Intelligent welding method, intelligent welding system and computer storage medium

Abstract

The present invention discloses an intelligent welding method, an intelligent welding system and a computer storage medium, which belong to the field of welding technology and are designed to solve the technical problem of low welding precision in the prior art. The method comprises: predicting linear data by a linear model to obtain a first prediction result; and / or predicting nonlinear data by a nonlinear model to obtain a second prediction result; obtaining target operating parameters based on the first prediction result and / or the second prediction result, and emitting a target laser to weld the current object to be welded according to the target operating parameters by a laser body. The present invention can improve welding precision.

Description

Technical Field

[0001] The invention belongs to the field of laser technology, and in particular relates to an intelligent welding method, an intelligent welding system and a computer storage medium. Background Art

[0002] At present, for the laser welding of non-ferrous metals, ordinary lasers and welding methods are not sufficient to meet the welding needs. When welding galvanized sheets, copper, aluminum and other highly reflective materials, laser welding defects such as cracks and pores are easily generated inside, and there is large spatter and poor surface forming quality. When welding highly reflective materials, there are technical problems such as high energy consumption and low welding efficiency. Summary of the invention

[0003] In view of this, the present invention aims to overcome the defect of low welding precision in the prior art and proposes an intelligent welding method, an intelligent welding system and a computer storage medium.

[0004] Provided is an intelligent welding method, comprising:

[0005] Predicting the linear data through the linear model to obtain a first prediction result;

[0006] and / or predicting the nonlinear data by using a nonlinear model to obtain a second prediction result;

[0007] Based on the first prediction result and / or the second prediction result, the target operating parameters are obtained, and the target laser is emitted by the laser body according to the target operating parameters to weld the current object to be welded, that is, the first operating parameter (i.e., the first prediction result) having a linear relationship with the expected welding effect data is obtained according to the expected welding effect data through a linear model, and the second operating parameter (i.e., the second prediction result) having a nonlinear relationship with the expected welding effect data is obtained according to the expected welding effect data through a nonlinear model, and then the target operating parameters are obtained according to the first operating parameters and the second operating parameters, and then the target laser is emitted by the laser body according to the target operating parameters to weld the current object to be welded; before this, different data can be marked and classified in advance to mark whether the relationship between data A and data B is linear or nonlinear, and then the data therein are predicted through different models.

[0008] The first linear model is constructed by an algorithm that can process linear data; the second nonlinear model is constructed by an algorithm that can process nonlinear data;

[0009] The linear data includes data that are linearly related under certain conditions in welding, which may be current and power data of the laser; the nonlinear data includes data that are nonlinearly related under certain conditions in welding, which may be the temperature of the object to be welded increased by the action of the laser beam and the stress of the object to be welded changed due to the temperature change.

[0010] Provided is an intelligent welding method, comprising:

[0011] receiving desired effect parameters;

[0012] Acquire technical parameters according to the effect parameters, and determine the category to which the technical parameters belong;

[0013] If the category to which the technical parameter belongs is linear data, the linear data is predicted by a linear model to obtain a first prediction result; if the category to which the technical parameter belongs is nonlinear data, the linear data is predicted by a nonlinear model to obtain a second prediction result;

[0014] Acquire target operating parameters based on the first prediction result and / or the second prediction result, and weld the current object to be welded by emitting target laser light according to the target operating parameters through the laser body;

[0015] The current data are technical parameters for implementing the welding process and effect parameters for achieving the welding results.

[0016] Provided is an intelligent welding method, comprising:

[0017] Step S10: constructing an initial linear model and an initial nonlinear model based on a preset algorithm;

[0018] Step S20: inputting the pre-collected original data set into the initial nonlinear model for prediction training, and obtaining the first linear model and the first nonlinear model after the training;

[0019] The original data includes the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the energy density of the laser beam, the spot size, and the angle between the laser beam and the object to be welded;

[0020] Step S30: collecting current operating condition data, the current operating condition data including first linear data and first nonlinear data, inputting the first linear data into the first linear model, and inputting the first nonlinear data into the first nonlinear model;

[0021] Predicting the first linear data by using the first linear model to obtain a first prediction result, and predicting the second nonlinear data by using the first nonlinear model to obtain a second prediction result;

[0022] The working condition data include the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the laser beam energy density, the spot size, and the angle between the laser beam and the object to be welded in actual welding;

[0023] Step S40: acquiring target operating parameters based on the first prediction result and the second prediction result, and emitting target laser light to weld the current object to be welded according to the target operating parameters through the laser body, collecting image information of the object to be welded after welding, and obtaining welding effect data; the actual operating parameters include power, pulse, laser beam swing angle, and angle and distance of the laser lens compared to the incident end of the laser beam of the laser system;

[0024] Step S50: determining whether the welding effect data is less than or equal to an expected value;

[0025] Step S60: If the welding effect data is less than or equal to the expected value, return to the step of inputting the pre-collected original data set into the initial nonlinear model for predictive training, and obtaining the first linear model and the first nonlinear model after training, until the welding effect data is greater than the expected value, and the welding effect data that meets the expected value is obtained, and the welding is completed.

[0026] The step of acquiring target operating parameters based on the first prediction result and / or the second prediction result, and emitting target laser light to weld the object to be welded according to the target operating parameters by the laser body comprises:

[0027] The weight of the first prediction result and / or the second prediction result is adjusted to obtain target operating parameters, and the laser body emits a target laser according to the target operating parameters to weld the current object to be welded.

[0028] Optionally, the step S40 includes:

[0029] Step S401: The step of welding the object to be welded by the intelligent welding system includes:

[0030] Step S402: a first laser beam is emitted from the laser body, the first laser beam is collimated by a collimator to obtain a collimated laser, and the collimated laser passes through a cylindrical lens with a hole to achieve rearrangement of laser energy to obtain a rearranged laser;

[0031] Step S403: the rearranged laser is reflected by a reflective mirror capable of displacement, thereby obtaining an oscillating laser whose spatial position can be changed;

[0032] Step S404: the oscillating laser is focused by the field lens to form a target laser, and the current object to be welded is welded by the target laser;

[0033] Alternatively, the step S40 includes:

[0034] Step S405: The laser body emits a first laser beam and a second laser beam, the first laser beam and the second laser beam are collimated by a collimator to obtain a collimated laser, and the collimated laser passes through a cylindrical lens with a hole to achieve rearrangement of laser energy to obtain a rearranged laser;

[0035] Step S406: The rearranged laser is reflected by a displacement mirror to obtain an oscillating laser whose spatial position can be changed;

[0036] Step S407: After the oscillating laser is focused by the field lens, a target laser is formed, and the current object to be welded is welded by the target laser.

[0037] Optionally, the step S404 includes:

[0038] Step S4041, the oscillating laser forms a target laser after passing through the field lens, and the focus position of the target laser can move along the axial direction or radial direction of the main optical axis of the field lens;

[0039] Step S4042, welding the current object to be welded by using the target laser.

[0040] Optionally, a filling wire is provided between the object to be welded and the target laser.

[0041] Provide an intelligent welding system:

[0042] A first prediction module, used to predict linear data through a linear model to obtain a first prediction result;

[0043] and / or, a second prediction module, used to predict the nonlinear data by using a nonlinear model to obtain a second prediction result;

[0044] The first welding module is used to obtain target operating parameters based on the first prediction result and / or the second prediction result, and to weld the current object to be welded by emitting target laser according to the target operating parameters through the laser body.

