Welding quality inspection method and system for an electric tricycle

Abstract

The present invention discloses a welding quality detection method and system for an electric tricycle, relating to the field of non-destructive testing. The method includes: determining the weld structure characteristics of the electric tricycle, excavating the defect types and constructing a weld detection framework; configuring one-step array element focusing parameters, controlling a phased array probe to perform one-step ultrasonic detection to determine a first detection signal; receiving the first detection signal, and based on a quality detection module, performing single-scale feature recognition and determination to locate the directional detection area of the weld detection framework; configuring two-step array element focusing parameters, performing two-step ultrasonic detection under full array element echo reception to determine a second detection signal; receiving the second detection signal, and based on the quality detection module, performing multi-scale feature recognition and determination to determine the welding quality detection result of the electric tricycle. It solves the technical problem of insufficient detection accuracy and efficiency existing in the welding quality detection of existing electric tricycles, and achieves the technical effect of improving the detection accuracy and efficiency.

Description

Technical Field

[0001] The present application relates to the field of non-destructive testing, and in particular to a welding quality testing method and system for an electric tricycle. Background Art

[0002] In the manufacturing process of electric tricycles, welding is a crucial link, and its quality directly affects the safety and service life of the vehicle. However, due to the complexity and diversity of weld structures, traditional welding quality inspection methods, such as visual inspection and magnetic particle inspection, can meet basic quality control needs to a certain extent, but they still have many limitations in terms of complex weld structures, small defect detection, and detection efficiency and accuracy. In particular, in the case of electric tricycles with complex weld structures and diverse defect types, it is difficult to meet the detection needs of high precision and high efficiency at the same time.

[0003] In the current related technologies, the welding quality inspection of electric tricycles has technical problems of insufficient inspection accuracy and efficiency. Summary of the invention

[0004] The present application provides a welding quality inspection method and system for an electric tricycle, which adopts the methods of determining the structural characteristics of the weld, exploring the defect types, constructing a weld inspection architecture, using phased array technology for multi-step ultrasonic detection, preliminarily locating the directional detection area through single-scale feature recognition and judgment, and then combining multi-scale feature recognition and judgment for precise detection. Through the combination of phased array ultrasonic technology and multi-scale feature recognition technology, defects in the weld, including tiny defects and deep defects, are accurately identified, the risk of missed detection and misjudgment is reduced, and the technical effect of improving detection accuracy and efficiency is achieved.

[0005] The present application provides a welding quality detection method for an electric tricycle, comprising:

[0006] Determine the weld structural features of the electric tricycle, explore the defect types and construct a weld detection architecture, wherein the weld structural features are determined according to the weld material, thickness and welding method; for the weld detection architecture, configure a one-step array element focusing parameter, control the phased array probe to perform a one-step ultrasonic detection, and determine a first detection signal, wherein the phased array probe has multiple array elements built in; receive the first detection signal, perform single-scale feature recognition and judgment based on a quality detection module, and locate the directional detection area of ​​the weld detection architecture, wherein the quality detection module is a multi-scale pyramid structure; for the directional detection area, configure a two-step array element focusing parameter, perform a two-step ultrasonic detection under full-array element echo reception, and determine a second detection signal; receive the second detection signal, perform multi-scale feature recognition and judgment based on the quality detection module, and determine the welding quality detection result of the electric tricycle.

[0007] In a possible implementation, for configuring the one-step array element focusing parameters, the following processing is performed:

[0008] Traverse the weld detection architecture to determine the first structural partition; for the first structural partition, configure the first focusing parameters of the first array element, where the first focusing parameters define the directional ultrasonic beam and the scanning mode, the directional ultrasonic beam defines the excitation time and phase definition, and the scanning mode is a linear scan; determine the Nth focusing parameters of the Nth structural partition within the weld detection architecture; integrate the first focusing parameters to the Nth focusing parameters, and establish an array element mapping to determine the one-step array element focusing parameters.

[0009] In a possible implementation, for configuring the two-step array element focusing parameters, the following processing is performed:

[0010] Traverse the directional detection area to determine the azimuth constraint parameters, where the azimuth constraint parameters are determined based on the area size, shape, and depth; traverse the one-step array element focusing parameters, and determine the matching focusing parameters through structural partition matching; fuse the azimuth constraint parameters and the matching focusing parameters to determine the two-step array element focusing parameters, where the two-step array element focusing parameters correspond one-to-one with the directional detection area.

