CN113441827B - Automatic generation method and system for resistance spot welding technological parameters - Google Patents
Automatic generation method and system for resistance spot welding technological parameters Download PDFInfo
- Publication number
- CN113441827B CN113441827B CN202110812975.4A CN202110812975A CN113441827B CN 113441827 B CN113441827 B CN 113441827B CN 202110812975 A CN202110812975 A CN 202110812975A CN 113441827 B CN113441827 B CN 113441827B
- Authority
- CN
- China
- Prior art keywords
- welding
- parameters
- value
- parameter
- result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K11/00—Resistance welding; Severing by resistance heating
- B23K11/10—Spot welding; Stitch welding
- B23K11/11—Spot welding
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K11/00—Resistance welding; Severing by resistance heating
- B23K11/36—Auxiliary equipment
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for automatically generating resistance spot welding technological parameters, and belongs to the technical field of automobile welding manufacturing. The method solves the problems that the welding parameters obtained in the prior art cannot be determined to be the optimal parameters and the welding quality of each welding spot cannot be guaranteed. The automatic generation method of the process parameters of the resistance spot welding comprises the following steps: acquiring plate information parameters, welding process parameters and welding result parameters; establishing a welding result evaluation function according to the welding result parameters, and further calculating according to the welding result evaluation function to obtain a welding result evaluation value; taking the plate information parameters and the welding process parameters as input, and taking the welding result evaluation value as output so as to establish a welding result prediction model; and predicting the welding result predicted value of the specific plate information parameter according to the welding result prediction model, and further generating the welding process parameter with the highest welding result predicted value corresponding to the specific plate information parameter. The invention can obtain high-quality welding process parameters and improve the welding quality.
Description
Technical Field
The invention belongs to the technical field of automobile welding manufacturing, and relates to a method and a system for determining parameters of a resistance spot welding process.
Background
At present, new vehicle models are more and more released by various automobile manufacturers, the release time interval is shorter and shorter, and meanwhile, the requirements of users on quality are higher and higher. The resistance spot welding (hereinafter referred to as welding) can realize high-strength connection between plates, 3000-5000 resistance spot welding spots are contained in one automobile in automobile manufacturing, and the welding quality of each spot welding spot directly influences the safety performance of the automobile. Therefore, welding is one of the most central processes in the whole vehicle manufacturing, and how to quickly adjust parameters with optimal process quality is a problem to be faced.
At present, the debugging of welding parameters is generally based on the experience of debugging personnel, after initial process parameters are input, the process parameters meeting the quality requirements are set finally through continuous debugging and improvement under the actual working condition on site and combining quality detection means such as multiple rounds of chisel detection and breaking detection, and the process needs to consume a large amount of time and resources.
In order to solve the existing problems, the existing chinese patent literature discloses a method, an apparatus, an electronic device, and a storage medium for determining welding parameters of resistance spot welding, wherein the method includes that a server obtains a welding spot information set of a vehicle, performs characteristic processing on the welding spot information set, that is, a non-numerical information set in the welding spot information set is converted into a 0-1 matrix data set, reduces a numerical information set in the welding spot information set according to a preset proportion so that the numerical values in the numerical information set are all within a preset interval, adds the numerical information in the reduced numerical information set to corresponding noise numerical values in the noise numerical set to obtain a processed welding spot information set, and determines the welding parameters of the processed welding spot information set based on a trained welding parameter determination model to obtain the welding parameters corresponding to the processed welding spot information set. Although the method does not need to manually refer to a large number of existing welding parameters, the training cost of training engineers can be saved, the time cost can also be saved, the welding parameters obtained through the method cannot be determined to be the optimal parameters, the welding quality of each welding point cannot be guaranteed during production and manufacturing, and the safety performance of the vehicle is influenced.
Disclosure of Invention
The invention aims to provide a method and a system for automatically generating process parameters of resistance spot welding aiming at the problems in the prior art, and the technical problems to be solved are as follows: how to obtain high-quality welding process parameters and improve the welding quality.
