CN-121979114-A - Thin-wall part processing deformation control method and system based on digital twin technology

CN121979114ACN 121979114 ACN121979114 ACN 121979114ACN-121979114-A

Abstract

The invention discloses a thin-wall part machining deformation control method and system based on a digital twin technology, wherein the method comprises the steps of calculating rigidity characterization quantities of all sub-processes of thin-wall part machining in different machining stages, inputting monitoring data of corresponding machining processes into a first prediction model, predicting deformation quantities of the thin-wall part, triggering early warning if the predicted deformation quantities exceed a preset threshold value and corresponding machining areas are linear milling areas, determining key constraint conditions of a second prediction model after triggering early warning, constructing an acquired parameter combination scheme into a pareto optimal parameter solution set through the second prediction model, selecting an optimal parameter combination scheme from the pareto optimal parameter solution set by adopting a multi-attribute decision method, and adjusting actual machining parameters of the next linear milling areas according to the optimal parameter scheme. The invention can effectively inhibit the processing deformation of the thin-wall part by fusing the digital twin algorithm and the intelligent algorithm, and improves the processing precision and quality.

Inventors

WANG XIAOWEI
CUI ZHENGMIN
GUO YANDONG
YUE CAIXU
LU YAO

Assignees

中国航发哈尔滨东安发动机有限公司
哈尔滨理工大学

Dates

Publication Date: 20260505
Application Date: 20251226

Claims (10)

1. The thin-wall part machining deformation control method based on the digital twin technology is characterized by comprising the following steps of: in a digital twin system which is built in advance and simulates the processing process of the thin-wall part, calculating the rigidity characterization values of each sub-process of the thin-wall part processing in different processing stages through finite element simulation based on a theoretical geometric model; inputting the rigidity characterization values of each sub-process of the thin-wall part processing at different processing stages and corresponding processing process monitoring data into a first prediction model according to the processing sequence to predict the deformation of the thin-wall part; after triggering early warning, determining key constraint conditions of a second prediction model according to the magnitude relation between the stiffness characterization quantity of the next sub-process and the stiffness characterization quantity of the current sub-process, optimizing all machining parameters of the thin-wall part by utilizing a multi-objective optimization algorithm through the second prediction model, and constructing an acquired parameter combination scheme as a pareto optimal parameter solution set; And selecting an optimal parameter combination scheme from the pareto solution set by adopting a multi-attribute decision method, and adjusting actual processing parameters of a next linear milling area according to the optimal parameter scheme.
2. The method for controlling the processing deformation of the thin-walled workpiece based on the digital twin technology according to claim 1 is characterized in that the stiffness characteristic quantity is characteristic stiffness, and the method for acquiring the stiffness characteristic quantity comprises the steps of applying a unit force to a theoretical maximum deformation position of a sub-process processing area and calculating the characteristic stiffness of the position according to the measured deformation quantity.
3. The thin-walled workpiece processing deformation control method based on the digital twin technology according to claim 1 is characterized in that the processing process monitoring data comprise processed milling force signals, the processing method of the milling force signals comprises the steps of performing multi-scale decomposition on the milling force signals by adopting a multi-component modal decomposition algorithm to obtain modal components, screening principal components from the modal components by adopting a correlation analysis method based on deformation correlation degrees of the modal components, and extracting time domain, frequency domain and time domain characteristics of the principal components.
4. The method for controlling the processing deformation of the thin-walled workpiece based on the digital twin technology according to claim 1 is characterized in that the method for judging whether the current processing area is a linear milling area comprises the steps of generating a dynamic bounding box at a geometric inflection point of the thin-walled workpiece, and judging whether the current processing area is the linear milling area according to the position change of a processing path relative to the bounding box.
5. The thin-walled workpiece processing deformation control method based on the digital twin technology according to claim 1, wherein the first prediction model and the second prediction model are regression models based on an online extreme learning machine, and the weight matrix from an input layer to a hidden layer and the bias vector of the hidden layer are optimized and selected through a dung beetle optimization algorithm.
6. The thin-walled workpiece processing deformation control method based on the digital twin technology according to claim 1, wherein the multi-objective optimization algorithm comprises a third generation non-dominant ordering genetic algorithm, and the optimization objectives comprise processing quality, processing efficiency and processing performance.
7. The thin-walled workpiece processing deformation control method based on the digital twin technology according to claim 1 is characterized in that the multi-attribute decision method comprises the steps of quantifying preference weights of decision makers on optimization targets through a hierarchical analysis method, sorting parameter combination schemes in a pareto optimal parameter solution set by utilizing a superior-inferior solution distance method based on the preference weights, and selecting an optimal parameter combination scheme.
8. The thin-walled workpiece processing deformation control method based on the digital twin technology according to claim 2, wherein the method for determining the key constraint condition of the second prediction model according to the magnitude relation between the stiffness characteristic of the next sub-process and the stiffness characteristic of the current sub-process specifically comprises: comparing the characteristic rigidity of the next sub-process with the characteristic rigidity of the current sub-process; If the characteristic rigidity of the next sub-process is smaller than that of the current sub-process, selecting the characteristic rigidity of the next sub-process as a rigidity characterization quantity for optimization; otherwise, the characteristic rigidity of the current sub-process is selected as a rigidity characterization quantity for optimization.
9. A thin-wall part processing deformation control system based on digital twin technology is characterized by comprising: the rigidity characterization quantity calculating module is used for calculating rigidity characterization quantities of all sub-processes of thin-wall piece processing at different processing stages through finite element simulation based on a theoretical geometric model in a digital twin system which is built in advance and simulates the thin-wall piece processing process; The deformation prediction and early warning module is used for inputting the rigidity characterization values of all sub-processes of the thin-wall part processing at different processing stages and corresponding processing process monitoring data into the first prediction model to predict the deformation of the thin-wall part according to the processing sequence; The multi-target parameter optimization module is used for determining key constraint conditions of a second prediction model according to the magnitude relation between the stiffness characteristic quantity of the next sub-process and the stiffness characteristic quantity of the current sub-process after triggering and early warning, optimizing all processing parameters of the thin-wall part by utilizing a multi-target optimization algorithm through the second prediction model, and constructing an acquired parameter combination scheme as a pareto optimal parameter solution set; The intelligent decision and feedback control module is used for selecting an optimal parameter combination scheme from the pareto solution set by adopting a multi-attribute decision method, and adjusting actual processing parameters of a next linear milling area according to the optimal parameter scheme.
10. A computer storage medium in which a computer program executable by a processor is stored, the computer program executing the thin-walled workpiece processing deformation control method based on the digital twin technique as set forth in any one of claims 1 to 8.