[0045] Provide an intelligent welding system:

[0046] A first receiving module, used for receiving a desired effect parameter and an initial technical parameter set;

[0047] A first receiving module, configured to obtain a target technical parameter set according to the effect parameter, traverse the target technical parameter set, and sequentially determine a corresponding relationship between each target technical parameter in the target technical parameter set and the effect parameter;

[0048] A second welding module is used for predicting the specific value of the target technical parameter through a linear model to obtain a first prediction result when there is a linear relationship between the target technical parameter and the effect parameter, adjusting the initial technical parameter according to the first prediction result, taking the adjusted initial technical parameter as the first target operating parameter, and emitting a target laser through the laser body according to the first target operating parameter to weld the current object to be welded;

[0049] When there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, the adjusted initial technical parameter is used as the second target operating parameter, and the target laser is emitted by the laser body according to the second target operating parameter to weld the current object to be welded.

[0050] Provide an intelligent welding system:

[0051] A second receiving module is used to receive a desired effect parameter and an initial technical parameter set, and obtain a target technical parameter set according to the effect parameter, wherein the initial technical parameter set is composed of a plurality of initial technical parameters, and the target technical parameter set is composed of a plurality of target technical parameters;

[0052] A second judgment module is used to compare each target technical parameter with the initial technical parameter to determine whether the target technical parameter meets the preset conditions;

[0053] A third judgment module, if the target technical parameter meets the preset condition, then judges the corresponding relationship between the target technical parameter and the effect parameter;

[0054] A third welding module is used for predicting the specific value of the target technical parameter through a linear model to obtain a first prediction result when there is a linear relationship between the target technical parameter and the effect parameter, adjusting the initial technical parameter according to the first prediction result, taking the adjusted initial technical parameter as the first target operating parameter, and emitting a target laser through the laser body according to the first target operating parameter to weld the current object to be welded;

[0055] When there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, the adjusted initial technical parameter is used as the second target operating parameter, and the target laser is emitted by the laser body according to the second target operating parameter to weld the current object to be welded.

[0056] Provided is an intelligent welding system capable of implementing the intelligent welding method as described above, comprising a hardware component and a software component, wherein the software component and the hardware component are electrically connected or communicatively connected, and both the software component and the hardware component are connected to a main control module;

[0057] The hardware components include a laser body for outputting a laser beam, a rotary solenoid, a collimating lens, a cylindrical lens with a hole, a reflector, and a field lens arranged in sequence along the transmission direction of the laser, the rotary solenoid is driven by an internal or external power supply, the rotary solenoid is used to drive the reflector to rotate or swing, the laser beam incident end of the cylindrical lens has a curvature, and the laser beam forms a focusing point and a laser spot after passing through the cylindrical lens;

[0058] The software components include:

[0059] A construction module, used for constructing an initial linear model and an initial nonlinear model based on a preset algorithm;

[0060] A training module, used for inputting the pre-collected original data set into the initial nonlinear model for prediction training, and obtaining the first linear model and the first nonlinear model after the training respectively;

[0061] The original data includes the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the energy density of the laser beam, the spot size, and the angle between the laser beam and the object to be welded;

[0062] An acquisition module, used for acquiring current operating condition data, wherein the current operating condition data includes first linear data and first nonlinear data, inputting the first linear data into the first linear model, and inputting the first nonlinear data into the first nonlinear model;

[0063] A prediction module, configured to predict the first linear data by using the first linear model to obtain a first prediction result, and predict the second nonlinear data by using the first nonlinear model to obtain a second prediction result;

[0064] The working condition data include the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the laser beam energy density, the spot size, and the angle between the laser beam and the object to be welded in actual welding;

[0065] an acquisition module, configured to acquire target operating parameters based on the first prediction result and the second prediction result, and to weld the current object to be welded by emitting a target laser through the laser body according to the target operating parameters, and to collect image information of the object to be welded after welding, so as to obtain welding effect data; the actual operating parameters include power, pulse, laser beam swing angle, and angle and distance of the laser lens compared to the incident end of the laser beam of the laser system;

[0066] An expected judgment module, used to judge whether the welding effect data is less than or equal to an expected value;

[0067] A prediction training module is used to input the pre-collected original data set into the initial nonlinear model for prediction training if the welding effect data is less than or equal to the expected value, and obtain the first linear model and the first nonlinear model after training respectively, until the welding effect data is greater than the expected value.

[0068] An intelligent welding system is provided, which can implement the intelligent welding method as described in claim 5, wherein the first welding module includes an adjustment unit, and the adjustment unit is used to adjust the weight of the first prediction result and / or the second prediction result to obtain target operating parameters, and the laser body emits a target laser according to the target operating parameters to weld the current object to be welded.

[0069] The component driven by the rotary solenoid can be limited by a spring to control its maximum swinging angle. One end of the spring is connected to the component, and the other end is connected to a drive motor. The spring is stretched by the drive motor to reduce the maximum angle. The drive motor is connected to a main control module.

[0070] Optionally, a collimating lens, a cylindrical lens with a hole, a reflector, and a field lens are arranged in sequence along the transmission direction of the laser. The reflector is connected to a rotating motor, and the rotating motor is used to drive the reflector to rotate at a preset angle. The side of the cylindrical lens opposite to the incident direction of the laser beam has a certain curvature, and the laser beam forms a focal point and a laser spot after passing through the cylindrical lens.

[0071] Optionally, the curved end of the cylindrical lens may be perpendicular to the optical path direction of the laser beam.

[0072] Optionally, the location with the curvature may be convex or concave.

[0073] When the position where the cylindrical lens has a curvature is concave, the laser beam can form a focal point through the empty hole part of the cylindrical lens, and the laser beam can form a laser spot through the non-empty hole part of the cylindrical lens. The focal point falls inside the laser spot, and the focal point is used for keyhole welding and preheating or slow cooling.

[0074] Optionally, it also includes a laser processing module, and the laser processing module also includes an output unit, and the output unit is used to supply a continuous sinusoidal voltage to the rotary solenoid according to the first prediction result and the second prediction result; the laser processing module is electrically connected to a laser source, and while the rotary solenoid drives the rotary axis of the rotary solenoid to rotate into a sine wave, the laser processing module controls the laser source to emit a laser beam, and the laser beam is first collimated by a collimator, and then passes through a cylindrical lens with an empty hole, a part of the laser beam passes through the empty hole, and another part passes through the surrounding of the empty hole, and then reaches the reflector, at which time the rotating shaft drives the reflector to rotate or swing, and the laser beam reflected by the reflector also swings and rotates accordingly, and is finally focused by a field lens into a light spot composed of central light and annular light, the maximum inner diameter of the center is smaller than the maximum inner diameter from the center of the circle to the outermost edge of the annular light, and the light spot acts on the object to be welded in a swinging or rotating manner;

[0075] Alternatively, the output unit is used to supply an intermittent sinusoidal voltage to the rotary solenoid; when the output unit is in operation, it includes:

[0076] The duration of each sine wave voltage supply is 0.1ms and the frequency is 300Hz;

[0077] Alternatively, when the object to be welded is an aluminum plate, the frequency of the supplied sine wave voltage is 50 Hz; when the object to be welded is stainless steel, the frequency of the supplied sine wave voltage is 100 Hz.

