Real-time detection method and system for arc welding defects

Abstract

This invention discloses a real-time detection method and system for welding defects in arc welding. The method includes: Step 1, collecting a set of welding voltage data during the arc welding process; Step 2, calculating at least one feature value of the welding voltage data set using probability distribution curves; Step 3, determining whether the feature value is less than a preset threshold. If any feature value is less than the preset threshold, a welding defect exists in the weld corresponding to the welding voltage data set, and a real-time alarm is issued. If all feature values ​​are not less than the preset threshold, steps 1 to 2 are repeated. Its advantages are: by collecting the welding voltage during the welding process in real time using a voltage sensor, and performing distribution curve fitting and feature value extraction on the welding voltage data, real-time detection and early warning of welding defects are achieved based on a threshold alarm method, enabling automated operation of the detection process.

Description

[Technical Field]

[0001] This invention relates to the field of real-time detection technology for welding defects, specifically, to a method and system for real-time detection of welding defects in arc welding. [Background Technology]

[0002] Welding is an important process in modern manufacturing. Among various welding methods, arc welding is widely used and is one of the most common processes, accounting for about 60% of the total welding production. Its working principle is to convert electrical energy into heat energy through an electric arc and supply it for welding, thereby connecting metals. Specifically, arc welding is a fusion welding method that uses an electric arc as a heat source, also known as arc welding. According to the process, it can be roughly divided into: gas tungsten inert gas welding (GTAW), gas metal arc welding (GMAW), submerged arc welding (SAW), etc. Arc welding has the following characteristics: (1) Arc welding forms an electric arc between the metal welding wire and the welding material to melt the metal and complete the welding; (2) The welding area usually needs to be protected by a shielding gas. Commonly used shielding gases include argon, carbon dioxide, etc.; (3) Weld porosity is a relatively common welding defect in arc welding. In the process of arc welding, welding defects are a primary problem that affects the quality of the weld and even the overall performance of the workpiece. Real-time detection of welding defects is crucial.

[0003] Currently, welding defect detection methods for arc welding primarily rely on post-weld inspection. Post-weld inspection equipment, such as ultrasonic flaw detectors, is relatively expensive and cannot meet the real-time and predictive production requirements of Industry 4.0, while also incurring high labor costs. Although some processes utilize real-time detection solutions using laser sensors or infrared thermal imagers, these solutions still suffer from high costs and insufficient anti-interference capabilities. Therefore, to enable the widespread adoption of real-time detection technology in arc welding defect detection, a low-cost and effective defect detection solution is a key challenge in this field. [Summary of the Invention]

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing a real-time detection method and system for arc welding defects. This system uses a voltage sensor to collect welding voltage data in real time during the welding process, performs distribution curve fitting and feature value extraction on the welding voltage data, and implements real-time detection and early warning of welding defects based on a threshold alarm method. This enables automated operation of the detection process, improves the accuracy and real-time performance of welding defect detection, and saves time and labor costs.

[0005] Other features and advantages of the invention will become apparent from the following detailed description, or may be learned in part by practice of the invention.

[0006] In a first aspect, the present invention provides a method for real-time detection of welding defects in arc welding, comprising the following steps:

[0007] Step 1: Collect a set of welding voltage data during the arc welding process;

[0008] Step 2: Calculate at least one characteristic value for each of the welding voltage data sets using probability distribution curves;

[0009] Step 3: Determine whether the feature value is less than a preset threshold. If any feature value is less than the preset threshold, the weld seam corresponding to the welding voltage data group has a welding defect, and a real-time alarm is issued. If all feature values ​​are not less than the preset threshold, repeat steps 1 to 2.

[0010] Preferably, the feature value includes at least one of the following: skewness, kurtosis, or low-frequency spectral density of the probability distribution curve.

[0011] Preferably, step two further includes constructing a probability distribution curve for the welding voltage data, comprising the following steps:

[0012] Sort a set of welding voltage data in ascending order of value;

[0013] The sorted data set is divided into several numerical intervals at equal intervals according to the range of data values.

