Welding robot self-adaptive wire feeding control method, platform and medium

Abstract

The invention discloses a welding robot self-adaptive wire feeding control method, a platform and a medium, relates to the technical field of welding automation process control, and is suitable for welding process management under complex working conditions. According to the method, an arc voltage sequence and a molten pool characteristic parameter sequence in the welding process are collected, a welding state characteristic value is obtained through calculation, a reference database containing welding process parameters, welding working condition parameters and the welding state characteristic value is constructed based on historical experimental data, and corresponding preset reference indexes are generated through a mapping relation function; and in real-time welding, the welding state characteristic value of the current wire feeding control period is obtained and compared with a preset reference index, the deviation value is calculated, the welding state is judged according to the grading threshold value interval, the wire feeding speed is dynamically adjusted, and self-adaptive matching of the wire feeding speed and the welding state is achieved. According to the method, instability caused by traditional adjustment depending on experience can be overcome, dynamic control over the welding process is achieved, and the stability of a welding pool and the welding seam forming quality are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of welding automation process control, in particular to a welding robot adaptive wire feeding control method, platform and medium. BACKGROUND

[0002] In the application of welding robots, the wire feeding speed is one of the key process parameters that affect the welding quality and the stability of the forming. In the existing welding adaptive wire feeding control methods, a common type is the feedback control based on the arc voltage or current signal. By monitoring the changes of the arc voltage and current, the arc length or the droplet transfer state is inferred, and the wire feeding speed is adjusted accordingly. This type of method is relatively simple to implement, but the feedback signal is easily affected by power fluctuations, environmental interference and working condition changes, resulting in unstable wire feeding adjustment, and often appearing response lag or excessive adjustment under complex welding conditions. Another type of method is based on visual or thermal imaging detection means, which directly observes the shape of the molten pool or droplet to realize wire feeding control. Although it can provide more intuitive welding state information, it is highly dependent on imaging equipment, is greatly limited by smoke, light and installation environment, and has high difficulty in real-time processing.

[0003] Overall, existing methods either rely on single electrical signal feedback, which lacks robustness, or rely on complex detection means, which is difficult to adapt to long-term stable operation in industrial sites. Therefore, how to ensure the implementability while realizing more stable and accurate adaptive wire feeding regulation of the welding process is still a technical problem to be solved. SUMMARY

[0004] To solve the above problems, the present application provides a welding robot adaptive wire feeding control method, platform and medium. First, a reference database of the mapping relationship between the welding state characteristic value and the welding process parameters and welding working condition parameters is constructed. The arc voltage sequence and the molten pool state characteristic parameters are calculated in real time during the actual welding process, and the deviation value corresponding to the welding state characteristic value is obtained by comparing with the mapping relationship function in the database. Then, the wire feeding speed is dynamically adjusted according to the deviation value belonging to the classified threshold interval, so as to realize real-time adaptive monitoring of the welding state and optimization of the wire feeding speed, and improve the welding quality, stability and production efficiency.

[0005] To achieve the above purpose, in a first aspect, the present application provides a welding robot adaptive wire feeding control method, comprising the following steps, Based on historical welding experimental data, the arc voltage sequence collected in each experiment and the characteristic parameter sequence reflecting the molten pool state in each wire feeding control period are calculated to obtain the welding state characteristic value corresponding to each wire feeding control period. The welding state characteristic value is used to represent the welding state, including the arc length deviation, the arc length fluctuation amplitude and the molten pool stability index. The mapping relationship function among the welding process parameters, the welding condition parameters and the welding state characteristic values corresponding to each wire feeding control period is fitted, and the mapping relationship function is stored as a database entry together with the corresponding welding process parameters and welding condition parameters to construct a reference database; In the current wire feeding control period, a database entry matching the welding process parameters and the welding condition parameters of the current wire feeding control period is searched in the reference database, and the mapping relationship function stored in the entry is called to calculate a preset reference index corresponding to the wire feeding control period, which is used for subsequent welding state evaluation. The welding state characteristic values of the current wire feeding control period are calculated according to the arc voltage sequence collected at a sampling frequency in the current wire feeding control period and the characteristic parameter sequence reflecting the molten pool state. After the welding state characteristic values of the current wire feeding control period are standardized, the welding state characteristic values are compared with the corresponding preset reference index one by one to obtain the deviation values corresponding to the welding state characteristic values. According to the deviation values corresponding to the welding state characteristic values, the classification threshold interval to which the welding state characteristic values belong is determined, and the classification threshold interval includes a stable interval, an adjustable interval and an abnormal processing interval, and the wire feeding speed is dynamically adjusted according to the classification adjustment strategy to realize the dynamic matching of the wire feeding speed and the welding state.

[0006] To achieve the above-mentioned purpose, another aspect of the present application provides a welding robot adaptive wire feeding control platform, comprising, A historical data calculation module is configured to calculate the arc voltage sequence collected in each experiment and the characteristic parameter sequence reflecting the molten pool state based on historical welding experiment data to obtain welding state characteristic values corresponding to each wire feeding control period, wherein the welding state characteristic values are used to represent the welding state, including the arc length deviation, the arc length fluctuation amplitude and the molten pool stability index. A reference database construction module is configured to fit the mapping relationship function among the welding process parameters, the welding condition parameters and the welding state characteristic values corresponding to each wire feeding control period, and store the mapping relationship function together with the corresponding welding process parameters and welding condition parameters as a database entry to construct a reference database. A reference index generation module is configured to search a database entry matching the welding process parameters and the welding condition parameters of the current wire feeding control period in the reference database in the current wire feeding control period, and call the mapping relationship function stored in the entry to calculate a preset reference index corresponding to the wire feeding control period, which is used for subsequent welding state evaluation. A real-time data calculation module is configured to calculate welding state characteristic values of a current wire feeding control period according to an arc voltage sequence collected at a sampling frequency in the current wire feeding control period and a characteristic parameter sequence reflecting a molten pool state in the welding process; A deviation calculation module is configured to compare each welding state characteristic value in the current wire feeding control period with a corresponding preset reference index after standardization processing to obtain a corresponding deviation value of each welding state characteristic value. A wire feeding adjustment module is configured to determine a corresponding hierarchical threshold interval of each welding state characteristic value according to the corresponding deviation value, wherein the hierarchical threshold interval includes a stable interval, an adjustable interval and an abnormal processing interval, and dynamically adjust a wire feeding speed according to a hierarchical adjustment strategy to realize dynamic matching of the wire feeding speed and the welding state.