The purpose of the invention can be realized by the following technical scheme: a method for automatically generating technological parameters of resistance spot welding comprises the following steps:
acquiring plate information parameters, welding process parameters and welding result parameters in a workshop resistance spot welding process;
establishing a welding result evaluation function according to the welding result parameters, and further calculating according to the welding result evaluation function to obtain a welding result evaluation value;
taking the plate information parameters and the welding process parameters as input, and taking the welding result evaluation value as output to carry out model training, thereby establishing a welding result prediction model;
and predicting the welding result predicted value of the specific plate information parameter according to the welding result prediction model, and further generating the welding process parameter with the highest welding result predicted value corresponding to the specific plate information parameter.
When the automatic generation method of the resistance spot welding technological parameters is used, actual welding data including plate information parameters, welding technological parameters, welding result parameters and the like in the workshop resistance spot welding process are firstly obtained, the plate information parameters, the welding technological parameters and the welding result parameters are in one-to-one correspondence, and the actual welding data includes actual welding data in the original resistance spot welding process of a workshop and actual welding data in the current resistance spot welding process of the workshop. The welding result parameters are calculated through a welding result evaluation function to obtain welding result evaluation values, then, a large number of welding result evaluation values obtained through calculation and corresponding plate information parameters and welding process parameters are subjected to data training, a welding result prediction model is further established, the established welding result prediction model can obtain predicted welding result prediction values when different plate information parameters and different welding process parameters are input, and optimal welding process parameters can be obtained by comparing the sizes of the welding result prediction values, namely the welding process parameters with the highest welding result evaluation values under the condition of specific plate information parameters are the optimal welding process parameters. The automatic generation method of the resistance spot welding process parameters adopts actual welding data in the workshop production process to carry out model training, the data is closer to the actual situation, the accuracy of the established model is high, and compared with the actual measurement work in a laboratory, the cost is reduced.
In the automatic generation method of the process parameters of the resistance spot welding, the welding result parameters comprise a spatter value, a quality factor and a process stability factor. The quality factor is a quality index established on resistance, voltage, current and time, whether a problem of welding core is small or insufficient welding exists can be judged through the quality index, the process stability factor is calculated based on the voltage, the current and the prolonged welding time, the display value relates to the difference between a sample curve and an actual curve, the reference value is 100, and the range is 0-200.
In the above method for automatically generating process parameters of resistance spot welding, the calculation formula of the welding result evaluation function is as follows:
wherein F is a welding result evaluation value; UIP is a quality factor; f is the spatter value; PSF is the process stability factor. The spattering value, the quality factor and the process stability factor can be converted into a numerical value through the welding result evaluation function, namely the welding result evaluation value, the quality of the welding result is represented according to the welding result evaluation value, and the welding quality is indicated to be better when the welding result evaluation value is higher.
In the automatic generation method of the resistance spot welding process parameters, the plate information parameters and the welding process parameters are used as input, and the algorithm used for carrying out model training by using the welding result evaluation value as output is a linear regression algorithm.
In the automatic generation method of the resistance spot welding process parameters, the operation of performing model training by adopting a linear regression algorithm comprises the following steps:
constructing a sample set according to the acquired plate information parameters, the acquired welding process parameters and the acquired welding result evaluation values;
dividing the sample set into a training set and a test set;
performing model training by using the training set, and establishing a welding result prediction model;
and testing the welding result prediction model by using the test set, evaluating the test result by using the root mean square error, and judging that the welding result prediction model is effective when the error meets the requirement.
In the automatic generation method of the process parameters of the resistance spot welding, the process parameters of the welding include welding current, welding pressure and welding time. Among the welding process parameters, the welding current, the welding pressure, and the welding time are core parameters, and include a pre-pressing time, a ramp current, a ramp time, and the like, in addition to these parameters.
In the above method for automatically generating parameters for a resistance spot welding process, the operation of automatically generating the welding process parameters with the highest evaluation value of the welding result under the condition of automatically generating the specific plate information parameters according to the welding result prediction model includes:
under the condition of specific plate information parameters, respectively selecting a minimum value and a maximum value of welding current, a minimum value and a maximum value of welding pressure and a minimum value and a maximum value of welding time from welding process parameters;
selecting a plurality of welding current values between the minimum value and the maximum value of the welding current, selecting a plurality of welding pressure values between the minimum value and the maximum value of the welding pressure, selecting a plurality of welding time values between the minimum value and the maximum value of the welding time, and combining the plurality of welding current values, the plurality of welding pressure values and the plurality of welding time values to construct and obtain various possible parameter combinations;
predicting a welding result prediction value corresponding to each parameter combination according to the welding result prediction model and by taking the constructed parameter combinations as input;
and selecting the parameter combination with the highest predicted value of the welding result as the optimal welding process parameter under the condition of the specific plate information parameter. The welding parameters acquired on site may not be the optimal parameters, and in the method, based on the welding result prediction model, the welding process parameter combination under the condition of complete specific plate information parameters can be traversed with a certain step length precision, the data is complete, and the relatively optimal welding process parameters can be obtained.
