Aluminum circuit board surface treatment quality detection system and method

Abstract

The present invention relates to the field of circuit board quality inspection, and in particular to a system and method for inspecting the surface treatment quality of aluminum circuit boards. The method comprises the following steps: obtaining a real-time thermal imaging view of an aluminum circuit board; performing multi-point sampling calculation on the real-time thermal imaging view of the aluminum circuit board, and performing temperature distribution analysis, thereby generating temperature distribution data of the aluminum circuit board; reconstructing the temperature rise distribution trend based on the temperature distribution data of the aluminum circuit board, and locating the abnormal area, thereby obtaining the coating abnormal area; performing multi-directional ultrasonic pulse detection on the coating abnormal area, and collecting multi-array sound wave signals; performing low-pass filtering on the multi-array sound wave signals, and then performing three-dimensional projection reshaping to construct a three-dimensional sound wave projection map; performing sound wave penetration circuit board attenuation inversion on the three-dimensional sound wave projection map, and then performing surface defect analysis, thereby obtaining the surface defect area of ​​the aluminum circuit board. The present invention realizes an accurate and efficient aluminum circuit board surface defect detection method.

Description

Technical Field

[0001] The present invention relates to the field of circuit board quality inspection, and in particular to a system and method for inspecting the surface treatment quality of an aluminum circuit board. Background Art

[0002] With the vigorous development of the electronic information industry, aluminum-based circuit boards are widely used in various electronic products due to their advantages of light weight, good thermal conductivity, and low cost. Compared with traditional copper circuit boards, the production of copper circuit boards can adopt conventional etching, tinning and other processes, while the production of aluminum circuit boards requires special processes, such as chemical etching and anodizing, to prevent aluminum surface defects or flaws caused by aluminum corrosion. In addition, the thermal expansion coefficient of aluminum circuit boards is also high, and attention should be paid to thermal matching with other materials to prevent welding joints from breaking. Therefore, aluminum-based circuit boards also face many challenges in surface treatment technology. During the production process, complex surface treatment processes are prone to various defects, such as uneven oxide layer, substandard surface roughness, cracks at sharp corners, etc. These problems will seriously affect the electromagnetic properties, welding quality and service life of the product.

[0003] Traditional circuit board surface quality inspection mostly relies on manual visual inspection or simple optical imaging technology. It is difficult to comprehensively and accurately identify and classify complex surface defects, especially for some microscopic detail defects. Existing detection methods often have problems such as low recognition rate and low detection efficiency. The electronic manufacturing industry has increasingly stringent requirements on product quality, and in order to meet the special physical properties of aluminum circuit boards, there is an urgent need for a surface treatment quality inspection method for aluminum circuit boards that can overcome the limitations of traditional detection methods and comprehensively perceive and accurately diagnose various surface quality problems of aluminum circuit boards. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a system and method for detecting the surface treatment quality of aluminum circuit boards to solve at least one of the above technical problems.

[0005] To achieve the above object, the present invention provides a method for detecting the surface treatment quality of an aluminum circuit board, comprising the following steps:

[0006] Step S1: obtaining a real-time thermal imaging view of an aluminum circuit board; performing multi-point sampling calculation on the real-time thermal imaging view of the aluminum circuit board, and performing temperature distribution analysis, thereby generating temperature distribution data of the aluminum circuit board;

[0007] Step S2: reconstructing the temperature rise distribution trend based on the temperature distribution data of the aluminum circuit board, and locating the abnormal area, thereby obtaining the coating abnormal area;

[0008] Step S3: performing multi-directional ultrasonic pulse detection on the coating abnormal area to collect multi-array sound wave signals; performing low-pass filtering on the multi-array sound wave signals, and then performing three-dimensional projection reconstruction to construct a three-dimensional sound wave projection map;

[0009] Step S4: performing attenuation inversion of the sound wave penetrating the circuit board on the three-dimensional sound wave projection image, and then performing surface defect analysis to obtain the surface defect area of ​​the aluminum circuit board;

[0010] Step S5: performing deep mining of defect diffusion on the surface defect area of ​​the aluminum circuit board, and performing three-dimensional defect structure evolution to construct a three-dimensional defect structure model;

[0011] Step S6: Perform a deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data; and generate a quality inspection and evaluation result of the aluminum circuit board to be inspected based on the deep defect repair simulation response data.