[0078] Optionally, the expected value judgment module is specifically used to detect the processing effect of the object to be welded and obtain actual welding effect data. The welding effect data can be a laser radar unit for ranging, an image acquisition unit for acquiring image information, or a temperature acquisition unit for sensing temperature information; it also includes a similarity judgment module, which is used to calculate the similarity between the current welding effect data and the preset data through a similarity algorithm.

[0079] Optionally, the cylindrical lens 41 has one or more holes.

[0080] Optionally, the similarity judgment module includes a sorting unit, and the sorting unit is used to prioritize the original data through a weight algorithm.

[0081] Optionally, the judgment module is used to detect the processing effect of the object to be welded and obtain actual welding effect data. The welding effect data can be a laser radar unit for ranging, an image acquisition unit for acquiring image information, or a temperature acquisition unit for sensing temperature information; and also includes a judgment module, which is used to calculate the similarity between the current welding effect data and the preset data through a similarity algorithm.

[0082] Optionally, the cylindrical lens has one or more holes.

[0083] Optionally, the judgment module includes a sorting unit, and the sorting unit is used to prioritize the original data through a weight algorithm.

[0084] A computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the intelligent welding method described above are implemented.

[0085] The beneficial effects of the present invention are as follows: linear models and nonlinear models can be used to predict and classify data with linear relationships and nonlinear relationships respectively, thereby avoiding mutual interference between different types of data, increasing the accuracy of prediction and classification effects, and providing data support for more accurate laser welding, thereby improving welding quality and achieving high-precision welding. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 It is a schematic diagram of the execution steps of the method of the present invention;

[0087] Figure 2 It is a schematic diagram of the modules of the system of the present invention;

[0088] Figure 3 is a schematic diagram of an embodiment of the system of the present invention;

[0089] Figure 4 It is a schematic diagram of a computer device of the present invention. Specific implementation plan

[0090] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and cannot be understood as limiting the present invention. In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "inside" and "upper" is based on the orientation or positional relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. Unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral body; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood in specific circumstances.

[0091] The present invention will be further described below in conjunction with the accompanying drawings:

[0092] In one embodiment of the present invention, there is provided an intelligent welding method, comprising:

[0093] The linear data is predicted by the linear model to obtain the first prediction result, which means that the linear data is predicted by the trained linear model, and the linear data refers to the data that is linearly related to the expected welding effect data. For example, the current laser output power is 1W, and the expected power, that is, the expected welding effect data is 10W. The power and current are linearly related, and the current working current, for example, 10A, can be obtained based on the expected power through the linear model;

[0094] And / or, the nonlinear data is predicted by a nonlinear model to obtain a second prediction result, wherein the nonlinear data refers to data having a nonlinear relationship with the desired welding effect data, for example, the temperature and stress have a nonlinear relationship, and the stress of the object to be welded is now expected to be 5Pa, and the nonlinear data 5Pa can be predicted by a nonlinear model to obtain 101°C, wherein the above 10A and 101°C are both target operating parameters. Since the magnitude of the stress can be controlled, cracks can be prevented from being generated inside the object to be welded and pores can be generated on the welding surface. Since the magnitude of the power can be controlled, the generation of welding spatter can be prevented, the energy consumption can be reduced, and the welding quality can be improved;

[0095] As the training ability of the model used for prediction or classification improves, its prediction ability will decrease due to learning too many details of the training samples, which is called "overfitting". In order to maintain a high level of prediction, it is necessary to first distinguish data with linear and nonlinear relationships from massive parameters with complex relationships, so as to achieve noise reduction for data with different relationships and improve the accuracy of prediction.

[0096] Based on the first prediction result and / or the second prediction result, a target operating parameter is obtained, and a target laser is emitted by the laser body according to the target operating parameter to weld the current object to be welded. The target operating parameter refers to the actual working parameter of the laser during welding, and the actual parameter directly affects the welding quality; that is, a first operating parameter (i.e., a first prediction result) having a linear relationship with the expected welding effect data is obtained according to the expected welding effect data through a linear model, a second operating parameter (i.e., a second prediction result) having a nonlinear relationship with the expected welding effect data is obtained according to the expected welding effect data through a nonlinear model, and then the first operating parameter and the second operating parameter are used to obtain the first operating parameter and the second operating parameter. The target operating parameters are obtained by the data, and then the target laser is emitted by the laser body according to the target operating parameters to weld the current object to be welded; before this, different data can be marked and classified in advance to mark whether the relationship between data A and data B is linear or nonlinear, and then the data therein is predicted through different models, that is, the target operating parameters are obtained through the expected effect data. For example, it is expected that the center distance between the two laser spots output by the laser is 0.6 microns. According to the linear relationship between the center distance between the light spots and the radial relative positions of the two laser sources, the linear data 0.6 microns can be predicted through the linear model to obtain the first prediction result 0.1 microns.

[0097] The first linear model is constructed by an algorithm that can process linear data; the second nonlinear model is constructed by an algorithm that can process nonlinear data;

[0098] The linear data includes data that are linearly related under certain conditions in welding, which may be current and power data of the laser; the nonlinear data includes data that are nonlinearly related under certain conditions in welding, which may be the temperature of the object to be welded increased by the action of the laser beam and the stress of the object to be welded changed due to the temperature change.

[0099] In one embodiment, a smart welding method is provided, comprising:

[0100] receiving desired effect parameters;

[0101] Acquire technical parameters according to the effect parameters, and determine the category to which the technical parameters belong;

[0102] If the category to which the technical parameter belongs is linear data, the linear data is predicted by a linear model to obtain a first prediction result; if the category to which the technical parameter belongs is nonlinear data, the linear data is predicted by a nonlinear model to obtain a second prediction result;

[0103] Acquire target operating parameters based on the first prediction result and / or the second prediction result, and weld the current object to be welded by emitting target laser light according to the target operating parameters through the laser body;

[0104] The current data are technical parameters for realizing the welding process and effect parameters for realizing the welding result;

[0105] The expected welding width is 1mm, then 1mm is the effect parameter. The above-mentioned first prediction result 10A and the second prediction result 101°C are both technical parameters for realizing the welding process. In this embodiment, the technical parameters can be directly used as target operating parameters, or the technical parameters can be processed to obtain the target operating parameters; there can be a one-to-many relationship between the effect parameters and the technical parameters. Therefore, in order to achieve the purpose of obtaining the technical parameters according to the effect parameters, it is necessary to establish a mapping relationship between the effect parameters and the technical parameters in advance.

[0106] In one embodiment, receiving a desired effect parameter and an initial set of technical parameters;

[0107] Acquire a target technical parameter set according to the effect parameter, traverse the target technical parameter set, and sequentially determine the corresponding relationship between each target technical parameter in the target technical parameter set and the effect parameter;

[0108] If there is a linear relationship between the target technical parameter and the effect parameter, a specific value of the target technical parameter is predicted by a linear model to obtain a first prediction result, the initial technical parameter is adjusted according to the first prediction result, the adjusted initial technical parameter is used as a first target operating parameter, and the laser body emits a target laser according to the first target operating parameter to weld the current object to be welded;

[0109] If there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, and the adjusted initial technical parameter is used as the second target operating parameter. The laser body emits a target laser according to the second target operating parameter to weld the current object to be welded.