In the automatic generation method of the process parameters of the resistance spot welding, the selection operations of a plurality of welding current values, a plurality of welding voltage values and a plurality of welding time values are as follows:
selecting a welding current value between the minimum value and the maximum value of the welding current according to the respective fixed step length, selecting a welding pressure value between the minimum value and the maximum value of the welding pressure, and selecting a welding time value between the minimum value and the maximum value of the welding time;
the calculation formulas of the respective fixed step lengths are as follows: (max-min)/10.
The values are taken between the minimum value and the maximum value of the respective process parameters, so that the optimal welding process parameters can be obtained more effectively and more accurately.
In the above method for automatically generating process parameters for resistance spot welding, after predicting the welding result prediction value of a specific plate information parameter according to a welding result prediction model and further generating a welding process parameter with the highest welding result prediction value corresponding to the specific plate information parameter, the method further comprises:
and integrating the welding process parameters with the highest predicted values of the welding results to form a database of the optimal welding process parameters under the condition of different plate information parameters. And the database is established, so that the optimal welding process parameters can be directly obtained according to the plate information parameters when the database is used at the later stage. The travel of the database can also be used as an initial set value of welding process parameters when the welding debugging work is started.
In the automatic generation method of the resistance spot welding technological parameters, the plate information parameters, the welding technological parameters and the welding result parameters in the workshop resistance spot welding process are obtained through the Internet of things technology.
An automatic generation system for resistance spot welding process parameters, comprising:
the data acquisition unit is used for acquiring plate information parameters, welding process parameters and welding result parameters in the workshop resistance spot welding process;
the evaluation function construction unit is used for establishing a welding result evaluation function according to the welding result parameters;
the model construction unit is used for inputting the plate information parameters and the welding process parameters, and performing model training by taking a welding result evaluation value obtained by calculation according to a welding result evaluation function as output so as to establish a welding result prediction model;
and the parameter automatic generation unit is used for predicting the welding result predicted value of the specific plate information parameter according to the welding result prediction model so as to generate the welding process parameter with the highest welding result predicted value corresponding to the specific plate information parameter.
The system for automatically generating the resistance spot welding process parameters acquires actual welding data including plate information parameters, welding process parameters, welding result parameters and the like in the workshop resistance spot welding process through the data acquisition unit, wherein the plate information parameters, the welding process parameters and the welding result parameters are in one-to-one correspondence, and the actual welding data includes actual welding data in the original workshop resistance spot welding process and actual welding data in the current workshop resistance spot welding process. The method comprises the steps that an evaluation function construction unit constructs a welding result evaluation function according to welding result parameters, welding result parameters are calculated through the welding result evaluation function to obtain a welding result evaluation value, then a model construction unit conducts data training on a large number of welding result evaluation values obtained through calculation and corresponding plate information parameters and welding process parameters, a welding result prediction model is further established, a parameter automatic generation unit can obtain a predicted welding result prediction value when different plate information parameters and different welding process parameters are input according to the established welding result prediction model, and the optimal welding process parameters can be obtained through comparison of the welding result prediction values, namely the welding process parameters when the welding result evaluation value is the highest under the condition of specific plate information parameters are the optimal welding process parameters. The automatic generation system for the resistance spot welding process parameters adopts actual welding data in a workshop production process to perform model training, the data is closer to the actual situation, the accuracy of the established model is high, and compared with the actual measurement work in a laboratory, the cost is reduced.
In the automatic generation system for the resistance spot welding technological parameters, the data acquisition unit is in communication connection with a data server of the resistance spot welding workshop through the internet of things and is used for acquiring the plate information parameters, the welding technological parameters and the welding result parameters which are stored in the data server.