[0014] The frequency of welding voltage data within each of the aforementioned numerical intervals was counted sequentially.

[0015] Preferably, the method for calculating the preset threshold includes:

[0016] Collect several sets of welding voltage data for known normal, defect-free welds during the arc welding process;

[0017] The data in each group of welding voltage data are sorted in ascending order and divided into several numerical intervals at equal intervals.

[0018] The frequency of welding voltage data within each numerical interval is counted sequentially, and a probability distribution curve is constructed for each group of welding voltage data.

[0019] Based on the probability distribution curve, calculate the characteristic value of each welding voltage data group, calculate the average and variance of all characteristic values, and substitute them into the threshold calculation formula: threshold=μ+nσ, where threshold represents the arc welding weld defect threshold, μ represents the average value of the characteristic value of the normal defect-free weld data group, σ represents the variance of the characteristic value of the normal defect-free weld data group, and n represents a variable with a value range of positive integers greater than 1.

[0020] Preferably, the step of acquiring the welding voltage data set includes:

[0021] Customize the data reading cycle;

[0022] The data value of the voltage sensor connected in parallel to the arc welding machine circuit is read according to the data reading cycle.

[0023] Preferably, steps two to three are replaced with:

[0024] Construct the probability distribution curve of the welding voltage data set;

[0025] If the number of peaks in the probability distribution curve is 1 within the numerical range of the welding voltage data set, then the weld corresponding to the welding voltage data set is a normal, defect-free weld, and the next set of welding voltage data sets is collected repeatedly; otherwise, the weld corresponding to the welding voltage data set has a welding defect, and a real-time alarm is issued.

[0026] Secondly, the present invention provides a real-time detection system for welding defects in arc welding, the system comprising:

[0027] The data acquisition module is used to collect a set of welding voltage data during the arc welding process;

[0028] The data processing module is used to construct a probability distribution curve and use the probability distribution curve to calculate at least one feature value of the welding voltage data set respectively;

[0029] The defect detection module is used to compare the feature value with the preset threshold value, and to determine whether there is a porosity defect in the weld corresponding to the welding voltage data group based on the comparison result.

[0030] Preferably,

[0031] The system also includes a threshold calculation module, used to calculate the characteristic values ​​of the welding voltage data set of a known normal, defect-free weld during the arc welding process.

[0032] Preferably, the data processing module includes:

[0033] The curve fitting unit is used to divide a set of welding voltage data into intervals according to their numerical values ​​and construct a probability distribution curve.

[0034] The curve analysis unit is used to analyze probability distribution curves and calculate the characteristic values ​​of welding voltage data sets based on the curve distribution characteristics.

[0035] Thirdly, the present invention provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the real-time detection method for arc welding defects as described in any of the first aspects above.

[0036] The advantages of this invention are:

[0037] 1. This invention extracts feature values ​​from welding voltage data sets collected by a voltage sensor connected in parallel in the welding machine to determine whether there are welding defects in the weld. The actual data acquisition is more stable, and the voltage measurement only requires connecting the voltage sensor in parallel to the positive and negative poles of the welding machine and periodically reading the voltage data. The accessory connection cost is low, the amount of data that can be collected is abundant, and the voltage data has less noise.

[0038] 2. This invention processes welding voltage data to construct a probability distribution curve, extracts feature values, and then uses a threshold alarm method to achieve real-time detection and early warning of weld seams. The skewness, kurtosis or low-frequency interval spectral density, number of peaks, and other data of the probability distribution curve are valuable data obtained through certain experimental comparisons and industrial field tests. The extracted feature values ​​of skewness, kurtosis or low-frequency interval spectral density, number of peaks, and other feature values ​​of the probability distribution curve are the best features suitable for welding defect detection, verified by experimental testing.

[0039] 3. This invention replaces the manual labor required by traditional detection methods, and eliminates the need for special equipment for post-weld inspection or online detection such as laser sensors and infrared thermal imagers, thus greatly reducing the cost of arc welding defect detection. [Attached Image Description]

[0040] To more clearly illustrate the embodiments of the present invention or the existing technical solutions, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Appendix Figure 1 This is a flowchart illustrating one embodiment of the real-time detection method for arc welding defects of the present invention.

