CN-121972759-A - Online monitoring and real-time control method and system for arc additive manufacturing

CN121972759ACN 121972759 ACN121972759 ACN 121972759ACN-121972759-A

Abstract

The invention provides an online monitoring and real-time control method and system for arc additive manufacturing, and relates to the technical field of monitoring and control of metal additive manufacturing processes. The invention discloses a multi-mode sensor in the manufacturing process, which synchronously collects arc voltage, current, molten pool morphology, temperature field and forming layer geometric characteristic data and forms an original multi-mode data set under a unified time reference. The characteristics of energy input, molten pool morphology, temperature distribution, geometric deviation and the like are extracted through signal processing and image preprocessing, a forming state vector is generated through fusion, and the forming state vector is matched with a preset process model to obtain the deviation. And on-line adjustment amounts of welding current, wire feeding speed, moving track and layer thickness are calculated by adopting a self-adaptive control algorithm based on the deviation amount, and are converted into control instructions to act on an executing mechanism, so that on-line closed-loop control of a forming process is realized, and forming stability, consistency and geometric precision are improved.

Inventors

ZHANG TIMING
CHEN YUHUA
Deng Yunfa
WANG SHANLIN
LONG WEIMIN
HUANG KE
WANG SUYU
XU YUXIN

Assignees

江西昌景航空制造有限公司
南昌航空大学

Dates

Publication Date: 20260505
Application Date: 20260206

Claims (10)