[0110] In one embodiment, the step of acquiring target operating parameters based on the first prediction result and / or the second prediction result, and emitting target laser light to weld the object to be welded according to the target operating parameters by the laser body includes:

[0111] The weight of the first prediction result and / or the second prediction result is adjusted to obtain target operating parameters, and the laser body emits a target laser according to the target operating parameters to weld the current object to be welded.

[0112] Reference Figure 1 In one embodiment of the present utility model, there is provided an intelligent welding method, comprising:

[0113] Step S10: constructing an initial linear model and an initial nonlinear model based on a preset algorithm;

[0114] Step S20: inputting the pre-collected original data set into the initial nonlinear model for prediction training, and obtaining the first linear model and the first nonlinear model after the training;

[0115] The original data includes the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the energy density of the laser beam, the spot size, and the angle between the laser beam and the object to be welded;

[0116] Step S30: collecting current operating condition data, the current operating condition data including first linear data and first nonlinear data, inputting the first linear data into the first linear model, and inputting the first nonlinear data into the first nonlinear model;

[0117] Predicting the first linear data by using the first linear model to obtain a first prediction result, and predicting the second nonlinear data by using the first nonlinear model to obtain a second prediction result;

[0118] The working condition data include the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the laser beam energy density, the spot size, and the angle between the laser beam and the object to be welded in actual welding;

[0119] Step S40: acquiring target operating parameters based on the first prediction result and the second prediction result, and emitting target laser light to weld the current object to be welded according to the target operating parameters through the laser body, collecting image information of the object to be welded after welding, and obtaining welding effect data; the actual operating parameters include power, pulse, laser beam swing angle, and angle and distance of the laser lens compared to the incident end of the laser beam of the laser system;

[0120] In this embodiment, the collected image information and the target image information can be feature-matched by the square difference matching method. For example, when the similarity is greater than 90%, it is considered that an ideal welding effect is achieved; the target image is an ideal welding object with a smooth surface. The collected image is processed into the same size as the target image. If the target image is composed of multiple small images, the collected image is also divided into multiple small images of the same size. Then, the average pixel value in each row or each small image between the target image and the collected image information is calculated, and the variance value of the above average value is calculated. The smaller the variance value, the greater the similarity.

[0121] The linear relationship and nonlinear relationship in this application are not concepts in the mathematical sense, but an approximate relationship that can be formed in welding. For example, the generation of a laser beam is nonlinearly related to voltage. When the voltage is small, the light is emitted in all directions. When the voltage increases to a certain value, monochromatic light with consistent phase and direction, i.e., a laser beam, is emitted. For another example, there is a linear relationship between pump current and output power. For another example, the focal length and the spot diameter are linear within a certain range. Under the premise that the output laser is a flat-top spot, the spot diameter and the welding width are linear, while Gaussian spots and annular spots are not applicable. For another example, under the influence of ambient temperature and impact force, the laser beam is not linear. The stress generated by the object to be welded is nonlinearly related to the above-mentioned conditions, and the stress intuitively corresponds to the deformation value of the object to be welded. For example, when the temperature of the object to be welded is irradiated by a 1000W power laser, the temperature gradually rises within 0-5 seconds, and there is an approximate linear relationship between time and temperature during the rising stage. A molten pool begins to form on the surface of the object to be welded within 5-10 seconds, and the temperature of the molten pool will remain in a relatively stable state and will no longer rise. If only the data within 0-5 seconds are taken, the temperature data within 0-5 seconds can be labeled as linear data, and if only the data within 0-10 seconds are taken, the temperature data within 0-10 seconds can be labeled as nonlinear data. Since different algorithm models have different classification capabilities for linear problems and nonlinear problems, in order to achieve a more accurate classification effect, in this embodiment, the initial linear model is trained with data with a linear relationship, and the initial nonlinear model is trained with data with a nonlinear relationship. The initial nonlinear model can be constructed by a nonlinear algorithm such as a gradient descent algorithm, and the initial linear model can be constructed by a linear algorithm such as a support vector machine. Therefore, the model constructed and trained by the gradient descent algorithm can be used to predict the ambient temperature to which the stress of the weld is 20Pa, and the trained support vector machine can predict the current to which it needs to be adjusted to achieve a power of 1000W. Specifically, the current weld width image can be collected by a CCD camera, and the square difference matching method can be used to determine whether the weld width and the target weld meet the preset conditions. If not, the support vector machine is used to obtain the spot diameter data according to the target weld width data, and then the focal length data is obtained according to the spot diameter data, or the focal length data is directly obtained according to the weld width data, and then the distance between the optical lens and the laser source is adjusted according to the focal length data to obtain the target weld width; specifically, the temperature of the welding point is collected by a temperature sensor, and the trained gradient descent algorithm model is used to predict from which second the temperature remains relatively stable;

[0122] When predicting data in the same time period and the same space, linear data and nonlinear data are processed independently to prevent interference between data of different natures. Compared with the solution of using a single model to classify or predict all data, the solution provided by this application can eliminate interference factors and obtain better classification or prediction results. The prediction results can be sent to the CPU main control system to provide decision-making data support for further improving welding quality and efficiency;

[0123] Step S50: determining whether the welding effect data is less than or equal to an expected value;

[0124] Step S60: If the welding effect data is less than or equal to the expected value, return to step S20 until the welding effect data is greater than the expected value, obtain welding effect data that meets the expected value, and complete welding.

[0125] In one embodiment, referring to Figure 2-3 , provides an intelligent welding system that can implement the intelligent welding method as described above, including hardware components and software components, the software components and the hardware components are electrically connected or communicatively connected, and the software components and the hardware components are both connected to the main control module 101.

[0126] In one embodiment, the step of acquiring target operating parameters based on the first prediction result and / or the second prediction result, and emitting target laser light to weld the object to be welded according to the target operating parameters through the laser body includes:

[0127] The weight of the first prediction result and / or the second prediction result is adjusted to obtain the target operating parameters. The target laser is emitted by the laser body according to the target operating parameters to weld the current object to be welded. In order to obtain a welding width of 1 mm, both temperature data and current data can play a role. Different weights can be set for temperature data and current data according to experimental results. For example, the weight of temperature data is 0.4, and the weight of current data is 0.6. After adjusting the weights, a more accurate welding width can be obtained.

[0128] In one embodiment, the step S40 includes:

[0129] Step S401: The step of welding the object to be welded by the intelligent welding system includes:

[0130] Step S402: The laser body emits a first laser beam, which is collimated by a collimator to obtain a collimated laser. The collimated laser passes through a cylindrical lens with an empty hole to achieve rearrangement of laser energy to obtain a rearranged laser. Since the first part of the collimated laser passes through the empty hole of the cylindrical lens, and the second part diverges through the curved surface of the cylindrical lens, the final focal position of the first part is farther than the focal position of the second part. Since the empty hole is located in the center of the cylindrical lens, the energy density of the obtained central beam is greater than the energy density of the periphery. Therefore, deep melting welding can be performed by the central beam, and preheating or slow cooling can be performed by the peripheral laser. The first part is the focal point, and the second part is the laser point.