[0011] In a possible implementation, the following processing is performed:

[0012] The one-step ultrasonic detection is a one-to-one reception of the array element and the echo signal, and the two-step ultrasonic detection is a full-array element reception under a one-to-many relationship between the array element and the echo signal.

[0013] In a possible implementation, for mining the defect types and constructing the weld detection architecture, the following processing is performed:

[0014] Traverse the weld structure features, perform similarity partitioning, and identify the structural partitions; for the structural partitions, mine the defect types under the partitioned weld structure features to determine the defect types, where the defect types are partition existence defects; establish the mapping between the defect types and the structural partitions as the weld detection architecture.

[0015] In a possible implementation, for performing multi-scale feature recognition and determination, the following processing is performed:

[0016] Identify the detection signals in the first directional detection area, identify the signal features and cross-check them to determine the first feature part and the second feature part, where the first feature part is the consistent part under full-array element reception, and the second feature part is the different part under full-array element reception; for the first feature part and the second feature part, perform multi-scale recognition determination to determine the welding quality detection result of the first directional detection area.

[0017] In a possible implementation, for the first feature part and the second feature part, multi-scale recognition and determination are performed, and the following processing is executed:

[0018] For the first feature part, based on the feature resolution, scale amplification and recognition are performed to determine a detection result; for the second feature part, ratio screening and scale amplification fusion are performed to determine two detection results, where screening is performed at a preset ratio based on the full-array element received signal; the one detection result and the two detection results are integrated as the welding quality detection result of the first directional detection area.

[0019] This application also provides a welding quality detection system for an electric tricycle, including:

[0020] A weld detection architecture construction module, which is used to determine the weld structure characteristics of the electric tricycle, excavate the defect types and construct a weld detection architecture, where the weld structure characteristics are determined according to the weld material, thickness and welding method; a one-step ultrasonic detection module, which is used to configure one-step array element focusing parameters for the weld detection architecture, control a phased array probe to perform one-step ultrasonic detection, and determine a first detection signal, and the phased array probe is internally provided with a plurality of array elements; a directional detection area positioning module, which is used to receive the first detection signal, perform single-scale feature recognition and determination based on a quality detection module, and locate the directional detection area of the weld detection architecture, where the quality detection module is a multi-scale pyramid structure; a two-step ultrasonic detection module, which is used to configure two-step array element focusing parameters for the directional detection area, perform two-step ultrasonic detection under full-array element echo reception, and determine a second detection signal; a welding quality detection result determination module, which is used to receive the second detection signal, perform multi-scale feature recognition and determination based on the quality detection module, and determine the welding quality detection result of the electric tricycle.

[0021] A welding quality detection method and system for an electric tricycle proposed in this application first determines the weld structure characteristics of the electric tricycle, excavates the defect types, and constructs a weld detection framework. Among them, the weld structure characteristics are determined according to the weld material, thickness, and welding method. Then, for the weld detection framework, one-step array element focusing parameters are configured, and a phased array probe is controlled to perform one-step ultrasonic detection to determine the first detection signal. The phased array probe has multiple array elements built in. Then, the first detection signal is received, and based on the quality detection module, single-scale feature recognition and determination are performed to locate the directional detection area of the weld detection framework. Among them, the quality detection module is a multi-scale pyramid structure. Furthermore, for the directional detection area, two-step array element focusing parameters are configured, and two-step ultrasonic detection under full array element echo reception is performed to determine the second detection signal. Finally, the second detection signal is received, and multi-scale feature recognition and determination are performed based on the quality detection module to determine the welding quality detection result of the electric tricycle, achieving the technical effect of improving the detection accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] 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 will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0023] Figure 1 It is a schematic flowchart of a welding quality detection method for an electric tricycle provided by an embodiment of the present application.

[0024] Figure 2 It is a schematic structural diagram of a welding quality detection system for an electric tricycle provided by an embodiment of the present application.

[0025] Description of the reference numerals: Weld detection framework construction module 10, one-step ultrasonic detection module 20, directional detection area positioning module 30, two-step ultrasonic detection module 40, welding quality detection result determination module 50. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] The above description is only an overview of the technical solutions of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically illustrates the specific embodiments of this application.

[0027] To make the objectives, technical solutions and advantages of this application clearer, the following will further describe this application in detail with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application.

