material magnetic memory signal detection result;
[0101] A surface crack roughness analysis is performed according to the material magnetic memory signal detection result to generate the crack roughness detection result.
[0102] Furthermore, the device also includes a crack detection report generation module to perform the following operation steps:
[0103] Constructing the crack-multi-parameter composite detection submodule, wherein the crack-multi-parameter composite detection submodule includes an ultrasonic-crack feature analysis channel, an eddy current-crack feature analysis channel and a crack analysis feature comparison channel;
[0104] Divide the target ferromagnetic material into regions according to the crack roughness detection result to obtain a plurality of surface crack partitions;
[0105] Based on the multiple surface crack partitions, the target ferromagnetic material is subjected to precise surface crack detection according to the ultrasonic detection component and the ultrasonic-crack feature analysis channel to obtain ultrasonic-surface crack detection results;
[0106] Based on the multiple surface crack partitions, the target ferromagnetic material is subjected to precise surface crack detection according to the eddy current detection component and the eddy current-crack feature analysis channel to obtain an eddy current-surface crack detection result;
[0107] Inputting the ultrasonic-surface crack detection result and the eddy current-surface crack detection result into the crack analysis feature comparison channel to obtain a crack detection comparison result;
[0108] The target surface crack detection report is generated according to the ultrasonic-surface crack detection result, the eddy current-surface crack detection result and the crack detection comparison result.
[0109] Furthermore, the device further comprises a construction module to perform the following operation steps:
[0110] Obtaining an ultrasonic-crack feature analysis record library, an eddy current-crack feature analysis record library, and a crack analysis feature comparison record library;
[0111] Obtaining a plurality of preset training operators, and randomly selecting the plurality of preset training operators to obtain a first preset training operator, a second preset training operator, and a third preset training operator;
[0112] Performing supervised training on the ultrasonic-crack feature analysis record library based on the first preset training operator to generate the ultrasonic-crack feature analysis channel;
[0113] Performing supervised training on the eddy current-crack feature analysis record library based on the second preset training operator to generate the eddy current-crack feature analysis channel;
[0114] Performing supervised training on the crack analysis feature comparison record library based on the third preset training operator to generate the crack analysis feature comparison channel;
[0115] The ultrasonic-crack feature analysis channel, the eddy current-crack feature analysis channel and the crack analysis feature comparison channel are connected to generate the crack-multi-parameter composite detection submodule.
[0116] Furthermore, the device also includes a crack ultrasonic detection module to perform the following operation steps:
[0117] Performing ultrasonic detection fitness analysis on the target ferromagnetic material and generating ultrasonic fitness instructions;
[0118] According to the ultrasonic adaptation instruction, activating the ultrasonic detection component and the ultrasonic-crack feature analysis channel;
[0119] Performing ultrasonic testing on the plurality of surface crack partitions according to the ultrasonic testing component to obtain partition ultrasonic testing results;
[0120] The partitioned ultrasonic detection results are input into the ultrasonic-crack feature analysis channel to generate the ultrasonic-surface crack detection results.
[0121] Furthermore, the device also includes an ultrasonic detection fitness analysis module to perform the following operation steps:
[0122] Obtaining basic material information of the target ferromagnetic material;
[0123] Obtaining a material ultrasonic attenuation record library, and performing data fusion on the material ultrasonic attenuation record library to generate an ultrasonic attenuation identification domain;
[0124] Inputting the material basic information into the ultrasonic attenuation identification domain to obtain a matching ultrasonic attenuation coefficient;
[0125] Determining whether the matched ultrasonic attenuation coefficient is less than a preset ultrasonic attenuation coefficient;
[0126] If the matched ultrasonic attenuation coefficient is less than the preset ultrasonic attenuation coefficient, obtaining the ultrasonic adaptation instruction;
[0127] If the matched ultrasonic attenuation coefficient is greater than / equal to the preset ultrasonic attenuation coefficient, an ultrasonic taboo instruction is generated, and ultrasonic detection of the target ferromagnetic material is tabooed according to the ultrasonic taboo instruction.
[0128] A device for detecting surface cracks of a ferromagnetic material provided in an embodiment of the present invention can execute a method for detecting surface cracks of a ferromagnetic material provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.
[0129] The modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of the functional modules are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.