[0042] Appendix Figure 2 This is a flowchart illustrating another embodiment of the real-time detection method for arc welding defects of the present invention.

[0043] Appendix Figure 3 This is a schematic block diagram of one embodiment of the real-time detection system for arc welding defects of the present invention;

[0044] Appendix Figure 4 This is a circuit diagram of the data acquisition module in one embodiment of the real-time detection method and system for arc welding defects of the present invention;

[0045] Appendix Figure 5 This is a comparison chart of the probability distribution curves of normal welds and defective welds.

Detailed Implementation Methods

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0047] The real-time detection method and system for arc welding defects of the present invention are mainly used for real-time detection of weld quality in arc welding. By processing the voltage data collected by the voltage sensor during the welding operation online, extracting the feature values ​​of the data, and comparing the feature values ​​with the threshold to issue warnings, the real-time online prediction and analysis of weld defects can be achieved. It is not necessary to download the data to the local machine for post-weld analysis and detection, thereby realizing the automated and periodic operation of the detection process.

[0048] See attached document Figure 1 The flowchart of the real-time detection method for arc welding defects of the present invention shown in this embodiment includes the following steps:

[0049] Step 1: Collect a set of welding voltage data during the arc welding process.

[0050] In practical applications, a voltage sensor is connected in parallel to the circuit of the arc welding operating equipment. By reading the voltage sensor values, welding voltage data for one detection cycle is collected. (See attached diagram) Figure 4 The schematic diagram of the hardware device corresponding to the data acquisition module shown shows that the negative terminal of the welding machine is connected to the welding operation table circuit, and the positive terminal of the welding machine is connected to the welding robot circuit. In order to obtain data with less noise, the voltage sensor is directly connected in parallel to the positive and negative terminals of the welding machine to collect welding voltage data information during the welding process and store it in the online database. Specifically, the data acquisition process in this step can include: setting the signal acquisition frequency of the voltage sensor to no less than 1000Hz, that is, 1000 times per second, and setting the frequency of reading data from the sensor to 1Hz, that is, the custom data reading period is 1 second. Each time, all the welding voltage data collected by the voltage sensor within one period (1 second) is read to form a welding voltage data group and stored in the database in real time for subsequent data processing steps, or all the collected welding voltage data are stored in a data group according to the numerical order.

[0051] Step 2: Calculate at least one characteristic value of the welding voltage data set using the probability distribution curve.

[0052] This step mainly processes the welding voltage data sets collected in step one. Within a data reading cycle, each set of voltage data includes multiple values. The feature values ​​of this set of welding voltage data are calculated and extracted using a probability distribution curve calculation method. In this embodiment, the extracted feature values ​​include at least one of the following: skewness, kurtosis, or low-frequency spectral density of the probability distribution curve. It should be noted that before extracting the feature values, the probability distribution curve of the welding voltage data can be constructed using the following method, including the following steps:

[0053] S21. The welding voltage data stored in the same group are sorted in ascending order of value. Usually, the data stored in the same group is all the data collected in one cycle.

[0054] S22. Divide the sorted data group into several numerical intervals at equal intervals according to the range of data values;

[0055] S23. Count the frequency of welding voltage data in each of the numerical intervals.

[0056] This yields a probability distribution curve. Based on observation and statistical analysis of multiple sets of experimental data, the inventors found that the probability distribution curves obtained using the above method show a significant difference in the distribution between the welding voltage data corresponding to normal, defect-free welds and the welding voltage data corresponding to defective welds. Typically, the curve for the welding voltage data corresponding to normal, defect-free welds has only one peak within the corresponding numerical range of the data set, while the curve for the welding voltage data corresponding to defective welds has two peaks within the corresponding numerical range of the data set. Therefore, the skewness and kurtosis can be calculated as two statistical values ​​to be used as characteristic values ​​for weld detection.