[0028] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first / second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0029] An embodiment of this application provides a method for detecting the welding quality of an electric tricycle, as Figure 1 shown. The method includes:

[0030] Step S100, determining the weld structure characteristics of the electric tricycle, excavating the defect types and constructing a weld detection framework, wherein the weld structure characteristics are determined according to the weld material, thickness and welding method. Specifically, collect basic information such as the material, thickness and welding method of the electric tricycle welds. According to the collected information, analyze the geometric shape, material properties and welding process of the welds to determine the weld structure characteristics. Based on the weld structure characteristics and combined with historical data, excavate the possible defect types, that is, the possible quality problems in the welds, such as cracks, slag inclusions, lack of fusion, etc. According to the excavated defect types and weld structure characteristics, construct a non-destructive testing framework applicable to the weld, including the detection path, detection parameters, defect identification method, etc.

[0031] In a possible implementation manner, for the step of mining defect types and constructing a weld detection architecture, step S100 further includes step S110 of traversing the weld structure features, performing similarity partitioning, and identifying structural partitions. Specifically, detailed structural information of the electric tricycle welds is comprehensively collected, including the material, thickness, shape, welding method, etc. of the welds. The collected weld structure information is analyzed to identify different structural features in the welds, such as the starting and ending positions of the welds, the degree of bending of the welds, the intersection points of the welds, etc. According to the structural features of the welds, the welds are divided into multiple regions with similarity, that is, structural partitions, and these regions have certain similarities in weld structure, material, or welding method. A unique identifier is assigned to each structural partition to enable accurate identification in the subsequent defect mining and detection processes. Step S120 is to perform defect type mining under the weld structure features of the partition for the structural partition to determine the defect type, where the defect type is a partition existence defect. Specifically, each structural partition is analyzed to identify the weld structure features within the partition, such as the geometric shape, material composition, welding process, etc. of the welds. Based on the weld structure features within the partition and combined with historical data, possible defect types are mined, and these defect types are existence defects closely related to the partition structure features, such as cracks, slag inclusions, lack of fusion, etc. The mined defect types are sorted out and summarized to ensure that each structural partition has a clear list of defect types. Step S130 is to establish a mapping between the defect type and the structural partition as the weld detection architecture. Specifically, a mapping relationship is established between each structural partition and its corresponding defect type, that is, the possible defect types in each structural partition are determined. Based on the mapping relationship between the defect type and the structural partition, a non-destructive testing architecture applicable to the weld is constructed, and this architecture is used to guide the subsequent ultrasonic detection and defect identification processes to ensure that defects in the weld can be accurately and efficiently detected. This implementation manner divides the weld into structural partitions with similarity, and performs defect type mining and identification for each partition, greatly reducing unnecessary detection work and improving the detection efficiency. At the same time, establishing a dedicated list of defect types for each structural partition can more accurately identify defects in the weld, reduce the possibility of misjudgment and missed judgment, and enhance the accuracy of detection.

[0032] Step S200: For the weld detection architecture, configure a one-step array element focusing parameter, control the phased array probe to perform one-step ultrasonic detection, and determine the first detection signal. The phased array probe has multiple array elements built in. Specifically, according to the weld detection architecture, configure the focusing parameter for each array element, including excitation time, phase limitation, etc., to form a directional ultrasonic beam. Using the configured focusing parameter, control the phased array probe to perform ultrasonic detection. The ultrasonic echo signal received by the phased array probe is the first detection signal. Among them, the phased array probe is a device that can dynamically control the direction and shape of the ultrasonic beam, has multiple array elements built in, and realizes the focusing and scanning of the ultrasonic beam by adjusting the excitation time and phase of each array element.

