An adaptive intelligent gas monitoring system for teaching-free welding robot

Abstract

The application provides a self-adaptive intelligent gas monitoring system for teaching-free welding robot, and relates to the technical field of industrial robots, comprising: a three-dimensional perception module, which is used for identifying the workpiece material category through a multispectral sensor integrated with a welding robot welding gun, and simultaneously acquiring three-dimensional point cloud data of a weld seam in real time by using a stereo vision camera and a line structure light scanning camera; a parameter adjustment module, which is used for calling a heat conductivity coefficient database based on the three-dimensional point cloud data, and correcting the initial flow of welding current, arc voltage and shielding gas in real time; and a matching control module, which is used for calculating the arc length variation based on the arc voltage, and controlling the opening degree of a gas flow valve through a dynamic regulator to realize dynamic matching of the gas flow and the changing arc length, and generate gas dynamic matching control parameters. Through the cooperative operation of self-adaptive posture compensation and real-time gas consumption monitoring, the application realizes teaching-free precise operation in the welding process, and improves the intelligent control of gas consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial robots, and in particular to an intelligent energy conservation monitoring system for a self-adaptive teaching-free welding robot. Background Art

[0002] In automobile chassis welding, traditional technology has some limitations in supporting adaptive teaching-free welding robots to achieve intelligent energy-saving monitoring. Traditional systems find it difficult to dynamically adjust the gas supply according to the specific structural characteristics of the weld and rely more on fixed flow settings. For example, when welding welds with different groove angles on the chassis, such as switching from a 30° groove longitudinal beam weld to a 60° groove crossbeam butt weld, the two grooves require different shielding gas coverage. The 60° groove requires a slightly larger flow rate to ensure adequate protection on both sides of the groove, while a 30° groove with too much flow rate is prone to overflow and cause waste. However, most traditional systems maintain the same flow rate, which either results in defects due to insufficient coverage when welding at a 60° groove, or ineffective gas consumption when welding at a 30° groove.

[0003] In addition, there is a lack of an instant flow response mechanism for dynamic changes in the arc. The gas flow regulation of the traditional system has a weak correlation with the arc state and cannot quickly match the real-time changes in the arc length. For example, when welding the transition area between the thick-walled casing and the thin-walled connecting plate of the chassis, the arc length will fluctuate due to the sudden change in the thickness of the workpiece. The traditional system finds it difficult to perceive this arc change in real time and often maintains a fixed flow rate. As a result, the thick-walled area may affect the protection effect due to insufficient flow, and the thin-walled area will have excess gas, making it impossible to achieve precise gas saving control. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an adaptive teaching-free welding robot intelligent energy conservation monitoring system, which dynamically adjusts the shielding gas flow by real-time sensing of the workpiece material, weld characteristics and arc state, thereby improving welding quality and reducing gas consumption.

[0005] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0006] In a first aspect, an adaptive teaching-free welding robot intelligent energy conservation monitoring system comprises:

[0007] The 3D perception module is used to identify the material type of the workpiece through the multispectral sensor integrated into the welding robot's welding gun, while using a stereo vision camera and a line structured light scanning camera to obtain 3D point cloud data of the weld in real time;

[0008] The parameter adjustment module is used to call the thermal conductivity database based on the 3D point cloud data to modify the welding current, arc voltage and initial flow rate of the shielding gas in real time;

[0009] The matching control module is configured to calculate an arc length variation based on the arc voltage, and control the opening degree of the gas flow valve through a dynamic regulator to realize dynamic matching of the gas flow and the varying arc length, and generate a gas dynamic matching control parameter;

[0010] The trajectory compensation module is configured to acquire a temperature gradient field of a welding area in a welding process by using an infrared thermal imager based on the gas dynamic matching control parameter, and calculate a thermal deformation of the workpiece according to the temperature distribution to generate a welding gun trajectory compensation instruction;

[0011] The posture control module is configured to calculate a final posture of the welding gun and an inclination angle of the gas nozzle in real time according to the welding gun trajectory compensation instruction and a preset three-dimensional welding path, and drive the robot to synchronously adjust the nozzle direction and the welding gun movement trajectory;

[0012] The posture compensation module is configured to set three spatially discrete detection points along the length direction of the weld according to the welding gun movement trajectory and the spatial orientation state of the nozzle, construct a spatial geometric configuration and perform grid segmentation to generate a welding gun posture compensation amount, and input the welding gun posture compensation amount into a welding gun movement control device to correct the final posture of the welding gun and the inclination angle of the nozzle in real time, thereby realizing intelligent gas monitoring.

[0013] Further, based on the three-dimensional point cloud data, a heat conduction coefficient database is called to correct the initial flow of the welding current, the arc voltage and the shielding gas in real time, including:

[0014] Based on the three-dimensional point cloud data, the geometric feature parameters of the weld cross section are acquired through point cloud registration and feature extraction algorithms, including the weld root gap width, the groove angle, the root face thickness and the base material thickness.

[0015] Based on the geometric feature parameters of the weld cross section, the workpiece material category identified by the multi-spectral sensor is synchronously associated, a preset dynamic database of heat conduction coefficients is queried, and the nonlinear heat conduction dynamic characteristic data of the material in the high-temperature welding zone are acquired;

[0016] Based on the geometric feature parameters of the weld cross section and the nonlinear heat conduction dynamic characteristic data, the theoretical heat input amount interval required for the welding pool to maintain a stable form is determined through dynamic balance calculation of heat input and heat conduction.

[0017] Based on the theoretical heat input amount interval, the welding current intensity reference value is dynamically adjusted and reduced according to the lower limit value, the upper limit value is adjusted to adjust the arc voltage fluctuation threshold, and the minimum initial set flow of the shielding gas is matched in combination with the predefined heat input-gas flow corresponding relationship.

[0018] Further, the arc length variation is calculated based on the arc voltage, and the opening degree of the gas flow valve is controlled through a dynamic regulator to realize dynamic matching of the gas flow and the varying arc length, and generate a gas dynamic matching control parameter, including:

[0019] Based on the corrected arc voltage signal, the real-time arc length dynamic change amount is calculated and input into the dynamic adjuster;

[0020] In the dynamic adjuster, according to the input real-time arc length dynamic change amount, the corresponding gas flow regulating valve opening degree proportional adjustment instruction is generated by mapping;

[0021] The gas flow regulating valve opening degree proportional adjustment instruction is sent to the actuator of the gas flow regulating valve, the opening degree of the gas flow regulating valve is adjusted in real time, and the current opening value of the gas flow regulating valve is obtained;

[0022] The real-time arc length dynamic change amount and the current opening value are integrated to generate a dynamic gas flow matching control parameter representing the real-time dynamic matching relationship between the protective gas flow and the arc length.

[0023] Further, based on the gas dynamic matching control parameter, the infrared thermal imager is used to obtain the temperature gradient field of the welding area during the welding process, and the thermal deformation amount of the workpiece is calculated according to the temperature distribution to generate a welding gun trajectory compensation instruction, including:

[0024] Based on the real-time heat input parameter represented by the gas dynamic matching result, the infrared thermal imager is triggered to synchronously obtain the infrared radiation intensity distribution data of the welding pool center and the surrounding area;

[0025] The infrared radiation intensity distribution data is subjected to spatial thermodynamic analysis to obtain three-dimensional temperature gradient distribution data of the welding area;

[0026] Based on the three-dimensional temperature gradient distribution data, the thermal expansion deformation amount of the current position of the workpiece is calculated in real time by quantifying the temperature non-uniformity difference in the vertical plane of the welding path;

[0027] The thermal expansion deformation amount is converted into the pose offset amount of the welding gun end in the three-dimensional space coordinate system to generate a real-time dynamic compensation instruction of the welding gun motion trajectory.

[0028] Further, based on the three-dimensional temperature gradient distribution data, the thermal expansion deformation amount of the current position of the workpiece is calculated in real time by quantifying the temperature non-uniformity difference in the vertical plane of the welding path, including:

[0029] Taking the weld centerline as a reference, the workpiece surface is symmetrically divided into a left temperature measurement area and a right temperature measurement area in the vertical plane of the welding path to generate area configuration data including distributed temperature monitoring points;

[0030] Based on the area configuration data, the temperature distribution of the two temperature measurement areas is obtained in real time, and the bilateral characteristic temperature difference is obtained by comparing the temperature distribution characteristics of the left temperature measurement area and the right temperature measurement area;

[0031] Based on the bilateral feature temperature difference, the pre-stored material thermal expansion coefficient is called to calculate the thermal expansion deformation variable of the current position of the workpiece through scalar operation;

[0032] Based on the thermal expansion deformation variable, the thermal expansion deformation variable is converted into the compensation displacement amount of the welding torch in the normal direction according to the space geometry of the welding path coordinate system.

[0033] Further, according to the welding torch trajectory compensation instruction and the preset three-dimensional welding path, the final pose of the welding torch and the inclination angle of the gas nozzle are calculated in real time, and the robot is driven to adjust the nozzle direction and the welding torch motion trajectory synchronously, including:

[0034] The space offset vector in the welding torch trajectory dynamic compensation instruction is subjected to real-time vector superposition operation with the target position coordinates of the preset three-dimensional welding base path at the current welding time, to obtain the corrected welding torch target position coordinates;

[0035] Based on the welding torch target position coordinates, the real-time pose Euler angle of the welding torch end in the three-dimensional space coordinate system is calculated in combination with the preset welding torch working pose constraint condition;

[0036] According to the real-time pose Euler angle of the welding torch end, the real-time space inclination angle of the gas nozzle center axis relative to the workpiece surface is determined through the pre-calibrated welding torch-nozzle geometric correlation;

[0037] The welding torch target position coordinates, real-time pose Euler angle and space inclination angle are synchronously input into the robot motion controller to generate real-time motion instructions of multi-joint linkage, and the robot is driven to synchronously execute the welding torch space trajectory adjustment and nozzle direction calibration.

