CN-121978109-A - Multi-sensor fusion laser shock peening quality monitoring device and method

CN121978109ACN 121978109 ACN121978109 ACN 121978109ACN-121978109-A

Abstract

The invention discloses a multi-sensor fused laser shock strengthening quality monitoring device and method, which relate to the technical field of laser surface strengthening and comprise the following steps that 1, laser energy spectrum characteristics, flow matching curves, acoustic signal characteristic templates and standard surface quality images under different materials and different process conditions are collected, and a process parameter reference database is constructed; step 2, collecting multi-source signals in the processing process, step 3, preprocessing the multi-source signals to generate a multi-dimensional characteristic matrix, and step 4, comprehensively analyzing according to the multi-dimensional characteristic matrix and a process parameter reference database to obtain a quality monitoring result. The invention realizes the omnibearing real-time monitoring of the laser shock strengthening process through the deep integration of the multi-sensor fusion and the intelligent control technology.

Inventors

GUO WEI
SHI JIAXIN
ZHANG HONGQIANG

Assignees

北京航空航天大学

Dates

Publication Date: 20260505
Application Date: 20260121

Claims (10)

1. The multi-sensor fusion laser shock peening quality monitoring method is characterized by comprising the following steps of: Step 1, collecting laser energy spectrum characteristics, flow matching curves, acoustic signal characteristic templates and standard surface quality images of different materials and under different process conditions, setting set values of laser energy, water film flow and acoustic frequency spectrum, and constructing a process parameter reference database by combining the standard surface quality images; step 2, collecting multi-source signals in the processing process; Step3, preprocessing the multi-source signals to generate a multi-dimensional feature matrix; And 4, comprehensively analyzing according to the multidimensional feature matrix and the process parameter reference database to obtain a quality monitoring result.
2. The method of claim 1, wherein the multi-source signals include laser energy signals, flow signals, surface images, and acoustic signals.
3. The multi-sensor fusion laser shock peening quality monitoring method according to claim 1, wherein the preprocessing firstly adopts a layered noise reduction and multi-dimensional standardization strategy to carry out noise reduction and standardization processing to achieve accurate time synchronization and spatial registration and generate a multi-dimensional feature matrix, and the layered noise reduction and multi-dimensional standardization strategy comprises Kalman filtering, moving average filtering, median filtering, gaussian filtering band-pass filtering and normalization processing.
4. The multi-sensor fusion laser shock peening quality monitoring method according to claim 2 is characterized by comprising the steps of calculating machining quality including laser quality, flow quality and noise level according to laser energy signals, flow signals and acoustic signals in a multi-dimensional feature matrix and combining set values of laser energy, water film flow and acoustic frequency spectrum in a process parameter reference database, detecting through Canny operator edge detection and region growing algorithm according to a surface image and a standard surface quality image, identifying a defect region in the surface image, calculating geometric feature parameters of the defect region, calculating workpiece surface quality scores according to the geometric feature parameters, and forming quality monitoring results by the machining quality and the workpiece surface quality scores.
5. The multi-sensor fused laser shock peening quality monitoring method according to claim 4, wherein energy deviation is calculated according to the collected laser energy signal and the laser energy set value, the laser quality is slightly deviated when the energy deviation is smaller than an energy lower threshold, the laser quality is moderately deviated when the energy deviation is within the range of the energy lower threshold and the energy upper threshold, and the laser quality is severely deviated when the energy deviation is larger than the energy upper threshold; calculating flow deviation according to the acquired flow signal and the water film flow set value, wherein when the flow deviation is smaller than a flow lower limit threshold, the flow quality is slightly deviated, when the flow deviation is in the range of the flow lower limit threshold and the flow upper limit threshold, the flow quality is moderately deviated, and when the flow deviation is larger than the flow upper limit threshold, the flow quality is severely deviated; The method comprises the steps of extracting spectral features of an acoustic signal by adopting a multi-domain feature extraction technology, and determining a system state according to the extracted spectral features and a spectral grading judgment standard constructed according to an acoustic spectral set value, wherein the spectral grading judgment standard is that if the similarity between the spectral features and the acoustic spectral set value is larger than a spectral upper threshold, the noise quality is normal, if the similarity between the spectral features and the acoustic spectral set value is within a spectral lower threshold and a spectral upper threshold, the noise quality is early-warning, and if the similarity between the spectral features and the acoustic spectral set value is smaller than a spectral lower threshold, the noise quality is abnormal.
6. The method for monitoring the laser shock peening quality by the multi-sensor fusion according to claim 5 is characterized by further comprising the step of 5 generating a response action according to quality monitoring results in combination with a grading response strategy, wherein the grading response strategy comprises the steps of executing a first-level response and adjusting signals when laser quality and flow quality are slightly deviated and noise level is early warning, executing a second-level response and suspending an alarm and performing manual intervention when the laser quality and the flow quality are moderately deviated and the noise level is early warning, executing a third-level response and performing emergency shutdown when the laser quality and the flow quality are severely deviated and the noise level is abnormal and the surface quality score of a workpiece is lower than a set surface threshold, and feeding the adjusting signals back to a process parameter reference database for recording.
7. The method for monitoring the quality of laser shock peening with multi-sensor fusion according to claim 6, wherein the laser energy signal is adjusted by improved fuzzy PID control, the expression is p_ newtarget =p_current× (1+k1×Δe+k2×Δq+k3×Δs), where Δe is energy deviation, Δq is flow deviation, Δs is acoustic signal deviation, k1=0.5, k2=0.3, k3=0.2, p_newtarget represents the target laser power, and p_current is the current laser power; The flow signal is regulated by using a PID control loop, the expression u (t) =kp×e (t) +ki×e (t) dt+kd×de (t)/dt, u (t) represents the target flow, e (t) represents the flow deviation, kp represents the proportional gain, ki represents the integral gain, and Kd represents the differential gain.
8. The multi-sensor fused laser shock peening quality monitoring device is characterized by comprising a multi-sensor array module and an intelligent control system, wherein the multi-sensor fused laser shock peening quality monitoring method is applied to any one of claims 1-7; The multi-sensor array module is used for collecting multi-source signals; The intelligent control system comprises a database, a signal preprocessing unit and a decision control unit, wherein the database stores a process parameter reference database, the signal preprocessing unit preprocesses a multi-source signal to generate a multi-dimensional feature matrix, and the decision control unit performs comprehensive analysis according to the multi-dimensional feature matrix and the process parameter reference database to obtain a quality monitoring result.
9. The multi-sensor fusion laser shock peening quality monitoring device according to claim 8, wherein the multi-sensor array module comprises a laser energy sensor, a flow rate sensor, a surface quality sensor and an acoustic wave sensor, wherein the laser energy sensor is arranged in a laser and used for collecting laser energy signals of a laser shock peening process, the laser energy sensor comprises laser power and pulse frequency, the flow rate sensor is used for collecting flow signals of a water film of the laser shock peening process, the surface quality sensor is used for collecting surface images of a workpiece, and the acoustic wave sensor is used for collecting acoustic signals of the laser shock peening process.
10. The multi-sensor fusion laser shock peening quality monitoring apparatus according to claim 8, wherein the intelligent control system further comprises an execution output unit for generating a response action according to the quality monitoring result in combination with the hierarchical response strategy, and adjusting the signal execution response by using a PID control algorithm.