[0033] In a possible implementation, the step of configuring the one-step array element focusing parameters, step S200 further includes step S210 of traversing the weld detection architecture to determine the first structural partition. Specifically, traverse the constructed weld detection architecture, which is constructed based on the weld structural features and the excavated defect types, and visit each part of the architecture one by one according to a certain logical order (such as from front to back, from top to bottom, etc.). First, determine and select the first structural partition, which is a basic unit in the weld detection architecture and represents a specific area or feature of the weld. Step S220, for the first structural partition, configure the first focusing parameters of the first array element, where the first focusing parameters define the directional ultrasonic beam and the scanning mode, the directional ultrasonic beam defines the excitation time and phase definition, and the scanning mode is a linear scan. Specifically, for the first structural partition, configure the focusing parameters for the corresponding first array element in the phased array probe. These focusing parameters include the excitation time, phase definition of the directional ultrasonic beam (an ultrasonic beam with a specific direction and shape formed by focusing and scanning of the phased array probe), and the scanning mode (linear scan, the ultrasonic beam moves in a straight line direction in the weld). Among them, the excitation time and phase definition of the directional ultrasonic beam determine the direction and shape of the ultrasonic beam, and the scanning mode determines the movement path of the ultrasonic beam in the weld. Step S230, determine the Nth focusing parameters of the Nth structural partition in the weld detection architecture. Specifically, continue to traverse the weld detection architecture and configure the corresponding focusing parameters for each structural partition. This step is a repetition of step 220, but for different structural partitions and array elements. When traversing to the Nth structural partition, configure the focusing parameters for the Nth array element of this partition. Step S240, integrate the first focusing parameters to the Nth focusing parameters and establish an array element mapping to determine the one-step array element focusing parameters. Specifically, integrate the focusing parameters configured for each structural partition and array element. These parameters together constitute the one-step array element focusing parameters for controlling the phased array probe to perform one-step ultrasonic detection. At the same time, establish an array element mapping, that is, record the relationship between each array element and its corresponding focusing parameters. This implementation method can more effectively detect defects in the weld by configuring specific focusing parameters for each structural partition and array element, improving the pertinence of detection. At the same time, by integrating these parameters and establishing an array element mapping, it can more efficiently control the phased array probe to perform ultrasonic detection, thereby improving the detection efficiency and accuracy.

[0034] Step S300: Receive the first detection signal, and based on the quality detection module, perform single-scale feature recognition and determination to locate the directional detection area of the weld detection architecture. Herein, the quality detection module is a multi-scale pyramid structure. Specifically, receive the initial ultrasonic echo signal from the phased array probe, and use the quality detection module (the module for feature extraction, recognition, and determination of the detection signal) to perform feature extraction and recognition on the first detection signal. Among them, single-scale (i.e., fixed scale) is used for feature recognition. According to the recognized features, determine the possible defects in the weld and locate the directional detection area where these defects are located. That is, the directional detection area is the weld area that needs to be further detected according to the preliminary determination result.

[0035] Step S400: For the directional detection area, configure two-step array element focusing parameters, perform two-step ultrasonic detection under full array element echo reception, and determine the second detection signal. Specifically, according to the characteristics of the directional detection area, reconfigure the focusing parameters of the array elements to improve the detection accuracy. In the two-step ultrasonic detection, all array elements participate in the reception of the echo signal, and the ultrasonic echo signal received by all array elements is the second detection signal, which is used for further feature recognition and determination.

[0036] In a possible implementation, for configuring the two-step array element focusing parameters, step S400 further includes step S410 of traversing the directional detection area to determine the azimuth constraint parameters, where the azimuth constraint parameters are determined based on the area size, shape, and depth. Specifically, traverse the directional detection areas determined in step S300, and each directional detection area represents a specific position or range where defects may exist in the weld. Analyze each area, and based on its characteristics such as size, shape, and depth, determine the azimuth constraint parameters, which are used to limit the direction and range of the ultrasonic beam in the subsequent two-step ultrasonic detection to ensure that the ultrasonic wave can accurately cover and detect potential defects. Step S420 is to traverse the one-step array element focusing parameters and determine the matching focusing parameters through structural partition matching. Specifically, traverse the one-step array element focusing parameters, which have configured specific focusing parameters for each structural partition in the weld detection architecture. By comparing the directional detection area with the structural partitions in the weld detection architecture, find the structural partition that best matches the directional detection area. Extract the focusing parameters corresponding to these structural partitions from the one-step array element focusing parameters as the matching focusing parameters. Step S430 is to fuse the azimuth constraint parameters and the matching focusing parameters to determine the two-step array element focusing parameters, where the two-step array element focusing parameters correspond one-to-one with the directional detection areas. Specifically, fuse the azimuth constraint parameters determined in step 410 and the matching focusing parameters determined in step 420. Fusing means adjusting and optimizing the two parameter sets to ensure that they can work together to provide accurate focusing parameters for the two-step ultrasonic detection. Finally, generate two-step array element focusing parameters corresponding one-to-one with each directional detection area for controlling the phased array probe to perform more accurate ultrasonic detection. This implementation method accurately controls the direction and range of the ultrasonic beam by traversing the directional detection area and determining the azimuth constraint parameters, reducing unnecessary detections and false alarms. At the same time, by traversing the one-step array element focusing parameters and performing structural partition matching, the focusing parameters that best match the directional detection area are found, further improving the detection accuracy.