[0007] To achieve the above object, the present application provides a computer readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the welding robot adaptive wire feeding control method.

[0008] The one or more technical solutions provided in the present application have at least the following technical effects or advantages: in view of the problems of difficulty in real-time self-adaptive adjustment of wire feeding speed in the welding process, inaccurate quantification of welding state, and untimely response of abnormal processing, a mapping relationship database of welding state characteristic values and welding process parameters and welding working condition parameters is established, and combined with real-time welding state monitoring and hierarchical adjustment strategy, dynamic identification of the welding state, deviation determination, and closed-loop self-adaptive control of the wire feeding speed are realized. Specifically, the welding state characteristic values in the wire feeding control period are obtained by calculating the arc voltage sequence and the molten pool state characteristic parameters based on historical experimental data, the multi-dimensional quantitative characterization of the welding state is realized, and the subsequent adjustment operation can rely on accurate data; the mapping relationship function of the welding process parameters and the welding working condition parameters is fitted and stored as a database entry, which is used to generate the preset reference index of each wire feeding control period, and provides a reliable basis for real-time welding state deviation determination; in the actual welding process, the arc voltage sequence and the molten pool state characteristic parameters are calculated in real time, the standardized welding state characteristic values are compared with the preset reference index corresponding to the mapping relationship function in the database, the deviation corresponding to the welding state characteristic value is obtained, and the hierarchical adjustment strategy is realized by dividing the deviation into a stable interval, an adjustable interval, and an abnormal processing interval; when the deviation value corresponding to the welding state characteristic value falls into the adjustable interval, the wire feeding speed is dynamically adjusted according to the dominant deviation value and the deviation range, the rapid response to the change of the welding state is realized, and when it falls into the abnormal interval, hierarchical abnormal processing measures are executed to ensure the safety and stability of the welding process under abnormal state and prevent the expansion of welding defects. Compared with the prior art, the welding state characteristic value database is established, the hierarchical deviation control and abnormal processing mechanism are introduced, and the closed-loop control from real-time state perception, deviation determination to dynamic wire feeding adjustment is realized; while improving the welding stability and the weld consistency, the frequency of manual intervention is reduced, and the welding automation level and production efficiency are improved. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0010] Figure 1 A welding robot adaptive wire feeding control method flowchart provided by the embodiment of the present application; Figure 2 A welding robot adaptive wire feeding control platform structure schematic diagram provided by the embodiment of the present application; Figure labeling: Historical data calculation module 11, benchmark database construction module 12, benchmark index generation module 13, real-time data calculation module 14, deviation calculation module 15, wire feeding adjustment module 16. Detailed Implementation