[0038] Further, based on the welding torch motion trajectory and the nozzle space orientation state, three spatially discrete detection points are set along the length direction of the weld, a space geometric configuration is constructed and grid segmentation is performed, and a welding torch pose compensation amount is generated, including:

[0039] Based on the welding torch motion trajectory and the nozzle space orientation, actual coordinate data of the welding torch end in the three-dimensional space are synchronously acquired at the head position, the middle position and the tail position along the length direction of the weld;

[0040] The first end, middle and tail three groups of actual coordinate data are taken as spatial vertices, and the spatial plane equation coefficient representing the current welding trajectory is determined through vector cross product operation;

[0041] Based on the spatial plane equation coefficient, the current plane normal vector is extracted, and dot product operation is performed with the preset ideal welding reference plane normal vector to calculate the spatial normal angle value;

[0042] According to the spatial normal angle value, a welding gun tool coordinate system rotation direction mapping relationship is combined to generate a welding gun end pose space rotation compensation vector around the tool coordinate system.

[0043] Further, the welding gun pose compensation quantity is input into a welding gun motion control device to correct the final pose of the welding gun and the nozzle inclination angle in real time, and intelligent gas saving monitoring is realized, including:

[0044] Based on the welding gun pose space rotation compensation component, real-time welding gun end position coordinates, pose Euler angles and gas nozzle space inclination angles are obtained.

[0045] The welding gun pose space rotation compensation component is superimposed on the real-time pose Euler angle parameter to generate target pose parameters of the compensated welding gun end effector, while maintaining the nozzle space inclination angle unchanged.

[0046] Based on the target pose parameters and real-time position coordinates, the angle compensation value sequence of the six-axis joint is calculated in real time through robot inverse kinematics.

[0047] The joint angle compensation value sequence is input into a servo driver to synchronously drive the welding gun six-axis motion device to perform pose adjustment and nozzle inclination angle maintenance actions.

[0048] During the execution process, the original data stream of the protective gas flow sensor is collected in real time, and the intelligent gas saving monitoring of the whole welding process is realized by calculating the change rate of the gas consumption per unit time.

[0049] In a second aspect, a computing device includes:

[0050] One or more processors;

[0051] A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, so that the one or more processors implement the system.

[0052] In a third aspect, a computer readable storage medium stores a program, which is executed by a processor to implement the system.

[0053] The above-mentioned scheme of the present application at least includes the following beneficial effects:

[0054] The high-efficiency utilization of the protective gas is realized through multi-dimensional accurate regulation, the workpiece material is identified by means of a multi-spectrum sensor, and the weld characteristics are extracted by three-dimensional point cloud, the minimum initial gas flow is matched by combining with a heat conductivity coefficient database, and the excessive supply of gas under the traditional fixed parameters is avoided; based on the dynamic adjustment of the flow valve opening degree according to the arc voltage, the gas flow and the arc length are matched in real time, and the invalid gas supply during the arc fluctuation is reduced; at the same time, the heat deformation compensation and the welding gun posture adjustment are combined, the gas is accurately applied to the welding area, the gas waste is further reduced, the gas utilization efficiency is improved, and the welding production cost is reduced; in terms of welding quality guarantee, the weld quality stability is improved through the whole-process dynamic adaptation, the welding current, the voltage and the initial gas flow are accurately controlled through the dynamic balance calculation of heat input and heat conduction, and the foundation conditions for the molten pool stability are provided; the real-time dynamic matching of the arc length and the gas flow effectively avoids the defects such as oxidation and inclusion caused by insufficient gas, or the porosity problem caused by excessive flow; based on the heat deformation monitoring and trajectory compensation of the infrared thermal imager, and the real-time correction of the welding gun posture and the nozzle inclination angle, it is ensured that the welding gun always welds at the final position and angle, the weld forming problem caused by the workpiece deformation or the posture deviation is reduced, and the consistency and reliability of the welding quality are improved; in the adaptive teaching-free scene, the welding process is highly automated and intelligent, manual pre-teaching parameters are not needed, the workpiece state and the welding environment are perceived through multi-sensor fusion, and a series of operations such as material identification, parameter correction, flow regulation and trajectory compensation are completed autonomously, the dependence on manual experience is reduced, and the manual intervention cost is reduced; the real-time dynamic parameter adjustment and posture correction capability can flexibly cope with various changes in the welding process, such as workpiece material difference, weld characteristic change, arc fluctuation and the like, the adaptability and operation efficiency of the welding robot under complex working conditions are improved, and the efficient, stable and energy-saving welding demand in the teaching-free scene is met. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 FIG. 1 is a schematic diagram of an adaptive teaching-free welding robot intelligent gas monitoring system provided by an embodiment of the present application.

[0056] Figure 2 FIG. 2 is a schematic diagram of the adaptive teaching-free welding robot intelligent gas monitoring system provided by an embodiment of the present application, which controls the parameters based on the dynamic matching of the gas, acquires the temperature gradient field of the welding area by means of an infrared thermal imager during the welding process, calculates the heat deformation amount of the workpiece according to the temperature distribution, and generates a welding gun trajectory compensation instruction. DETAILED DESCRIPTION

[0057] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0058] As shown in Figure 1 An embodiment of the present application proposes an adaptive intelligent gas-saving monitoring system for teaching-free welding robot, comprising:

[0059] A three-dimensional perception module for identifying the workpiece material category through the multi-spectral sensor integrated with the welding robot's welding torch, and simultaneously using a stereo vision camera and a line structured light scanning camera to obtain real-time three-dimensional point cloud data of the weld;

[0060] A parameter adjustment module for calling a thermal conductivity database based on the three-dimensional point cloud data, and real-time correcting the initial welding current, arc voltage and protective gas flow rate;

[0061] A matching control module for calculating the arc length variation based on the arc voltage, and controlling the gas flow valve opening degree through a dynamic regulator to achieve dynamic matching of the gas flow and the varying arc length, and generating gas dynamic matching control parameters;

[0062] A trajectory compensation module for obtaining the welding area temperature gradient field using an infrared thermal imager during the welding process based on the gas dynamic matching control parameters, and calculating the thermal deformation of the workpiece according to the temperature distribution to generate welding torch trajectory compensation instructions;

[0063] A posture control module for real-time calculating the final posture of the welding torch and the inclination angle of the gas nozzle based on the welding torch trajectory compensation instructions and the preset three-dimensional welding path, and driving the robot to adjust the nozzle direction and the welding torch movement trajectory synchronously;

[0064] A posture compensation module for setting three spatially discrete detection points along the length direction of the weld based on the welding torch movement trajectory and the nozzle spatial orientation state, constructing a spatial geometric configuration and performing grid segmentation to generate welding torch posture compensation amount; inputting the welding torch posture compensation amount into the welding torch movement control device to real-time correct the final posture of the welding torch and the inclination angle of the nozzle, and achieving intelligent gas-saving monitoring.

[0065] In the embodiment of the present application, the precise identification of the workpiece material by the multispectral sensor, combined with the real-time capture of the weld three-dimensional point cloud data by the stereo vision camera and the line structured light scanning camera, provides comprehensive data support for the entire welding process from material characteristics to weld morphology. This basic capability enables the welding robot to quickly adapt to different material welding requirements (such as carbon steel, aluminum alloy, stainless steel, etc.), reducing welding defects such as porosity and cracks caused by material mismatch from the source. Based on the three-dimensional point cloud data, the thermal conductivity database is called to real-time correct the welding current, arc voltage and initial flow of the shielding gas, so that the core welding parameters can be dynamically adapted according to the actual weld morphology and workpiece material characteristics, avoiding over-welding and under-welding problems caused by parameter fixation. At the same time, by dynamically adjusting the gas flow valve opening degree through the arc length change, the real-time matching of the shielding gas flow and the arc state is realized, which not only prevents the oxidation of the weld caused by insufficient gas protection during arc fluctuation, but also avoids waste caused by excessive gas, improving the uniformity and stability of the weld formation.

[0066] During the welding process, the real-time monitoring of the temperature gradient field of the welding area by the infrared thermal imager, combined with the calculation of the workpiece thermal deformation amount based on the temperature distribution, provides a key basis for trajectory compensation. The trajectory compensation instruction can accurately offset the deformation of the workpiece caused by heating, such as warping and shrinkage, effectively avoiding the deviation of the welding torch trajectory caused by workpiece deformation, ensuring that the weld is always formed along the preset path, and the posture control combines the trajectory compensation instruction with the preset three-dimensional path to real-time adjust the welding torch posture and nozzle inclination angle, ensuring that the welding torch always aligns with the weld center at the final angle during the welding process, reducing the problem of uneven penetration caused by posture deviation. The posture compensation further corrects the welding torch posture and nozzle orientation by setting spatial discrete detection points to construct a geometric configuration, which can flexibly cope with complex spatial scenarios such as weld direction changes and workpiece curved surface transitions, reducing the risk of collision between the welding torch and the workpiece, and improving the safety and adaptability of the welding operation. From the dynamic matching of gas flow and arc length to the accurate gas supply control based on posture compensation, the entire system forms an intelligent gas-saving mechanism that adjusts the nozzle inclination angle and gas flow in real-time, avoiding the problem of continuous excessive gas output in traditional welding, reducing the invalid consumption of shielding gas, and at the same time, the automatic parameter adjustment and error compensation throughout the process reduce the rework rate caused by welding defects, reducing material loss and secondary processing costs.