In the automatic generation system of the resistance spot welding process parameters, the evaluation function construction unit constructs a welding result evaluation function through the following calculation formula;
wherein, F is a welding result evaluation value; UIP is a quality factor; f is the spatter value; PSF is the process stability factor.
In the above automatic generation system of resistance spot welding process parameters, the automatic parameter generation unit includes:
the parameter acquisition module is used for acquiring the minimum value and the maximum value of welding current, the minimum value and the maximum value of welding pressure and the minimum value and the maximum value of welding time in the welding process parameters under the condition of acquiring the specific plate information parameters;
the value-taking module is used for selecting a plurality of welding current values between the minimum value and the maximum value of the welding current, selecting a plurality of welding pressure values between the minimum value and the maximum value of the welding pressure, and selecting a plurality of welding time values between the minimum value and the maximum value of the welding time;
the parameter combination construction module is used for combining a plurality of welding current values, a plurality of welding pressure values and a plurality of welding time values so as to construct and obtain various possible parameter combinations;
and the parameter generation module is used for predicting the welding result predicted value corresponding to each parameter combination according to the welding result prediction model and taking the constructed parameter combination as input, and further selecting the parameter combination with the highest welding result predicted value as the optimal welding process parameter under the condition of the specific plate information parameter. In the system, the parameter automatic generation unit can traverse the welding process parameter combination under the condition of complete specific plate information parameters with certain step length precision based on a welding result prediction model, the data is complete, and the relatively optimal welding process parameters can be obtained.
Compared with the prior art, the automatic generation method and the system for the resistance spot welding process parameters have the following advantages:
1. according to the invention, firstly, the welding result parameters are evaluated to obtain the welding result evaluation value, and then the welding result prediction model is established according to the welding result evaluation value, the plate information parameters and the welding process parameters, and the welding result prediction model can be used for quickly and accurately obtaining the optimal welding process parameters, so that the welding quality and efficiency of resistance spot welding are improved.
2. The method adopts the existing welding data in the actual workshop, does not need a large amount of additional welding actual measurement work on one hand, greatly reduces the cost, and on the other hand, the welding working condition of the data source is closer to the actual condition relative to the laboratory, thereby improving the guarantee for obtaining high-quality welding process parameters.
3. According to the invention, the optimal welding process parameter database can be formed through the welding result prediction model under the condition of different plate information parameters, and can be used as an initial set value of the welding process parameter when a welding debugging project is started, so that the operation is more convenient, and the data processing process can be repeated for the increase of new welding equipment data in the follow-up process, so that a further perfect optimal welding process parameter database is finally obtained, and the prediction model is more accurate along with the increase of the acquired data.
Drawings
Fig. 1 is a control flow chart according to a first embodiment of the present invention.
Fig. 2 is a control flow chart of a second embodiment of the present invention.
Fig. 3 is a schematic structural diagram of the present invention.
In the figure, 1, a data acquisition unit; 2. an evaluation function construction unit; 3. a model construction unit; 4. a parameter automatic generation unit; 41. a parameter acquisition module; 42. a value taking module; 43. a parameter combination construction module; 44. and a parameter generation module.
Detailed Description
The following are specific embodiments of the present invention and are further described with reference to the drawings, but the present invention is not limited to these embodiments.