[0011] This invention introduces welding state feature quantification, benchmark database construction, and hierarchical adaptive control mechanism, and proposes an adaptive wire feeding control method, platform, and medium for welding robots. It solves the problems of wire feeding speed relying on experience adjustment, insufficient welding state quantification, and untimely response under abnormal working conditions in the existing welding process, thereby achieving dynamic matching between wire feeding speed and welding state, and improving the stability of the welding process and weld quality.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that the terms "first," "second," etc., used in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, platform, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown, the adaptive wire feeding control method for welding robots includes the following steps: S1. Based on historical welding experiment data, the arc voltage sequence and the characteristic parameter sequence reflecting the molten pool state collected in each wire feeding control cycle of each experiment are calculated to obtain the welding state characteristic value corresponding to each wire feeding control cycle. The welding state characteristic value is used to characterize the welding state, including arc length deviation, arc length fluctuation amplitude, and molten pool stability index. Specifically During historical welding experiments, for each wire feed control cycle of each experiment, arc voltage sequences and characteristic parameter sequences reflecting the molten pool state were collected at a set sampling frequency. The collected data comprehensively reflects the dynamic characteristics of the arc and molten pool during the welding process. The collected data sequences were processed, including noise reduction and outlier removal. Based on this, statistical analysis was performed on each sequence to calculate the welding state characteristic values ​​for each wire feed control cycle. These welding state characteristic values ​​characterize the welding state, including arc length deviation, arc length fluctuation amplitude, and molten pool stability index. The obtained welding state characteristic values ​​serve as the data source for subsequently establishing a benchmark database to support the fitting and storage of the relationship between welding process parameters, welding condition parameters, and welding state characteristic values. Furthermore, the calculation of the arc length deviation specifically includes, 1) Preprocess the arc voltage sequence obtained during the wire feeding control cycle. The specific processing steps include: first, sorting the voltage samples collected during the wire feeding control cycle by timestamp, ensuring the correct time order of the data and the equal time interval between each data point, with at least 50 voltage samples; second, using a high-pass filter to remove low-frequency drift in the arc voltage sequence, with the cutoff frequency of the high-pass filter set to 0.1Hz; next, removing outliers from the arc voltage sequence, eliminating isolated pulses and transient spikes in the voltage data; then, applying a low-pass filter to the processed arc voltage sequence to suppress high-frequency noise and electromagnetic interference, with the filter cutoff frequency set to 200Hz; and finally, applying a smoothing process to the filtered arc voltage sequence using a moving average filter with a window size of 5 to reduce high-frequency jitter, ensuring smoother data and avoiding the impact of short-term fluctuations on subsequent calculations. 2) Based on the conversion relationship between welding arc voltage and arc length, the processed arc voltage sequence is converted into the corresponding arc length value sequence. Specifically, based on the approximately linear relationship between arc voltage and arc length in the welding process, an arc voltage ( ) and arc length ( Linear formula ,in, It is the proportionality coefficient between the change in arc voltage and the change in arc length, representing the degree of influence of the change in arc voltage on the arc length, and reflecting the sensitivity of arc voltage to arc length; This is an offset value used to represent the arc length when the arc voltage is zero. It is typically used to correct for zero-point offset in equipment or the effects of environmental factors. and The calibration is obtained through experiments. The calibration process is performed according to specific welding processes, equipment parameters and environmental conditions. The calibration process follows a recognized calibration method in the welding field. Specifically, ① calibration conditions are set, including setting a fixed welding current, a fixed welding voltage, using a welding wire specification consistent with actual production, and using the same protective gas and gas flow as actual production; ② arc length measurement: using accurate measurement tools to obtain actual arc length data during multiple welding tests; ③ corresponding the set arc voltage under multiple test conditions with the actual measured arc length data, using regression analysis method for fitting, obtaining the specific values of and Based on the linear formula of arc voltage ( ) and arc length ( ), each arc voltage value in the processed arc voltage sequence is calculated to obtain the corresponding arc length value. After conversion, a complete arc length value sequence is obtained; 3) the arithmetic mean of the arc length value sequence is obtained, and the average arc length of the wire feeding control period is obtained, Specifically, in the wire feeding control period, the obtained arc length value sequence is denoted as , wherein is the number of sampling points in the wire feeding control period and ≥ 50. The average arc length in the wire feeding control period is obtained by adding all the arc length values and dividing by the number of sampling points; 4) the difference between each sampling point in the arc length value sequence and the average arc length forms a deviation sequence, and the arc length deviation is obtained by calculating the standard deviation of the deviation sequence, In the wire feeding control period, the arc length of each sampling point in the arc length value sequence is subtracted from the average arc length in the current wire feeding control period to obtain a set of difference values, forming a deviation sequence. The square of each difference value in the deviation sequence is calculated and summed, and then the sum is divided by the number of sampling points to obtain the mean square value of the deviation. Finally, the square root of the mean square value is taken to obtain the arc length deviation in the wire feeding control period, which is used to reflect the fluctuation degree of the arc length in the wire feeding control period; Further, the calculation of the arc length fluctuation amplitude specifically includes, The maximum value and the minimum value of the arc length value sequence are determined based on the arc length value sequence, and the arc length fluctuation amplitude is obtained by calculating the difference between the maximum value and the minimum value, Specifically, based on the arc length value sequence obtained after conversion according to the linear formula of arc voltage ( ) and arc length ( ), the maximum arc length value and the minimum arc length value are determined by comparing the arc length values of each sampling point in the sequence, and the arc length fluctuation amplitude is obtained by calculating the difference between the two values, which is used to reflect the degree of fluctuation of the arc length in the wire feeding control period; Further, the calculation of the molten pool stability index includes, 1) The sequence of characteristic parameters reflecting the molten pool state obtained in the wire feeding control period is sorted, denoised and outlier removed, and the sequence of characteristic parameters reflecting the molten pool state includes the sequence of molten pool geometric shape change characteristic parameters, the sequence of surface fluctuation characteristic parameters and the sequence of temperature change characteristic parameters, The specific processing process includes: first, the collected sequence of molten pool geometric shape change characteristic parameters, the sequence of surface fluctuation characteristic parameters and the sequence of temperature change characteristic parameters are sorted according to the time stamp to ensure the correct time sequence of the data, and the sampling interval is checked for consistency. If missing points are found, linear interpolation is used to fill them in; then, the noise of each characteristic parameter sequence is removed, and the moving average filtering method is used with a window size of 5-7 to smooth out random jitter and retain the main trend characteristics, avoiding misjudgment caused by high-frequency noise interference; finally, the outlier of each characteristic parameter sequence is removed, and the statistical method is used to judge each data point. If a point deviates from the mean value of the characteristic parameter sequence by more than 3 times the standard deviation, it is determined to be abnormal and is removed. If necessary, the mean value of the adjacent data points before and after the position is used to replace the position to ensure the continuity and effectiveness of the sequence; 2) Based on the processed sequence of molten pool geometric shape change characteristic parameters, the sequence of surface fluctuation characteristic parameters and the sequence of temperature change characteristic parameters, statistical analysis is performed respectively, and the standard deviation of each sequence is calculated to obtain the molten pool stability index reflecting the stability of the molten pool in the wire feeding control period, including the molten pool geometric shape change amplitude, the molten pool surface fluctuation amplitude and the molten pool temperature change amplitude, Specifically, the processed sequence of molten pool geometric shape change characteristic parameters, the sequence of molten pool surface fluctuation characteristic parameters and the sequence of molten pool temperature change characteristic parameters are respectively denoted as , , wherein is the number of sampling points in the wire feeding control period and ≥50, the standard deviations of the three sequences are calculated respectively, and the three standard deviation results are respectively taken as the quantitative indicators of the molten pool geometric shape change amplitude, the molten pool surface fluctuation amplitude and the molten pool temperature change amplitude.