[0131] Step S403: the rearranged laser is reflected by a reflective mirror capable of displacement, thereby obtaining an oscillating laser whose spatial position can be changed;

[0132] Step S404: the oscillating laser is focused by the field lens to form a target laser, and the current object to be welded is welded by the target laser;

[0133] Alternatively, the step S40 includes:

[0134] Step S405: The laser body emits a first laser beam and a second laser beam, the first laser beam and the second laser beam are collimated by a collimator to obtain a collimated laser, and the collimated laser passes through a cylindrical lens with a hole to achieve rearrangement of laser energy to obtain a rearranged laser;

[0135] Step S406: The rearranged laser is reflected by a displacement mirror to obtain an oscillating laser whose spatial position can be changed;

[0136] Step S407: After the oscillating laser is focused by the field lens, a target laser is formed, and the current object to be welded is welded by the target laser.

[0137] The step S404 includes:

[0138] Step S4041, the oscillating laser forms a target laser after passing through the field lens, and the focus position of the target laser can move along the axial direction or radial direction of the main optical axis of the field lens;

[0139] Step S4042, welding the current object to be welded by using the target laser.

[0140] A filling wire is arranged between the object to be welded and the target laser.

[0141] Provide an intelligent welding system:

[0142] A first prediction module, used to predict linear data through a linear model to obtain a first prediction result;

[0143] and / or, a second prediction module, used to predict the nonlinear data by using a nonlinear model to obtain a second prediction result;

[0144] The first welding module is used to obtain target operating parameters based on the first prediction result and / or the second prediction result, and to weld the current object to be welded by emitting target laser according to the target operating parameters through the laser body.

[0145] Provide an intelligent welding system:

[0146] A first receiving module, used for receiving a desired effect parameter and an initial technical parameter set;

[0147] A first receiving module, configured to obtain a target technical parameter set according to the effect parameter, traverse the target technical parameter set, and sequentially determine a corresponding relationship between each target technical parameter in the target technical parameter set and the effect parameter;

[0148] A second welding module is used for predicting the specific value of the target technical parameter through a linear model to obtain a first prediction result when there is a linear relationship between the target technical parameter and the effect parameter, adjusting the initial technical parameter according to the first prediction result, taking the adjusted initial technical parameter as the first target operating parameter, and emitting a target laser through the laser body according to the first target operating parameter to weld the current object to be welded;

[0149] When there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, the adjusted initial technical parameter is used as the second target operating parameter, and the target laser is emitted by the laser body according to the second target operating parameter to weld the current object to be welded.

[0150] Provide an intelligent welding system:

[0151] A second receiving module is used to receive a desired effect parameter and an initial technical parameter set, and obtain a target technical parameter set according to the effect parameter, wherein the initial technical parameter set is composed of a plurality of initial technical parameters, and the target technical parameter set is composed of a plurality of target technical parameters;

[0152] A second judgment module is used to compare each target technical parameter with the initial technical parameter to determine whether the target technical parameter meets the preset conditions;

[0153] A third judgment module, if the target technical parameter meets the preset condition, then judges the corresponding relationship between the target technical parameter and the effect parameter;

[0154] A third welding module is used for predicting the specific value of the target technical parameter through a linear model to obtain a first prediction result when there is a linear relationship between the target technical parameter and the effect parameter, adjusting the initial technical parameter according to the first prediction result, taking the adjusted initial technical parameter as the first target operating parameter, and emitting a target laser through the laser body according to the first target operating parameter to weld the current object to be welded;

[0155] When there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, the adjusted initial technical parameter is used as the second target operating parameter, and the target laser is emitted by the laser body according to the second target operating parameter to weld the current object to be welded.

[0156] An intelligent welding system, capable of implementing the intelligent welding method as described above, comprises a hardware component and a software component, wherein the software component and the hardware component are electrically connected or communicatively connected, and the software component and the hardware component are both connected to a main control module;

[0157] The hardware components include a laser body for outputting a laser beam, a rotary solenoid 10, a collimating lens 5, a cylindrical lens 41 with a hole, a reflector 8, and a field lens 11 arranged in sequence along the transmission direction of the laser, the rotary solenoid 10 is driven by an internal or external power supply, the rotary solenoid 10 is used to drive the reflector 8 to rotate or swing, the laser beam incident end of the cylindrical lens 41 has a curvature, and the laser beam forms a focal point and a laser point after passing through the cylindrical lens 41, the focal point is located in the laser point, and the energy density of the focal point is greater than the energy density of the laser point;

[0158] In this embodiment, the laser beam first passes through a cylindrical lens 41 with an empty hole, a part of the laser beam passes through the empty hole, and the other part passes around the empty hole, and then reaches the reflector 8. At this time, the rotating shaft drives the reflector 8 to rotate or swing, and the laser beam reflected by the reflector also swings and rotates accordingly, and finally is focused by the field lens 11 into a light spot composed of central light and annular light. The maximum inner diameter of the center is smaller than the maximum inner diameter from the center of the circle to the outermost edge of the annular light, and the light spot continuously swings or rotates to act on the object to be welded; an intermittent sinusoidal wave voltage can be supplied to the rotary solenoid, and while the rotary solenoid drives the rotary shaft of the rotary solenoid to rotate into a sinusoidal wave, the laser source is controlled to emit a laser beam, and the laser beam is first collimated by a collimator, and then passes through a A cylindrical lens 41 with an empty hole, a part of the laser beam passes through the empty hole, and the other part passes around the empty hole, and then reaches the reflector 8. At this time, the rotating shaft drives the reflector 8 to rotate or swing, and the laser beam reflected by the reflector also swings and rotates accordingly, and finally focuses through the field lens 11 into a light spot composed of central light and annular light, the maximum inner diameter of the center is smaller than the maximum inner diameter from the center of the circle to the outermost edge of the annular light, and the light spot intermittently swings or rotates to act on the object to be welded, wherein the specific parameters include: the duration of each supply of the sinusoidal wave voltage is 0.1ms, the frequency is 300Hz, or, the frequency of each supply of the sinusoidal wave voltage is 50Hz, and when the object to be processed is stainless steel, the frequency of each supply of the sinusoidal wave voltage is 100Hz.

[0159] The software components include:

[0160] A construction module, used for constructing an initial linear model and an initial nonlinear model based on a preset algorithm;

[0161] A training module, used for inputting the pre-collected original data set into the initial nonlinear model for prediction training, and obtaining the first linear model and the first nonlinear model after the training respectively;

[0162] The original data includes the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the energy density of the laser beam, the spot size, and the angle between the laser beam and the object to be welded;

[0163] An acquisition module, used for acquiring current operating condition data, wherein the current operating condition data includes first linear data and first nonlinear data, inputting the first linear data into the first linear model, and inputting the first nonlinear data into the first nonlinear model;

[0164] A prediction module, configured to predict the first linear data by using the first linear model to obtain a first prediction result, and predict the second nonlinear data by using the first nonlinear model to obtain a second prediction result;

[0165] The working condition data include the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the laser beam energy density, the spot size, and the angle between the laser beam and the object to be welded in actual welding;

[0166] an acquisition module, configured to acquire target operating parameters based on the first prediction result and the second prediction result, and to weld the current object to be welded by emitting a target laser through the laser body according to the target operating parameters, and to collect image information of the object to be welded after welding, so as to obtain welding effect data; the actual operating parameters include power, pulse, laser beam swing angle, and angle and distance of the laser lens compared to the incident end of the laser beam of the laser system;

[0167] An expected judgment module, used to judge whether the welding effect data is less than or equal to an expected value;

[0168] A prediction training module is used to input the pre-collected original data set into the initial nonlinear model for prediction training if the welding effect data is less than or equal to the expected value, and obtain the first linear model and the first nonlinear model after training respectively, until the welding effect data is greater than the expected value.