1. An arc additive manufacturing on-line monitoring and real-time control method is characterized by comprising the following steps: Arranging a multi-mode sensor in the arc additive manufacturing process, synchronously acquiring arc voltage data, arc current data, molten pool morphology data, temperature field data and forming layer geometric characteristic data, and writing a time stamp with a unified time reference into the data acquired by the multi-mode sensor to form an original multi-mode data set; performing signal processing and image preprocessing on the original multi-mode data set to obtain a preprocessed data set for feature calculation; Extracting a feature set representing process parameters and forming states based on the preprocessing data set, wherein the feature set at least comprises energy input features obtained by the arc voltage data and the arc current data, molten pool morphology features obtained by the molten pool morphology data, temperature distribution features obtained by the temperature field data and geometric deviation features obtained by the forming layer geometric feature data; Performing feature alignment and fusion on the feature set to generate a forming state vector, and matching the forming state vector with a target state corresponding to a preset process model to obtain a deviation amount for control; Calculating on-line adjustment quantity by adopting a self-adaptive control algorithm based on the deviation quantity, wherein the on-line adjustment quantity comprises welding current adjustment quantity, wire feeding speed adjustment quantity, moving track adjustment quantity and layer thickness adjustment quantity; And converting the online adjustment quantity into a control instruction and sending the control instruction to an executing mechanism, so that the executing mechanism correspondingly adjusts the welding current, the wire feeding speed, the moving track and the layer thickness online.
2. The method according to claim 1, wherein the multi-mode sensor comprises an arc voltage sensor, an arc current sensor, a bath topography sensor, a temperature field sensor and a shaping layer geometry sensor, wherein the arc voltage sensor is used for acquiring the arc voltage data, the arc current sensor is used for acquiring the arc current data, the bath topography sensor is used for acquiring the bath topography data, the temperature field sensor is used for acquiring the temperature field data, and the shaping layer geometry sensor is used for acquiring the shaping layer geometry data.
3. The arc additive manufacturing online monitoring and real-time control method of claim 1, wherein writing a time stamp of a uniform time reference for data collected by the multi-modal sensor to form an original multi-modal dataset comprises: acquiring a clock signal and determining the clock signal as the unified time reference; Triggering the multi-mode sensor to complete one-time synchronous acquisition at the same acquisition time according to a preset synchronous triggering mechanism to obtain the arc voltage data, the arc current data, the molten pool morphology data, the temperature field data and the forming layer geometric characteristic data corresponding to the same acquisition time; And respectively writing the time stamp corresponding to the acquisition time under the unified time reference into the arc voltage data, the arc current data, the molten pool morphology data, the temperature field data and the forming layer geometric characteristic data, and collecting all the data after the time stamp is written into the original multi-mode data set.
4. The method of claim 1, wherein performing signal processing and image preprocessing on the raw multi-modal dataset to obtain a preprocessed dataset for feature computation comprises: Performing outlier removal processing and filtering processing on the arc voltage data and the arc current data to obtain arc voltage preprocessing data and arc current preprocessing data; Carrying out denoising treatment and molten pool region extraction treatment on the molten pool morphology data to obtain molten pool morphology pretreatment data; Denoising and temperature calibration are carried out on the temperature field data to obtain temperature field preprocessing data; Denoising and geometric calibration are carried out on the geometric feature data of the forming layer, so that geometric feature preprocessing data are obtained; And collecting the arc voltage pretreatment data, the arc current pretreatment data, the molten pool morphology pretreatment data, the temperature field pretreatment data and the geometric feature pretreatment data to obtain the pretreatment data set.
5. The arc additive manufacturing on-line monitoring and real-time control method according to claim 1, the method is characterized in that the method for determining the energy input characteristics comprises the following steps: calculating an instantaneous power sequence based on the arc voltage data and the arc current data corresponding to the same time stamp to obtain an instantaneous power characteristic; And integrating the instantaneous power sequence in a preset time window corresponding to the time stamp to obtain the energy input characteristic.
6. The arc additive manufacturing online monitoring and real-time control method of claim 4, wherein extracting a feature set characterizing process parameters and formation conditions based on the pre-processed dataset comprises: extracting a molten pool width characteristic, a molten pool length characteristic and a molten pool area characteristic based on the molten pool morphology pretreatment data, and taking the molten pool width characteristic, the molten pool length characteristic and the molten pool area characteristic as the molten pool morphology characteristic; Extracting a highest temperature characteristic and a temperature gradient characteristic based on the temperature field preprocessing data as the temperature distribution characteristic; Extracting a measured layer thickness and a measured forming width based on the geometric feature pretreatment data, and determining a target layer thickness and a target forming width based on a target state corresponding to the preset process model; And obtaining a layer thickness deviation feature based on the difference between the measured layer thickness and the target layer thickness, and obtaining a forming width deviation feature based on the difference between the measured forming width and the target forming width, wherein the layer thickness deviation feature and the forming width deviation feature are used as the geometric deviation feature.
7. The arc additive manufacturing online monitoring and real-time control method of claim 1, wherein performing feature alignment and fusion on the feature set to generate a forming state vector comprises: performing time alignment on each feature in the feature set according to the time stamp to obtain a synchronous feature vector corresponding to the same acquisition time; performing normalization processing on the synchronous feature vector to obtain a normalized synchronous feature vector; And carrying out weighted fusion on the normalized synchronous feature vector based on a preset fusion weight, and outputting the forming state vector, wherein the preset fusion weight is a weight coefficient corresponding to each feature in the weighted fusion.
8. The method of claim 1, wherein matching the forming state vector with a target state corresponding to a predetermined process model to obtain a deviation amount for control comprises: determining a target state vector corresponding to the current timestamp based on the preset process model, wherein the target state vector is a vectorized representation of the target state; Performing distance measurement calculation on the forming state vector and the target state vector to obtain a deviation vector; And determining the deviation amount according to the deviation vector, wherein the deviation amount comprises an energy input deviation amount, a molten pool morphology deviation amount, a temperature distribution deviation amount and a geometric deviation amount.
9. The method of claim 1, wherein calculating an on-line adjustment using an adaptive control algorithm based on the deviation amount comprises: Establishing a control mapping relation between the deviation amount and the welding current adjustment amount, the wire feeding speed adjustment amount, the moving track adjustment amount and the layer thickness adjustment amount based on the deviation amount; Performing self-adaptive updating on control parameters in the control mapping relation according to the deviation amount, and outputting the online adjustment amount; And applying a preset amplitude constraint and a preset change rate constraint to the online adjustment quantity to obtain the online adjustment quantity meeting the constraint.
10. An arc additive manufacturing on-line monitoring and real-time control system, comprising: the multi-mode synchronous acquisition and timestamp calibration unit is used for arranging a multi-mode sensor in the arc additive manufacturing process, synchronously acquiring arc voltage data, arc current data, molten pool morphology data, temperature field data and forming layer geometric characteristic data, and writing a timestamp with a unified time reference into the data acquired by the multi-mode sensor to form an original multi-mode data set; The signal processing and image preprocessing unit is used for performing signal processing and image preprocessing on the original multi-mode data set to obtain a preprocessed data set for feature calculation; A process-state feature extraction unit for extracting a feature set characterizing process parameters and a forming state based on the preprocessing dataset, wherein the feature set at least comprises an energy input feature obtained from the arc voltage data and the arc current data, a molten pool feature obtained from the molten pool feature data, a temperature distribution feature obtained from the temperature field data, and a geometric deviation feature obtained from the forming layer geometric feature data; The feature alignment fusion and state deviation calculation unit is used for executing feature alignment and fusion on the feature set to generate a forming state vector, and matching the forming state vector with a target state corresponding to a preset process model to obtain a deviation amount for control; The self-adaptive control on-line adjustment amount calculating unit is used for calculating on-line adjustment amounts based on the deviation amount by adopting a self-adaptive control algorithm, wherein the on-line adjustment amounts comprise welding current adjustment amounts, wire feeding speed adjustment amounts, movement track adjustment amounts and layer thickness adjustment amounts; The control instruction generation and execution mechanism driving unit is used for converting the online adjustment quantity into a control instruction and sending the control instruction to the execution mechanism, so that the execution mechanism correspondingly adjusts the welding current, the wire feeding speed, the moving track and the layer thickness online.