[0015] S2. Based on the welding process parameters, welding working condition parameters and welding state characteristic values corresponding to each wire feeding control period, a mapping relationship function between the three is fitted, and the mapping relationship function is stored together with the corresponding welding process parameters and welding working condition parameters as a database entry to construct a reference database, Specifically, Firstly, the welding process parameters and welding condition parameters in each experiment in the historical welding experiments are obtained, wherein the welding process parameters include arc voltage, arc current, wire feeding speed, wire material and diameter, and protective gas type and flow, and the welding condition parameters include welding position, welding torch angle and environmental conditions. The welding process parameters, welding condition parameters and calculated welding state characteristic values corresponding to each wire feeding control period are correspondingly associated to form a data record containing the corresponding relationship of the parameters and characteristic values, and the data record is arranged, including removing abnormal values, completing missing values and normalizing data with inconsistent dimensions, to ensure the consistency and integrity of the data distribution, thereby ensuring the reliability of the subsequent fitting results; Secondly, a multiple regression method is used to establish the corresponding relationship between the welding process parameters, welding condition parameters and welding state characteristic values, and a mapping relationship function representing the relationship among the three is obtained by fitting, Further, the fitting of the mapping relationship function specifically includes, 1) Quantizing and standardizing the welding process parameters, welding condition parameters and welding state characteristic values so that parameters with different dimensions can directly participate in multiple linear regression fitting, Specifically, a multiple linear regression method is used to establish a mapping function between welding process parameters, welding condition parameters, and welding state characteristic values. The arc voltage, arc current, wire feed speed, wire material, wire diameter, and shielding gas flow rate from the welding process parameters, and the welding position, welding torch angle, and environmental conditions from the welding condition parameters, are denoted as parameter vectors. These parameter vectors and welding state characteristic values ​​are then quantified and standardized. Arc voltage, arc current, wire feed speed, wire diameter, and shielding gas flow rate are directly taken from measured values. Wire material is quantified using a material coding system, with each material corresponding to a unique code value; for example, carbon steel welding wire is coded as 1, stainless steel as 2, and aluminum alloy as 3. The welding position is represented by standardized spatial coordinate values, standardized according to the workpiece size ratio. The welding torch angle is directly mapped from the measured posture angle. Environmental conditions... A comprehensive index is constructed by weighted summation of measurable parameters such as temperature, humidity, and wind speed, and quantified into a numerical value. The comprehensive environmental index is calculated as follows: Comprehensive Environmental Index = 0.5 × Temperature Value + 0.3 × Humidity Value + 0.2 × Wind Speed ​​Value. The weighting coefficients are allocated based on the sensitivity of environmental parameters in welding technology to welding state characteristic values. Welding state characteristic values ​​are calculated using arc voltage sequences and characteristic parameter sequences reflecting the molten pool state. The nine parameters in the quantified parameter vector and the welding state characteristic values ​​are standardized. Specifically, the numerical sets of each parameter vector and the numerical sets of various welding state characteristic values ​​in historical welding experimental data are statistically analyzed, and the mean and standard deviation of each parameter and various welding state characteristic values ​​are calculated. Then, the mean is subtracted from the original value of each sample and divided by the standard deviation to obtain a normalized dimensionless value, so that parameters of different dimensions can directly participate in multiple linear regression fitting. 2) For each type of welding condition characteristic value, establish the following mapping relationship function: , Specifically, before the regression coefficients are determined, this mapping function is merely a linear combination expression of the independent variables and regression coefficients, reflecting the functional form of the fit, where, Indicates the first k The standardized values ​​of the characteristic values ​​of the welding state ( k= 1-5). Indicates the deviation in arc length. Arc length fluctuation amplitude This indicates the magnitude of change in the geometry of the molten pool. Indicates the amplitude of surface fluctuations in the molten pool. Indicates the range of temperature change in the molten pool. This represents the quantized and standardized values ​​of each parameter in the parameter vector. This represents the regression coefficient corresponding to each parameter after quantization and standardization. is a constant term; 3) Based on the mapping relationship function, the residual between the normalized parameter calculation value and the normalized welding state characteristic value is compared, and the least square method is used for optimization to obtain the optimal constant term and regression coefficient, Specifically, the determination process of the regression coefficient and the constant term includes: First, the historical experimental data is sorted into a sample matrix, wherein each sample is composed of a group of parameters and corresponding normalized welding state characteristic values after quantization and normalization of the parameters; then, for each sample, the normalized welding state characteristic value is retained, and the quantized and normalized parameters are substituted into the mapping relationship function to obtain a function value expression dependent on the regression coefficient, and then the function value expression is subtracted from the corresponding normalized welding state characteristic value to define the residual of the sample; then, the squares of the residuals of all samples are accumulated to form the overall residual sum of squares; finally, the least square method is used for matrix operation to minimize the residual sum of squares, thereby obtaining the optimal constant term and regression coefficient combination. The constant term represents the baseline offset of the welding state characteristic value under the action of all parameters, and the regression coefficient reflects the contribution degree of each parameter to the welding state characteristic value; 4) Statistical test is performed on the established regression function to ensure the effectiveness of the mapping relationship function, Specifically, the statistical test is performed on the established regression function, which specifically includes: t testing each regression coefficient to eliminate insignificant parameters and avoid redundancy; performing F testing on the overall regression equation to verify whether the overall fitting has statistical significance; and using the coefficient of determination R 2 to evaluate the fitting accuracy, R 2 The closer to 1, the better the fitting effect, according to the requirement of the welding process on stability, the threshold is set to 0.9, when R² ≥0.9, it is considered that the function fitting accuracy meets the requirements, and it is confirmed as an effective mapping relationship function, Subsequently, after the fitting of the mapping relationship function is completed, the fitting results corresponding to each wire feeding control period are organized into a standardized database entry, which specifically includes the following steps: 1) For each wire feeding control period, the corresponding welding process parameters and welding condition parameters are collected and stored in the form of standardized numerical values; 2) The constant term and the regression coefficient of each parameter of the fitted mapping relationship function are extracted, and the welding state characteristic value category corresponding to the function is identified; 3) the collected welding process parameters, welding working condition parameters and corresponding mapping relationship function information are organized as complete database entries, each entry containing parameter information, function information and characteristic value identification, wherein the parameter information represents the process parameters and working condition parameters of the wire feeding control period, the function information represents the function form, constant term and regression coefficient for calculating the specific welding state characteristic value, and the characteristic value identification represents the corresponding welding state characteristic value category, and a unique index number is generated for each database entry, Finally, the obtained database entries are organized and stored according to a predetermined data structure format to form a corresponding reference database, and the reference database also stores the mean and standard deviation calculated when the numerical set of each parameter vector and the numerical set of each welding state characteristic value in the historical welding experiment data are counted.

[0016] S3. In the current wire feeding control period, find the database entry in the reference database that matches the welding process parameters and welding working condition parameters of the current wire feeding control period, and call the mapping relationship function stored in the entry to calculate the preset reference index corresponding to the wire feeding control period, which is used for subsequent welding state evaluation, Specifically, First, the welding process parameters and welding working condition parameters of the current wire feeding control period are obtained, including arc voltage, arc current, wire feeding speed, welding wire material, welding wire diameter, protective gas flow, welding position, welding gun angle and environmental conditions. To ensure consistency with the database entries in the reference database, the above parameters are processed according to the numerical standardization rules used when generating the database entries, specifically, the current parameter vector is standardized by combining the mean and standard deviation calculated from the numerical set of each parameter vector when the reference database counts the historical welding experiment data, and is converted into a dimensionless value, so that its data format is consistent with the parameter vector in the database entry; Subsequently, the standardized current parameter vector is used as a search condition to find the database entry in the reference database that completely matches the current parameter vector. When there is a completely matched database entry, the database entry is directly selected as the optimal matching entry. If there is no completely matched database entry, the similarity calculation link is entered, and the Euclidean distance between the current parameter vector and the parameter vector stored in each candidate database entry is calculated. The Euclidean distance is used as a similarity index to quantify the closeness of the current period parameters and the database entry parameters. When there are multiple candidate entries, the database entry with the smallest Euclidean distance is selected as the optimal matching database entry for the current wire feeding control period, so as to ensure that the selected database entry is as close as possible to the actual situation of the current wire feeding control period in terms of welding process parameters and welding working condition parameters; Finally, the mapping function stored in the optimal matching entry is called to calculate the dimensionless preset reference index corresponding to the current wire feeding control period according to the normalized current parameter vector, which is used for subsequent welding state evaluation and wire feeding adjustment decision.