[0169] In this embodiment, the component driven by the rotary solenoid 10 can be limited by a spring to control the maximum swing angle thereof. One end of the spring is connected to the component, and the other end is connected to a drive motor. The spring is stretched by the drive motor to reduce the maximum angle. The drive motor is connected to the main control module 101.

[0170] When it is predicted that in order to obtain a weld width of 4mm, the laser beam swing angle needs to be adjusted from the current 60° to 30°. The swing amplitude of the component can be reduced by stretching the spring until the laser beam swing angle reaches 30°. In order to determine whether the adjustment is successful, the position information of the laser beam entering the object to be welded can be obtained by infrared temperature detection and other devices to determine whether the angle is qualified. Similarly, the voltage parameters applied to the rotary solenoid 10 can be adjusted to obtain a predetermined laser swing frequency;

[0171] When galvanizing steel sheets, since the boiling point of zinc is 907°C, which is much lower than the melting point of iron, 1538°C, zinc evaporates and becomes powdered zinc oxide, changing from metal to a mixture like soil. It is difficult to control the gap for zinc vapor discharge within 0.5mm. In order to solve this problem, this embodiment can add a hollow cylindrical lens 401 to the optical path. A part of the collimated laser beam passes through the hollow hole, and the other part passes through the hollow hole. After being focused by the field lens 11, the formed laser beam consists of a focal point and a laser point linearly extending along the focal point. The laser point has low energy, while the focal point has high energy, which can preheat or slow cool the workpiece.

[0172] In one embodiment, the first welding module includes an adjustment unit, which is used to adjust the weight of the first prediction result and / or the second prediction result to obtain target operating parameters, and to weld the current object to be welded by emitting a target laser through the laser body according to the target operating parameters.

[0173] In one embodiment, referring to Figure 2 , another intelligent welding system is provided, in which the software component ( Figure 3 ), further comprising: a collimating lens 5, a cylindrical lens 41 with an empty hole, a reflector 8, and a field lens 11 are sequentially arranged along the transmission direction of the laser. Since the field lens 11 can integrate lasers with different focal points into the same plane, the laser beam energy is more stable and the welding effect is more stable. The reflector 8 is connected to a rotating motor 102, and the rotating motor 102 is used to drive the reflector 8 to rotate according to a preset angle. The side of the cylindrical lens 41 opposite to the incident direction of the laser beam has a certain curvature. After the laser beam passes through the cylindrical lens 41, a focusing point and a laser point are formed; the laser beam is first collimated by the collimating lens 5, and then passes through all the above lenses in sequence. The rotating motor 102 can play the same role as the rotary solenoid 10 in the previous embodiment. One of them can be selected according to the actual scenario. For example, in a high-power laser, the heat generation speed is fast and it is desired to reduce energy consumption. The rotary solenoid 10 is used to intermittently drive the component. In the scenario where the swing angle of the component needs to be accurately obtained, the rotating motor 102 can be used to drive the component to swing.

[0174] In one embodiment, the curved end of the cylindrical lens 41 may be perpendicular to the optical path direction of the laser beam.

[0175] In one embodiment, the location having the curvature may be convex or concave.

[0176] In this embodiment, when the position where the cylindrical lens 41 has a curvature is concave, the laser beam can form a focal point through the empty hole portion of the cylindrical lens 41, and the laser beam can form a laser spot through the non-empty hole portion of the cylindrical lens 41. The focal point falls inside the laser spot, and the focal point is used for keyhole welding, and the focal point is used for preheating or slow cooling.

[0177] In one embodiment, it also includes a laser processing module, and the laser processing module also includes an output unit, and the output unit is used to supply a continuous sinusoidal voltage to the rotary solenoid according to the first prediction result and the second prediction result; the laser processing module is electrically connected to a laser source, and while the rotary solenoid drives the rotary shaft of the rotary solenoid to rotate into a sine wave, the laser processing module controls the laser source to emit a laser beam, and the laser beam is first collimated by a collimator, and then passes through a cylindrical lens 41 with an empty hole, a part of the laser beam passes through the empty hole, and another part passes through the surrounding of the empty hole, and then reaches the reflector 8, at which time the rotating shaft drives the reflector 8 to rotate or swing, and the laser beam reflected by the reflector also swings and rotates accordingly, and finally is focused by the field lens 11 into a light spot composed of central light and annular light, the maximum inner diameter of the center is smaller than the maximum inner diameter from the center of the circle to the outermost edge of the annular light, and the light spot swings or rotates to act on the object to be welded;

[0178] Alternatively, the output unit is used to supply an intermittent sinusoidal voltage to the rotary solenoid; when the output unit is in operation, it includes:

[0179] The duration of each sine wave voltage supply is 0.1ms and the frequency is 300Hz;

[0180] Alternatively, when the object to be welded is an aluminum plate, the frequency of the supplied sine wave voltage is 50 Hz; when the object to be welded is stainless steel, the frequency of the supplied sine wave voltage is 100 Hz.

[0181] In one embodiment, the expected value judgment module is specifically used to detect the processing effect of the object to be welded and obtain actual welding effect data. The welding effect data can be a laser radar unit for ranging, an image acquisition unit for acquiring image information, or a temperature acquisition unit for sensing temperature information; it also includes a similarity judgment module, and the similarity judgment module is used to calculate the similarity between the current welding effect data and the preset data through a similarity algorithm.

[0182] In one embodiment, the cylindrical lens 41 has one or more holes.

[0183] In one embodiment, the similarity judgment module includes a sorting unit, and the sorting unit is used to prioritize the original data through a weight algorithm.

[0184] In one embodiment, the judgment module is used to detect the processing effect of the object to be welded and obtain actual welding effect data. The welding effect data can be a laser radar unit for ranging, an image acquisition unit for acquiring image information, or a temperature acquisition unit for sensing temperature information; and also includes a judgment module, which is used to calculate the similarity between the current welding effect data and the preset data through a similarity algorithm.

[0185] In this embodiment, the judgment module is used to determine whether the similarity is less than or equal to a preset threshold; when the similarity is less than or equal to the preset threshold, it means that the first linear model or the first nonlinear model fails to achieve the ideal prediction or classification effect, and therefore more groups of original data need to be input into the first linear model or the first nonlinear model for prediction training until the similarity is greater than the preset threshold. When the similarity is greater than the preset threshold, the accuracy of the prediction will also be improved accordingly.

[0186] In one embodiment, the cylindrical lens 41 has one or more holes.

[0187] In one embodiment, the judgment module includes a sorting unit, and the sorting unit is used to prioritize the original data through a weight algorithm.

[0188] In this embodiment, since there are many parameters that determine the welding width, a sorting mechanism is required to screen out the most relevant parameters, thereby avoiding the influence of other parameters with low relevance, thereby improving the model training efficiency and thus improving the welding quality.

