CN-122017013-A - Method, device and equipment for rapidly analyzing processing defects of remanufactured parts

CN122017013ACN 122017013 ACN122017013 ACN 122017013ACN-122017013-A

Abstract

The invention relates to the field of acoustic nondestructive testing, and provides a method, a device and equipment for rapidly analyzing processing defects of a remanufactured part, wherein the method comprises the steps of establishing a finite element simulation model of the part to be analyzed; the method comprises the steps of dividing a finite element simulation model into a finite element grid and an infinite element grid, applying a preset ultrasonic excitation signal to an excitation point, calculating the propagation process of the ultrasonic excitation signal in a part to be analyzed through the finite element simulation model, collecting ultrasonic echo signals at a signal collecting point, extracting a plurality of characteristic parameters related to a defect state from the collected ultrasonic echo signals, inputting the extracted characteristic parameters into a pre-trained deep learning prediction model, and obtaining a residual service duration prediction result of the part to be analyzed. The invention solves the problem of insufficient ultrasonic detection accuracy in the prior art, and realizes the quick and accurate prediction of the residual service length of the remanufactured part.

Inventors

SUN YEWANG
FU HUIWEN
YANG LUWEI
ZHANG GUILAN

Assignees

中国科学院理化技术研究所

Dates

Publication Date: 20260512
Application Date: 20260106

Claims (10)

1. A rapid analysis method for processing defects of a remanufactured part is characterized by comprising the following steps: Establishing a finite element simulation model of a part to be analyzed, wherein the simulation model comprises an excitation point, a signal acquisition point, an intermediate grid part used for representing an intermediate solid area of the part and a boundary grid part surrounding the intermediate grid part; performing grid division on the finite element simulation model, wherein the middle grid part is divided into finite element grids, and the boundary grid part is divided into infinite element grids; Applying a preset ultrasonic excitation signal to the excitation point; calculating the propagation process of the ultrasonic excitation signal in the part to be analyzed through the finite element simulation model, and collecting an ultrasonic echo signal at the signal collecting point; extracting a plurality of characteristic parameters related to the defect state from the acquired ultrasonic echo signals; And inputting the extracted characteristic parameters into a pre-trained deep learning prediction model to obtain a residual service duration prediction result of the part to be analyzed, wherein the deep learning prediction model is a neural network model fused with a time sequence convolutional network TCN module and a transducer module.
2. The method for rapid analysis of manufacturing defects of remanufactured parts according to claim 1, wherein the establishing a finite element simulation model of the parts to be analyzed comprises: Presetting a crack model in the central area of the geometric model of the part to be analyzed; Dividing the geometric model into areas on two sides of the crack model to obtain a middle area containing the crack model and boundary areas on two sides of the middle area; And respectively endowing the intermediate region and the boundary region with material properties, and setting excitation points and signal acquisition points on the intermediate region to obtain the constructed finite element simulation model.
3. The method for rapid analysis of manufacturing defects of remanufactured parts of claim 1, wherein meshing the finite element simulation model comprises: Performing finite element meshing on the intermediate mesh component with a global dimension of no more than one quarter of an ultrasonic wavelength; Carrying out local seed arrangement on the boundary grid parts by adopting grid sizes larger than the global size, and then carrying out first grid division; performing second grid division on the boundary grid component based on a sweeping mode to generate a quadrilateral leading grid with a sweeping direction pointing to the outside of the model; And configuring the grid cell type of the boundary grid component obtained after the second grid division into two-dimensional acoustic infinite element cells.
4. A remanufactured part tooling defect rapid analysis method according to claim 3 wherein the grid cell type of the boundary grid member obtained after the second meshing is configured as a two-dimensional acoustic infinite element cell comprising: in finite element analysis software, preliminarily defining the boundary grid part as a two-dimensional acoustic quadrilateral unit and generating a corresponding input file; editing the input file, and locating a grid cell definition portion associated with the boundary grid member; modifying the grid cell type identification in the grid cell definition part into a preset code for representing two-dimensional acoustic infinite elements; And calculating based on the preset codes of the two-dimensional acoustic infinite elements, so that the boundary grid component takes effect as a two-dimensional acoustic infinite element unit in simulation.
5. The method for rapid analysis of a manufactured part machining defect of claim 1, wherein the acquiring ultrasonic echo signals at the signal acquisition point comprises: Applying the ultrasonic excitation signal in the form of a concentrated force at the excitation point; setting an explicit dynamics analysis step, and configuring a tiny time increment step which is fixed and smaller than the excitation signal period; defining displacement data sequences of the signal acquisition points at intervals of the small time increment steps in the course output; after finite element calculation is performed, a time domain waveform is extracted from the displacement data sequence and used as an ultrasonic echo signal containing crack reflection and diffraction information.
6. The method for rapid analysis of a remanufactured part tooling defect of claim 1, wherein the extracting a plurality of defect-state related feature parameters from the acquired ultrasonic echo signals comprises: performing time-frequency transformation on the time domain waveform of the ultrasonic echo signal to obtain a corresponding frequency domain signal; Calculating a plurality of statistical features of the time domain signal and the frequency domain signal as energy feature parameters reflecting the energy of the signals; Calculating a nonlinear coefficient generated by the interaction of ultrasonic waves and defects based on harmonic analysis, wherein the nonlinear coefficient is used as a nonlinear characteristic parameter for reflecting material damage; based on a crack propagation theoretical model, calculating waveform distortion parameters between the time domain waveform and a defect-free reference waveform as damage related parameters reflecting the crack propagation degree; And combining the energy characteristic parameters, the nonlinear characteristic parameters and the damage related parameters into a plurality of characteristic parameters related to the defect state.
7. A rapid analysis device for processing defects of a remanufactured part, comprising: The defect integrated modeling module is used for establishing a finite element simulation model of the part to be analyzed, and the simulation model comprises an excitation point, a signal acquisition point, an intermediate grid part used for representing an intermediate solid area of the part and a boundary grid part surrounding the intermediate grid part; the grid division module is used for carrying out grid division on the finite element simulation model, wherein the middle grid part is divided into finite element grids, and the boundary grid part is divided into infinite element grids; The ultrasonic excitation loading module is used for applying a preset ultrasonic excitation signal to the excitation point; the echo acquisition module is used for calculating the propagation process of the ultrasonic excitation signal in the part to be analyzed through the finite element simulation model and acquiring an ultrasonic echo signal at the signal acquisition point; The feature extraction module is used for extracting a plurality of feature parameters related to the defect state from the acquired ultrasonic echo signals; The life prediction module is used for inputting the extracted characteristic parameters into a pre-trained deep learning prediction model to obtain a residual service duration prediction result of the part to be analyzed, wherein the deep learning prediction model is a neural network model fused with a time sequence convolutional network TCN module and a transducer module.
8. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the rapid analysis method of manufacturing part tooling defects of any one of claims 1 to 6 when the computer program is executed by the processor.
9. A non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of rapid analysis of a machining defect of a remanufactured part according to any one of claims 1 to 6.
10. A computer program product comprising a computer program which, when executed by a processor, implements a method for rapid analysis of machining defects of a remanufactured part according to any one of claims 1 to 6.