The first embodiment is as follows:
as shown in fig. 3, the automatic generation system of resistance spot welding process parameters includes: the data acquisition unit 1 is in communication connection with a data server of the resistance spot welding workshop through the Internet of things and is used for acquiring plate information parameters, welding process parameters and welding result parameters stored in the data server;
the evaluation function construction unit 2 is used for establishing a welding result evaluation function according to the welding result parameters; specifically, a welding result evaluation function is constructed through the following calculation formula;
wherein F is a welding result evaluation value; UIP is a quality factor; f is the spatter value; PSF is a process stability factor;
the model construction unit 3 is used for inputting the plate information parameters and the welding process parameters, calculating a welding result evaluation value according to a welding result evaluation function, and outputting the welding result evaluation value to perform model training so as to establish a welding result prediction model;
the parameter automatic generation unit 4 is configured to predict a welding result prediction value of the specific plate information parameter according to the welding result prediction model, and further generate a welding process parameter with a highest welding result prediction value corresponding to the specific plate information parameter, and specifically includes the following modules:
the parameter obtaining module 41 is configured to obtain a minimum value and a maximum value of a welding current, a minimum value and a maximum value of a welding pressure, and a minimum value and a maximum value of a welding time in the welding process parameters under the condition of obtaining the information parameters of the specific plate;
the value-taking module 42 is used for selecting a plurality of welding current values between the minimum value and the maximum value of the welding current, selecting a plurality of welding pressure values between the minimum value and the maximum value of the welding pressure, and selecting a plurality of welding time values between the minimum value and the maximum value of the welding time;
a parameter combination construction module 43, configured to combine a plurality of welding current values, a plurality of welding pressure values, and a plurality of welding time values to construct various possible parameter combinations;
and the parameter generating module 44 is configured to predict a welding result predicted value corresponding to each parameter combination according to the welding result prediction model and using the constructed parameter combination as input, and further select a parameter combination with a highest welding result predicted value as an optimal welding process parameter under the condition of a specific plate information parameter. The parameter obtaining module 41 is connected to the value taking module 42, the value taking module 42 is connected to the parameter combination building module 43, and the parameter combination building module 43 is connected to the parameter generating module 44.
The data acquisition unit 1 is connected with the evaluation function construction unit 2, the model construction unit 3 and the parameter acquisition module 41 in the parameter automatic generation unit 4, respectively, the evaluation function construction unit 2 is connected with the model construction unit 3, and the model construction unit 3 is connected with the parameter generation module 44 in the parameter automatic generation unit 4.
The automatic generation system of the resistance spot welding process parameters realizes the automatic generation method of the resistance spot welding process parameters by setting various functional parts to respectively correspond. The working principle of the automatic generation system of the resistance spot welding technological parameters is further explained by the automatic generation method of the resistance spot welding technological parameters.
As shown in fig. 1, the automatic generation method of the resistance spot welding process parameters comprises the following steps:
the method comprises the following steps that a data acquisition unit 1 acquires plate information parameters, welding process parameters and welding result parameters in a workshop resistance spot welding process through the Internet of things technology; the welding result parameters comprise a spatter value, a quality factor and a process stability factor; in the embodiment, the welding current, the welding pressure and the welding time are taken as core parameters, and the core parameters are used for predicting the optimal welding parameter combination under the specific plate condition; the panel information parameters include parameters such as the thickness of the panel 1, the material of the panel 1, the thickness of the panel 2, the material of the panel 2, the thickness of the panel 3, the material of the panel 3, and the presence or absence of glue application.
After the plate information parameters, the welding process parameters and the welding result parameters, the evaluation function construction unit 2 establishes a welding result evaluation function according to the welding result parameters, and the specific construction process is as follows: firstly, sigmoid function transformation is performed on the spatter value f, and the formula is as follows:
Then, the dispersion calculation conversion of the quality factor UIP relative to 100 is carried out, and the formula is as follows:
And finally, multiplying the constant value of the process stability factor PSF by the constant value of the process stability factor PSF to obtain an evaluation function calculation formula of the final welding result evaluation value:
wherein F is a welding result evaluation value; UIP is a quality factor; f is the spatter value; PSF is the process stability factor.
Then, the model construction unit 3 takes the plate information parameters and the welding process parameters as input, takes the welding result evaluation value as output to perform model training, and further establishes a welding result prediction model, wherein a linear regression algorithm is adopted to perform model training, and the specific operations comprise: firstly, acquiring historical data from a welding machine in a workshop, wherein each record comprises more than 10 characteristic values such as plate information parameters, welding process parameters and the like, a label value is a welding result evaluation value (which integrates a spattering value, a PSF (safety warning message) and a UIP (unified emission warning message)) obtained by a welding result evaluation function, and a sample set is constructed according to the acquired plate information parameters, the welding process parameters and the corresponding welding result evaluation values thereof; randomly dividing each parameter data in the sample set into a test set and a sample set, namely dividing the training set and the test set according to the number of the data, for example, dividing 20% of the sample set into the test set, and dividing 80% of the sample set into the training set; selecting a least square-based linear regression algorithm and fitting by using a training set, wherein a fitting straight line y = w x Tx + b, tx is a group of vectors consisting of more than 10 characteristic values such as plate information parameters, welding process parameters and the like, y is a label value, and weight parameters w and b are obtained through training. And then testing the welding result prediction model by using the test set, namely inputting the plate information parameters and the welding process parameters in the test set into the welding result prediction model, calculating by using the welding result prediction model to obtain a welding result prediction value, comparing the welding result prediction value with a welding result evaluation value corresponding to the plate information parameters and the welding process parameters in the test set to obtain a test result, judging the test result, presetting an error acceptable range in the system, evaluating the test result by using a root mean square error, and judging that the welding result prediction model is effective when the error meets the requirement of the error acceptable range.