[0189] In one embodiment, referring to Figure 4 , provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the intelligent welding method in the above embodiment are implemented, such as Figure 1 The steps shown, or the processor executes the computer program to implement the functions of each module / unit of the intelligent welding system in the above embodiment, for example Figure 2 The function of the module.

[0190] In one embodiment, a computer storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned intelligent welding method are implemented. The memory in the embodiment of the present invention is used to store various types of data to support the operation of the intelligent welding system. The computer storage medium stores a computer program, and the computer storage medium can be a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a magnetic random access memory (FRAM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); it can also be a variety of devices including one or any combination of the above memories.

[0191] Although the present invention is disclosed through the above embodiments, the scope of the present invention is not limited thereto, and the above components may be replaced with similar or equivalent elements known to those skilled in the art without departing from the concept of the present invention.

Claims

1. An intelligent welding method, characterized in that: include: Predicting the linear data through the linear model to obtain a first prediction result; and / or, predicting the nonlinear data using a nonlinear model to obtain a second prediction result; Acquire target operating parameters based on the first prediction result and / or the second prediction result, and weld the current object to be welded by emitting target laser according to the target operating parameters through the laser body; receive a set of expected effect parameters and initial technical parameters; Acquire a target technical parameter set according to the effect parameter, traverse the target technical parameter set, and sequentially determine the corresponding relationship between each target technical parameter in the target technical parameter set and the effect parameter; If there is a linear relationship between the target technical parameter and the effect parameter, a specific value of the target technical parameter is predicted by a linear model to obtain a first prediction result, the initial technical parameter is adjusted according to the first prediction result, the adjusted initial technical parameter is used as a first target operating parameter, and the laser body emits a target laser according to the first target operating parameter to weld the current object to be welded; If there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, and the adjusted initial technical parameter is used as the second target operating parameter. The laser body emits a target laser according to the second target operating parameter to weld the current object to be welded.

2. An intelligent welding method, characterized in that: include: Receiving a desired effect parameter and an initial technical parameter set, and acquiring a target technical parameter set according to the effect parameter, wherein the initial technical parameter set is composed of a plurality of initial technical parameters, and the target technical parameter set is composed of a plurality of target technical parameters; Compare each target technical parameter with the initial technical parameter to determine whether the target technical parameter meets the preset conditions; If yes, then determining the correspondence between the target technical parameter and the effect parameter; If there is a linear relationship between the target technical parameter and the effect parameter, a specific value of the target technical parameter is predicted by a linear model to obtain a first prediction result, the initial technical parameter is adjusted according to the first prediction result, the adjusted initial technical parameter is used as a first target operating parameter, and the laser body emits a target laser according to the first target operating parameter to weld the current object to be welded; If there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, and the adjusted initial technical parameter is used as the second target operating parameter. The laser body emits a target laser according to the second target operating parameter to weld the current object to be welded.

3. The intelligent welding method according to claim 1, characterized in that The step of obtaining target operating parameters based on the first prediction result and / or the second prediction result, and emitting target laser light to weld the object to be welded according to the target operating parameters by the laser body includes: The weight of the first prediction result and / or the second prediction result is adjusted to obtain target operating parameters, and the laser body emits a target laser according to the target operating parameters to weld the current object to be welded.

4. An intelligent welding method, characterized in that: include: Step S10: constructing an initial linear model and an initial nonlinear model based on a preset algorithm; Step S20: inputting the pre-collected original data set into the initial nonlinear model for prediction training, and obtaining a first linear model and a first nonlinear model after training; The original data includes the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the energy density of the laser beam, the spot size, and the angle between the laser beam and the object to be welded; Step S30: collecting current operating condition data, the current operating condition data including first linear data and first nonlinear data, inputting the first linear data into the first linear model, and inputting the first nonlinear data into the first nonlinear model; Predicting the first linear data by using the first linear model to obtain a first prediction result, and predicting the first nonlinear data by using the first nonlinear model to obtain a second prediction result; The working condition data include the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the laser beam energy density, the spot size, and the angle between the laser beam and the object to be welded in actual welding; Step S40: acquiring target operating parameters based on the first prediction result and / or the second prediction result, and emitting target laser light to weld the current object to be welded according to the target operating parameters through the laser body, collecting image information of the object to be welded after welding, and obtaining welding effect data; the target operating parameters include power, pulse, laser beam swing angle, and angle and distance of the laser lens compared to the incident end of the laser beam of the laser system; Step S50: determining whether the welding effect data is less than or equal to an expected value; Step S60: If yes, return to step S20 until the welding effect data is greater than the expected value, and the welding effect data that meets the expected value is obtained, and the welding is completed.

5. The intelligent welding method according to claim 4, characterized in that: The step S40 comprises: Step S401: The step of welding the object to be welded by the intelligent welding system includes: Step S402: a first laser beam is emitted from the laser body, the first laser beam is collimated by a collimator to obtain a collimated laser, and the collimated laser passes through a cylindrical lens with a hole to achieve rearrangement of laser energy to obtain a rearranged laser; Step S403: the rearranged laser is reflected by a reflective mirror capable of displacement, thereby obtaining an oscillating laser whose spatial position can be changed; Step S404: the oscillating laser is focused by the field lens to form a target laser, and the current object to be welded is welded by the target laser; Alternatively, the step S40 includes: Step S405: The laser body emits a first laser beam and a second laser beam, the first laser beam and the second laser beam are collimated by a collimator to obtain a collimated laser, and the collimated laser passes through a cylindrical lens with a hole to achieve rearrangement of laser energy to obtain a rearranged laser; Step S406: The rearranged laser is reflected by a displacement mirror to obtain an oscillating laser whose spatial position can be changed; Step S407: After the oscillating laser is focused by the field lens, a target laser is formed, and the current object to be welded is welded by the target laser.

6. The intelligent welding method according to claim 5, characterized in that: The step S404 includes: Step S4041, the oscillating laser forms a target laser after passing through the field lens, and the focus position of the target laser can move along the axial direction or radial direction of the main optical axis of the field lens; Step S4042, welding the current object to be welded by using the target laser.

7. The intelligent welding method according to any one of claims 1 to 6, characterized in that: include: A filling wire is arranged between the object to be welded and the target laser.

8. An intelligent welding system, characterized in that: The intelligent welding method as claimed in claim 1 can be realized: A first prediction module, used to predict linear data through a linear model to obtain a first prediction result; and / or, a second prediction module, used to predict the nonlinear data by using a nonlinear model to obtain a second prediction result; A first welding module, used for obtaining target operating parameters based on the first prediction result and / or the second prediction result, and emitting target laser light according to the target operating parameters to weld the current object to be welded by the laser body; A first receiving module is used to receive a desired effect parameter and an initial technical parameter set; the first receiving module is used to obtain a target technical parameter set according to the effect parameter, traverse the target technical parameter set, and sequentially determine the corresponding relationship between each target technical parameter in the target technical parameter set and the effect parameter; A second welding module is used for predicting the specific value of the target technical parameter through a linear model to obtain a first prediction result when there is a linear relationship between the target technical parameter and the effect parameter, adjusting the initial technical parameter according to the first prediction result, taking the adjusted initial technical parameter as the first target operating parameter, and emitting a target laser through the laser body according to the first target operating parameter to weld the current object to be welded; When there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, the adjusted initial technical parameter is used as the second target operating parameter, and the target laser is emitted by the laser body according to the second target operating parameter to weld the current object to be welded.