[0057] Step 3: Determine whether the feature value is less than a preset threshold. If any feature value is less than the preset threshold, the weld corresponding to the welding voltage data group has a welding defect, and a real-time alarm is issued. If all feature values ​​are not less than the preset threshold, repeat steps 1 and 2. The preset threshold here is pre-set based on experimental results, and the threshold type is consistent with the feature value type. For example, when skewness is selected as the feature value, the threshold is the skewness value calculated from the data of known normal, defect-free welds; when kurtosis is selected as the feature value, the threshold is the kurtosis value calculated from the data of known normal, defect-free welds.

[0058] The calculation method using the preset threshold in this embodiment includes the following specific steps:

[0059] S31. For welds that have been confirmed to be free of porosity defects by existing post-weld inspection methods or other inspection methods, read the welding voltage data set collected by the voltage sensor during the arc welding process.

[0060] S32. Sort the data in each group of welding voltage data in ascending order of value, and divide them into several numerical intervals at equal intervals.

[0061] S33. Statistically count the frequency of welding voltage data in each of the numerical intervals, and construct the probability distribution curve for each group of welding voltage data.

[0062] S34. Calculate the characteristic value of each welding voltage data group according to the probability distribution curve, calculate the average and variance of all characteristic values, and substitute them into the threshold calculation formula: threshold=μ+nσ, where threshold represents the arc welding weld defect threshold, μ represents the average value of the characteristic value of the normal defect-free weld data group, σ represents the variance of the characteristic value of the normal defect-free weld data group, and n represents a variable with a value range of positive integers greater than 1.

[0063] Here, the types of eigenvalues ​​are the same as those described in step two, so they will not be repeated here.

[0064] See attached document Figure 2 The flowchart of the real-time detection method for arc welding defects of the present invention in the illustrated embodiment is shown. In this example, before real-time detection of arc welding defects, a preset threshold under the same welding parameter settings is first calculated, including the following steps:

[0065] Step 101. Collect welding voltage data sets of several known normal and defect-free welds during the welding process using a voltage sensor.

[0066] Step 102. Calculate the time-domain and frequency-domain characteristics of the welding voltage respectively. In this embodiment, the time-domain characteristics are the skewness and kurtosis of the probability distribution curve, and the frequency-domain characteristics are the spectral density in the low-frequency range. Both the time-domain and frequency-domain characteristics are the best characteristics suitable for welding defect detection obtained through experimental testing.

[0067] Step 103. Calculate the welding defect detection threshold using the feature values ​​extracted from the welding voltage data set of known normal and defect-free welds.

[0068] Step 104. Collect welding voltage data sets in real time during the robot welding process using a voltage sensor, with the same welding parameter settings as those used in the threshold calculation above.

[0069] Step 105. Calculate the time-domain and frequency-domain characteristics of the welding voltage data set respectively.

[0070] Step 106. Determine whether the time-domain and frequency-domain features of the real-time acquired voltage data set are less than the detection threshold. Specifically, only the time-domain feature can be compared with the detection threshold, and the corresponding numerical types of the time-domain feature and the detection threshold are the same; only the frequency-domain feature can be compared with the detection threshold, and the corresponding numerical types of the frequency-domain feature and the detection threshold are the same; or two detection thresholds can be calculated, corresponding to the numerical types of the time-domain feature and the frequency-domain feature, respectively. In this case, both the time-domain feature and the frequency-domain feature must be not less than the corresponding detection threshold to determine that there is no weld defect in the corresponding weld segment. Otherwise, if either the time-domain feature or the frequency-domain feature is less than the detection threshold, it is determined that there is a weld defect in the weld corresponding to the welding voltage data set, and a real-time alarm is issued.

[0071] In another feasible embodiment, the real-time detection method for arc welding defects of the present invention can also be implemented through the following specific steps:

[0072] Step 201. Collect a set of welding voltage data during the arc welding process. The data collection method is the same as in the above embodiment, and will not be repeated here.

[0073] Step 202. Construct the probability distribution curve of the welding voltage data set. The method of constructing the probability distribution curve is the same as in the above embodiment, and will not be repeated here.