[0017] S4. Calculate the welding state characteristic value of the current wire feeding control period according to the arc voltage sequence collected at the sampling frequency and the characteristic parameter sequence reflecting the molten pool state in the current wire feeding control period of the welding process, wherein the welding state characteristic value includes the arc length deviation, the arc length fluctuation amplitude and the molten pool stability index, Specifically, in the current wire feeding control period of the welding process, the arc voltage sequence and the characteristic parameter sequence reflecting the molten pool state are collected at the set sampling frequency, and the collected data sequences are processed, such as sorting, denoising and outlier rejection, wherein, 1) The arc length deviation and the arc length fluctuation amplitude are calculated based on the arc voltage sequence; 2) The molten pool geometry change amplitude, the molten pool surface fluctuation amplitude and the molten pool temperature change amplitude are calculated based on the characteristic parameter sequence reflecting the molten pool state, including the molten pool geometry change characteristic parameter sequence, the surface fluctuation characteristic parameter sequence and the temperature change characteristic parameter sequence, The arc length deviation, the arc length fluctuation amplitude, the molten pool geometry change amplitude, the molten pool surface fluctuation amplitude and the molten pool temperature change amplitude are used to represent the welding state in the current welding process, and provide a basis for subsequent determination of welding arc stability and adjustment of wire feeding speed. Each wire feeding control period is processed in this way to ensure that the characteristic value can accurately reflect the real-time change of the welding state.

[0018] S5. After standardizing each type of welding state characteristic value in the current wire feeding control period, compare it with the corresponding preset reference index one by one to obtain the deviation value corresponding to each type of welding state characteristic value, Specifically, when the reference database is used to statistically analyze historical welding experimental data, the mean and standard deviation calculated from the numerical set of each type of welding state characteristic value are used to standardize the welding state characteristic value of the current wire feeding control period, which is converted into a dimensionless value to form a standardized welding state characteristic vector of the current control period. Compare the standardized welding state characteristic vectors of each type with the corresponding preset reference index one by one to calculate the deviation value corresponding to each type of welding state characteristic value.

[0019] S6. Determine the classification threshold interval to which each type of welding state characteristic value belongs according to the deviation value corresponding to each type of welding state characteristic value, wherein the classification threshold interval includes a stable interval, an adjustable interval and an abnormal processing interval, and dynamically adjust the wire feeding speed according to the classification adjustment strategy to realize the dynamic matching of the wire feeding speed and the welding state, Specifically, according to the deviation value corresponding to each type of welding state characteristic value, the classification threshold interval is determined, when the deviation value corresponding to each type of welding state characteristic value is within ±0.5σ range of the corresponding preset reference index, it is indicated that the welding state is stable, and it is determined as a stable interval; when the deviation value corresponding to each type of welding state characteristic value exceeds ±0.5σ but does not exceed ±2σ of the corresponding preset reference index, it is indicated that the welding state fluctuates, and it is determined as an adjustable interval; when the deviation value corresponding to each type of welding state characteristic value exceeds ±2σ of the corresponding preset reference index, it is indicated that the welding state is abnormal, and it is determined as an abnormal processing interval, wherein σ is the standard deviation of each type of welding state characteristic value in the historical welding experimental data, in the current wire feeding control period, based on the classification threshold interval in which the deviation value corresponding to each type of welding state characteristic value is located, the wire feeding speed is dynamically adjusted according to the classification adjustment strategy, Further, the classification adjustment strategy specifically includes, In the current wire feeding control period, the deviation values corresponding to the arc length deviation, the arc length fluctuation amplitude, the molten pool geometric shape change amplitude, the molten pool surface fluctuation amplitude and the molten pool temperature change amplitude are determined, when the deviation value corresponding to any one welding state characteristic value falls into the value range corresponding to the abnormal processing interval, the abnormal processing strategy is executed, so as to adjust the wire feeding control process in time when the abnormal state occurs, thereby avoiding further deterioration of the welding quality; When all the deviation values corresponding to the welding state characteristic values do not enter the abnormal processing interval, and at least one deviation value falls into the adjustable interval, the wire feeding speed of the current wire feeding control period is adjusted according to the deviation amplitude, so as to realize the adaptive correction of the welding process; When all the deviation values corresponding to the welding state characteristic values are in the stable interval, the wire feeding speed of the current wire feeding control period is maintained unchanged, so as to maintain the stability of the welding process, Further, the adjustment of the wire feeding speed of the current wire feeding control period includes, 1) The deviation values corresponding to each welding state characteristic value in the current wire feeding control period are determined one by one, and the deviation position of the deviation value in the adjustable interval is determined, the deviation position includes low deviation position, medium deviation position and high deviation position, Specifically, in the current wire feeding control period, the deviation values corresponding to each welding state characteristic value are obtained, which are the differences between each welding state characteristic value and the corresponding preset reference index calculated in the previous step, for each deviation value falling into the adjustable interval, the deviation position of the deviation value is determined according to the deviation amplitude of the corresponding preset reference index, and the determination rule is, Low deviation position: the deviation value exceeds ±0.5σ but does not exceed ±1σ; Medium deviation position: the deviation value exceeds ±1σ but does not exceed ±1.5σ; High deviation range: deviation value exceeds ±1.5σ, but does not exceed ±2σ, The low deviation range indicates that the welding state slightly deviates from the preset reference index, the medium deviation range indicates that the deviation degree is moderate, and the high deviation range indicates that the deviation is larger but still within the adjustable range; 2) Among all the deviation values corresponding to the welding state characteristic values in the adjustable interval, the absolute value of the largest deviation value is taken as the dominant deviation value of the current wire feeding control period. When there are two or more deviation values with the same absolute value and the maximum value, the dominant deviation value is determined according to the characteristic priority, Specifically, all the deviation values corresponding to the welding state characteristic values determined to fall within the adjustable interval are extracted to form a candidate deviation value set. First, the absolute value of each deviation value in the candidate deviation value set is taken. The size of the absolute value reflects the influence intensity of the welding state characteristic on the deviation degree of the welding state. Therefore, the absolute value of the largest deviation value is selected as the dominant deviation value of the current wire feeding control period to ensure that the most significant deviation affecting the welding stability is corrected first. The sign of the determined dominant deviation value is used to indicate the speed adjustment direction. When the deviation value is positive, it indicates that the wire feeding speed is too fast and needs to be reduced. When the deviation value is negative, it indicates that the wire feeding speed is too slow and needs to be increased. When there are two or more deviation values with the same absolute value and the maximum value, the characteristic priority is determined according to the characteristic priority order recorded in the reference database. The characteristic priority order is set based on process sensitivity and historical experimental results, and the order is as follows: arc length deviation > arc length fluctuation amplitude > molten pool geometry change amplitude > molten pool surface fluctuation amplitude > molten pool temperature change amplitude. If the arc length deviation is the largest among the other deviations, the arc length deviation is selected as the dominant deviation value. If the arc length deviation is not the largest, the above order is followed to select the dominant deviation value to ensure that the dominant deviation value can be determined in order and stability when multiple deviations coexist, 1) The deviation range corresponding to the dominant deviation value is taken as the adjustment deviation range of the current wire feeding control period, and the wire feeding speed is adjusted according to the speed adjustment amount corresponding to the deviation range. The speed adjustment amounts corresponding to the low deviation range, the medium deviation range and the high deviation range are set to ±5%, ±10% and ±15% of the current wire feeding speed, respectively, Specifically, the implementation process is as follows: the adjustable interval is divided into low deviation range, medium deviation range and high deviation range according to the amplitude, and the corresponding speed adjustment amount is set. The setting of the speed adjustment amount follows the following principles: on the one hand, the performance of the welding forming quality and the molten pool stability under different deviation amplitudes in the historical welding experimental data is considered, and on the other hand, the stability and operability of the wire feeding speed adjustment are considered to avoid excessive adjustment causing arc fluctuation; The speed adjustment amount corresponding to the low deviation range is set to ±5% of the current wire feeding speed. The reason is that the deviation has a small amplitude and has limited impact on the welding state. The stability of the molten pool and the quality of the seam formation do not show obvious abnormalities. Small adjustments are sufficient to fine-tune the molten pool state to avoid excessive adjustments that cause arc disturbances. The speed adjustment amount corresponding to the medium deviation range is set to ±10% of the current wire feeding speed. The reason is that the deviation starts to affect the welding continuity and the molten pool fluctuation. If not adjusted in time, it may cause defects. Medium amplitude adjustment can quickly correct the deviation while ensuring dynamic stability. The speed adjustment amount corresponding to the high deviation range is set to ±15% of the current wire feeding speed. The reason is that the deviation is obvious and the risk of welding defects is significant. It must be corrected first. Large adjustments ensure that the deviation is pulled back to a controllable range in the shortest time. The deviation range corresponding to the dominant deviation value is set as the adjustment deviation range of the current wire feeding control period. The speed adjustment amount corresponding to the corresponding range is adjusted. When the dominant deviation value is positive, it means that the current wire feeding speed is higher than the optimal matching state, and the wire feeding speed should be reduced. When the dominant deviation value is negative, it means that the current wire feeding speed is lower than the optimal matching state, and the wire feeding speed should be increased.