9. An intelligent welding system, characterized in that: The intelligent welding method as claimed in claim 2 can be realized: A second receiving module is used to receive a desired effect parameter and an initial technical parameter set, and obtain a target technical parameter set according to the effect parameter, wherein the initial technical parameter set is composed of a plurality of initial technical parameters, and the target technical parameter set is composed of a plurality of target technical parameters; A second judgment module is used to compare each target technical parameter with the initial technical parameter to determine whether the target technical parameter meets the preset conditions; A third judgment module, if the target technical parameter meets the preset condition, then judges the corresponding relationship between the target technical parameter and the effect parameter; A third welding module is used for predicting the specific value of the target technical parameter through a linear model to obtain a first prediction result when there is a linear relationship between the target technical parameter and the effect parameter, adjusting the initial technical parameter according to the first prediction result, taking the adjusted initial technical parameter as the first target operating parameter, and emitting a target laser through the laser body according to the first target operating parameter to weld the current object to be welded; When there is a nonlinear relationship between the target technical parameter and the effect parameter, the specific value of the target technical parameter is predicted by a nonlinear model to obtain a second prediction result, the initial technical parameter is adjusted according to the second prediction result, the adjusted initial technical parameter is used as the second target operating parameter, and the target laser is emitted by the laser body according to the second target operating parameter to weld the current object to be welded.

10. An intelligent welding system, characterized in that: A smart welding method as claimed in claim 4 can be implemented, comprising a hardware component and a software component, wherein the software component and the hardware component are electrically connected or communicatively connected, and the software component and the hardware component are both connected to a main control module; The hardware components include a laser body for outputting a laser beam, a rotary solenoid, a collimating lens, a cylindrical lens with a hole, a reflector, and a field lens arranged in sequence along the transmission direction of the laser, the rotary solenoid is driven by an internal or external power supply, the rotary solenoid is used to drive the reflector to rotate or swing, the laser beam incident end of the cylindrical lens has a curvature, and the laser beam forms a focusing point and a laser spot after passing through the cylindrical lens; The software components include: A construction module, used for constructing an initial linear model and an initial nonlinear model based on a preset algorithm; A training module, used for inputting the pre-collected original data set into the initial nonlinear model for prediction training, and obtaining the first linear model and the first nonlinear model after the training respectively; The original data includes the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the energy density of the laser beam, the spot size, and the angle between the laser beam and the object to be welded; An acquisition module, used for acquiring current operating condition data, wherein the current operating condition data includes first linear data and first nonlinear data, inputting the first linear data into the first linear model, and inputting the first nonlinear data into the first nonlinear model; A prediction module, configured to predict the first linear data by using the first linear model to obtain a first prediction result, and predict the first nonlinear data by using a first nonlinear model to obtain a second prediction result; The working condition data include the deformation of the object to be welded, the temperature at the junction of the laser beam and the object to be welded, the welding width, the laser beam energy density, the spot size, and the angle between the laser beam and the object to be welded in actual welding; an acquisition module, used to acquire the first prediction result and the second prediction result, and adjust the actual operation parameters of the intelligent welding system based on the first prediction result and the second prediction result, and weld the current object to be welded by the target laser emitted by the intelligent welding system, collect image information of the object to be welded after welding, and obtain welding effect data; the actual operation parameters include the power, pulse, laser beam swing angle of the laser system, and the angle and distance of the laser lens compared to the incident end of the laser beam; An expected value judgment module, used to judge whether the welding effect data is less than or equal to the expected value; A prediction training module is used to input the pre-collected original data set into the initial nonlinear model for prediction training if the welding effect data is less than or equal to the expected value, and obtain the first linear model and the first nonlinear model after training respectively, until the welding effect data is greater than the expected value.

11. An intelligent welding system, characterized in that: The intelligent welding method as described in claim 3 can be implemented, wherein the first welding module includes an adjustment unit, and the adjustment unit is used to adjust the weight of the first prediction result and / or the second prediction result to obtain target operating parameters, and the laser body emits a target laser according to the target operating parameters to weld the current object to be welded.

12. The intelligent welding system according to any one of claims 8 to 11, characterized in that: A collimating lens (5), a cylindrical lens with a hole (41), a reflector (8), and a field lens (11) are sequentially arranged along the transmission direction of the laser; the reflector (8) is connected to a rotating motor, and the rotating motor is used to drive the reflector (8) to rotate according to a preset angle; the side of the cylindrical lens (41) opposite to the incident direction of the laser beam has a certain curvature; and the laser beam forms a focusing point and a laser point after passing through the cylindrical lens (41).

13. The intelligent welding system according to claim 12, characterized in that: One end of the cylindrical lens (41) which is in the form of a curved surface may be perpendicular to the optical path direction of the laser beam.

14. The intelligent welding system according to claim 13, characterized in that: The location having the curvature may be convex or concave.

15. The intelligent welding system according to claim 13, characterized in that: It also includes a laser processing module, which also includes an output unit, and the output unit is used to supply a continuous sinusoidal voltage to the rotating solenoid according to the first prediction result and the second prediction result; the laser processing module is electrically connected to a laser source, and when the rotating solenoid drives the rotating shaft of the rotating solenoid to rotate into a sine wave shape, the laser processing module controls the laser source to emit a laser beam, the laser beam is first collimated by a collimator, and then passes through a cylindrical lens (41) with an empty hole, a part of the laser beam passes through the empty hole, and another part passes through the surrounding of the empty hole, and then reaches the reflector (8), at which time the rotating shaft drives the reflector (8) to rotate or swing, and the laser beam reflected by the reflector also swings and rotates accordingly, and finally is focused by the field lens (11) into a light spot composed of central light and annular light, the maximum inner diameter of the center is smaller than the maximum inner diameter from the center of the circle to the outermost edge of the annular light, and the light spot swings or rotates to act on the object to be welded; Or, the output unit is used to supply an intermittent sinusoidal wave voltage to the rotary solenoid; When the output unit is working, it includes: The duration of each sine wave voltage supply is 0.1ms and the frequency is 300Hz; Alternatively, when the object to be welded is an aluminum plate, the frequency of the supplied sinusoidal voltage is 50 Hz; when the object to be welded is stainless steel, the frequency of the supplied sinusoidal voltage is 100 Hz.

16. The intelligent welding system according to claim 10, characterized in that: The expected value judgment module is specifically used to detect the processing effect of the object to be welded and obtain actual welding effect data. The welding effect data can be a laser radar unit for ranging, an image acquisition unit for acquiring image information, or a temperature acquisition unit for sensing temperature information; it also includes a similarity judgment module, which is used to calculate the similarity between the current welding effect data and the preset data through a similarity algorithm.

17. The intelligent welding system according to claim 12, characterized in that: The cylindrical lens (41) has one or more holes.

18. The intelligent welding system according to claim 16, characterized in that: The similarity judgment module includes a sorting unit, and the sorting unit is used to prioritize the original data through a weight algorithm.

19. A computer storage medium, characterized in that: The computer storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the intelligent welding method according to any one of claims 1 to 8 are implemented.

Images