[0067] In a preferred embodiment of the present application, based on the three-dimensional point cloud data, the thermal conductivity database is called to real-time correct the welding current, arc voltage and initial flow of the shielding gas, which can include:

[0068] Based on the three-dimensional point cloud data, the weld cross-section geometric feature parameters are obtained through point cloud registration and feature extraction algorithms, including weld root gap width, groove angle, blunt edge thickness and base metal thickness.

[0069] Based on the weld cross-section geometric feature parameters, the workpiece material category identified by the synchronous correlation multispectral sensor is inquired into the preset dynamic database of the thermal conductivity coefficient to obtain the nonlinear heat conduction dynamic characteristic data of the material in the welding high temperature zone.

[0070] Based on the weld cross-section geometric feature parameters and the nonlinear heat conduction dynamic characteristic data, the required theoretical heat input amount interval for maintaining a stable form of the welding molten pool is determined through dynamic balance calculation of heat input and heat conduction.

[0071] Based on the theoretical heat input amount interval, the welding current intensity reference value is dynamically adjusted and reduced according to the lower limit value, the upper limit value is adjusted to adjust the arc voltage fluctuation threshold value, and the minimum initial setting flow of the shielding gas is matched in combination with the predefined heat input-gas flow corresponding relationship.

[0072] In the embodiment of the application, after obtaining the three-dimensional point cloud data, the improved hawk optimization algorithm first removes the noise points from the data. The algorithm analyzes the spatial distribution density of each point in the point cloud data, just like a hawk overlooking the terrain from a high altitude. Points that are obviously abnormal in distance from surrounding points, like small stones in the terrain, are determined as noise points and removed. Then, the point cloud data is smoothed. The algorithm simulates the smooth flight trajectory of a hawk, fine-tunes the distance between adjacent points, and makes the point cloud data smoother, just like polishing rough ground. Then, the data is down-sampled. The algorithm retains the point cloud data of key areas according to the importance of the weld area, just like a hawk selecting the most important prey for tracking among many targets. The point cloud data of less important areas is appropriately simplified to reduce the computational workload.

[0073] In the point cloud registration process, the improved hawk optimization algorithm first quickly finds some points with obvious features in the vast amount of point cloud data, such as sharp points on the weld edge and flat area points at the bottom of the groove. These points are like landmark buildings on a map. Then, the algorithm preliminarily matches the feature points in the current point cloud data with the corresponding feature points in the preset standard weld, just like aligning the landmark buildings on two maps. Then, the algorithm calculates the rotation and translation parameters between the current point cloud and the standard weld, just like adjusting the angle and position of a painting to make it completely coincide with another painting. The algorithm iteratively optimizes these parameters by calculating the distance error between points, just like measuring the deviation between corresponding points in two paintings. The algorithm continuously adjusts until the error is minimized, thereby achieving accurate registration of the point cloud data and determining the accurate position of the weld in space.

[0074] In the extraction of the root gap width, the improved eagle optimization algorithm will scan along the direction of the weld root with a small step, just like the eagle scans the ground line by line with sharp eyes. During the scanning process, the algorithm records the distribution of point cloud at each position, finds the sparsest place of the root area point cloud, which is the position of the root gap. Then, the algorithm finds the nearest points on the left and right sides of the gap, measures the distance between the two points to get the preliminary width of the root gap. In order to improve the accuracy, the algorithm will repeatedly measure near the position, just like the eagle observes the same target multiple times. Then, the algorithm performs statistical analysis on the measurement results, eliminates outliers, and takes the average value as the final root gap width. For the extraction of the groove angle, the algorithm first identifies the edge regions of the groove on both sides, which is like the eagle identifying the outline of the prey. The algorithm analyzes the curvature change of the point cloud data, and the place with larger curvature is the edge of the groove. Then, the algorithm selects multiple points on both edge regions and determines the direction of the edge by fitting straight lines, just like determining the direction of a straight line by multiple points. Finally, the algorithm calculates the included angle between the two straight lines to get the groove angle. During the calculation process, the algorithm filters out abnormal points caused by surface unevenness and other factors to ensure the accuracy of the angle calculation.

[0075] For the acquisition of the root gap width, the algorithm will perform horizontal scanning at the bottom of the groove, just like the eagle flies parallel to the ground at low altitude to observe the ground conditions. During the scanning process, the algorithm records the distribution range of the point cloud data in the vertical direction, and the size of this range is the thickness of the root gap. In order to more accurately determine the root gap thickness, the algorithm will perform multiple scans at different positions at the bottom of the groove, and take the average value as the final result. At the same time, the algorithm will analyze the scanning data to exclude abnormal points that may be caused by splashes and other factors, ensuring the reliability of the root gap thickness measurement.

[0076] For the measurement of the base material thickness, the algorithm will select multiple measurement points in the base material area away from the weld, just like the eagle selects multiple observation points on the vast prairie. At each measurement point, the algorithm will perform vertical measurement from the upper surface point cloud to the lower surface point cloud of the base material to get the base material thickness at that point. Then, the algorithm performs statistical analysis on the thickness values of all measurement points, excludes outliers, and takes the average value to get the average thickness of the base material. The algorithm will intelligently adjust the number and position of the measurement points according to the density and distribution of the point cloud data to ensure that the measurement result accurately reflects the actual thickness of the base material.

[0077] The multispectral sensor will collect the spectral information of the workpiece surface, which contains the reflection intensity of light at different wavelengths, like the "light signal" emitted by the workpiece surface. The improved eagle optimization algorithm will extract features from these spectral information and analyze the change rule of the reflection intensity of light at different wavelengths to extract the spectral feature vector that can represent the material characteristics. This process is like extracting key information from complex signals. Then, the algorithm will compare the extracted spectral feature vector with the standard spectral feature vectors of various materials stored in the database. During the comparison process, the algorithm will calculate the similarity between them, just like comparing the similarity of two people's fingerprints. The material category corresponding to the standard spectral feature vector with the highest similarity is the recognized workpiece material category. To improve the accuracy of recognition, the algorithm will use multiple similarity calculation methods for cross-validation, just like observing the same object from multiple angles to gain a more accurate understanding.

[0078] After determining the material category of the workpiece, the improved eagle optimization algorithm will query the dynamic database of thermal conductivity coefficients combined with the obtained weld cross-sectional geometric feature parameters, such as root gap width and bevel angle. The algorithm will first narrow the query range according to the material category, just like the eagle searching for prey in a specific area. Then, the algorithm will analyze the influence of weld geometric feature parameters on the heat conduction process. For example, when the root gap is wider, the heat conduction path will change, and the algorithm will search for corresponding thermal conductivity data in the database based on this change. The algorithm will perform multidimensional search in the database, considering factors such as temperature and pressure that affect the thermal conductivity coefficient, just like the eagle considering terrain and climate when searching for prey. During the search process, the algorithm will continuously adjust the search strategy, predict the next possible search direction based on the existing query results, and improve the query efficiency. Finally, the algorithm will find the nonlinear heat conduction dynamic characteristic data of the material in the high-temperature welding area, which reflects the change of the material's heat conduction ability at different temperatures.

[0079] The improved eagle optimization algorithm will first analyze the weld cross-sectional geometric feature parameters in depth. For example, when the bevel angle is larger, the heat dissipation area of the weld is relatively larger, and heat is more easily dissipated. When the root gap is wider, more arc energy is needed to fill the gap, and the utilization efficiency of heat input will change. The algorithm will simulate the heat transfer path and distribution in the weld based on these geometric feature parameters, just like simulating the flow of water in a river. At the same time, combined with the nonlinear heat conduction dynamic characteristic data, the algorithm will analyze the influence of the change of the material's heat conduction ability at different temperatures on the heat distribution. The algorithm will divide the weld area into multiple small areas, each with different temperatures and heat conduction characteristics, just like dividing a map into multiple small areas, each with different terrain and climate.

[0080] In the dynamic balance calculation of heat input and heat conduction, the improved eagle optimization algorithm sets an initial heat input range, then for each heat input value in this range, the algorithm calculates the temperature distribution and heat conduction rate of the weld area under this heat input, and simulates the heat conduction process for a period of time, just like observing the temperature change of an object during heating; during the simulation, the algorithm continuously updates the temperature value of each small area, calculates the heat transfer rate according to the nonlinear heat conduction dynamic characteristic data, and checks whether the shape of the molten pool is stable and the temperature of the molten pool is within the appropriate range. If the molten pool shape is unstable or the temperature is too high or too low, the algorithm will adjust the heat input and re-simulate the calculation, just like adjusting the power of the heating equipment to control the temperature of the object. Through continuous iterative calculation, the algorithm will find a heat input range, in which the molten pool can maintain a stable shape and the temperature distribution is reasonable. This range is the theoretical heat input range.

