CN-120791072-B - Quality monitoring method and equipment for surfacing process

CN120791072BCN 120791072 BCN120791072 BCN 120791072BCN-120791072-B

Abstract

The invention relates to the technical field of welding quality detection, in particular to a quality monitoring method and equipment for a surfacing process, wherein the method comprises the steps of acquiring a dynamic parameter set of the surfacing process in real time, wherein the dynamic parameter set comprises arc characteristic parameters, temperature field distribution parameters and mechanical state parameters; setting an edge computing node, accessing a welding quality prediction model, inputting a dynamic parameter set, outputting a quality evaluation result comprising a defect probability and a penetration predicted value, generating a welding parameter adjustment instruction based on the quality evaluation result, adjusting the running states of a welding power supply and a wire feeding mechanism in real time, updating the dynamic parameter set according to the welding parameter adjustment instruction, and transmitting the dynamic parameter set to an industrial control switchboard based on the edge computing node. The invention effectively solves the problem that instantaneous quality fluctuation caused by multi-physical field intensity coupling in the surfacing process is uncontrollable.

Inventors

WANG JIAWEN
ZHANG LIJIAN
YU QUN
WANG ZE
JIA YANXUN

Assignees

中国电建集团长春发电设备有限公司

Dates

Publication Date: 20260508
Application Date: 20250714

Claims (6)

1. A method for monitoring the quality of a build-up welding process, the method comprising: acquiring a dynamic parameter set of a surfacing process in real time, wherein the dynamic parameter set comprises arc characteristic parameters, temperature field distribution parameters and mechanical state parameters; setting an edge computing node, accessing a welding quality prediction model, inputting the dynamic parameter set, and outputting a quality evaluation result comprising defect probability and penetration predicted value; generating a welding parameter adjustment instruction based on the quality evaluation result, and adjusting the running states of a welding power supply and a wire feeding mechanism in real time; updating the dynamic parameter set according to the welding parameter adjustment instruction and transmitting the dynamic parameter set to an industrial personal computer based on the edge computing node; outputting a quality assessment result comprising a defect probability and a penetration predicted value, including: acquiring quality influence characteristics of each parameter in the dynamic parameter set and establishing a real-time interaction influence relation according to the quality influence characteristics; performing cross-domain time sequence association analysis on the real-time interaction influence relationship, and dynamically distributing weight coefficients of the influence of each parameter in the dynamic parameter set on the quality of the surfacing; Synchronously generating the defect probability and the penetration predicted value based on the distribution result of the weight coefficient; performing edge computation according to the edge computation node, including: The space-time alignment processing of the dynamic parameter set is completed locally at the edge computing node, and a real-time analysis window synchronous with the dynamic evolution of the molten pool is established; Executing the dynamic allocation of the weight coefficients based on the real-time analysis window, and simultaneously caching historical weight allocation sequences of at least three continuous windows; when the communication delay of the industrial personal computer exceeds a preset threshold, starting the historical weight distribution sequence to generate the quality evaluation result, and transmitting the quality evaluation result to the industrial personal computer in parallel to be regulated with the welding power supply; Enabling the historical weight distribution sequence to generate the quality assessment result, comprising: Extracting fluctuation characteristics in the historical weight distribution sequence, and identifying weight deviation rules of the arc characteristic parameters and the mechanical state parameters; performing mode similarity matching on the coupling strength change trend of the current real-time analysis window and the weight deviation rule; when the matching degree is lower than a preset tolerance interval, generating a compensation factor based on solidification of the molten pool; and correcting the average value calculation result of the historical weight distribution sequence by adopting the compensation factor to generate the quality evaluation result.
2. The method of claim 1, wherein when a high frequency vibration signature indicative of wire feeder anomalies is detected in the machine state parameter, an equipment failure command is generated and the welding process is interrupted.
3. The method of claim 1, wherein generating a welding parameter adjustment command based on the quality assessment result comprises: analyzing the coupling relation between the defect probability and the penetration predicted value, and generating an arc energy adjusting instruction when the defect probability exceeds a preset value; dynamically calculating a wire feeding speed compensation quantity and generating a mechanical state correction instruction according to the degree of deviation of the penetration predicted value from a process reference; and fusing the arc energy adjusting instruction and the mechanical state correcting instruction to generate a cooperative control instruction set which simultaneously acts on pulse waveform parameters of the welding power supply and a servo motor of the wire feeding mechanism.
4. The method of claim 3, wherein adjusting the operating conditions of the welding power source and the wire feeder in real time comprises: Disassembling the cooperative control instruction set into a welding power supply execution subset and a wire feeding mechanism execution subset; generating pulse waveforms according to the hardware abstraction layer interface by using the welding power supply execution subset, and dynamically adjusting the current rising rate and the pulse frequency to match the molten pool oscillation suppression requirement; performing gradient compensation of wire feeding speed within a preset time window based on the torque ring controller of the subset driving servo motor by the wire feeding mechanism; And synchronously collecting feedback data of the molten pool state after execution, and verifying the cooperative effectiveness of the arc energy adjusting instruction and the mechanical state correcting instruction.
5. The overlay welding process quality monitoring method according to claim 1, wherein a cross-regional welding quality statistical analysis is performed based on the updated dynamic parameter sets transmitted by the plurality of edge computing nodes, and a global quality process optimization strategy is generated and issued to the corresponding edge computing nodes.
6. A quality monitoring device for a build-up welding process for implementing a quality monitoring method for a build-up welding process according to any one of claims 1 to 5.