Description

Method, device and equipment for rapidly analyzing processing defects of remanufactured parts Technical Field The invention relates to the field of acoustic nondestructive testing, in particular to a rapid analysis method, device and equipment for processing defects of remanufactured parts. Background Acoustic non-destructive testing, particularly ultrasonic testing, is a critical means of assessing internal defects and structural integrity of mechanical parts, particularly parts in the remanufacturing field. The method realizes the positioning and quantification of the defects and the prediction of the residual service life of the component by analyzing reflection, diffraction and nonlinear effect signals generated by the interaction of ultrasonic waves with the defects such as cracks and the like when the ultrasonic waves propagate in the material. At present, an ultrasonic detection method based on finite element simulation has become an important tool for supplementing and guiding experimental research. In the prior art, a pure finite element model is generally adopted to simulate the ultrasonic wave propagation process, and a low reflection or absorption boundary condition is set at the outer boundary of the model so as to approximate the propagation of sound waves in an infinite medium, and avoid the interference of boundary reflection waves on defect echo signals. Meanwhile, after the simulated or experimental ultrasonic signals are obtained, an empirical correlation model is generally established according to single or a few acoustic characteristic parameters and the defect size or fatigue life, so as to evaluate and predict. However, in the aspect of simulation modeling, the calculation precision and efficiency of the traditional low-reflection boundary processing method are difficult to be compatible, the integrity of sound wave reflection interference signal acquisition is not easy to be caused due to improper setting, the real dissipation behavior of ultrasonic waves at the boundary of the component cannot be accurately simulated, and the reliability of a simulation result is affected. In the aspects of signal analysis and life prediction, the analysis method which depends on a single or a small number of characteristic parameters cannot fully utilize the rich multidimensional information contained in the ultrasonic echo signals, and the characterization capability of the ultrasonic echo signals on complex defect states is limited, so that the prediction accuracy rate of the residual service life of the parts is insufficient. Disclosure of Invention The invention provides a rapid analysis method, a rapid analysis device and rapid analysis equipment for processing defects of remanufactured parts, which solve the problem of insufficient ultrasonic detection accuracy in the prior art and realize rapid and accurate prediction of the residual service length of the remanufactured parts. The invention provides a rapid analysis method for processing defects of a remanufactured part, which comprises the following steps: Establishing a finite element simulation model of a part to be analyzed, wherein the simulation model comprises an excitation point, a signal acquisition point, an intermediate grid part used for representing an intermediate solid area of the part and a boundary grid part surrounding the intermediate grid part; performing grid division on the finite element simulation model, wherein the middle grid part is divided into finite element grids, and the boundary grid part is divided into infinite element grids; Applying a preset ultrasonic excitation signal to the excitation point; calculating the propagation process of the ultrasonic excitation signal in the part to be analyzed through the finite element simulation model, and collecting an ultrasonic echo signal at the signal collecting point; extracting a plurality of characteristic parameters related to the defect state from the acquired ultrasonic echo signals; And inputting the extracted characteristic parameters into a pre-trained deep learning prediction model to obtain a residual service duration prediction result of the part to be analyzed, wherein the deep learning prediction model is a neural network model fused with a time sequence convolutional network TCN module and a transducer module. The method for rapidly analyzing the processing defects of the remanufactured part comprises the steps of presetting a crack model in a central area of a geometric model of the part to be analyzed, dividing the geometric model into areas on two sides of the crack model to obtain a middle area containing the crack model and boundary areas on two sides of the middle area, respectively endowing the middle area and the boundary areas with material properties, and setting excitation points and signal acquisition points on the middle area to obtain the constructed finite element simulation model. The rapid analysis method for the remanufactured part mac