[0081] The improved eagle optimization algorithm will adjust the welding current intensity reference value based on the lower limit value of the theoretical heat input range. The algorithm will first analyze the relationship between the current welding current intensity reference value and the lower limit value. If the current reference value is much higher than the lower limit value, the algorithm will use a larger step size to adjust it down, just like quickly reducing the speed of the car. If the difference between the two is small, the algorithm will use a smaller step size to fine-tune it, just like slowly adjusting the speed of the car. During the adjustment process, the algorithm will monitor the changes in heat input in real time to ensure that the heat input is always above the lower limit value. At the same time, the algorithm will consider the stability of the welding process to avoid unstable arc caused by rapid current adjustment. The algorithm will simulate the welding process under different current intensities to evaluate the stability of the molten pool and the quality of the weld formation, just like testing the driving performance of the vehicle under different road conditions. Through multiple adjustments and evaluations, the algorithm will determine a suitable welding current intensity reference value that meets the requirements of heat input and ensures the stability of the welding process.

[0082] For the adjustment of the arc voltage fluctuation threshold, the improved eagle optimization algorithm will take the upper limit value of the theoretical heat input range as a reference. The algorithm will first analyze the current arc voltage fluctuation, record the maximum, minimum and fluctuation frequency of the voltage, etc., just like recording the heart rate changes of a person. Then, according to the upper limit value, the algorithm will calculate the maximum fluctuation range allowed for the arc voltage, which is the new fluctuation threshold. During the adjustment process, the algorithm will consider the stability of the arc and the welding quality. If the fluctuation threshold is set too small, the arc may be unstable, resulting in defects in the weld. If the fluctuation threshold is set too large, the heat input may exceed the upper limit value, affecting the welding effect. The algorithm will simulate the welding process under different fluctuation thresholds to evaluate the stability of the arc and the quality of the weld, just like testing the performance of equipment under different environmental conditions. Finally, the algorithm will determine an arc voltage fluctuation threshold that can ensure the stability of the arc and the heat input does not exceed the upper limit value.

[0083] In matching the minimum initial set flow of the shielding gas, the improved eagle optimization algorithm will rely on the predefined heat input-gas flow correspondence relationship. The algorithm will first find the gas flow range corresponding to the theoretical heat input range in the correspondence relationship, just like looking up the definition of a word in a dictionary. Then, within this range, the algorithm will find the minimum initial set flow. During the search process, the algorithm will consider the protection effect and cost factors of the shielding gas. If the flow is too small, it may not be able to effectively protect the molten pool, resulting in weld oxidation. If the flow is too large, it will cause waste of shielding gas and increase costs. The algorithm will simulate the distribution of shielding gas around the weld under different flow rates to evaluate its protection effect on the molten pool, just like simulating the diffusion of smoke under different wind speeds. At the same time, the algorithm will calculate the cost of shielding gas consumption under different flow rates for comprehensive evaluation. Through continuous comparison and analysis, the algorithm will determine a minimum initial set flow that can meet the protection requirements and reduce costs.

[0084] The improved eagle optimization algorithm plays an important role in various aspects. In the point cloud data processing and feature extraction process, the fine search and analysis capability of the algorithm makes the acquisition of the geometric feature parameters of the weld cross section more accurate and less error. In the material identification and thermal conductivity coefficient query link, the intelligent matching and multi-dimensional search function of the algorithm improves the accuracy of material identification and the matching accuracy of thermal conductivity coefficient data. In the heat input interval determination and parameter adjustment process, the iterative optimization and dynamic balance calculation capability of the algorithm ensures the accuracy of the theoretical heat input interval and the accuracy of the welding parameter adjustment, thereby effectively avoiding welding defects caused by inaccurate parameter setting and improving welding quality. In terms of welding process stability, based on the theoretical heat input interval and the dynamically adjusted welding current and arc voltage, the welding pool can always maintain a stable form. In the parameter adjustment process, the improved eagle optimization algorithm fully considers the dynamic changes and various interference factors in the welding process. Through real-time monitoring and adjustment, the welding parameters can adapt to different welding conditions, reducing fluctuations and instability in the welding process. The reasonable matching of the initial flow of the protective gas also provides a good protection environment for the welding process, reducing external factors that interfere with the molten pool, further improving the stability of the welding process, reducing the probability of problems such as weld deviation, incomplete fusion, and porosity, and improving the efficiency and reliability of welding production. In terms of resource utilization efficiency, the improved eagle optimization algorithm determines the minimum initial setting flow of the protective gas through accurate calculation, avoiding waste of the protective gas. At the same time, the dynamic adjustment of the welding current and arc voltage controls the heat input within a reasonable range, reducing unnecessary consumption of electrical energy and improving energy utilization efficiency. In welding production, protective gas and electrical energy are important cost factors. By optimizing the setting of these parameters, welding production costs can be effectively reduced, and the economic benefits of enterprises can be improved. In addition, the efficient search and optimization capability of the algorithm shortens the parameter adjustment time, improves the pace of welding production, and improves resource utilization efficiency. In terms of adaptability, it can simultaneously associate with the workpiece material category and obtain nonlinear heat conduction dynamic characteristic data, and correct the parameters in combination with the weld geometric features.

[0085] In a preferred embodiment of the present application, the arc length change is calculated based on the arc voltage, and the gas flow valve opening is controlled by a dynamic regulator to dynamically match the gas flow with the changing arc length, generating gas dynamic matching control parameters, which can include:

[0086] Based on the corrected arc voltage signal, the real-time arc length dynamic change is calculated and input into the dynamic regulator.

[0087] In the dynamic regulator, the corresponding gas flow regulating valve opening proportional adjustment instruction is generated based on the input real-time arc length dynamic change.

[0088] Send the gas flow regulating valve opening proportion adjusting instruction to the actuator of the gas flow regulating valve, adjust the opening of the gas flow regulating valve in real time, and obtain a current opening value of the gas flow regulating valve;

[0089] Integrate the real-time arc length dynamic change amount and the current opening value to generate a dynamic gas flow matching control parameter representing a real-time dynamic matching relationship between the protective gas flow and the arc length.

[0090] In the embodiment of the present application, the original arc voltage signal is modified. The original signal is collected by a voltage sensor. In the collection process, the signal may fluctuate or deviate due to electromagnetic interference in the welding environment, sensor error and other factors. In the modification, the original voltage signal data within a certain time is collected, and the abnormal values, such as values exceeding the normal voltage fluctuation range, are counted. These abnormal values may be caused by instantaneous interference. By eliminating these abnormal values, a preliminary purified voltage signal is obtained. Then, the purified signal is further processed by using a sliding average method. A plurality of continuous voltage data within a recent period of time is selected, and the average value thereof is calculated to smooth the slight fluctuation in the signal, and a modified arc voltage signal is obtained.

[0091] Then, the real-time arc length is calculated according to the modified arc voltage signal. In the welding process, the arc voltage and the arc length have a certain corresponding relationship. When the arc length changes, the arc voltage will change accordingly. Under stable welding conditions, the relationship has a reference regularity. Through the accumulation of experimental data in the early stage, the corresponding proportional relationship between the arc voltage and the arc length under the current welding material, welding current and other parameters is determined. For example, in a certain welding, it is found that the arc length increases by a certain length when the arc voltage increases by a certain value. Based on this corresponding relationship, the modified real-time arc voltage signal value is converted into the corresponding real-time arc length value. Then, the dynamic change amount of the arc length is calculated. The real-time arc length values at different times are continuously collected. The arc length value at the current time is subtracted from the arc length value at the previous time to obtain the arc length change value between the adjacent two times. With the passage of time, such calculation is continuously performed, so that the real-time arc length dynamic change amount is obtained. The change amount can reflect the increase and decrease of the arc length in the welding process in real time.

[0092] The dynamic adjuster is internally preset with a mapping relationship table of the arc length dynamic change amount and the proportional adjustment instruction of the gas flow regulating valve opening degree, simulates different arc length change conditions, and records the proportion of the gas flow regulating valve opening degree that needs to be adjusted in each change condition to enable the protection gas flow to be well matched with the arc length. When the real-time arc length dynamic change amount is input to the dynamic adjuster, the adjuster analyzes the change amount. First, the positive and negative of the change amount is judged. A positive value indicates that the arc length is increasing, and a negative value indicates that the arc length is decreasing. Then, the size of the change amount, that is, the amplitude of the arc length change, is determined. According to the positive and negative and size of the change amount, the corresponding adjustment direction and adjustment proportion in the preset mapping relationship table are found. For example, when it is detected that the arc length dynamic change amount is positive and the value is large, it is indicated that the arc length increases a lot. At this time, the gas flow needs to be increased to better protect the arc. The corresponding proportional adjustment instruction of the gas flow regulating valve opening degree in the mapping table is that the opening degree is increased.

[0093] After the generated proportional adjustment instruction of the gas flow regulating valve opening degree is sent to the actuator, the actuator acts according to the instruction. The actuator has a driving component inside and can convert the proportional adjustment instruction into a mechanical action. For example, if the instruction is to increase the opening degree by 10%, the actuator will drive the valve core to move, so that the flow cross section of the valve is increased, thereby increasing the opening degree. During the adjustment process, the opening degree sensor installed on the valve will monitor the position of the valve core in real time. The opening degree sensor determines the current opening degree value by detecting the displacement of the valve core. The sensor converts the displacement signal into an electrical signal and transmits it to the control system. The control system processes the electrical signal, such as converting the strength of the electrical signal into the corresponding opening degree percentage. For example, the sensor detects that the valve core moves to the position of 60% of the maximum displacement. After signal processing, it is determined that the current opening degree value is 60%, and this value is fed back to the control system in real time.