[0037] In a possible implementation, the one-step ultrasonic detection is one-to-one reception of array elements and echo signals, and the two-step ultrasonic detection is full-array element reception under one-to-many of array elements and echo signals.

[0038] Specifically, in one-step ultrasonic detection, each element of the phased array probe sequentially acts as a transmitter to emit an ultrasonic beam, and the reception is completed by a single element corresponding to the transmitting element (usually the transmitting element itself, but in some configurations it can be an adjacent or specific other element). This one-to-one transmit-receive mode is used to preliminarily locate abnormal areas in the weld. The transmitting element emits an ultrasonic beam according to preset focusing parameters (such as excitation time, phase limitation, and scanning mode), and the received echo signal (i.e., the signal reflected back after the ultrasonic beam encounters the weld structure or defect) is used to generate the first detection signal.

[0039] After determining the directional detection area, two-step ultrasonic detection is carried out. In this step, all elements of the phased array probe participate in the transmission, but the reception is completed simultaneously by all elements, that is, full-element reception. When the transmitting elements emit ultrasonic beams, all elements act as receivers to receive echo signals from the weld structure and defects. This one-to-many transmit-receive mode provides rich data for precisely quantifying the characteristics of the weld, such as the size, shape, and location of the defects. The multiple received echo signals are used to generate the second detection signal to obtain the final welding quality detection result. This implementation method quickly narrows down the area where defects may exist in the weld through the preliminary positioning of one-step ultrasonic detection, reducing unnecessary detection work. The full-element reception mode of two-step ultrasonic detection is used to provide rich data, enabling more precise analysis of the characteristics and defects of the weld, thereby improving the accuracy of detection.

[0040] Step S500: Receive the second detection signal, perform multi-scale feature recognition and determination based on the quality detection module, and determine the welding quality detection result of the electric tricycle. Specifically, receive the ultrasonic echo signals from all elements, and use the quality detection module to perform feature extraction and recognition on the second detection signal at multiple scales, that is, perform feature extraction and recognition on the detection signal at different scales to improve the accuracy and comprehensiveness of recognition. According to the recognized features, combined with the preset determination criteria, determine the welding quality detection result of the electric tricycle. The embodiments of the present application adopt technical means such as determining the weld structure characteristics, excavating the defect types, constructing a weld detection architecture, using phased array technology for multi-step ultrasonic detection, preliminarily locating the directional detection area through single-scale feature recognition and determination, and then combining multi-scale feature recognition and determination for precise detection. Through the combination of phased array ultrasonic technology and multi-scale feature recognition technology, the defects in the weld, including micro-defects and deep defects, are accurately identified, the risks of missed detection and misjudgment are reduced, and the technical effects of improving the detection accuracy and efficiency are achieved.

[0041] In a possible implementation, for the multi-scale feature recognition and determination, step S500 further includes step S510 of recognizing the detection signals in the first directional detection area, identifying signal features and cross-checking them with each other to determine the first feature part and the second feature part, where the first feature part is the consistent part under full-array element reception, and the second feature part is the part with differences under full-array element reception. Specifically, in two-step ultrasonic detection, full-array elements receive echo signals from the weld, and these signals contain information about the weld structure and potential defects. For the first directional detection area, features are extracted from the received detection signals, including the amplitude, phase, frequency, arrival time, etc. of the signals, which reflect the physical characteristics of the weld. The extracted features are cross-checked in the data received by the full-array elements. Since full-array elements receive signals, each element will receive echo signals from the same weld area. Therefore, the consistency and differences can be verified by comparing the signal features received by different elements. According to the cross-check results, the features are classified into two categories: the first feature part (the consistent part under full-array element reception) and the second feature part (the part with differences under full-array element reception). The first feature part represents the uniform and defect-free part of the weld, and the second feature part indicates abnormalities or defects in the weld.

[0042] Step S520, for the first feature part and the second feature part, perform multi-scale recognition and determination to determine the welding quality detection result of the first directional detection area. Specifically, for the first feature part and the second feature part, a quality detection module with a multi-scale pyramid structure is used for further analysis. Multi-scale analysis can capture feature information at different scales, thereby more comprehensively evaluating the quality of the weld. Based on the results of multi-scale analysis, the first feature part and the second feature part are recognized and determined. For the first feature part, if its features remain highly consistent under full-array element reception, it indicates that the weld quality in this area is good. For the second feature part, further analyze whether the feature differences are caused by defects, and determine the type, size, and location of the defects. Integrate the welding quality detection results of the first directional detection area to form a final detection report, including information such as the quality status of the weld, the type, size, and location of the defects. This implementation captures rich information in the weld, including the uniform part and potential defect part, through full-array element reception and two-step ultrasonic detection, and further verifies and refines this information through cross-checking and multi-scale analysis, thereby more accurately evaluating the quality of the weld and improving the comprehensiveness and accuracy of the detection.