[0074] Step 203. Determine whether the number of peaks in the probability distribution curve is 1 within the numerical range of the welding voltage data set. If yes, the weld corresponding to the welding voltage data set is a normal, defect-free weld, and the next set of welding voltage data sets is collected repeatedly. If no, the weld corresponding to the welding voltage data set has a welding defect, and a real-time alarm is issued.

[0075] To make the technical solution of the present invention clearer, the following description is provided in conjunction with the appendix. Figure 5 The probability distribution curves and specific implementation examples shown in the diagram provide a detailed explanation of the practical application of this invention.

[0076] The characteristic values ​​of the welding voltage data set are calculated using the probability distribution curve method. The process involves dividing the collected welding voltage data set into equally spaced small intervals within its data value range, and then counting the frequency of data occurrence in each interval, i.e., the number of data values ​​in the data set within the current interval range, thereby obtaining the probability distribution curve.

[0077] In practical applications, the above-mentioned numerical range refers to the value range of the collected data set. Under the same welding parameter settings, the numerical range is the same; under different welding process parameter settings, the data value range of the collected data set is different. This range is usually related to the input voltage. This embodiment includes... Figure 5The curve shown corresponds to a data set with a value range of 40–70V used in the test. When dividing the data set into intervals, the number of intervals and the interval spacing are inversely proportional. That is, for a fixed data value range, the larger the interval spacing, the fewer the number of intervals. Experimentally, the optimal interval spacing in this example was set to 0.4V, therefore the optimal number of intervals is 175.

[0078] Based on the above method of constructing probability distribution curves, as shown in the attached figure. Figure 5 The figure shows four different probability distribution curves corresponding to four sets of voltage data for different arc weld segments collected under the same welding parameter settings: curve 20, curve 30, curve 60, and curve 71. Curves 20 and 30 represent the probability distribution curves corresponding to the welding voltage data sets of normal, defect-free welds during the arc welding process, while curves 60 and 71 represent the probability distribution curves corresponding to the voltage data sets of defective welds. Each curve represents a data set collected in one second. Within the range of 40–70V, the probability distribution curve corresponding to normal, defect-free welds has one peak, while the probability distribution curve corresponding to defective welds has two peaks. The calculated skewness values ​​of the probability distribution curves corresponding to curves 20, 30, 60, and 71 are 0.5394, 0.3100, 0.0302, and 0.0864, ​​respectively. The threshold calculated based on the threshold setting method is adjusted to 0.1. The skewness values ​​of the four probability distribution curves are compared with the threshold. If the skewness is less than the threshold, it is determined that there is a welding defect. Therefore, the welds corresponding to curves 60 and 71 can be determined as defective welds.

[0079] Based on the same concept as the above embodiments of the real-time detection method for arc welding defects of the present invention, the real-time detection system for arc welding defects provided by the embodiments of the present invention will be described below. The real-time detection system for arc welding defects described below can be referred to in correspondence with the real-time detection method for arc welding defects described above. Please refer to... Figure 3 The schematic block diagram of the real-time arc welding defect detection system of the present invention shown in the attached figure is as follows. Figure 3 The real-time detection of arc welding defects in this embodiment includes: a data acquisition module, a data processing module, and a defect detection module.

[0080] The data acquisition module is used to acquire a set of welding voltage data during the arc welding process; the data processing module is used to construct a probability distribution curve and use the probability distribution curve to calculate at least one feature value of the welding voltage data set; the defect detection module is used to compare the feature value with a preset threshold and determine whether there is a porosity defect in the weld corresponding to the welding voltage data set based on the comparison result.

[0081] In a preferred embodiment, the real-time detection system for arc welding defects of the present invention further includes a threshold calculation module for calculating the characteristic values ​​of the welding voltage data set of a known normal and defect-free weld during the arc welding process.