[0094] The real-time arc length dynamic change amount and the corresponding current opening degree value in a period of time are collected, classified and arranged in time sequence, and the corresponding relationship between the two is recorded. For example, at a certain time, the arc length dynamic change amount is +2 mm, and the corresponding current opening degree value is 55%. At the next time, the change amount is +1 mm, and the opening degree value is 58%. Then, the corresponding relationship is integrated and analyzed to find the adjustment rule of the opening degree value under different arc length change conditions. Through statistical analysis, it is clear that when the arc length increases or decreases by a certain amount, the opening degree value increases or decreases by a certain amplitude. Therefore, the dynamic gas flow matching control parameter that can represent the real-time dynamic matching relationship between the protection gas flow and the arc length is generated. These parameters include the opening degree adjustment corresponding to different arc length changes and completely reflect the dynamic matching state between the two.

[0095] By calculating the dynamic change amount of the arc length in real time and adjusting the gas flow, the change of the arc length can be responded in time. When the arc length increases, the protective gas flow is increased in time, which can effectively isolate the interference of air on the arc and the molten pool, and avoid welding defects caused by insufficient gas protection. When the arc length decreases, the gas flow is correspondingly reduced to prevent excessive impact of the gas flow on the arc, and maintain the stable combustion of the arc, thereby improving the stability of the entire welding process. In traditional welding, the protective gas flow is often set as a fixed value, which remains unchanged regardless of the change of the arc length. The gas flow may be too large to cause waste, or the flow may be too small to cause insufficient protection. The technology adjusts the gas flow in real time according to the dynamic change of the arc length, so that the gas flow is always matched with the arc length, which ensures good protection effect, avoids unnecessary gas consumption, improves the utilization efficiency of the protective gas, and reduces the welding cost. Since the protective gas flow and the arc length are dynamically matched in real time, the protective gas can always provide appropriate protection effect regardless of the fluctuation of the arc length during the welding process. This makes the weld uniform in heating and stable in molten pool state, reduces the generation of welding defects such as pores and cracks caused by improper protection, reduces the quality difference of the weld at different positions and different times, and enhances the overall welding quality.

[0096] As shown in Figure 2 In a preferred embodiment of the present application, based on the gas dynamic matching control parameter, the temperature gradient field of the welding area is obtained by using an infrared thermal imager during the welding process, and the thermal deformation amount of the workpiece is calculated according to the temperature distribution to generate a welding gun trajectory compensation instruction, which can include:

[0097] Based on the real-time heat input parameter represented by the gas dynamic matching result, the infrared thermal imager is triggered to synchronously obtain the infrared radiation intensity distribution data of the welding pool center and the surrounding area;

[0098] The infrared radiation intensity distribution data is subjected to spatial thermodynamic analysis to obtain three-dimensional temperature gradient distribution data of the welding area;

[0099] Based on the three-dimensional temperature gradient distribution data, the thermal expansion deformation of the current position of the workpiece is calculated in real time by quantifying the temperature non-uniformity difference in the vertical plane of the welding path, specifically including: taking the weld centerline as the reference, the workpiece surface is symmetrically divided into left and right temperature measurement regions in the vertical plane of the welding path, and region configuration data including distributed temperature monitoring points are generated; based on the region configuration data, the temperature distribution of the two temperature measurement regions is obtained in real time, and the bilateral characteristic temperature difference is obtained by comparing the temperature distribution characteristics of the left and right temperature measurement regions; based on the bilateral characteristic temperature difference, the pre-stored material thermal expansion coefficient is called, and the thermal expansion deformation of the current position of the workpiece is calculated by scalar operation; based on the thermal expansion deformation, the thermal expansion deformation is converted into the compensation displacement amount of the welding torch in the normal direction according to the space geometry of the welding path coordinate system.

[0100] The thermal expansion deformation is converted into the pose offset of the welding torch end in the three-dimensional space coordinate system to generate real-time dynamic compensation instructions for the welding torch motion trajectory.

[0101] In the embodiment of the present application, when welding is performed, the real-time heat input parameter formed by dynamic gas matching is continuously tracked, which reflects the matching state between the welding arc light energy, gas flow and heat conversion efficiency in real time. When the heat input rate, energy density and other indicators in the heat input parameter reach the preset starting threshold, such as when the heat input rate exceeds a certain fixed value, a synchronous trigger signal is sent to the infrared thermal imager. After receiving the signal, the infrared thermal imager immediately starts the internal scanning mechanism, and according to the fixed shooting frame number per second, the lens is aimed at the center of the welding pool and the area within a certain range around it. In the shooting process, the infrared detector of the thermal imager receives the infrared radiation of the welding area point by point. The temperature at different positions is different, and the intensity of the infrared energy radiated is also different. The high-temperature area radiates strong energy, and the low-temperature area radiates weak energy. The detector converts the received infrared radiation signal into a weak electric signal, which is amplified by the internal circuit to a signal strength that can be identified, and then the electromagnetic interference signal in the environment is removed through filtering, and finally a set of original data containing the infrared radiation intensity information of each specific position in the welding area is formed, that is, the infrared radiation intensity distribution data.

[0102] After obtaining the infrared radiation intensity distribution data, first, data cleaning is performed, each data point in the data is checked one by one, and the radiation intensity values of the data point and the surrounding adjacent data points are compared. If the value of a certain data point exceeds the preset normal fluctuation range from the surrounding data points, it is determined as a noise point and is removed, and the average value of the surrounding data points is used to fill the vacancy at this position. Then, according to the installation position and shooting angle of the infrared thermal imager, combined with the actual size of the welding area measured in advance, the correspondence between the data coordinates and the actual spatial position is established, and each radiation intensity data point is accurately corresponded to the actual three-dimensional spatial position of the welding area. Then, according to the infrared radiation characteristic curve of the welding material, which is determined by experiment in advance, the corresponding infrared radiation intensity of the material at different temperatures is indicated, and the radiation intensity data of each spatial position is converted into the corresponding temperature value. After that, for the spatial interval existing in the temperature data, an interpolation method is used to calculate the temperature value of the interval position according to the values of the adjacent known temperature points, so that the temperature distribution data is continuous and complete in the three-dimensional space. Finally, the temperature difference between each point in the three-dimensional space and the adjacent point in the x, y and z directions is calculated, and then divided by the spatial distance between the two points to obtain the temperature change rate in each direction. Integrating these change rates forms the three-dimensional temperature gradient distribution data of the welding area.

[0103] Based on the three-dimensional temperature gradient distribution data, the welding path is first determined, and then a vertical plane range perpendicular to the welding path is demarcated. This plane covers the current welding position and a certain length of the area before and after. In this vertical plane, a plurality of detection points are selected in a uniform grid form to ensure that these detection points can cover the corners and key areas of the plane. The temperature values of each detection point are recorded respectively, the difference between the highest temperature value and the lowest temperature value in the plane is calculated, and the difference in average temperature in different areas is calculated to quantify the unevenness of the temperature distribution in the plane. According to the thermal expansion characteristics of the welding material, the expansion ratio of the material per temperature rise is known, the temperature rise value of each detection point is obtained by subtracting the initial temperature of the position before welding from the temperature value of each detection point, and the influence degree of temperature rise at different positions on the overall deformation is judged combined with the temperature non-uniformity difference. The temperature difference is larger, the influence weight on the deformation is higher. According to the temperature rise value of each detection point, the material expansion ratio and the corresponding influence weight, the expansion amount of each detection point is calculated, and the thermal expansion deformation amount of the entire workpiece at the current position in the vertical plane in different directions is obtained by comprehensively considering the expansion of all detection points.

[0104] First, set up a three-dimensional coordinate system, determine the origin position of the coordinate system, usually with the welding starting point or a certain fixed reference point of the workpiece as the origin, clear the corresponding space direction of x, y, z axis, measure and record the position coordinates and attitude angle of the welding torch end in the initial state in the coordinate system as the reference pose, the calculated thermal expansion deformation variable corresponds to the three axes of the coordinate system, analyze how much the workpiece will be elongated or shortened in the x-axis direction due to thermal expansion, and correspond to the distance the welding torch needs to move in the x-axis direction; Similarly, analyze the expansion influence in the y-axis and z-axis directions, for attitude shift, according to the thermal expansion difference of different positions of the workpiece, judge whether the workpiece has deformation such as bending and tilting, calculate the angle that the welding torch end needs to adjust, such as the rotation angle around the x-axis, y-axis and z-axis, integrate the position adjustment distance in the x-axis, y-axis and z-axis directions and the three rotation angles to form the pose offset of the welding torch end in the three-dimensional space, finally, according to the movement control logic of the welding torch, convert the pose offset into specific movement instructions, and clear the direction and distance that the welding torch needs to move in each axis, as well as the direction and angle that needs to be rotated. These instructions are dynamic compensation instructions for real-time adjustment of the welding torch trajectory.