After the model is built, the parameter automatic generation unit 4 predicts the welding result prediction value of the specific plate information parameter according to the welding result prediction model, and further generates the welding process parameter with the highest welding result prediction value corresponding to the specific plate information parameter, and the specific operation is as follows:
under the condition that the parameter obtaining module 41 obtains the information parameters of the specific plate, respectively selecting the minimum value and the maximum value of welding current, the minimum value and the maximum value of welding pressure and the minimum value and the maximum value of welding time from the corresponding welding process parameters;
a value module 42 selects a plurality of welding current values between the minimum value and the maximum value of the welding current according to respective fixed step lengths, a plurality of welding pressure values between the minimum value and the maximum value of the welding pressure, a plurality of welding time values between the minimum value and the maximum value of the welding time, and a parameter combination construction module 43 combines the welding current values, the welding pressure values and the welding time values to construct various possible parameter combinations, wherein the calculation formulas of the respective fixed step lengths are as follows: (max-min)/10, taking the value at this fixed step size, the possible parameter combinations have 10 3 And (4) seed preparation.
And predicting a welding result predicted value corresponding to each parameter combination by the parameter generation module 44 according to the welding result prediction model and taking the constructed parameter combination as input, and finally selecting the parameter combination with the highest welding result predicted value as the optimal welding process parameter under the condition of the specific plate information parameter. In actual use, an operator can quickly select the optimal welding process parameters according to the invention under the condition of knowing the plate information parameters, so that convenience is provided for process production, the welding quality of resistance spot welding is improved, and the safety of vehicles is guaranteed.
Example two:
as shown in fig. 2, the technical solution in this embodiment is substantially the same as the technical solution in the first embodiment, except that the method further includes, after predicting the welding result prediction value of the specific plate information parameter according to the welding result prediction model, and further generating the welding process parameter with the highest welding result prediction value corresponding to the specific plate information parameter:
and integrating the welding process parameters with the highest predicted values of the welding results to form a database of the optimal welding process parameters under the condition of different plate information parameters. And the database is established, and when the database is used at the later stage, the optimal welding process parameters can be directly obtained according to the plate information parameters. The travel of the database can also be used as an initial set value of welding process parameters when the welding debugging work is started.
The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments, or alternatives may be employed, by those skilled in the art, without departing from the spirit or ambit of the invention as defined in the appended claims.
Claims (8)
1. A method for automatically generating technological parameters of resistance spot welding is characterized by comprising the following steps:
acquiring plate information parameters, welding process parameters and welding result parameters in a workshop resistance spot welding process, wherein the welding result parameters comprise a spattering value, a quality factor and a process stability factor;
establishing a welding result evaluation function according to the welding result parameters, and further calculating according to the welding result evaluation function to obtain a welding result evaluation value, wherein the calculation formula of the welding result evaluation function is as follows:
wherein F is a welding result evaluation value; UIP is a quality factor; f is the spatter value; PSF is a process stability factor;
taking the plate information parameters and the welding process parameters as input, and taking the welding result evaluation value as output to carry out model training, thereby establishing a welding result prediction model;
and predicting the welding result predicted value of the specific plate information parameter according to the welding result prediction model, and further generating the welding process parameter with the highest welding result predicted value corresponding to the specific plate information parameter.
2. The method for automatically generating parameters for a resistance spot welding process according to claim 1, wherein an algorithm used for model training with the plate information parameters and the welding process parameters as inputs and the welding result evaluation values as outputs is a linear regression algorithm.