[0043] In a possible implementation, for the first feature part and the second feature part, multi-scale recognition and determination are performed. Step S520 further includes step S521. For the first feature part, based on the feature resolution, scale amplification and recognition are performed to determine a detection result. Specifically, the feature resolution of the first feature part is determined. The feature resolution refers to the smallest unit or scale that can distinguish different features. In the first feature part, since it is the consistent part under full-array element reception, the feature resolution is relatively high, that is, the difference between features is small. In order to more accurately identify and analyze these subtle feature differences, the first feature part is scale-amplified by increasing the sampling rate, improving the signal processing accuracy, etc. After scale amplification, the multi-scale pyramid structure in the quality detection module is used to perform feature recognition on the first feature part, including recognizing the shape, size, position of the feature and its relationship with other features, etc. Based on the result of feature recognition, the welding quality detection result of the first feature part is determined. If the difference between features is within the acceptable range, it indicates that the weld quality in this area is good; if the difference exceeds the acceptable range, it indicates that there are potential quality problems. Step S522, for the second feature part, ratio screening and scale amplification fusion are performed to determine two detection results, where screening is performed according to a preset ratio based on the full-array element received signal. Specifically, in the second feature part, since it is the distinguishable part under full-array element reception and there may be multiple echo signals, in order to reduce noise and interference, first, a ratio is preset based on the full-array element received signal for screening, which is used to screen out the discrete parts with low probability. For the remaining signal part after screening, signal amplification is performed to highlight its features, and feature fusion is performed to integrate the information between different signals. Feature fusion can be achieved through methods such as weighted average, maximum / minimum value selection, and feature vector splicing. Based on the result of signal amplification and fusion, the welding quality detection result of the second feature part is determined. If the fused features indicate obvious abnormalities or defects, it indicates that the weld quality in this area is poor. Step S523, integrate the one detection result and the two detection results as the welding quality detection result of the first directional detection area. Specifically, the two detection results obtained in step 521 and step 522 are integrated, and based on the integrated result, the welding quality detection result of the first directional detection area is formed, which is used for subsequent weld quality evaluation and improvement. This implementation method screens signals according to a preset ratio, reducing misjudgment and missed detection caused by noise or interference, ensuring that only the signals most likely to represent the true features are retained for subsequent analysis. Amplifying and fusing the screened signals enhances the visibility and recognizability of features, which helps to discover potential defects that may be ignored due to weak signals, improving the accuracy and reliability of detection.

[0044] In the foregoing, with reference to Figure 1A welding quality detection method for an electric tricycle according to an embodiment of the present invention is described in detail. Next, reference will be made to Figure 2 Describe a welding quality detection system for an electric tricycle according to an embodiment of the present invention.

[0045] A welding quality detection system for an electric tricycle according to an embodiment of the present invention is used to solve the technical problems of insufficient detection accuracy and efficiency in the existing welding quality detection of electric tricycles, and achieve the technical effects of improving detection accuracy and efficiency. A welding quality detection system for an electric tricycle includes: a weld detection architecture construction module 10, a one-step ultrasonic detection module 20, a directional detection area positioning module 30, a two-step ultrasonic detection module 40, and a welding quality detection result determination module 50.

[0046] The weld detection architecture construction module 10 is used to determine the weld structure characteristics of the electric tricycle, excavate the defect types and construct the weld detection architecture, wherein the weld structure characteristics are determined according to the weld material, thickness and welding method; the one-step ultrasonic detection module 20 is used to configure the one-step element focusing parameters for the weld detection architecture, control the phased array probe to perform one-step ultrasonic detection, and determine the first detection signal, and a plurality of elements are built in the phased array probe; the directional detection area positioning module 30 is used to receive the first detection signal, perform single-scale feature recognition and determination based on the quality detection module, and position the directional detection area of the weld detection architecture, wherein the quality detection module is a multi-scale pyramid structure; the two-step ultrasonic detection module 40 is used to configure the two-step element focusing parameters for the directional detection area, perform two-step ultrasonic detection under the full-element echo reception, and determine the second detection signal; the welding quality detection result determination module 50 is used to receive the second detection signal, perform multi-scale feature recognition and determination based on the quality detection module, and determine the welding quality detection result of the electric tricycle.