[0105] Through detailed processing of infrared radiation intensity data and temperature gradient calculation, the temperature changes at each position of the welding area can be accurately captured, and the thermal expansion deformation variable can be accurately obtained. The welding torch pose offset is converted into compensation instructions, which can make the welding torch adjust in real time following the thermal deformation of the workpiece, reduce the position deviation caused by thermal deformation during welding, ensure the accuracy and consistency of the welding seam, improve the quality of the welding joint, and the whole process from infrared data acquisition to compensation instruction generation is real-time, which can quickly capture the dynamic changes of the workpiece thermal deformation during welding. When the temperature distribution of the welding area changes, temperature analysis, deformation variable calculation and instruction generation can be completed in a short time, the welding torch trajectory is adjusted in time to avoid large-area welding defects caused by deformation accumulation. For complex structures and variable welding paths, through three-dimensional temperature gradient analysis and detailed calculation of thermal expansion deformation variable, the deformation state of each part of the workpiece can be mastered, and targeted compensation instructions can be generated, which makes the technology adapt to the welding needs of various complex workpieces, expands the application range of welding equipment, and reduces the dependence on specific welding experience. Because it can compensate the influence of thermal deformation in real time, it avoids common welding defects such as incomplete penetration, welding deviation and undercut caused by inaccurate welding torch position, reduces the repair work after welding, saves rework time and cost, improves the pass rate of welding products, and improves the overall welding production efficiency.

[0106] In a preferred embodiment of the present application, the final pose of the welding torch and the inclination angle of the gas nozzle are calculated in real time according to the welding torch trajectory compensation instruction and the preset three-dimensional welding path, the nozzle direction and the welding torch movement trajectory are synchronously adjusted by driving the robot, which can include:

[0107] The space offset vector in the welding torch trajectory dynamic compensation instruction is subjected to real-time vector superposition operation with the target position coordinates of the preset three-dimensional welding basic path at the current welding time, to obtain the corrected welding torch target position coordinates.

[0108] Based on the welding torch target position coordinates, the real-time pose Euler angle of the welding torch end in the three-dimensional space coordinate system is calculated in combination with the preset welding torch working pose constraint condition.

[0109] According to the real-time pose Euler angle of the welding torch end, the real-time space inclination angle of the gas nozzle center axis relative to the workpiece surface is determined through the pre-calibrated welding torch-nozzle geometric correlation.

[0110] The welding torch target position coordinates, the real-time pose Euler angle and the space inclination angle are synchronously input into the robot motion controller to generate real-time motion instructions of multi-joint linkage, so as to drive the robot to synchronously execute the welding torch space trajectory adjustment and the nozzle direction calibration.

[0111] In the embodiment of the present application, first, the space offset data is extracted from the welding torch trajectory dynamic compensation instruction, which contains the adjustment distances in three directions, corresponding to the lengths of movement in the front-back, left-right and up-down directions, and the target position information corresponding to the current welding time in the three-dimensional welding basic path planned in advance is called out, which is the theoretical welding point without compensation. Next, the position correction calculation is performed. In the front-back direction, the front-back adjustment distance in the space offset data and the front-back coordinates of the basic path target position are added. In the left-right direction, the left-right adjustment distance in the space offset data and the left-right coordinates of the basic path target position are added. In the up-down direction, the up-down adjustment distance in the space offset data and the up-down coordinates of the basic path target position are added. Through the superposition calculation in the three directions, the new position information obtained is the corrected welding torch target position coordinates.

[0112] Based on the corrected welding torch target position coordinates, first determine the specific paragraph of this position in the welding path, whether it is a straight line part or a curve part, then call out the preset welding torch working posture restriction conditions, including the angle range between the welding torch and the welding path, the maximum degree of vertical inclination of the welding torch, etc., according to the corrected target position and the welding path, first calculate the rotation angle in the first direction: determine how much the welding torch needs to tilt up or down to ensure that the welding torch head is directly opposite the molten pool, while not exceeding the maximum limit of vertical inclination, then calculate the rotation angle in the second direction: adjust the degree of inclination of the welding torch in the horizontal direction according to the bending direction of the welding path to ensure that the angle between the welding torch and the path meets the restriction range, and finally calculate the rotation angle in the third direction: fine-tune the rotation angle of the welding torch in the horizontal plane to accurately align the welding torch wire outlet with the center of the welding seam. After such adjustment, the real-time posture angle that meets the restriction conditions is obtained.

[0113] First, call out the geometric relationship data of the welding torch and the nozzle measured in advance, which records the synchronous change rule of the nozzle when the welding torch rotates, such as the angle of rotation of the nozzle when the welding torch rotates in a certain direction. After obtaining the real-time posture angle of the welding torch end, first determine the inclination change of the nozzle center axis in the vertical direction relative to the workpiece surface according to the rotation angle in the first direction, and then determine the inclination change of the nozzle in the horizontal direction relative to the workpiece surface according to the rotation angle in the second direction. By comprehensively considering the inclination changes in the two directions, the included angle between the gas nozzle center axis and the workpiece surface is calculated, which is the real-time spatial inclination angle and can reflect the current inclination state of the nozzle.

[0114] First, check whether the three groups of data of the welding torch target position coordinates, the real-time posture angle and the gas nozzle spatial inclination angle are reasonable to ensure that the data is within the normal working range of the equipment. After checking that the data is correct, synchronize these data to the robot motion controller. After receiving the data, the motion controller analyzes the difference between the data and the current state of each joint of the robot, calculates the distance each joint needs to move according to the welding torch target position coordinates to make the welding torch reach the target position, determines the angle each joint needs to rotate to adjust the welding torch posture according to the real-time posture angle, and further fine-tunes the motion parameters of the related joints according to the spatial inclination angle of the nozzle to ensure that the direction of the nozzle is correct. During the calculation process, the motion range and speed limit of each joint of the robot are considered to ensure safe and reasonable motion. Then, real-time motion instructions for multi-joint linkage are generated to clearly indicate the motion direction, speed and movement of each joint. Finally, the robot motion controller sends the instructions to the drive device to drive the joints of the robot to move synchronously according to the instructions, completing the welding torch trajectory adjustment and nozzle direction calibration.

[0115] By superimposing the target position coordinates, the position deviation caused by thermal deformation and other factors can be accurately offset, so that the welding gun always moves near the preset welding path, avoids the weld from deviating from the joint, ensures the accuracy of the welding position, improves the weld forming effect, calculates the attitude angle according to the limit conditions, so that the welding gun maintains a reasonable attitude during welding and does not appear excessive inclination or deviation, which ensures the stability of the arc, reduces the increase of spatter and insufficient penetration caused by attitude problems, and ensures the stable welding process. According to the geometric relationship, the nozzle inclination angle is determined, so that the nozzle covers the molten pool area at a suitable angle, so that the protective gas uniformly wraps the molten pool and effectively isolates the air, which reduces the porosity and slag inclusion defects caused by poor gas protection, improves the quality and performance of the welded joint, generates multi-joint linkage command to drive the robot to move synchronously, avoids the trajectory problem caused by uncoordinated joint movement, improves the efficiency of welding gun adjustment and nozzle calibration, reduces equipment energy consumption and wear, prolongs the service life, and reduces manual intervention and improves the level of welding automation.

[0116] In a preferred embodiment of the application, based on the welding gun motion trajectory and the spatial orientation state of the nozzle, three spatial discrete detection points are arranged along the length direction of the weld, the spatial geometric configuration is constructed and grid segmentation is performed, and the welding gun attitude compensation amount is generated, which can include:

[0117] Based on the welding gun motion trajectory and the spatial orientation of the nozzle, the actual coordinate data of the welding gun end in three-dimensional space is synchronously obtained at the first end position, the middle position and the tail end position along the length direction of the weld;

[0118] The first end, middle and tail end three groups of actual coordinate data are taken as spatial vertices, and the spatial plane equation coefficient representing the current welding trajectory is determined by vector cross product operation;

[0119] Based on the spatial plane equation coefficient, the current plane normal vector is extracted, and point product operation is performed with the preset ideal welding reference plane normal vector to calculate the spatial normal angle value;

[0120] According to the spatial normal angle value, combined with the pre-marked welding gun tool coordinate system rotation direction mapping relationship, the attitude space rotation compensation vector of the welding gun end around the tool coordinate system is generated.

[0121] In the embodiment of the present application, during the welding process, the high-precision position detection device carried on the welding torch is used to collect coordinate data at the head end position, the middle position and the tail end position along the length direction of the weld. When the detection device is aligned with the head end position, the actual values of the three directions in the three-dimensional space are recorded in real time to form the actual coordinate data of the head end. Then, the detection device is moved to the middle position of the weld, and the coordinate values of the three directions at this position are recorded as the actual coordinate data of the middle position. Finally, the detection device reaches the tail end position, and the coordinate data of the three directions at the tail end position are collected and stored to obtain three complete sets of actual coordinate data of the head end, the middle position and the tail end.

[0122] The three sets of actual coordinate data of the head end, the middle position and the tail end are regarded as three vertices in space. First, two space vectors are constructed from the head end coordinate as the starting point to the middle position coordinate and the tail end coordinate. Then, the vector cross product operation is performed on the two vectors. During the operation, a new vector is obtained by cross-multiplying and then subtracting the components of the vectors. The three components of the new vector correspond to the coefficients of the space plane equation representing the current welding trajectory, i.e., the coefficients of the three direction terms and the constant term in the plane equation.

[0123] Based on the obtained coefficients of the space plane equation, the normal vector of the current plane is extracted, and the three components of the normal vector correspond to the coefficients of the plane equation. Then, the normal vector of the preset ideal welding reference plane is obtained, and the dot product operation is performed on the two normal vectors. During the dot product operation, the corresponding components of the two normal vectors are multiplied and then added to obtain a product result. Then, the product result is divided by the product of the module length of the normal vector of the current plane and the module length of the normal vector of the ideal reference plane to obtain a ratio. According to the ratio, the corresponding angle value is determined, which is the space normal angle value.