3. The method of automatically generating resistance spot welding process parameters according to claim 2, wherein the operation of performing model training using a linear regression algorithm comprises:
constructing a sample set according to the acquired plate information parameters, the welding process parameters and the corresponding welding result evaluation values;
dividing the sample set into a training set and a test set;
performing model training by using the training set, and establishing a welding result prediction model;
and testing the welding result prediction model by using the test set, evaluating the test result by using the root mean square error, and judging that the welding result prediction model is effective when the error meets the requirement.
4. The method of automatically generating resistance spot welding process parameters according to claim 1, wherein the welding process parameters include welding current, welding pressure, and welding time; the operation of generating the welding process parameter with the highest welding result predicted value corresponding to the specific plate information parameter includes:
under the condition of specific plate information parameters, respectively selecting a minimum value and a maximum value of welding current, a minimum value and a maximum value of welding pressure and a minimum value and a maximum value of welding time from welding process parameters;
selecting a plurality of welding current values between the minimum value and the maximum value of the welding current, selecting a plurality of welding pressure values between the minimum value and the maximum value of the welding pressure, selecting a plurality of welding time values between the minimum value and the maximum value of the welding time, and combining the plurality of welding current values, the plurality of welding pressure values and the plurality of welding time values to construct and obtain various possible parameter combinations;
predicting a welding result prediction value corresponding to each parameter combination according to the welding result prediction model and by taking the constructed parameter combinations as input;
and selecting the parameter combination with the highest predicted value of the welding result as the optimal welding process parameter under the condition of the specific plate information parameter.
5. The automatic generation method of the process parameters of resistance spot welding according to claim 4, wherein the selection of the plurality of welding current values, the plurality of welding voltage values and the plurality of welding time values is performed by:
selecting a welding current value between the minimum value and the maximum value of the welding current according to the respective fixed step length, selecting a welding pressure value between the minimum value and the maximum value of the welding pressure, and selecting a welding time value between the minimum value and the maximum value of the welding time;
the calculation formulas of the respective fixed step lengths are as follows: (max-min)/10.
6. The automatic generation method of the resistance spot welding process parameters according to claim 1, wherein after predicting the welding result prediction value of the specific plate information parameter according to the welding result prediction model and further generating the welding process parameter with the highest welding result prediction value corresponding to the specific plate information parameter, the method further comprises:
and integrating the welding process parameters with the highest predicted values of the welding results to form a database of the optimal welding process parameters under the condition of different plate information parameters.
7. An automatic generation system for resistance spot welding process parameters, characterized in that the system comprises:
the welding system comprises a data acquisition unit (1) and a welding control unit, wherein the data acquisition unit is used for acquiring plate information parameters, welding process parameters and welding result parameters in the workshop resistance spot welding process, and the welding result parameters comprise a spattering value, quality factors and process stability factors;
the evaluation function construction unit (2) is used for establishing a welding result evaluation function according to the welding result parameters, and the evaluation function construction unit (2) constructs the welding result evaluation function through the following calculation formula;
wherein F is a welding result evaluation value; UIP is a quality factor; f is the spatter value; PSF is a process stability factor;
the model construction unit (3) is used for inputting the plate information parameters and the welding process parameters, calculating a welding result evaluation value according to a welding result evaluation function, and outputting the welding result evaluation value to perform model training so as to establish a welding result prediction model;
and the parameter automatic generation unit (4) is used for predicting the welding result predicted value of the specific plate information parameter according to the welding result prediction model so as to generate the welding process parameter with the highest welding result predicted value corresponding to the specific plate information parameter.