Description

Online monitoring and real-time control method and system for arc additive manufacturing Technical Field The invention relates to the technical field of monitoring and controlling of metal additive manufacturing processes, in particular to an online monitoring and real-time controlling method and system for arc additive manufacturing. Background The existing arc additive manufacturing generally adopts preset process parameters to carry out forming control, the process monitoring is stopped at a single signal or manual observation level, for example, only arc voltage and current are monitored or only molten pool images are observed, and it is difficult to synchronously describe arc energy input, molten pool morphology evolution, temperature field distribution and forming layer geometric deviation under the same time reference, so that forming state fluctuation and defect aura are difficult to identify in time. Along with the improvement of the requirements of complex metal components on geometric precision and consistency, the development trend of arc additive manufacturing is to introduce vision, thermal imaging, electric signals and other multi-source sensing, combine signal processing and image recognition to realize online quality evaluation, and further use monitoring results for real-time closed-loop control so as to dynamically adjust welding current, wire feeding speed and movement path in the forming process, thereby realizing stable deposition and controllable size. However, the existing scheme generally has the problems of asynchronous time of multi-source data, lack of uniform state characterization for feature extraction and fusion and disconnection of monitoring and control, so that control quantity is difficult to generate in real time according to a forming state, parameter adjustment is delayed or excessive, layer height accumulated errors, unstable molten pool, forming defects and consistency are easily caused, and stable forming and efficiency improvement of complex components are restricted. Disclosure of Invention In order to overcome the defects of the prior art, the invention aims to provide an arc additive manufacturing on-line monitoring and real-time control method and system, which realize quantitative characterization and on-line closed-loop control of the forming state of the arc additive manufacturing based on multi-mode data fusion and improve the stability, consistency and geometric precision of the forming process. In order to achieve the above object, the present invention provides the following solutions: an arc additive manufacturing on-line monitoring and real-time control method comprises the following steps: Arranging a multi-mode sensor in the arc additive manufacturing process, synchronously acquiring arc voltage data, arc current data, molten pool morphology data, temperature field data and forming layer geometric characteristic data, and writing a time stamp with a unified time reference into the data acquired by the multi-mode sensor to form an original multi-mode data set; performing signal processing and image preprocessing on the original multi-mode data set to obtain a preprocessed data set for feature calculation; Extracting a feature set representing process parameters and forming states based on the preprocessing data set, wherein the feature set at least comprises energy input features obtained by the arc voltage data and the arc current data, molten pool morphology features obtained by the molten pool morphology data, temperature distribution features obtained by the temperature field data and geometric deviation features obtained by the forming layer geometric feature data; Performing feature alignment and fusion on the feature set to generate a forming state vector, and matching the forming state vector with a target state corresponding to a preset process model to obtain a deviation amount for control; Calculating on-line adjustment quantity by adopting a self-adaptive control algorithm based on the deviation quantity, wherein the on-line adjustment quantity comprises welding current adjustment quantity, wire feeding speed adjustment quantity, moving track adjustment quantity and layer thickness adjustment quantity; And converting the online adjustment quantity into a control instruction and sending the control instruction to an executing mechanism, so that the executing mechanism correspondingly adjusts the welding current, the wire feeding speed, the moving track and the layer thickness online. Preferably, the multi-mode sensor comprises an arc voltage sensor, an arc current sensor, a molten pool morphology sensor, a temperature field sensor and a forming layer geometric feature sensor, wherein the arc voltage sensor is used for collecting arc voltage data, the arc current sensor is used for collecting arc current data, the molten pool morphology sensor is used for collecting molten pool morphology data, the temperature field sensor is used for collec