[0124] First, the inclination degree of the current welding plane and the ideal reference plane is determined according to the calculated space normal angle value. Then, the rotation direction mapping relationship of the tool coordinate system of the welding torch, which is calibrated through experiments in advance, is combined. The relationship clearly shows the welding torch rotation direction rules corresponding to different angle changes. According to the size and direction of the angle value, the direction and angle range of the rotation of the end of the welding torch around the tool coordinate system are determined according to the mapping relationship, and then the posture space rotation compensation vector that can accurately describe the rotation compensation of the end of the welding torch is generated. The vector contains the direction and amplitude information of the rotation.

[0125] By collecting coordinate data at the three key positions of the weld joint head, middle and tail, the spatial form of the welding trajectory can be fully captured, and the deviation of the trajectory description caused by single point data error can be avoided. The spatial plane equation is determined based on the three vertices, which makes the representation of the welding trajectory more consistent with the actual welding situation. The current plane normal vector is extracted and multiplied with the ideal reference plane normal vector to calculate the included angle, which can accurately reflect the deviation degree of the actual welding plane from the ideal plane. The rotation compensation vector is generated according to the included angle value, which can adjust the welding gun posture in a targeted manner to make the actual welding plane as close as possible to the ideal reference plane, ensuring the consistency of the welding plane and reducing the welding defects caused by the deviation of the plane. With the help of the pre-calibrated rotation direction mapping relationship, the spatial normal included angle value can be quickly converted into a specific posture rotation compensation vector, which saves the complex manual calculation and adjustment process. The whole calculation process is clear in logic and clear in steps, which can realize the automatic compensation and adjustment of the welding gun posture, shorten the posture adjustment time, and improve the overall efficiency of the welding operation. Through accurate coordinate collection, plane equation determination, angle calculation and compensation vector generation, the welding gun can always maintain the ideal posture during the welding process, reducing the problems of insufficient welding strength and poor weld formation caused by posture deviation. The stable welding gun posture ensures the uniformity of the welding energy input, improves the stability and reliability of the welding quality, and reduces the welding scrap rate.

[0126] In a preferred embodiment of the present application, the welding gun posture compensation amount is input into the welding gun motion control device to real-time correct the final posture of the welding gun and the nozzle inclination angle, realize intelligent gas saving monitoring, which can include:

[0127] Based on the welding gun posture space rotation compensation component, the real-time welding gun end position coordinates, posture Euler angles and gas nozzle space inclination angle are obtained;

[0128] The welding gun posture space rotation compensation component is superimposed on the real-time posture Euler angle parameter to generate the target posture parameter of the compensated welding gun end effector, while keeping the nozzle space inclination angle unchanged;

[0129] Based on the target posture parameter and real-time position coordinates, the angle compensation value sequence of the six-axis joint is calculated in real time through the inverse kinematics of the robot;

[0130] The joint angle compensation value sequence is input into the servo driver to synchronously drive the welding gun six-axis motion device to perform the position adjustment and nozzle inclination angle maintenance action;

[0131] During the execution process, the original data stream of the protective gas flow sensor is collected in real time, and the intelligent gas saving monitoring of the whole welding process is realized by calculating the change rate of gas consumption per unit time.

[0132] In the embodiment of the present application, based on the generated welding gun attitude space rotation compensation component, the integrated multiple groups of sensors of the welding gun are started, the position detection sensor captures the position of the welding gun end in the three-dimensional space in real time through laser positioning or mechanical ranging, records the specific values of three mutually perpendicular directions every 0.1 seconds, and forms continuously updated real-time welding gun end position coordinates; the attitude detection element senses the rotation state of the welding gun around the three direction shafts through the internal gyroscope and accelerometer, converts the rotation angle into a specific value, and outputs real-time attitude Euler angle, including the rotation angles around the transverse shaft, the longitudinal shaft and the vertical shaft; the inclination sensor on the gas nozzle continuously monitors the inclination angle of the nozzle with the horizontal reference plane through the gravity sensing principle, records a value every 0.2 seconds, ensures the real-time of the gas nozzle space inclination angle data, and all sensor data are synchronously transmitted to the control terminal through the data line to form a complete parameter data set.

[0133] The rotation compensation values around the three direction shafts are extracted from the welding gun attitude space rotation compensation component, for example, the transverse shaft compensation value, the longitudinal shaft compensation value and the vertical shaft compensation value, and these compensation values are respectively superimposed with the corresponding angle parameters in the real-time attitude Euler angle. The real-time transverse shaft rotation angle is added to the transverse shaft compensation value to obtain the target angle of the transverse shaft after compensation. Similarly, the real-time longitudinal shaft rotation angle is added to the longitudinal shaft compensation value to obtain the target angle of the longitudinal shaft, and the real-time vertical shaft rotation angle is added to the vertical shaft compensation value to obtain the target angle of the vertical shaft. The three target angles are combined to form the target attitude parameters of the welding gun end effector after compensation. In this process, the output data of the nozzle inclination sensor are locked and do not participate in any superposition operation, so as to ensure that the space inclination angle of the gas nozzle maintains the initial measured value unchanged.

[0134] The obtained target attitude parameters (three direction target angles) and real-time position coordinates (three direction position values) are input into the robot control algorithm. The algorithm first determines the installation position, rotation range and transmission ratio of each joint according to the mechanical design parameters of the six-axis motion device, calculates the angle change amount of the first shaft according to the difference between the target position and the current position, and combines the joint kinematics relationship; then, the angle compensation value of the second shaft is deduced based on the adjustment angle of the first shaft to ensure that the movement of the second shaft can cooperate with the first shaft to realize the target position; according to the same logic, the angle compensation values of the third to sixth shafts are gradually calculated, and the calculation of each joint needs to consider the influence of the movement of the previous joint. The angle compensation values of the six joints are arranged in sequence according to the shaft number to form a continuous six-axis joint angle compensation value sequence.

[0135] The six-axis joint angle compensation value sequence is transmitted to the servo driver through the data interface. The signal conversion inside the driver converts the digital compensation value into an electrical signal. After the first-axis servo motor receives the electrical signal corresponding to the compensation value, it drives the motor rotor to rotate by a corresponding angle, which drives the first-axis joint to rotate through the speed reducer. The second-axis to the sixth-axis servo motors receive the signals in turn and perform the rotation action. The movement of all joints is started synchronously, ensuring that all adjustment actions are completed within 0.5 seconds. During the adjustment process, the nozzle inclination sensor feeds back the inclination data in real time. If there is a slight fluctuation, the driver will fine-tune the movement amplitude of the corresponding joint to maintain the nozzle inclination stable within the initial set value ±0.5° range.

[0136] After the welding starts, the protective gas flow sensor collects gas instantaneous flow data every 0.5 seconds, recording it as a raw data stream, such as the flow value at 1 second, the flow value at 1.5 seconds, etc. When calculating the unit time gas consumption rate, first select the flow values of two adjacent time points, such as the flow value at 1.5 seconds minus the flow value at 1 second, to get the gas consumption difference in 0.5 seconds. Then divide this difference by the time interval of 0.5 seconds to get the average consumption rate for this time period. Repeat the calculation every 1 second to generate continuous rate data. The control terminal compares the real-time rate with the pre-set normal range value. If it exceeds the range for three consecutive times, it will immediately trigger the gas monitoring prompt and display the gas consumption abnormal information.

[0137] By synchronously collecting position, attitude, and inclination data through multiple sensors and setting a high-frequency sampling interval, it is ensured that each parameter can reflect the actual state of the welding torch in real time. The attitude Euler angle real-time tracks the rotation changes, and the nozzle inclination is stably monitored, avoiding calculation deviations caused by data lag or missing. The rotation compensation component is added to the real-time Euler angle one by one, making the attitude correction of each direction axis accurate and in place. The compensation values of the horizontal, vertical, and perpendicular axes are applied to the corresponding rotation angles, ensuring that the target attitude of the welding torch end effector is highly consistent with the ideal state, reducing the welding trajectory deviation problem caused by attitude deviation. Based on the inverse kinematics solution of the mechanical structure parameters, the linkage relationship between the six-axis joints is fully considered, and the angle compensation value of each joint can cooperate with the previous joint movement, avoiding joint action conflicts or motion interference. The compensation value sequence in order ensures that the six-axis motion device is coordinated and consistent, ensuring smooth and accurate welding torch attitude adjustment and shortening the response time of attitude correction. After the servo driver receives the compensation value sequence, it synchronously drives the six-axis movement, avoiding nozzle inclination fluctuations caused by individual joint actions. High-frequency gas flow data collection and unit time rate calculation can capture subtle fluctuations in gas consumption in real time.

[0138] The embodiment of the present application also provides a computing device, comprising a processor, a memory storing a computer program, the computer program being executed by the processor to implement the system as described above. All implementation manners in the above system embodiment are suitable for this embodiment and can achieve the same technical effects.

[0139] The embodiment of the present application also provides a computer readable storage medium storing instructions, which, when executed on a computer, cause the computer to implement the system as described above. All implementation manners in the above system embodiment are suitable for this embodiment and can achieve the same technical effects.