[0082] In a further embodiment, the data processing module includes a curve fitting unit and a curve analysis unit. The curve fitting unit is used to divide a set of welding voltage data into intervals according to their numerical values ​​to construct a probability distribution curve. The curve analysis unit is used to analyze the probability distribution curve and calculate the feature values ​​of the welding voltage data set based on the curve distribution characteristics. In practical applications, to achieve threshold alarms, the defect detection module includes a judgment unit, a result analysis unit, and an early warning unit. The judgment unit is used to compare the feature values ​​with a preset threshold to obtain a comparison result. The result analysis unit is used to determine whether there are welding defects in the arc weld corresponding to the welding voltage data set based on the comparison result of the feature values ​​and the preset threshold. The early warning unit is used to issue a real-time alarm to the welding machine when the weld is determined to have welding defects.

[0083] The real-time detection system for arc welding defects in this embodiment is used to implement the aforementioned real-time detection method for arc welding defects. Therefore, the specific implementation of the real-time detection system for arc welding defects can be referred to the embodiment section of the real-time detection method for arc welding defects mentioned above, and will not be repeated here.

[0084] This invention also proposes a computer-readable storage medium that stores one or more programs, the programs including instructions that, when executed by an electronic device including multiple applications, enable the electronic device to perform the real-time detection method for arc welding defects provided in any embodiment of this invention.

[0085] The systems, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0086] For ease of description, the above apparatus is described by dividing it into various units or modules according to their functions. Of course, in implementing this invention, the functions of each unit or module can be implemented in one or more software and / or hardware.

[0087] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] This invention is described with reference to flowchart illustrations and block diagrams of methods and systems according to embodiments of the invention. It will be understood that each block and step in the flowchart illustrations and block diagrams, as well as combinations of blocks and steps in the flowchart illustrations and block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing one or more blocks and steps in the flowchart illustrations. Figure 1 A device that provides the functions specified in one or more boxes.

[0089] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and boxes Figure 1 The function specified in one or more boxes.

[0090] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and boxes Figure 1 The steps of the function specified in one or more boxes.

[0091] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0092] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0093] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for real-time detection of arc welding defects, characterized in that, The method comprises the following steps: Step 1: collecting a set of welding voltage data during arc welding; Step 2: sorting the set of welding voltage data in ascending order, dividing the sorted data set into several numerical intervals according to the data value range at equal intervals, and sequentially counting the frequency of welding voltage data in each numerical interval to construct a probability distribution curve of the welding voltage data set; Step 3: determining whether the number of wave crests of the probability distribution curve in the numerical range of the welding voltage data set is 1, if yes, the welding seam corresponding to the welding voltage data set is normal and defect-free, and the next set of welding voltage data is collected repeatedly; if no, the welding seam corresponding to the welding voltage data set has welding defects, and a real-time alarm is issued.

2. The method of real-time detection of arc welding defects according to claim 1, characterized in that, The step of collecting the welding voltage data set comprises: Customizing a data reading period; Reading the data value of the voltage sensor connected in parallel to the arc welding machine circuit according to the data reading period.

3. An electric arc welding defect real-time detection system for implementing the real-time detection method according to claim 1 or 2, characterized in that, The system comprises: A data collection module for collecting a set of welding voltage data during arc welding; A data processing module for constructing a probability distribution curve and calculating at least one characteristic value of the welding voltage data set using the probability distribution curve; A defect detection module for comparing the characteristic value with a preset threshold value, and determining whether there is a porosity defect in the welding seam corresponding to the welding voltage data set according to the comparison result.

4. The real-time detection system for arc welding defects according to claim 3, characterized in that, Further comprising: A threshold value calculation module for calculating the characteristic value of the welding voltage data set of a known normal and defect-free welding seam during arc welding.

5. The real-time detection system for electric arc welding defects according to claim 3 or 4, characterized in that, The data processing module comprises: A curve fitting unit for interval division of a set of welding voltage data according to the numerical size to construct a probability distribution curve; A curve analysis unit for analyzing the probability distribution curve and calculating the characteristic value of the welding voltage data set according to the curve distribution characteristics.

6. A computer readable medium having stored thereon a computer program, characterized in that, The program is executed by the processor to realize the arc welding defect real-time detection method as claimed in claim 1 or 2.

Images