Description

Quality monitoring method and equipment for surfacing process Technical Field The invention relates to the technical field of welding quality detection, in particular to a quality monitoring method and equipment for a surfacing process. Background The build-up welding is used as an important additive manufacturing technology and is widely applied to the field of surface strengthening and repairing of mechanical parts. The core process is essentially that an arc heat source is used for melting a metal welding wire into molten drops, and a metallurgically bonded surfacing layer is formed on the surface of a base material. The process involves strong coupling of multiple physical fields such as arc physics, molten pool fluid dynamics, solid phase transformation and the like, wherein arc energy fluctuation directly affects the temperature gradient of the molten pool, mechanical vibration of a wire feeding mechanism can interfere with melt drop transition stability, and finally, instantaneous generation of microscopic defects such as penetration mutation, air holes, cracks and the like is shown. The current industry mainstream technical proposal has monitoring hysteresis, the traditional method relies on X-ray or ultrasonic detection after welding, the result can not be obtained after a period of time is finished, the defects caused on a short time scale can not be intercepted, the on-line monitoring only carries out separation analysis on parameters such as arc characteristics, temperature fields, mechanical states and the like, in fact, the fluctuation of wire feeding speed can change a molten pool flow field through the impact force of molten drops, and further the shrinkage effect of arc plasma is influenced, and the multi-parameter chain reaction is not modeled. The defects are commonly pointed to that the quality fluctuation under the transient coupling action of multiple physical fields is not controllable. This directly results in high-end equipment core component build-up welding qualification rate that cannot be improved for a long time, and repair costs are high. The information disclosed in this background section is only for enhancement of understanding of the general background of the disclosure and is not to be taken as an admission or any form of suggestion that this information forms the prior art that is well known to a person skilled in the art. Disclosure of Invention The invention provides a quality monitoring method and equipment for a surfacing process, which can effectively solve the problems in the background technology. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: a quality monitoring method of a build-up welding process comprises the following steps: Acquiring a dynamic parameter set of a surfacing process in real time, wherein the dynamic parameter set comprises arc characteristic parameters, temperature field distribution parameters and mechanical state parameters; Setting edge computing nodes, accessing a welding quality prediction model, inputting a dynamic parameter set, and outputting a quality evaluation result comprising defect probability and fusion depth predicted values; generating a welding parameter adjustment instruction based on a quality evaluation result, and adjusting the running states of a welding power supply and a wire feeding mechanism in real time; And updating the dynamic parameter set according to the welding parameter adjustment instruction and transmitting the dynamic parameter set to the industrial personal computer based on the edge computing node. Further, outputting a quality evaluation result including the defect probability and the penetration prediction value, including: acquiring quality influence characteristics of each parameter in the dynamic parameter set and establishing a real-time interaction influence relation according to the quality influence characteristics; performing cross-domain time sequence association analysis on the real-time interaction influence relationship, and dynamically distributing weight coefficients of the influence of each parameter in the dynamic parameter set on the quality of the surfacing; And synchronously generating the defect probability and the penetration predicted value based on the distribution result of the weight coefficient. Further, when a dither characteristic indicative of wire feeder anomalies is detected in the machine state parameter, an equipment failure command is generated and the welding process is interrupted. Further, performing edge computation to obtain quality results includes: The space-time alignment processing of the dynamic parameter set is completed locally at the edge computing node, and a real-time analysis window synchronous with the dynamic evolution of the molten pool is established; Executing the dynamic allocation of the weight coefficients based on the real-time analysis window, and simultaneously caching historical weight all