Description

Multi-sensor fusion laser shock peening quality monitoring device and method Technical Field The invention relates to the technical field of laser surface strengthening, in particular to a multi-sensor fusion laser impact strengthening quality monitoring device and method. Background The laser shock strengthening technology generates plastic deformation on the surface layer of the material through shock waves induced by high-energy pulse laser, so that the fatigue resistance and the wear resistance of the part are improved. The process has the remarkable effect of multi-parameter coupling, namely, in the dimension of processing parameters, the laser energy directly determines the pressure of an impact wave crest value, the fluctuation of the pressure can lead to the depth deviation of a reinforced layer, the focusing quality and the focus position of a light spot influence the energy space distribution, the small deviation can cause the energy density gradient distortion, the matching relation between the impact frequency and the light spot overlapping rate directly determines the plastic deformation uniformity, and the water film recovery hysteresis of a constraint layer is easily caused by the too high frequency, so that the energy attenuation of the impact wave energy is caused. In the dimension of environmental parameters, the thickness of the water film of the constraint layer needs to be accurately controlled, the plasma shielding effect is easy to be caused by excessive thickness, and the pressure wave propagation loss rate is greatly increased due to excessive thickness. The current commercial laser impact equipment has the common technical bottlenecks of 3 major technical bottlenecks that firstly, quality fluctuation is uncontrollable, laser energy drifting, constrained layer water film fluctuation and other abnormal working conditions such as absorption layer aluminum foil breakage are difficult to capture in time, residual stress distribution is uneven, secondly, a detection means is lagged, manual visual inspection or off-line detection after processing is relied on, a defect transmission chain cannot be blocked, high repair rate is caused, thirdly, the monitoring dimension is single, the existing system is mostly dependent on a single sensor, and the response to key quality characteristics such as acoustic emission signals, surface morphology mutation and the like is absent, so that the misjudgment rate is high when a complex curved surface workpiece is processed. These drawbacks severely limit the large-scale application of laser shock technology in high-end manufacturing fields. Therefore, how to realize comprehensive quality detection of laser shock peening, improve control accuracy, and avoid abnormal response lag is a problem that needs to be solved by those skilled in the art. Disclosure of Invention In view of the above problems, the present invention is directed to a multi-sensor fusion laser shock peening quality monitoring apparatus and method for overcoming or at least partially solving the above problems, to achieve real-time monitoring of a machining process, and to facilitate adaptive quality control. In order to achieve the above purpose, the present invention adopts the following technical scheme: in a first aspect, an embodiment of the present invention provides a method for monitoring quality of laser shock peening with multi-sensor fusion, including the steps of: Step 1, collecting laser energy spectrum characteristics, flow matching curves, acoustic signal characteristic templates and standard surface quality images of different materials and under different process conditions, setting set values of laser energy, water film flow and acoustic frequency spectrum, and constructing a process parameter reference database by combining the standard surface quality images; step 2, collecting multi-source signals in the processing process; Step3, preprocessing the multi-source signals to generate a multi-dimensional feature matrix; And 4, comprehensively analyzing according to the multidimensional feature matrix and the process parameter reference database to obtain a quality monitoring result. Preferably, the multi-source signal comprises a laser energy signal, a flow signal, a surface image, and an acoustic signal, and the laser energy signal comprises a laser power and a pulse frequency. Preferably, the preprocessing adopts a layered noise reduction and multidimensional standardization strategy to carry out noise reduction and standardization processing to realize accurate time synchronization and spatial registration and generate a multidimensional feature matrix, wherein the layered noise reduction and multidimensional standardization strategy comprises Kalman filtering, moving average filtering, median filtering, gaussian filtering, band-pass filtering and normalization processing. Preferably, processing quality including laser quality, flow quality and noise level is calculate