Description

Thin-wall part processing deformation control method and system based on digital twin technology Technical Field The invention relates to the technical field of mechanical manufacturing, in particular to a thin-wall part processing deformation control method and system based on a digital twin technology. Background With the continuous progress of aerospace technology, the industrial field has put increasingly stringent requirements on the machining precision of parts. The thin-wall structural member is widely applied to the fields of aerospace equipment, automobile industry and the like by virtue of the excellent performance of the thin-wall structural member. At present, the parts are mainly processed through a milling process, but all key factors are mutually coupled, so that remarkable deformation is generated in the processing process, the final precision of the parts is affected, and even the parts are scrapped. Meanwhile, the new generation information technology is rapidly evolved and deeply fused with the manufacturing industry, so that intelligent collaboration and interconnection of the physical domain and the information domain of the manufacturing system are promoted. In the technological innovation background, digital twin has been developed as an innovative technological paradigm. The technology builds a system with high integration of the physical entity and the virtualization model, and realizes symbiotic evolution of the physical entity and the virtualization model through a bidirectional data interaction and collaborative optimization mechanism. In view of the deformation problem in the thin-wall part processing process, only the problem is solved, the stability of the processing process can be effectively improved, and the whole level of the manufacturing industry is promoted. In the field of thin-wall part manufacturing, in particular to an aluminum alloy thin-wall part, the cutting performance is poor, local deformation is very easy to occur in the processing process, the size precision of a workpiece is difficult to control, and the production efficiency and the product qualification rate are seriously affected. More serious, the excessive processing deformation not only can cause the scrapping of parts, but also can cause safety accidents such as instability and falling of workpieces, and the like, thereby forming a great potential safety hazard to the production field. Aiming at the problem of thin-wall component processing deformation, the prior art mainly adopts two research paradigms, namely a mechanism analysis method based on a mechanical principle and a simulation of a processing process by using a finite element numerical simulation technology. Both methods have the problems of high computational complexity and long time consumption, and real-time prediction of processing deformation is difficult to realize, and the characteristic limits the practical application of the methods in online error control. In the aspect of thin-wall piece deformation control, the method for simulating the thin-wall piece deformation by using the external clamp and other external auxiliary mechanisms is high in cost, and the finite element simulation method is long in time consumption and low in satisfaction degree of simulation results. Therefore, most of the deformation control of the thin-wall parts still adopts a mode of changing milling parameters at present, and the method for optimizing the milling parameters has higher reliability. However, aiming at the frame beam thin-wall workpiece, the multi-target optimization research on the milling force which is the most critical to the deformation influence of the frame beam thin-wall workpiece is still rare, and particularly in different milling layers, the influence caused by the rigidity change of the workpiece is not fully considered in the existing research. Disclosure of Invention The invention mainly aims to provide a thin-wall part processing deformation control method and system based on a digital twin technology, which effectively inhibit deformation and improve processing precision and quality through digital twin and intelligent algorithm fusion. The technical scheme adopted by the invention is that the thin-wall part processing deformation control method based on the digital twin technology comprises the following steps: in a digital twin system which is built in advance and simulates the processing process of the thin-wall part, calculating the rigidity characterization values of each sub-process of the thin-wall part processing in different processing stages through finite element simulation based on a theoretical geometric model; inputting the rigidity characterization values of each sub-process of the thin-wall part processing at different processing stages and corresponding processing process monitoring data into a first prediction model according to the processing sequence to predict the deformation of the thin-wall part; after t