8. The automatic generation system of resistance spot welding process parameters according to claim 7, characterized in that the automatic parameter generation unit (4) comprises:
the parameter acquisition module (41) is used for acquiring the minimum value and the maximum value of welding current, the minimum value and the maximum value of welding pressure and the minimum value and the maximum value of welding time in the welding process parameters under the condition of specific plate information parameters;
the value-taking module (42) is used for selecting a plurality of welding current values between the minimum value and the maximum value of the welding current, selecting a plurality of welding pressure values between the minimum value and the maximum value of the welding pressure, and selecting a plurality of welding time values between the minimum value and the maximum value of the welding time;
the parameter combination construction module (43) is used for combining a plurality of welding current values, a plurality of welding pressure values and a plurality of welding time values so as to construct and obtain various possible parameter combinations;
and the parameter generation module (44) is used for predicting a welding result predicted value corresponding to each parameter combination according to the welding result prediction model and taking the constructed parameter combination as input, and further selecting the parameter combination with the highest welding result predicted value as the optimal welding process parameter under the condition of specific plate information parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110812975.4A CN113441827B (en) | 2021-07-19 | 2021-07-19 | Automatic generation method and system for resistance spot welding technological parameters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110812975.4A CN113441827B (en) | 2021-07-19 | 2021-07-19 | Automatic generation method and system for resistance spot welding technological parameters |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113441827A CN113441827A (en) | 2021-09-28 |
| CN113441827B true CN113441827B (en) | 2022-12-02 |
Family
ID=77816561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110812975.4A Active CN113441827B (en) | 2021-07-19 | 2021-07-19 | Automatic generation method and system for resistance spot welding technological parameters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113441827B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114918581B (en) * | 2022-05-13 | 2024-05-10 | 中科云谷科技有限公司 | Welding parameter processing method and device, storage medium and processor |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1826340A1 (en) * | 1989-04-03 | 1998-07-10 | Волжское объединение по производству легковых автомобилей АвтоВАЗ | Apparatus for testing spot welding quality |
| JP4667678B2 (en) * | 2001-09-20 | 2011-04-13 | 中央精機株式会社 | Arc welding quality evaluation system |
| JP2008164482A (en) * | 2006-12-28 | 2008-07-17 | Daido Steel Co Ltd | Method for evaluating welding quality |
| DE102009016798A1 (en) * | 2009-04-07 | 2010-10-14 | Daimler Ag | Method and control unit for monitoring a quality of welding points of a resistance welding gun |
| US10632570B2 (en) * | 2009-08-27 | 2020-04-28 | Ihi Inspection & Instrumentation Co. Ltd. | Laser welding quality determination method and apparatus |
| EP3412397B1 (en) * | 2017-06-06 | 2020-11-18 | Robert Bosch GmbH | Method of predicting welding spatters during a resistance welding process |
| CN112487708A (en) * | 2020-11-24 | 2021-03-12 | 山西三友和智慧信息技术股份有限公司 | Resistance spot welding quality prediction method based on ensemble learning |
-
2021
- 2021-07-19 CN CN202110812975.4A patent/CN113441827B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113441827A (en) | 2021-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113441827B (en) | Automatic generation method and system for resistance spot welding technological parameters | |
| CN109675935B (en) | Online fault diagnosis method for IPCA rolling process with variable control limit | |
| CN111015006B (en) | Online monitoring method for laser-arc hybrid welding quality based on spectral information | |
| CN112867580A (en) | Method and device for machining workpieces | |
| CN116560327B (en) | Electric tool production control method and system based on Internet of things | |
| CN110614416B (en) | Method for optimizing welding parameters of dual-phase high-strength steel | |
| CN112001125A (en) | Failure rate prediction method for industrial robot | |
| CN105302123A (en) | Online data monitoring method | |
| CN110817694B (en) | Load hoisting weight calculation method and device and storage medium | |
| CN111992869B (en) | Predictive maintenance method for electron beam welding equipment based on edge calculation | |
| CN112597435B (en) | Thermal power equipment quality data processing method and device based on equipment supervision | |
| CN116252039B (en) | Intelligent control method and system for inertia friction welding machine | |
| DE102017115766A1 (en) | Method and system for operating a storage unit | |
| CN113161599A (en) | Method and device for monitoring lithium separation of battery | |
| CN113435794B (en) | Intelligent monitoring method for post-processing of nozzle casting based on image processing | |
| CN110108239B (en) | Pole piece quality information acquisition method, system and equipment | |
| CN110781615A (en) | CAE simulation accuracy evaluation method | |
| CN110954349A (en) | Crane structure health state monitoring method based on residual stress distortion rate | |
| DE102015111316A1 (en) | Hydraulic unit for screw process control | |
| CN116183111B (en) | Dynamic balance uniformity optimization method and system for tire | |
| CN116735235A (en) | Axle load and load state detection control system thereof | |
| KR102089152B1 (en) | Spot welding method | |
| CN117341261B (en) | Intelligent control method and system for servo direct-drive screw press | |
| CN108250342A (en) | A kind of perfluoroethylene-propylene product processes and system | |
| CN117094072A (en) | Whole vehicle parameter scheme optimizing method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |