Intelligent multi-parameter collaborative control method and system for welding process

Abstract

The application belongs to the technical field of intelligent welding and industrial robot control, and discloses a kind of intelligent welding process multi-parameter collaborative control method and system, method includes: obtaining the working condition information of component to be welded;Establish process parameter vector, and each constraint;Combining data, determine initial process parameter combination by parameter matching and rule screening mode;Collect welding process state information;Extract process state characteristic quantity and calculate deviation and risk quantity;Perform multi-parameter collaborative linkage correction;Compare preset judgment condition.System includes: each module and process database for executing the intelligent welding process multi-parameter collaborative control method.Affinity effect: the application can simultaneously consider heat input stability, penetration forming capacity, weld width consistency and equipment execution stability under complex working conditions, avoid control oscillation caused by disordered change of multiple parameters, and improve the self-adaptive ability and engineering implementability of welding process.

Description

Technical Field

[0001] This invention relates to the field of intelligent welding and industrial robot control technology, specifically to a multi-parameter collaborative control method and system for intelligent welding processes. Background Technology

[0002] During automated welding, the thickness of the component plate, joint type, welding position, assembly gap, and local heat dissipation conditions are usually not constant, but rather vary continuously or segmentally along the welding path. This variation directly affects the heat input per unit length, the molten pool, and the weld formation. If static process parameter control methods such as fixed current, fixed voltage, and fixed welding speed are still used, it is difficult to simultaneously ensure the formation of penetration depth, weld width consistency, and equipment stability, which can easily lead to problems such as local burn-through, incomplete penetration, and uncontrolled weld width.

[0003] To address the aforementioned issues, existing technologies employ two main approaches. One approach relies on offline process templates or preset parameters from experience databases. While simple to implement, this approach typically only solves the initial condition matching problem and struggles to compensate for dynamic deviations caused by assembly errors, heat accumulation, or trajectory changes during welding. Another approach introduces online adjustment mechanisms, but often focuses on correcting only a single parameter, such as adjusting only the current, voltage, or speed. Since welding heat input, penetration depth, and weld appearance are inherently influenced by multiple coupled parameters, isolated correction of a single parameter can easily lead to the problem of compensating for one parameter while compensating for another.

[0004] Some solutions, while involving multiple process parameters, often employ empirical serial adjustment methods for their control logic. They fail to provide a unified model of the relationships between process state characteristics, parameter deviations, quality risks, and parameter executable boundaries, and lack mechanisms for determining dominant states, prioritizing adjustments, and limiting single-cycle amplitude. When abnormal heat input, insufficient penetration, and weld width deviation occur simultaneously, the disordered changes in multiple parameters can easily lead to control overshoot, execution jitter, and closed-loop oscillations. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for multi-parameter collaborative control of intelligent welding processes.

[0006] The present invention achieves the above-mentioned technical objectives through the following technical means.

[0007] A multi-parameter collaborative control method for an intelligent welding process includes the following steps:

[0008] Step S1: Obtain the working condition information of the component to be welded: Collect the component type, material type, plate thickness, joint type, welding position, equipment capacity parameters and target forming index of the component to be welded;

[0009] Step S2: Establish the set of process parameters and constraints: Based on the working condition information of the component to be welded in step S1, establish the process parameter vector. In addition to heat input constraints, weld formation constraints, and equipment operation constraints;

[0010] Step S3: Generate initial process parameter combinations: Combining the process database, rule base, and empirical parameter templates, determine the initial process parameter combinations through parameter matching and rule filtering. ;

[0011] Step S4: Collect welding process status information: During the welding process, collect the arc status, current and voltage signals, wire feeding status, running speed, oscillation status, weld formation status and temperature in real time;

[0012] Step S5: Extract process state characteristic quantities and calculate deviation and risk quantities: Based on the process state information described in step S4, extract the arc length fluctuation quantity. Heat input deviation Speed ​​deviation Wire feeding matching degree deviation and weld formation deviation And further calculate the comprehensive quality risk index. ;

[0013] Step S6: Perform multi-parameter collaborative correction: First, construct the original collaborative correction amount based on the heat input deviation, weld formation deviation, and comprehensive quality risk index. Secondly, considering the contribution ratio and dominant weight of each risk source. and secondary weights The original linkage correction values ​​are then weighted hierarchically to obtain the hierarchically weighted parameter correction vector. Finally, a single-cycle amplitude constraint and equipment feasible region projection processing are applied to the hierarchically weighted parameter corrections to obtain the final updated combination of process parameters. ;

[0014] Step S7: Compare preset judgment conditions: Judge the preset stability conditions, forming consistency conditions, and equipment executability conditions of the updated process parameter combination. When all judgment conditions are met, drive the welding execution unit to complete the welding operation based on the process parameter combination; when any judgment condition is not met, repeat steps S4 to S7.

[0015] Furthermore, the comprehensive quality risk index The formula is expressed as:

[0016]

[0017] in, , , , and These are the reference thresholds for heat input deviation, weld formation deviation, speed deviation, wire feeding matching degree deviation, and arc length fluctuation, respectively, and λ1~λ5 are the risk fusion weights corresponding to each feature item.

[0018] Furthermore, the original linkage correction amount The formula is expressed as:

[0019]

[0020] in, and These represent the original correction values ​​for welding current, welding voltage, welding speed, oscillation frequency, and oscillation amplitude within the current sampling period, respectively. , , , , , , , , , All are linkage adjustment coefficients; This is the cross-coupling compensation coefficient.

[0021] Furthermore, the specific process of performing hierarchical weighting on the original linkage correction amount is as follows:

[0022] First, through analysis The contribution percentage of each internal risk source is used to construct the severity of heat input anomalies. Severity of weld formation abnormalities and the severity dominated by complex risks Secondly, based on , and The severity level with the highest value is selected through comparison. The abnormal state corresponding to the highest severity level is determined as the dominant state of the current sampling period, and the hot input anomaly correction weight is set according to the dominant state. Weighting of weld formation abnormality correction Finally, the original linkage correction amount is optimized based on the dominant state and correction weight. The parameter correction vector after hierarchical weighting is obtained. .

[0023] Furthermore, the severity of the heat input anomaly Severity of weld formation abnormalities and the severity dominated by complex risks The formula is expressed as:

[0024] , ,

[0025] in, , , , and These are the reference thresholds for heat input deviation, weld formation deviation, speed deviation, wire feeding matching degree deviation, and arc length fluctuation, respectively, and λ1~λ5 are the risk fusion weights corresponding to each feature item.

[0026] Furthermore, the determination of the dominant state specifically involves:

[0027] when At its maximum, it is determined to be a state dominated by abnormal heat input. At this time, let =Dominant weight, =Secondary weight; when At its maximum, it is determined to be the dominant state of abnormal weld formation, at which point... =Secondary weights, =Dominant weight; when At its maximum, it is determined to be in a state dominated by composite risk, at which point let =Dominant weight.

[0028] Furthermore, the dominant weight and secondary weights The following nonlinear mapping function is used for real-time calculation:

[0029]

[0030]

[0031] in, This represents the severity value of the dominant state within the current sampling period; The dominant weight benchmark value This is the sensitivity gain coefficient; This is a secondary correction and regulation factor.

[0032] Furthermore, the hierarchically weighted parameter correction vector The system of equations for calculation is as follows:

[0033]

[0034] in, , , , , These represent the original correction values ​​of welding current, welding voltage, welding speed, oscillation frequency, and oscillation amplitude after layered weighting within the current sampling period.

[0035] Furthermore, the final updated combination of process parameters The formula is expressed as:

[0036]

[0037]

[0038] in, For the current sampling period t, the process parameter combination, I t For welding current, U t For welding voltage, For welding speed, The oscillation frequency, The amplitude of the swing. (·) is the projection operator for projecting onto the device feasible region Ω. To satisfy the amplitude limit parameter correction vector, This is the amplitude limiting function.

[0039] A multi-parameter collaborative control system for an intelligent welding process, used to execute the multi-parameter collaborative control method for the intelligent welding process, characterized in that it includes:

[0040] The sensing and input module is used to acquire the component type, plate thickness, joint type, welding position, equipment capacity parameters and target forming index of the component to be welded;

[0041] The process parameter initialization module is used to establish a set of process parameters and output an initial combination of process parameters based on the process database.

[0042] The collaborative control decision module is used to receive the process status assessment results and, in combination with the dominant status judgment results, preset parameter linkage relationships, and various constraints, calculate the parameter correction amount.

[0043] The robot controller is used to convert the modified combination of process parameters into control instructions that can be executed by the welding power source, wire feeding mechanism, oscillating mechanism and robot motion control unit.

[0044] Welding execution unit, used to actually complete welding operations;

[0045] The process status assessment module is used to collect, extract features, and calculate deviations for the arc status, current and voltage signals, wire feeding status, running speed, oscillation status, and weld formation status during the welding process.

[0046] The quality and anomaly determination module is used to determine whether the current corrected parameter combination meets the stability and consistency requirements based on a preset threshold.

[0047] The process database stores reference process parameter templates, constraints, and parameter linkage correction rules for different component types, material types, plate thickness grades, joint types, and welding positions.

[0048] The present invention has the following beneficial effects:

[0049] 1. This invention achieves coordinated adjustment of welding current, voltage, speed, and oscillation parameters by constructing a cross-coupling compensation mechanism. This method not only provides feedback correction based on deviation but also counteracts the correlation interference of single parameter adjustments on other indicators by introducing a cross-coupling compensation factor, effectively reducing the risk of new heat input imbalances or forming deviations caused by local corrections.

[0050] 2. This invention transforms abnormal states in the welding process from empirical descriptions into quantifiable evaluation quantities by constructing thermal input deviation, weld width deviation, and comprehensive quality risk indicators. This elevates the control strategy from phenomenon-driven to deviation and risk-driven, improving the pertinence and interpretability of online correction under complex working conditions.

[0051] 3. In the linkage correction process, the present invention introduces dominant state determination, priority constraint, amplitude constraint and equipment feasible domain constraint, which can control the parameter update rhythm and execution boundary while ensuring welding effect, avoid overshoot, jitter and closed loop oscillation caused by disordered changes of multiple parameters, and improve the stability of engineering implementation.

[0052] 4. This invention is particularly suitable for welding scenarios of battery trays, battery pack boxes, vehicle door rings and other thin-walled complex components for new energy vehicles. It can maintain good heat input stability, forming consistency and welding adaptability even when there are changes in plate thickness, changes in heat dissipation conditions and assembly deviations. Attached Figure Description

[0053] Figure 1 This is a flowchart of the multi-parameter collaborative control method for the intelligent welding process described in this invention.

[0054] Figure 2 This is a schematic diagram of the multi-parameter collaborative control of the intelligent welding process described in this invention.

[0055] Figure 3 This is a flowchart of the intelligent welding process multi-parameter collaborative control system described in this invention. Detailed Implementation

[0056] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the scope of protection of the present invention is not limited thereto.

[0057] A multi-parameter collaborative control method for intelligent welding processes, such as Figure 1 , Figure 2 As shown, it includes the following steps:

[0058] Step S1: Obtain the working condition information of the component to be welded: Collect one or more of the following: component type, material type, plate thickness, joint type, welding position, equipment capability parameters, and target forming index of the component to be welded. The equipment capability parameters include welding current, welding voltage, welding speed range, and oscillation frequency and oscillation amplitude of the oscillation mechanism. The target forming index includes one or more of the following: target weld width, target penetration depth grade, allowable reinforcement height range, and allowable heat input range.

[0059] Step S2: Establish the set of process parameters and constraints: Based on the working condition information of the component to be welded in step S1, on the one hand, establish the process parameter vector. , among which, I t For welding current, U t For welding voltage, For welding speed, The oscillation frequency, For the swing amplitude; on the other hand, establish thermal input constraints (thermal input constraints are used to limit the thermal input per unit length to be within the target allowable range, which can be expressed as) ,in, This represents the heat input per unit length during the current sampling period. and These are the lower and upper limits of allowable heat input, respectively, and weld formation constraints (weld formation constraints are used to limit the weld formation state to meet target requirements; when weld width is used as the forming control index, it can be expressed as...). ,in The width of the weld obtained from the current inspection. and These are the lower and upper limits of the allowable weld width, respectively, and equipment operation constraints (equipment operation constraints are used to limit welding current, welding voltage, welding speed, oscillation frequency, and oscillation amplitude to within the allowable range of the equipment, and can be expressed as follows). , ,in, These are the allowable value boundaries for the corresponding process parameters, used to limit the feasible range of subsequent parameter linkage corrections.

[0060] Step S3: Generate initial process parameter combinations: Combining the process database, rule base, and empirical parameter templates, determine the initial process parameter combinations through parameter matching and rule filtering. , The initial welding current, The initial welding voltage, The initial welding speed, The initial oscillation frequency, This represents the initial oscillation amplitude. The process database stores the range of basic process parameters corresponding to different component types, material types, plate thicknesses, joint types, and welding positions. The rule base describes the matching relationships and constraint logic between various process parameters. The experience parameter template provides reference parameter combinations for typical operating conditions. The initial process parameter combination is obtained by matching the current component operating condition information with the parameter ranges in the process database, and then filtering and correcting it under rule constraints, ensuring that the initial process parameter combination matches the component type, material type, plate thickness, joint type, and welding position.

[0061] Step S4: Acquire welding process status information: During the welding process, collect one or more of the following in real time: arc status, current and voltage signals, wire feeding status, running speed, oscillation status, weld formation status, and temperature-related information, to form a process status information set for the current sampling period. .

[0062] Step S5: Extract process state characteristic quantities and calculate deviation and risk quantities: Based on the process state information described in Step S4, extract process state characteristic quantities such as arc length fluctuation, heat input deviation, speed deviation, wire feeding matching deviation, and weld formation deviation, and further calculate the comprehensive quality risk index. Specifically, this includes:

[0063] Arc length fluctuation By analyzing the arc length signals of n original arc length sampling points within the current sampling period The standard deviation was calculated to obtain:

[0064]

[0065] in, It is the arithmetic mean of the arc length signal within the sampling period;

[0066] speed deviation According to the formula The calculation determines that, At the current welding speed, Target welding speed;

[0067] Wire feeding matching deviation By calculating the actual wire feed speed Compared with the current current Ideal wire feeding mapping function The residual between them is obtained by the formula:

[0068]

[0069] in, The current-to-wire-speed matching empirical curve function is preset;

[0070] The thermal input per unit length in the current sampling period can be expressed as: Where η is the thermal efficiency coefficient, U t I t v t These represent the current welding voltage, welding current, and welding speed, respectively. Based on the target heat input Q... ref The thermal input deviation is defined as: ;

[0071] When weld width is used as the forming control index, the weld forming deviation is defined as: ;in, The width of the weld obtained from the current inspection. The target weld width.

[0072] To avoid the problem that a single deviation measure can only reflect local anomalies and is difficult to characterize the overall welding quality status under conditions of multiple concurrent disturbances, this embodiment constructs a comprehensive quality risk index based on multiple process state characteristic quantities. This method is used to uniformly quantify the overall risk level caused by abnormal heat input, weld formation deviation, speed fluctuation, wire feed mismatch, and arc length fluctuation within the current sampling period. It should be noted that the heat input deviation, weld formation deviation, speed deviation, wire feed mismatch deviation, and arc length fluctuation differ in physical meaning, dimensional form, and range of variation. Directly superimposing these values ​​could lead to a certain type of feature with a larger amplitude unreasonably dominating the comprehensive evaluation, thus affecting the stability and interpretability of subsequent control decisions. Therefore, this embodiment first performs dimensionless processing according to the reference thresholds corresponding to each feature, converting abnormal quantities from different sources into comparable risk contributions. Then, based on the different degrees of influence of each feature on welding quality, a risk fusion weight is introduced for weighted summation to form a unified risk assessment result. Specifically, based on the heat input deviation... Weld forming deviation Speed ​​deviation Wire feeding matching degree deviation and arc length fluctuation Construct a comprehensive quality risk index :

[0073]

[0074] in, , , , and These are reference thresholds corresponding to heat input deviation, weld formation deviation, speed deviation, wire feed matching deviation, and arc length fluctuation. These reference thresholds are determined based on typical process parameter ranges for the corresponding material type, plate thickness, and joint type in the process database, or based on historical experimental data statistics, and are used to characterize the normal fluctuation range of each feature. λ1~λ5 are the risk fusion weights corresponding to each feature item, satisfying λ1+λ2+λ3+λ4+λ5=1, and can be determined based on offline process experiments or historical quality data statistics. Through the above processing, a comprehensive quality risk index is generated. It is no longer limited to a local description of a single anomaly, but can uniformly reflect the overall instability trend, the decline in penetration capability, and the deterioration trend of weld formation quality in the current sampling period of the welding process, providing a unified risk measurement basis for subsequent determination of the dominant state and multi-parameter coordinated linkage correction.

[0075] Step S6: Perform multi-parameter coordinated linkage correction: This embodiment constructs a coordinated linkage model under the dominant state that combines deviation driving and cross compensation, so as to achieve unified scheduling of welding current, voltage, speed and oscillation parameters within the same control cycle.

[0076] (1) Real-time offsetting of associated quality risks caused by parameter changes using cross terms. The original linkage correction vectors for each process parameter within the current sampling period. Determined by the following system of concerted equations:

[0077]

[0078] in, and These represent the original correction values ​​for welding current, welding voltage, welding speed, oscillation frequency, and oscillation amplitude within the current sampling period, respectively. , , , , , , , , , All are linkage adjustment coefficients greater than zero, mainly used to characterize the main channel deviation. The cross-coupling compensation coefficient is used to characterize the correlation compensation strength across channels, so as to achieve deep decoupling between thermal input and forming quality.

[0079] The physical meaning and mapping relationship of the above-mentioned concerted equations are mainly reflected in the following three aspects:

[0080] Firstly, there is active pre-compensation for the heat input channel (to counteract lateral forming interference). When the heat input deviation... When the value is positive, the welding current and welding voltage tend to decrease, while the welding speed tends to increase, causing the heat input to return to the target range; when When it is negative, the opposite is true. Simultaneously, to suppress weld width fluctuations caused by arc pressure changes when adjusting current or speed to correct heat input deviations, this channel synchronously introduces... The determined corresponding cross-compensation terms enable the swing parameters to undergo preliminary reverse compensation while the heat input is adjusted, thus offsetting the lateral forming interference caused by the heat adjustment.

[0081] Second, the energy consistency and coordination of the forming channel (counteracting indirect interference from weld penetration). When the weld forming deviation... When the width deviates from the target, the oscillation frequency and amplitude correction amount mainly tend to be actively adjusted to correct the lateral heat distribution and weld spread morphology. To prevent changes in the lateral heat coverage rhythm from disrupting the stability of heat input per unit length, this channel introduces a coupling term from... The determined heat input deviation feedback enables synchronous, pre-emptive reverse adjustments to the welding current, voltage, or speed, offsetting the indirect interference of the oscillation motion on the melt depth.

[0082] Thirdly, it involves risk-based adaptive adjustment and dynamic performance synergy. Considering the large inertia and hysteresis of the welding process, the cross-coupling coefficient allows the system to utilize other parameter channels to provide auxiliary support when a single indicator shows an abnormal trend. Furthermore, it integrates quality risk indicators... Used to uniformly characterize the overall risk intensity and adjust the correction amplitude of each channel: when When it increases, in the form of or Under the premise that the determined correction direction remains unchanged, through the positive gain factor By increasing the absolute value of the original correction amount of the corresponding parameter channel, the system can adaptively enhance its ability to suppress the overall instability trend and the deterioration trend, thereby improving the dynamic response speed and control stability of the system.

[0083] (2) Through analysis The contribution percentage of each internal risk source is used to construct the severity of heat input anomalies. Severity of weld formation abnormalities and the severity dominated by complex risks This is to distinguish between two different control scenarios: one dominated by local deviations and the other by combined risks. The formula is as follows:

[0084] , ,

[0085] when , and When a certain item reaches its maximum value, its corresponding abnormal state is determined as the dominant state of the current sampling period, and a hot input anomaly correction weight is set based on the dominant state. Weighting of weld formation abnormality correction . Specifically, when At its maximum, it is determined to be a state dominated by abnormal heat input, indicating that the current overall risk is mainly caused by heat input deviation. Caused, at this time =Dominant weight, =Secondary weight. When At its maximum, it is determined to be a state dominated by abnormal weld formation, indicating that the current overall risk is mainly caused by deviations in weld formation (weld width). Caused, at this time =Secondary weights, =Dominant weight. When At its maximum, the condition is determined to be dominated by composite risk, indicating that the current welding process is under a state of enhanced coupling of multiple abnormal factors. To comprehensively enhance the anti-disturbance capability and increase the correction weight of each deviation source, at this point, let... =Dominant weight.