[0130] The present application provides a method for detecting surface cracks of ferromagnetic materials, wherein the method is applied to a device for detecting surface cracks of ferromagnetic materials, and the method includes: performing surface cleanliness verification on the target ferromagnetic material through a material pre-verification component to obtain a surface cleanliness verification result; if the surface cleanliness verification result is passed, activating the metal magnetic memory detection component, performing surface crack roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component, and obtaining a crack roughness detection result; performing feature recognition according to the crack roughness detection result to obtain a crack existence feature judgment operator; if the crack existence feature judgment operator is that the target ferromagnetic material has a surface crack, generating a crack precision detection instruction; activating the crack-multi-parameter composite detection submodule based on the crack precision detection instruction, combining the crack roughness detection result, the ultrasonic detection component and the eddy current detection component to perform surface crack precision detection on the target ferromagnetic material, and obtaining a target surface crack detection report. The present application solves the technical problem that the surface crack detection accuracy of ferromagnetic materials in the prior art is poor, which in turn causes the surface crack detection effect of ferromagnetic materials to be poor. By performing multi-dimensional surface crack detection on ferromagnetic materials, the accuracy and reliability of surface crack detection of ferromagnetic materials are improved, and the quality of surface crack detection of ferromagnetic materials is improved.
[0131] Although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention. The scope of the present invention is determined by the scope of the appended claims.
Claims
1. A method for detecting surface cracks of ferromagnetic materials, characterized in that: The method is applied to a device for detecting surface cracks of ferromagnetic materials, the device comprising a material pre-verification component, a metal magnetic memory detection component, an ultrasonic detection component and an eddy current detection component, and the method comprises: Performing surface cleanliness verification on the target ferromagnetic material according to the material pre-verification component to obtain a surface cleanliness verification result; If the surface cleanliness verification result is passed, activating the metal magnetic memory detection component; Performing surface crack and roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain a crack and roughness detection result; Performing feature recognition according to the crack roughness detection result to obtain a crack existence feature judgment operator; If the crack existence characteristic judgment operator indicates that the target ferromagnetic material has a surface crack, a crack precision detection instruction is generated; Based on the crack precision detection instruction, the crack-multi-parameter composite detection submodule is activated, and the surface crack precision detection of the target ferromagnetic material is performed in combination with the crack rough detection result, the ultrasonic detection component and the eddy current detection component to obtain a target surface crack detection report; Performing surface crack roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain a crack roughness detection result, including: Obtaining crack roughness detection accuracy of the target ferromagnetic material; Inputting the crack roughness detection accuracy into a probe lift-off feature table to obtain a detection probe lift-off operator; Based on the detection probe lift-off operator, the target ferromagnetic material is detected according to the metal magnetic memory detection component to obtain a material magnetic memory signal detection result; Performing surface crack roughness analysis according to the material magnetic memory signal detection result to generate the crack roughness detection result; Based on the crack precision detection instruction, the crack-multi-parameter composite detection submodule is activated, and the surface crack precision detection of the target ferromagnetic material is performed in combination with the crack rough detection result, the ultrasonic detection component and the eddy current detection component to obtain a target surface crack detection report, including: Constructing the crack-multi-parameter composite detection submodule, wherein the crack-multi-parameter composite detection submodule includes an ultrasonic-crack feature analysis channel, an eddy current-crack feature analysis channel and a crack analysis feature comparison channel; Divide the target ferromagnetic material into regions according to the crack roughness detection result to obtain a plurality of surface crack partitions; Based on the multiple surface crack partitions, the target ferromagnetic material is subjected to precise surface crack detection according to the ultrasonic detection component and the ultrasonic-crack feature analysis channel to obtain ultrasonic-surface crack detection results; Based on the multiple surface crack partitions, the target ferromagnetic material is subjected to precise surface crack detection according to the eddy current detection component and the eddy current-crack feature analysis channel to obtain an eddy current-surface crack detection result; Inputting the ultrasonic-surface crack detection result and the eddy current-surface crack detection result into the crack analysis feature comparison channel to obtain a crack detection comparison result; Generate the target surface crack detection report according to the ultrasonic-surface crack detection result, the eddy current-surface crack detection result and the crack detection comparison result; Constructing the crack-multi-parameter composite detection submodule includes: Obtaining an