[0020] The wire feeding speed adjustment amount corresponding to each range is a fixed percentage of the current wire feeding speed. This is obtained through a large number of welding experiments. The influence of the dominant deviation value on the welding quality and the stability of the molten pool under different wire feeding speeds is recorded. Statistical analysis is obtained. The relationship between the speed adjustment amount and the experimental weld quality effect is shown in Table 1. Table 1 shows the relationship between the speed adjustment amount and the experimental weld quality effect The experimental results show that the fixed adjustment amplitudes of ±5%, ±10%, and ±15% corresponding to the low, medium, and high deviation ranges can quickly correct the deviation without causing arc disturbances or excessive adjustments while ensuring weld quality. It can effectively control the deviation within an acceptable range under different welding conditions.

[0021] Further, the execution of the abnormal processing strategy includes, The abnormal processing interval is divided into a first abnormal processing interval, a second abnormal processing interval, and a third abnormal processing interval. The abnormal processing measures corresponding to the three abnormal processing intervals are executed. Specifically, the abnormal processing interval level to which each type of welding state characteristic value belongs is divided according to the deviation amplitude of the deviation value of the welding state characteristic value exceeding the preset reference index corresponding thereto. The determination rule is that, The first abnormal processing interval: the deviation value exceeds ±2σ but does not exceed ±2.5σ. Secondary exception handling interval: deviation value exceeds ±2.5σ, but not more than ±3σ; Tertiary exception handling interval: deviation value exceeds ±3σ, The specific condition triggering the first-level exception handling measure is that when the deviation value corresponding to any welding state characteristic value falls within the first-level exception handling interval, the wire feeding speed is restored to the initial wire feeding speed of the welding process, and in the subsequent two consecutive wire feeding control periods, the speed adjustment amount in a single wire feeding control period is limited to ±5% of the current wire feeding speed, and it is monitored whether the deviation value corresponding to the welding state characteristic value falling within the exception handling interval returns to the stable interval or the adjustable interval, if it does not return in the two consecutive wire feeding control periods, it is upgraded to the exception handling measure corresponding to the secondary exception handling interval for processing. Because it is not appropriate to try and error by large continuous adjustment in the exception handling stage, large speed adjustment mutation itself may cause greater disturbance to the arc or molten pool, so the speed adjustment amount is limited, and it is observed in a short period of time, which is a protection priority and gradual repair strategy; The specific condition triggering the second-level exception handling measure is that when the deviation value corresponding to any welding state characteristic value falls within the secondary exception handling interval, the wire feeding speed is restored to the initial wire feeding speed of the welding process, the welding current is reduced to 90% of the initial welding current of the welding process, the shielding gas flow is increased to 110% of the initial shielding gas flow of the welding process, and in the subsequent two consecutive wire feeding control periods, the speed adjustment amount in a single wire feeding control period is limited to ±5% of the current wire feeding speed, and it is monitored whether the deviation value corresponding to the welding state characteristic value falling within the exception handling interval returns to the stable interval or the adjustable interval, if it does not return in the two consecutive wire feeding control periods, it is upgraded to the exception handling measure corresponding to the tertiary exception handling interval for processing. The specific condition triggering the third-level exception handling measure is that when the deviation value corresponding to any welding state characteristic value falls within the tertiary exception handling interval, the wire feeding is immediately stopped, the welding current is reduced to a safe standby value, and the shielding gas flow is maintained or increased to protect the weld area, and an alarm is triggered and relevant operation data is recorded, after manual inspection and confirmation, the wire feeding speed is restored or rework measures are taken according to manual instructions.