[0086] Among them, dominant weight and secondary weights The following nonlinear mapping function is used for real-time calculation, and satisfies...

[0087]

[0088]

[0089] in, This represents the severity value of the dominant state within the current sampling period; The dominant weight benchmark value has a range of values. The sensitivity gain coefficient is preset based on the sensitivity of each parameter channel to the welding quality. This is a secondary adjustment factor, with a value range of [value range missing]. 3. When the dominant abnormality severity Or comprehensive quality risk indicators When it increases, the dominant weight Follow The function increases smoothly, achieving dynamic focusing of the correction force on the dominant channel.

[0090] Will and Substitute the original linkage correction values ​​for each process parameter In the process, the parameter correction vector after hierarchical weighting is obtained. The system of equations for calculation is as follows:

[0091]

[0092] in, , , , , These represent the original correction values ​​of welding current, welding voltage, welding speed, oscillation frequency, and oscillation amplitude after layered weighting within the current sampling period.

[0093] (3) After obtaining the parameter correction amount after layered weighting, directly applying it to the process parameter combination may cause sudden changes in welding power output or drastic fluctuations in robot operation. Therefore, a single-cycle amplitude constraint is applied to the parameter correction amount after layered weighting to limit the maximum allowable change of each parameter in a single sampling period, thus obtaining a parameter correction vector that meets the amplitude constraint. :

[0094]

[0095] in, This is the amplitude limiting function. For any parameter correction... :

[0096]

[0097] in, This represents the maximum permissible correction amplitude for the corresponding process parameter within a single sampling period. After obtaining the parameter correction vector that satisfies the single-period amplitude constraint, to ensure that the updated process parameter combination remains within the equipment's allowable execution range, further equipment feasible region projection processing is performed to obtain the final updated process parameter combination:

[0098]

[0099] in, For the current sampling period t, the process parameter combination For the updated combination of process parameters, (·) is the projection operator projected onto the feasible region Ω of the device, which is jointly determined by the welding power output range, the robot running speed range, and the oscillating mechanism execution range. When When the projection falls within the feasible region Ω of the device, the projection result remains unchanged; when When the parameters exceed the equipment's allowable boundaries, boundary truncation or the projection of the nearest feasible point is used to remap them into the executable region. Through the above-mentioned single-cycle limiting and equipment feasible region projection processing, this embodiment ensures that the parameter linkage correction results are not only logically sound but also feasible at the engineering execution level, thereby forming a multi-parameter collaborative linkage correction relationship that can be directly used for online closed-loop adjustment of automatic welding systems.

[0100] Step S7: Compare preset judgment conditions: Based on the comparison results of the arc stability index, heat input fluctuation index, weld formation deviation index, equipment execution stability index, and penetration depth risk index with their respective preset thresholds, determine whether the updated process parameter combination simultaneously meets the following conditions: preset stability condition, used to characterize that the arc state, heat input state, and welding process fluctuations are within the allowable range; formation consistency condition, used to characterize that the weld width, reinforcement height, or penetration depth level meets the target formation requirements; equipment executability condition, used to characterize that the updated welding current, welding voltage, welding speed, oscillation frequency, and oscillation amplitude do not exceed the equipment's allowable execution range; when all judgment conditions are met, the process parameter combination is converted into control instructions executable by the welding robot to drive the welding execution unit to complete the welding operation; when any judgment condition is not met, return to step S4 to continue the process status acquisition, deviation evaluation, and linkage correction for the next sampling cycle.

[0101] Furthermore, steps S4-S7 are executed cyclically during the welding process according to the sampling cycle, so that the process status information collection, deviation and risk assessment, multi-parameter linkage correction and threshold judgment output form a continuous closed loop, thereby forming an online adaptive adjustment mechanism that revolves around the real-time changes in the welding status.

[0102] A multi-parameter collaborative control system for intelligent welding processes, such as Figure 3 As shown, it is composed of a sensing and input module, a process parameter initialization module, a collaborative control decision module, a robot controller, a welding execution unit, a process status evaluation module, a quality and anomaly judgment module, and a process database.

[0103] The sensing and input module is used to acquire the component type, plate thickness, joint type, welding position, equipment capacity parameters and target forming index of the component to be welded;

[0104] The process parameter initialization module is used to establish a set of process parameters and output an initial combination of process parameters based on the process database.

[0105] The collaborative control decision module is used to receive the process status assessment results and, in combination with the dominant status determination results, preset parameter linkage relationships, and various constraints, calculate the parameter correction amount.

[0106] The robot controller is used to convert the modified combination of process parameters into control instructions that can be executed by the welding power source, wire feeding mechanism, oscillating mechanism and robot motion control unit.

[0107] The welding execution unit is used to actually complete the welding operation;

[0108] The process status assessment module is used to collect, extract features, and calculate deviations for the arc status, current and voltage signals, wire feeding status, running speed, oscillation status, and weld formation status during the welding process.

[0109] The quality and anomaly determination module is used to determine whether the current corrected parameter combination meets the stability and consistency requirements based on preset thresholds;

[0110] The process database is used to store reference process parameter templates, constraints, and parameter linkage correction rules corresponding to different component types, material types, plate thickness grades, joint types, and welding positions.

[0111] In this embodiment, the process database is used not only to generate initial process parameter combinations but also to provide reference correction rules for corresponding operating conditions to the collaborative control decision module during online control. After the welding execution unit performs welding within the current sampling period, it feeds back the welding process status information to the process status evaluation module. The process status evaluation module extracts process status features and forms deviation and risk quantities, then sends the results to the collaborative control decision module. The collaborative control decision module generates new parameter correction quantities and outputs them to the robot controller. The quality and anomaly judgment module judges the corrected results. When the preset requirements are not met, the closed loop returns to the process status evaluation and parameter linkage correction stage to continue execution, thus forming a closed-loop control architecture of database support - online perception - status evaluation - linkage decision - execution feedback - threshold judgment.

[0112] The embodiments described above are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Any obvious improvements, substitutions or modifications that can be made by those skilled in the art without departing from the essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A multi-parameter collaborative control method for an intelligent welding process, characterized in that, Includes the following steps: Step S1: Obtain the working condition information of the component to be welded: Collect the component type, material type, plate thickness, joint type, welding position, equipment capacity parameters and target forming index of the component to be welded; Step S2: Establish the set of process parameters and constraints: Based on the working condition information of the component to be welded in step S1, establish the process parameter vector. In addition to heat input constraints, weld formation constraints, and equipment operation constraints; Step S3: Generate initial process parameter combinations: Combining the process database, rule base, and empirical parameter templates, determine the initial process parameter combinations through parameter matching and rule filtering. ; Step S4: Collect welding process status information: During the welding process, collect the arc status, current and voltage signals, wire feeding status, running speed, oscillation status, weld formation status and temperature in real time; Step S5: Extract process state characteristic quantities and calculate deviation and risk quantities: Based on the process state information described in step S4, extract the arc length fluctuation quantity. Heat input deviation Speed ​​deviation Wire feeding matching degree deviation and weld formation deviation And further calculate the comprehensive quality risk index. ; Step S6: Perform multi-parameter collaborative correction: First, construct the original collaborative correction amount based on the heat input deviation, weld formation deviation, and comprehensive quality risk index. ; Secondly, considering the contribution ratio and dominant weight of each risk source. and secondary weights The original linkage correction values ​​are then weighted hierarchically to obtain the hierarchically weighted parameter correction vector. Finally, a single-cycle amplitude constraint and equipment feasible region projection processing are applied to the hierarchically weighted parameter corrections to obtain the final updated combination of process parameters. ; Step S7: Compare preset judgment conditions: Judge the preset stability conditions, forming consistency conditions, and equipment executability conditions of the updated process parameter combination. When all judgment conditions are met, drive the welding execution unit to complete the welding operation based on the process parameter combination; when any judgment condition is not met, repeat steps S4 to S7. The comprehensive quality risk indicators The formula is expressed as: ； in, , , , and These are the reference thresholds for heat input deviation, weld formation deviation, speed deviation, wire feeding matching degree deviation, and arc length fluctuation, respectively, and λ1~λ5 are the risk fusion weights corresponding to each feature item. The original linkage correction amount The formula is expressed as: ； in, and These represent the original correction values ​​for welding current, welding voltage, welding speed, oscillation frequency, and oscillation amplitude within the current sampling period, respectively. , , , , , , , , , All are linkage adjustment coefficients; This is the cross-coupling compensation coefficient.

2. The multi-parameter collaborative control method for intelligent welding process according to claim 1, characterized in that, The specific process of performing hierarchical weighting on the original linkage correction amount is as follows: First, through analysis The contribution percentage of each internal risk source is used to construct the severity of heat input anomalies. Severity of weld formation abnormalities and the severity dominated by complex risks Secondly, based on , and The severity level with the highest value is selected through comparison. The abnormal state corresponding to the highest severity level is determined as the dominant state of the current sampling period, and the hot input anomaly correction weight is set according to the dominant state. Weighting of weld formation abnormality correction Finally, the original linkage correction amount is optimized based on the dominant state and correction weight. The parameter correction vector after hierarchical weighting is obtained. .

3. The multi-parameter collaborative control method for intelligent welding process according to claim 2, characterized in that, The severity of the heat input anomaly Severity of weld formation abnormalities and the severity dominated by complex risks The formula is expressed as: ， ， ； in, , , , and These are the reference thresholds for heat input deviation, weld formation deviation, speed deviation, wire feeding matching degree deviation, and arc length fluctuation, respectively, and λ1~λ5 are the risk fusion weights corresponding to each feature item.

4. The multi-parameter collaborative control method for intelligent welding process according to claim 3, characterized in that, The determination of the dominant state is specifically as follows: when At its maximum, it is determined to be a state dominated by abnormal heat input. At this time, let =Dominant weight, =Secondary weight; when At its maximum, it is determined to be the dominant state of abnormal weld formation, at which point... =Secondary weights, =Dominant weight; when At its maximum, it is determined to be in a state dominated by composite risk, at which point let =Dominant weight.

5. The multi-parameter collaborative control method for intelligent welding process according to claim 4, characterized in that, The dominant weight and secondary weights The following nonlinear mapping function is used for real-time calculation: ； ； in, This represents the severity value of the dominant state within the current sampling period; The dominant weight benchmark value This is the sensitivity gain coefficient; This is a secondary correction and regulation factor.

6. The multi-parameter collaborative control method for intelligent welding process according to claim 5, characterized in that, The parameter correction vector after hierarchical weighting The system of equations for calculation is as follows: ； in, , , , , These represent the original correction values ​​of welding current, welding voltage, welding speed, oscillation frequency, and oscillation amplitude after layered weighting within the current sampling period.

7. The multi-parameter collaborative control method for intelligent welding process according to claim 6, characterized in that, The final updated combination of process parameters The formula is expressed as: ； ； in, For the current sampling period t, the process parameter combination, I t For welding current, U t For welding voltage, For welding speed, The oscillation frequency, The amplitude of the swing. (·) is the projection operator projected onto the device feasible region Ω. To satisfy the amplitude limit parameter correction vector, This is the amplitude limiting function.

8. A multi-parameter collaborative control system for an intelligent welding process, used to execute the multi-parameter collaborative control method for an intelligent welding process as described in claims 1-7, characterized in that, include: The sensing and input module is used to acquire the component type, plate thickness, joint type, welding position, equipment capacity parameters and target forming index of the component to be welded; The process parameter initialization module is used to establish a set of process parameters and output an initial combination of process parameters based on the process database. The collaborative control decision module is used to receive the process status assessment results and, in combination with the dominant status judgment results, preset parameter linkage relationships, and various constraints, calculate the parameter correction amount. The robot controller is used to convert the modified combination of process parameters into control instructions that can be executed by the welding power source, wire feeding mechanism, oscillating mechanism and robot motion control unit. Welding execution unit, used to actually complete welding operations; The process status assessment module is used to collect, extract features, and calculate deviations for the arc status, current and voltage signals, wire feeding status, running speed, oscillation status, and weld formation status during the welding process. The quality and anomaly determination module is used to determine whether the current corrected parameter combination meets the stability and consistency requirements based on a preset threshold. The process database stores reference process parameter templates, constraints, and parameter linkage correction rules for different component types, material types, plate thickness grades, joint types, and welding positions.

Images