ultrasonic-crack feature analysis record library, an eddy current-crack feature analysis record library, and a crack analysis feature comparison record library; Obtaining a plurality of preset training operators, and randomly selecting the plurality of preset training operators to obtain a first preset training operator, a second preset training operator, and a third preset training operator; Performing supervised training on the ultrasonic-crack feature analysis record library based on the first preset training operator to generate the ultrasonic-crack feature analysis channel; Performing supervised training on the eddy current-crack feature analysis record library based on the second preset training operator to generate the eddy current-crack feature analysis channel; Performing supervised training on the crack analysis feature comparison record library based on the third preset training operator to generate the crack analysis feature comparison channel; Connecting the ultrasonic-crack feature analysis channel, the eddy current-crack feature analysis channel and the crack analysis feature comparison channel to generate the crack-multi-parameter composite detection submodule; Based on the multiple surface crack partitions, the target ferromagnetic material is subjected to precise surface crack detection according to the ultrasonic detection component and the ultrasonic-crack feature analysis channel to obtain ultrasonic-surface crack detection results, including: Performing ultrasonic detection fitness analysis on the target ferromagnetic material and generating ultrasonic fitness instructions; According to the ultrasonic adaptation instruction, activating the ultrasonic detection component and the ultrasonic-crack feature analysis channel; Performing ultrasonic testing on the plurality of surface crack partitions according to the ultrasonic testing component to obtain partition ultrasonic testing results; The partitioned ultrasonic detection results are input into the ultrasonic-crack feature analysis channel to generate the ultrasonic-surface crack detection results.
2. The method according to claim 1, characterized in that Performing surface cleanliness verification on the target ferromagnetic material according to the material pre-verification component to obtain a surface cleanliness verification result includes: According to the material pre-verification component, obtaining material surface state information of the target ferromagnetic material; Evaluate the surface cleanliness of the target ferromagnetic material according to the surface state information of the material to obtain the surface cleanliness of the material; Determining whether the surface cleanliness of the material is less than a preset cleanliness; If the surface cleanliness of the material is less than the preset cleanliness, the surface cleanliness verification result obtained is failure, and a surface cleanliness warning signal is generated.
3. The method according to claim 1, characterized in that Performing ultrasonic detection fitness analysis on the target ferromagnetic material includes: Obtaining basic material information of the target ferromagnetic material; Obtaining a material ultrasonic attenuation record library, and performing data fusion on the material ultrasonic attenuation record library to generate an ultrasonic attenuation identification domain; Inputting the material basic information into the ultrasonic attenuation identification domain to obtain a matching ultrasonic attenuation coefficient; Determining whether the matched ultrasonic attenuation coefficient is less than a preset ultrasonic attenuation coefficient; If the matched ultrasonic attenuation coefficient is less than the preset ultrasonic attenuation coefficient, obtaining the ultrasonic adaptation instruction; If the matched ultrasonic attenuation coefficient is greater than / equal to the preset ultrasonic attenuation coefficient, an ultrasonic taboo instruction is generated, and ultrasonic detection of the target ferromagnetic material is tabooed according to the ultrasonic taboo instruction.
4. A device for detecting surface cracks of ferromagnetic materials, characterized in that: The device is used to perform the method according to any one of claims 1 to 3, the device comprises a material pre-verification component, a metal magnetic memory detection component, an ultrasonic detection component and an eddy current detection component, and the device further comprises: A surface cleanliness verification module, the surface cleanliness verification module is used to perform surface cleanliness verification on the target ferromagnetic material according to the material pre-verification component to obtain a surface cleanliness verification result; An activation module, the activation module is used to activate the metal magnetic memory detection component if the surface cleanliness verification result is passed; A crack roughness detection module, the crack roughness detection module is used to perform surface crack roughness detection on the target ferromagnetic material according to the metal magnetic memory detection component to obtain a crack roughness detection result; A judgment operator acquisition module, the judgment operator acquisition module is used to perform feature recognition according to the crack roughness detection result to obtain a crack existence feature judgment operator; A precision detection instruction generation module, the precision detection instruction generation module is used to generate a crack precision detection instruction if the crack existence feature judgment operator is that the target ferromagnetic material has a surface crack; A crack precision detection module, the crack precision detection module is used to activate the crack-multi-parameter composite detection submodule based on the crack precision detection instruction, combine the crack roughness detection result, the ultrasonic detection component and the eddy current detection component to perform surface crack precision detection on the target ferromagnetic material, and obtain a target surface crack detection report.