[0022] In the second embodiment, based on the same inventive concept as the welding robot adaptive wire feeding control method in the foregoing embodiments, as shown in Figure 2 The welding robot adaptive wire feeding control platform is provided, and the platform and the method embodiments in the present application are based on the same inventive concept. The platform comprises: The historical data calculation module 11 is configured to calculate the arc voltage sequence and the characteristic parameter sequence reflecting the molten pool state collected in each experiment based on historical welding experiment data, to obtain welding state characteristic values corresponding to each wire feeding control period, and the welding state characteristic values are used to represent the welding state, including the arc length deviation, the arc length fluctuation amplitude and the molten pool stability index. The reference database construction module 12 is configured to fit a mapping relationship function among the welding process parameters, the welding condition parameters and the welding state characteristic values corresponding to each wire feeding control period, and store the mapping relationship function together with the corresponding welding process parameters and welding condition parameters as a database entry to construct a reference database. The reference index generation module 13 is configured to find a database entry matching the welding process parameters and the welding condition parameters of the current wire feeding control period in the reference database in the current wire feeding control period, and calculate the preset reference index corresponding to the wire feeding control period by calling the mapping relationship function stored in the entry, which is used for subsequent welding state evaluation. The real-time data calculation module 14 is configured to calculate the welding state characteristic values of the current wire feeding control period according to the arc voltage sequence and the characteristic parameter sequence reflecting the molten pool state collected in the current wire feeding control period in the welding process according to the sampling frequency. The deviation calculation module 15 is configured to compare each type of welding state characteristic value after standardization with the corresponding preset reference index one by one to obtain the deviation value corresponding to each type of welding state characteristic value. The wire feeding adjustment module 16 is configured to determine the hierarchical threshold interval to which the deviation value corresponding to each type of welding state characteristic value belongs, the hierarchical threshold interval includes a stable interval, an adjustable interval and an abnormal processing interval, and dynamically adjust the wire feeding speed according to the hierarchical adjustment strategy to realize the dynamic matching of the wire feeding speed and the welding state.

[0023] The welding robot adaptive wire feeding control platform provided by the embodiment of the application can execute the welding robot adaptive wire feeding control method provided by any embodiment of the application, and has the corresponding function modules and beneficial effects of the execution method.

[0024] The embodiment of the application further provides a computer readable storage medium, In some embodiments, the computer readable storage medium stores computer executable instructions for executing the welding robot adaptive wire feeding control method mentioned in embodiment one.

[0025] Although the present application makes various references to certain modules in the platform according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, the various units and modules included are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific name of each functional unit is only for the convenience of mutual differentiation, and does not limit the protection scope of the present application.

[0026] The above detailed description does not constitute a limitation on 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 principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method of adaptive wire feeding control for a welding robot, characterized in that, The method comprises the following steps, Based on historical welding experiment data, the arc voltage sequence collected in each experiment and the characteristic parameter sequence reflecting the molten pool state in each wire feeding control period are calculated to obtain the welding state characteristic value corresponding to each wire feeding control period, which is used to represent the welding state, including the arc length deviation, the arc length fluctuation amplitude and the molten pool stability index; Based on the welding process parameters, welding condition parameters and welding state characteristic value corresponding to each wire feeding control period, the mapping relationship function among the three is fitted, and the mapping relationship function is stored together with the corresponding welding process parameters and welding condition parameters as a database entry to construct a reference database; In the current wire feeding control period, the database entry matching the welding process parameters and welding condition parameters of the current wire feeding control period is searched in the reference database, and the mapping relationship function stored in the entry is called to calculate the preset reference index corresponding to the wire feeding control period, which is used for subsequent welding state evaluation; According to the arc voltage sequence collected at the sampling frequency in the current wire feeding control period during the welding process and the characteristic parameter sequence reflecting the molten pool state, the welding state characteristic value of the current wire feeding control period is calculated. After the welding state characteristic values in the current wire feeding control period are standardized respectively, they are compared with the corresponding preset reference index one by one to obtain the deviation value corresponding to each welding state characteristic value. According to the deviation value corresponding to each welding state characteristic value, the classification threshold interval to which the welding state characteristic value belongs is determined, and the classification threshold interval includes the stable interval, the adjustable interval and the abnormal processing interval, and the wire feeding speed is dynamically adjusted according to the classification adjustment strategy to realize the dynamic matching of the wire feeding speed and the welding state.

2. The welding robot adaptive wire feeding control method of claim 1, wherein, The calculation of the arc length deviation comprises, The arc voltage sequence obtained in the wire feeding control period is preprocessed; Based on the conversion relationship between the welding arc voltage and the arc length, the processed arc voltage sequence is converted into the corresponding arc length value sequence; The arc length value sequence is calculated to obtain the average arc length in the wire feeding control period; The difference between each sampling point in the arc length value sequence and the average arc length forms a deviation sequence, and the arc length deviation is obtained by calculating the standard deviation of the deviation sequence.

3. The adaptive wire feeding control method for a welding robot according to claim 2, characterized by, The calculation of the arc length fluctuation amplitude comprises determining the maximum value and the minimum value of the arc length value sequence based on the arc length value sequence, and calculating the difference between the maximum value and the minimum value to obtain the arc length fluctuation amplitude.