[0047] Next, the specific configuration of the one-step ultrasonic detection module 20 will be described in detail. As described above, by configuring the one-step element focusing parameters, the one-step ultrasonic detection module 20 may further include: a first structure partition determination unit for traversing the weld detection architecture to determine the first structure partition; a first focusing parameter configuration unit for configuring the first focusing parameters of the first element for the first structure partition, wherein the first focusing parameters define the directional ultrasonic beam and the scanning method, the directional ultrasonic beam defines the excitation time and phase definition, and the scanning method is a linear scanning; an Nth focusing parameter determination unit for determining the Nth focusing parameter of the Nth structure partition in the weld detection architecture; a one-step element focusing parameter determination unit for integrating the first focusing parameter to the Nth focusing parameter, and establishing an element mapping to determine the one-step element focusing parameters.

[0048] Next, the specific configuration of the two-step ultrasonic detection module 40 will be described in detail. As described above, by configuring the two-step array element focusing parameters, the two-step ultrasonic detection module 40 may further include: an azimuth constraint parameter determination unit for traversing the directional detection area to determine the azimuth constraint parameters, where the azimuth constraint parameters are determined based on the area size, shape, and depth; a structure partition matching unit for traversing the one-step array element focusing parameters and determining the matching focusing parameters through structure partition matching; and a two-step array element focusing parameter determination unit for fusing the azimuth constraint parameters and the matching focusing parameters to determine the two-step array element focusing parameters, where the two-step array element focusing parameters correspond one-to-one with the directional detection area.

[0049] Among them, the system may further include: the one-step ultrasonic detection is a one-to-one reception of the array element and the echo signal, and the two-step ultrasonic detection is a full-array element reception under a one-to-many relationship between the array element and the echo signal.

[0050] Next, the specific configuration of the weld detection architecture construction module 10 will be described in detail. As described above, by mining the defect types and constructing the weld detection architecture, the weld detection architecture construction module 10 may further include: a similarity partition unit for traversing the weld structure features, performing similarity partitioning, and identifying the structure partitions; a defect type mining unit for mining the defect types under the weld structure features of the partitions for the structure partitions to determine the defect types, where the defect types are partition existence defects; and a mapping establishment unit for establishing the mapping between the defect types and the structure partitions as the weld detection architecture.

[0051] Next, the specific configuration of the welding quality detection result determination module 50 will be described in detail. As described above, by performing multi-scale feature recognition and determination, the welding quality detection result determination module 50 may further include: a feature part determination unit for identifying the detection signals in the first directional detection area, identifying the signal features and cross-checking each other to determine the first feature part and the second feature part, where the first feature part is the consistent part under full-array element reception, and the second feature part is the part with differences under full-array element reception; and a multi-scale recognition determination unit for performing multi-scale recognition determination on the first feature part and the second feature part to determine the welding quality detection result of the first directional detection area.

[0052] Among them, for the first feature part and the second feature part, multi-scale recognition and determination are performed. The multi-scale recognition and determination unit may further include: a scale amplification and recognition sub-unit for performing scale amplification and recognition on the first feature part based on the feature resolution to determine a detection result; a ratio screening and scale amplification fusion sub-unit for performing ratio screening and scale amplification fusion on the second feature part to determine two detection results, where screening is performed based on a preset ratio with respect to the full-array element received signal; and a detection result integration sub-unit for integrating the one detection result and the two detection results as the welding quality detection result of the first directional detection area.

[0053] The welding quality detection system for an electric tricycle provided by an embodiment of the present invention can execute the welding quality detection method for an electric tricycle provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0054] Although various references are made to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on a user terminal and / or a server. The various units and 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 achieved; in addition, the specific names of the functional units are only for facilitating mutual distinction and do not limit the protection scope of the present invention.

[0055] The above specific embodiments do not constitute a limitation to the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be executed in a different order from that in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In certain embodiments, multi-tasking and parallel processing are also possible or may be advantageous.