[0140] The above is the preferred embodiment of the present application, it should be pointed out that, for those skilled in the art, without departing from the principles of the present application, can make a number of improvements and refinements, these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. An adaptive, self-teaching, intelligent gas monitoring system for a welding robot, comprising: The method comprises the following steps: A three-dimensional perception module is used to identify the material category of the workpiece through a multi-spectral sensor integrated with a welding robot welding gun, and simultaneously use a stereo vision camera and a line structured light scanning camera to obtain three-dimensional point cloud data of the weld in real time; A parameter adjustment module is used to call a thermal conductivity database based on the three-dimensional point cloud data, and correct the initial flow of the welding current, arc voltage and shielding gas in real time, which specifically comprises: Based on the three-dimensional point cloud data, the cross-sectional geometric feature parameters of the weld are obtained through point cloud registration and feature extraction algorithm, including the root gap width, bevel angle, root face thickness and base metal thickness; Based on the cross-sectional geometric feature parameters of the weld, the material category of the workpiece identified by the multi-spectral sensor is synchronously associated, the pre-set dynamic database of thermal conductivity is queried, and the non-linear heat conduction dynamic characteristic data of the material in the high temperature zone of welding is obtained; Based on the cross-sectional geometric feature parameters of the weld and the non-linear heat conduction dynamic characteristic data, the theoretical heat input range required for the stable shape of the welding pool is determined through the dynamic balance calculation of heat input and heat conduction; Based on the theoretical heat input range, the welding current intensity reference value is dynamically adjusted and reduced according to the lower limit value, the upper limit value is adjusted to adjust the arc voltage fluctuation threshold, and the minimum initial set flow of the shielding gas is matched according to the pre-defined heat input-gas flow corresponding relationship; 2. The adaptive, teach-less welding robot intelligent gas monitoring system of claim 1, wherein, A matching control module is used to calculate the arc length variation based on the arc voltage, and control the gas flow valve opening degree through a dynamic adjuster to realize the dynamic matching of the gas flow and the changing arc length, and generate gas dynamic matching control parameters; A trajectory compensation module is used to obtain the temperature gradient field of the welding area by using an infrared thermal imager during the welding process based on the gas dynamic matching control parameters, and calculate the thermal deformation of the workpiece according to the temperature distribution to generate welding gun trajectory compensation instructions; An attitude control module is used to calculate the final attitude of the welding gun and the inclination angle of the gas nozzle in real time according to the welding gun trajectory compensation instructions and the pre-set three-dimensional welding path, and drive the robot to synchronously adjust the nozzle direction and the welding gun movement trajectory; An attitude compensation module is used to set three spatially discrete detection points along the length direction of the weld based on the welding gun movement trajectory and the spatial orientation state of the nozzle, construct a spatial geometric configuration and perform grid segmentation to generate welding gun attitude compensation amount; input the welding gun attitude compensation amount into the welding gun movement control device to correct the final attitude of the welding gun and the inclination angle of the nozzle in real time, and realize intelligent gas monitoring. The arc length variation is calculated based on the arc voltage, and the gas flow valve opening degree is controlled through a dynamic adjuster to realize the dynamic matching of the gas flow and the changing arc length, and generate gas dynamic matching control parameters, which comprises: Based on the corrected arc voltage signal, the real-time arc length dynamic variation is calculated and input into the dynamic adjuster; In the dynamic adjuster, the corresponding gas flow adjustment valve opening degree proportional adjustment instruction is generated according to the input real-time arc length dynamic variation; The gas flow adjustment valve opening degree proportional adjustment instruction is sent to the actuator of the gas flow adjustment valve to adjust the opening degree of the gas flow adjustment valve in real time, and the current opening value of the gas flow adjustment valve is obtained. The real-time arc length dynamic change amount and the current opening value are integrated to generate a dynamic gas flow matching control parameter representing a real-time dynamic matching relationship between the protective gas flow and the arc length.

3. The adaptive, teach pendant-free, intelligent gas monitoring system for a welding robot of claim 2, wherein, Based on the gas dynamic matching control parameter, the temperature gradient field of the welding area is obtained by using an infrared thermal imager during the welding process, and the thermal deformation amount of the workpiece is calculated according to the temperature distribution to generate a welding gun trajectory compensation instruction, including: Based on the real-time heat input parameter represented by the gas dynamic matching result, the infrared thermal imager is triggered to synchronously obtain the infrared radiation intensity distribution data of the welding pool center and the surrounding area; The infrared radiation intensity distribution data is subjected to spatial thermodynamic analysis to obtain three-dimensional temperature gradient distribution data of the welding area; Based on the three-dimensional temperature gradient distribution data, the thermal expansion deformation amount of the current position of the workpiece is calculated in real time by quantifying the temperature non-uniformity difference in the vertical plane of the welding path; The thermal expansion deformation amount is converted into the pose offset amount of the welding gun end in the three-dimensional space coordinate system to generate a real-time dynamic compensation instruction of the welding gun motion trajectory.

4. The adaptive, show-and-tell welding robot smart gas monitoring system of claim 3, wherein, Based on the three-dimensional temperature gradient distribution data, the thermal expansion deformation amount of the current position of the workpiece is calculated in real time by quantifying the temperature non-uniformity difference in the vertical plane of the welding path, including: The workpiece surface is symmetrically divided into a left temperature measurement area and a right temperature measurement area in the vertical plane of the welding path with the weld centerline as a reference to generate area configuration data including distributed temperature monitoring points; Based on the area configuration data, the temperature distribution of the two temperature measurement areas is obtained in real time, and the bilateral characteristic temperature difference is obtained by comparing the temperature distribution characteristics of the left temperature measurement area and the right temperature measurement area; Based on the bilateral characteristic temperature difference, the pre-stored material thermal expansion coefficient is called to calculate the thermal expansion deformation amount of the current position of the workpiece through scalar operation; Based on the thermal expansion deformation amount, the thermal expansion deformation amount is converted into the compensation displacement amount of the welding gun in the normal direction according to the spatial geometry of the welding path coordinate system.

5. The adaptive, show-and-tell welding robot smart gas monitoring system of claim 4, wherein, According to the welding gun trajectory compensation instruction and the preset three-dimensional welding path, the final pose of the welding gun and the inclination angle of the gas nozzle are calculated in real time, and the robot is driven to synchronously adjust the nozzle direction and the welding gun motion trajectory, including: The space offset vector in the welding gun trajectory dynamic compensation instruction is subjected to real-time vector superposition operation with the target position coordinates of the preset three-dimensional welding basic path at the current welding time to obtain the corrected welding gun target position coordinates; Based on the welding gun target position coordinates, the real-time attitude Euler angle of the welding gun end in the three-dimensional space coordinate system is calculated in combination with the preset welding gun working pose constraint condition; According to the real-time attitude Euler angle of the welding gun end, the real-time spatial inclination angle of the gas nozzle center axis relative to the workpiece surface is determined through the pre-calibrated welding gun-nozzle geometric correlation; The welding gun target position coordinates, the real-time attitude Euler angle and the spatial inclination angle are synchronously input into the robot motion controller to generate real-time motion instructions of multi-joint linkage, and the robot is driven to synchronously execute the welding gun space trajectory adjustment and the nozzle direction calibration.

6. The adaptive, teach pendant-free, intelligent gas monitoring system for a welding robot of claim 5, wherein, Based on the welding gun motion trajectory and the nozzle spatial position state, three spatial discrete detection points are arranged along the length direction of the weld, a spatial geometric configuration is constructed and grid segmentation is performed to generate a welding gun pose compensation amount, including: Based on the welding gun motion trajectory and the nozzle spatial orientation, the actual coordinate data of the welding gun end in three-dimensional space is synchronously acquired at the first end position, the middle position and the tail end position along the length direction of the weld; The first end, middle and tail end three groups of actual coordinate data are taken as spatial vertices, and the spatial plane equation coefficient representing the current welding trajectory is determined through vector cross product operation; Based on the spatial plane equation coefficient, the current plane normal vector is extracted, and point product operation is performed with the preset ideal welding reference plane normal vector to calculate the spatial normal angle value; According to the spatial normal angle value, the welding gun tool coordinate system rotation direction mapping relationship is combined to generate the welding gun end pose space rotation compensation vector around the tool coordinate system.

7. The adaptive, teach pendant-free, intelligent gas monitoring system for a welding robot of claim 6, wherein, The welding gun pose compensation amount is input into the welding gun motion control device to correct the final pose of the welding gun and the nozzle inclination angle in real time, and intelligent gas saving monitoring is realized, including: Based on the welding gun pose space rotation compensation component, the real-time welding gun end position coordinates, pose Euler angle and gas nozzle space inclination angle are acquired; The welding gun pose space rotation compensation component is superimposed on the real-time pose Euler angle parameter to generate the target pose parameter of the compensated welding gun end executor, while keeping the nozzle space inclination angle unchanged; Based on the target pose parameter and the real-time position coordinates, the angle compensation value sequence of the six-axis joint is calculated in real time through robot inverse kinematics; The joint angle compensation value sequence is input into the servo driver to synchronously drive the welding gun six-axis motion device to execute the pose adjustment and nozzle inclination angle maintenance action; During the execution process, the original data stream of the protective gas flow sensor is collected in real time, and the intelligent gas saving monitoring of the whole welding process is realized by calculating the gas consumption change rate per unit time.

8. A computing device, comprising: comprise: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, so that the one or more processors implement the system as claimed in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a program which, when executed by a processor, implements the system as claimed in any one of claims 1 to 7.

Images