4. The welding robot adaptive wire feeding control method of claim 3, wherein, The calculation of the molten pool stability index comprises, The characteristic parameter sequence reflecting the molten pool state obtained in the wire feeding control period is sorted, denoised and outlier removed, and the characteristic parameter sequence reflecting the molten pool state includes the molten pool geometric shape change characteristic parameter sequence, the surface fluctuation characteristic parameter sequence and the temperature change characteristic parameter sequence; Based on the processed molten pool geometric shape change characteristic parameter sequence, the surface fluctuation characteristic parameter sequence and the temperature change characteristic parameter sequence, statistical analysis is performed respectively to calculate the standard deviation of each sequence, and the molten pool stability index reflecting the stability of the molten pool in the wire feeding control period is obtained, including the molten pool geometric shape change amplitude, the molten pool surface fluctuation amplitude and the molten pool temperature change amplitude.

5. The welding robot adaptive wire feeding control method of claim 4, wherein, The fitting process of the mapping relationship function comprises, The welding process parameters, welding condition parameters and welding state characteristic values are quantized and standardized, so that parameters of different dimensions can directly participate in multiple linear regression fitting; For each type of welding state characteristic value, a mapping relationship function is established as follows: wherein, represents the value of the normalized welding state characteristic value of the i-th type (i=1, 2, 3, 4, 5), k k= 1-5), represents each parameter value of the parameter vector after quantization and normalization, represents the corresponding regression coefficient of each parameter after quantization and normalization, is a constant term;​ Based on the mapping relationship function, the residual between the calculated value of the standardized parameter and the standardized welding state characteristic value is compared, and the least square method is used for optimization to obtain the optimal constant term and regression coefficient; The established mapping relationship function is statistically tested to ensure the effectiveness of the mapping relationship function.

6. The welding robot adaptive wire feeding control method of claim 5, wherein, The hierarchical adjustment strategy comprises, in the current wire feeding control period, when any welding state characteristic value corresponding deviation value falls into the abnormal processing interval, executing the abnormal processing strategy; when all welding state characteristic value corresponding deviation values are not in the abnormal processing interval, and at least one deviation value falls into the adjustable interval, adjusting the wire feeding speed of the current wire feeding control period according to the deviation amplitude; when all welding state characteristic value corresponding deviation values are in the stable interval, maintaining the wire feeding speed of the current wire feeding control period.

7. The welding robot adaptive wire feeding control method of claim 6, wherein, The adjustment of the wire feeding speed of the current wire feeding control period comprises, The deviation values corresponding to each welding state characteristic value in the current wire feeding control period are determined one by one to determine the deviation range of the deviation value in the adjustable interval, and the deviation range comprises a low deviation range, a medium deviation range and a high deviation range; Among all the deviation values corresponding to the welding state characteristic values in the adjustable interval, the deviation value with the largest absolute value is taken as the dominant deviation value of the current wire feeding control period, and when there are two or more deviation values with the same maximum absolute value, the dominant deviation value is determined according to the characteristic priority; The deviation range corresponding to the dominant deviation value is taken as the adjustment deviation range of the current wire feeding control period, and the wire feeding speed is adjusted according to the speed adjustment amount corresponding to the deviation range, wherein the speed adjustment amounts corresponding to the low deviation range, the medium deviation range and the high deviation range are set to ±5%, ±10% and ±15% of the current wire feeding speed respectively.

8. The welding robot adaptive wire feeding control method of claim 7, wherein, The execution of the abnormal processing strategy comprises, The abnormal processing interval is divided into a first-level abnormal processing interval, a second-level abnormal processing interval and a third-level abnormal processing interval, and the abnormal processing measures corresponding to the three abnormal processing intervals are executed; When the first-level abnormal processing measure is triggered, the wire feeding speed is restored to the initial wire feeding speed of the welding process, and the limited speed adjustment amount processing is performed in the next two continuous wire feeding control periods; When the second-level abnormal processing measure is triggered, on the basis of the first-level abnormal processing measure, the welding current of the welding process is reduced, and the protective gas flow is increased to enhance the arc stability and prevent the welding defects from further expanding; When the third-level abnormal processing measure is triggered, the wire feeding is immediately stopped, the welding current is reduced to a safety standby value, and the protective gas flow is maintained or increased to protect the weld area, and an alarm is triggered and relevant operation data is recorded.

9. A welding robot adaptive wire feeding control platform, characterized by, A welding robot adaptive wire feeding control platform for implementing the steps of the welding robot adaptive wire feeding control method in any one of claims 1 to 8 comprises, a historical data calculation module, configured to calculate, based on historical welding experiment data, an arc voltage sequence collected in each experiment in each wire feeding control period and a characteristic parameter sequence reflecting a molten pool state, to obtain a welding state characteristic value corresponding to each wire feeding control period, the welding state characteristic value being used to represent a welding state, including an arc length deviation, an arc length fluctuation amplitude, and a molten pool stability index; a reference database construction module, configured to fit a mapping relationship function among welding process parameters, welding condition parameters, and welding state characteristic values corresponding to each wire feeding control period, and store the mapping relationship function together with corresponding welding process parameters and welding condition parameters as a database entry to construct a reference database; a reference index generation module, configured to find, in the reference database, a database entry matching welding process parameters and welding condition parameters of a current wire feeding control period in a current wire feeding control period, and calculate a preset reference index corresponding to the wire feeding control period by calling the mapping relationship function stored in the entry, for subsequent welding state evaluation; a real-time data calculation module, configured to calculate a welding state characteristic value of a current wire feeding control period according to an arc voltage sequence collected in the current wire feeding control period in a welding process at a sampling frequency and a characteristic parameter sequence reflecting a molten pool state; a deviation calculation module, configured to compare each type of welding state characteristic value after standardized processing with a corresponding preset reference index one by one to obtain a deviation value corresponding to each type of welding state characteristic value; a wire feeding adjustment module, configured to determine a hierarchical threshold interval to which each type of welding state characteristic value corresponds according to a deviation value corresponding to each type of welding state characteristic value, the hierarchical threshold interval including a stable interval, an adjustable interval, and an abnormal processing interval, and dynamically adjust a wire feeding speed according to a hierarchical adjustment strategy to realize dynamic matching of the wire feeding speed and the welding state.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 9. The computer program is executed by the processor to implement the welding robot adaptive wire feeding control method in any one of claims 1 to 8.

Images