Claims

1. A welding quality detection method for an electric tricycle, characterized in that: The method comprises: Determine the weld structural features of the electric tricycle, explore the defect types and build a weld detection framework, wherein the weld structural features are determined according to the weld material, thickness and welding method; According to the weld detection architecture, a step-array element focusing parameter is configured to control a phased array probe to perform a step of ultrasonic detection to determine a first detection signal, wherein the phased array probe has a plurality of array elements built therein; receiving the first detection signal, performing single-scale feature recognition and determination based on a quality detection module, and locating a directional detection area of ​​the weld detection architecture, wherein the quality detection module is a multi-scale pyramid structure; For the directional detection area, configure two-step array element focusing parameters, perform two-step ultrasonic detection under full array element echo reception, and determine a second detection signal; receiving the second detection signal, performing multi-scale feature recognition and determination based on the quality detection module, and determining a welding quality detection result of the electric tricycle; The step of configuring the focusing parameters of the one-step array element includes: Traversing the weld detection framework to determine a first structural partition; For the first structural partition, configure a first focusing parameter of the first array element, wherein the first focusing parameter defines a directional ultrasonic beam and a scanning mode, the directional ultrasonic beam defines an excitation time and a phase, and the scanning mode is a linear scan; Determining an Nth focusing parameter of an Nth structural partition within the weld inspection architecture; Integrate the first focusing parameter up to the Nth focusing parameter, establish array element mapping, and determine the one-step array element focusing parameter; The configuring of two-step array element focusing parameters includes: Traversing the directional detection area to determine an orientation constraint parameter, wherein the orientation constraint parameter is determined based on the size, shape and depth of the area; Traversing the one-step array element focusing parameters, and determining matching focusing parameters through structural partition matching; Fusion of the orientation constraint parameter and the matching focus parameter to determine the two-step array element focus parameter, wherein the two-step array element focus parameter corresponds to the directional detection area one by one; The method of mining defect types and building a weld inspection architecture includes: Traversing the weld structural features, performing similarity partitioning, and identifying structural partitions; For the structural partition, the defect type mining under the structural characteristics of the partition weld is performed to determine the defect type, wherein the defect type is a partition existence defect; A mapping between the defect type and the structural partition is established as the weld detection framework.

2. A welding quality detection method for an electric tricycle as claimed in claim 1, characterized in that: The one-step ultrasonic detection is a one-to-one reception of array elements and echo signals, and the two-step ultrasonic detection is a full-array element reception under a one-to-many relationship between array elements and echo signals.

3. The welding quality detection method of an electric tricycle as claimed in claim 1, characterized in that: The multi-scale feature recognition and determination includes: Identify the detection signal of the first directional detection area, identify the signal characteristics and verify them with each other, and determine the first characteristic part and the second characteristic part, wherein the first characteristic part is the consistent part under the reception of all array elements, and the second characteristic part is the different part under the reception of all array elements; A multi-scale recognition and determination is performed on the first characteristic portion and the second characteristic portion to determine a welding quality detection result of a first directional detection area.

4. A welding quality detection method for an electric tricycle as claimed in claim 3, characterized in that: Performing multi-scale recognition and determination on the first characteristic portion and the second characteristic portion includes: For the first characteristic part, scale magnification and identification are performed based on the characteristic resolution to determine a detection result; For the second characteristic part, ratio screening and scale amplification fusion are performed to determine two detection results, wherein the screening is performed based on a preset ratio based on the received signals of all array elements; The one detection result and the two detection results are integrated as the welding quality detection result of the first directional detection area.

5. A welding quality inspection system for an electric tricycle, characterized in that: The system is used to implement the welding quality detection method of an electric tricycle according to any one of claims 1 to 4, and the system comprises: A weld detection framework building module, wherein the weld detection framework building module is used to determine the weld structural characteristics of the electric tricycle, mine defect types and build a weld detection framework, wherein the weld structural characteristics are determined according to the weld material, thickness and welding method; A one-step ultrasonic detection module, wherein the one-step ultrasonic detection module is used to configure a one-step array element focusing parameter for the weld detection architecture, control a phased array probe to perform one-step ultrasonic detection, and determine a first detection signal, wherein the phased array probe has a plurality of array elements built therein; A directional detection area positioning module, the directional detection area positioning module is used to receive the first detection signal, perform single-scale feature recognition and determination based on the quality detection module, and locate the directional detection area of ​​the weld detection architecture, wherein the quality detection module is a multi-scale pyramid structure; A two-step ultrasonic detection module, the two-step ultrasonic detection module is used to configure the two-step array element focusing parameters for the directional detection area, perform two-step ultrasonic detection under full array element echo reception, and determine a second detection signal; A welding quality detection result determination module is used to receive the second detection signal, perform multi-scale feature recognition and judgment based on the quality detection module, and determine the welding quality detection result of the electric tricycle.

Images