[0012] The present invention acquires real-time thermal imaging views and temperature distribution data of aluminum circuit boards, monitors the temperature changes of aluminum circuit boards in real time, analyzes the temperature distribution of aluminum circuit boards, discovers potential abnormal areas and problems of uneven heat distribution, and accurately locates abnormal coating areas through temperature rise distribution trend reconstruction and abnormal area positioning, providing accurate information for subsequent detection and repair. After locating the abnormal coating area, subsequent analysis of defects on the surface of the aluminum circuit board is performed, and detection of abnormal coating areas is achieved through multi-directional ultrasonic pulse detection and three-dimensional acoustic wave projection map construction, while maintaining the integrity of the aluminum circuit board. A more accurate three-dimensional acoustic wave projection map is obtained through the processing and reshaping of acoustic wave signals, thereby improving the accuracy of detection, and attenuation of acoustic waves penetrating the circuit board is reduced. Inversion and surface defect analysis can accurately identify the surface defect area of ​​aluminum circuit boards, and further process and repair them. After locating the surface defects, the nature and extent of the defects are deeply analyzed to provide guidance for subsequent repair work. Deep mining and three-dimensional defect structure evolution of the defect area are carried out to establish a comprehensive three-dimensional defect structure model to help understand the formation and evolution process of defects. After understanding the depth and structure of the defects, targeted repair plans are planned to improve repair efficiency and quality. Through deep defect repair simulation and quality inspection and evaluation, comprehensive quality assessment results of aluminum circuit boards are generated to help formulate subsequent processing and improvement plans. Based on the results generated by the repair simulation response data, the quality status of the aluminum circuit board to be inspected is predicted to provide guidance for subsequent inspection and repair.

[0013] Preferably, step S1 comprises the following steps:

[0014] Step S11: performing global heating treatment on the aluminum circuit board to be inspected, and continuously performing infrared thermal imaging scanning to generate a real-time thermal imaging view of the aluminum circuit board;

[0015] Step S12: performing time-series segmentation processing on the real-time aluminum circuit board thermal imaging view to obtain a time-series thermal imaging view sequence;

[0016] Step S13: extracting the first thermal imaging view based on the time-series thermal imaging view sequence;

[0017] Step S14: performing multi-point sampling calculation on the first thermal imaging view, thereby obtaining multi-point initial temperature values ​​of the aluminum circuit board;

[0018] Step S15: performing temperature distribution analysis on the initial temperature values ​​at multiple points of the aluminum circuit board, thereby generating temperature distribution data of the aluminum circuit board.

[0019] The present invention promotes the manifestation of potential defects and improves the accuracy of detection by performing global heating treatment on the aluminum circuit board to be detected. Continuous infrared thermal imaging scanning can generate real-time thermal imaging views to realize real-time monitoring of temperature changes of the aluminum circuit board. The real-time thermal imaging views are subjected to time-series segmentation processing to cut continuous thermal imaging data into a time-series thermal imaging view sequence. In subsequent data processing and analysis, the first view in the time-series thermal imaging view sequence is extracted as the basis for initial analysis, key information is obtained from the sequence data, and initial temperature values ​​of multiple points of the aluminum circuit board are obtained through multi-point sampling calculation to provide basic data for subsequent temperature distribution analysis. Accurate initial temperature values ​​help to establish an accurate temperature distribution model to improve the accuracy of detection and analysis. Temperature distribution analysis is performed on the initial temperature values ​​of multiple points of the aluminum circuit board to generate temperature distribution data of the aluminum circuit board, so as to understand the overall temperature condition and the abnormal areas that exist. The temperature distribution of the aluminum circuit board is identified to help discover potential problems and abnormalities, and provide guidance for subsequent detection and repair.

[0020] Preferably, the specific steps of step S2 are:

[0021] Step S21: performing temperature change analysis on the time-series thermal imaging view sequence to generate temperature change data of the aluminum circuit board;

[0022] Step S22: Perform dynamic temperature rise fitting on the temperature change data of the aluminum circuit board to construct a dynamic temperature rise curve;

[0023] Step S23: reconstructing the temperature rise distribution trend of the dynamic temperature rise curve based on the temperature distribution data of the aluminum circuit board, and constructing a temperature rise distribution trend diagram;

[0024] Step S24: performing temperature rise distribution deviation detection on the temperature rise distribution trend diagram based on a preset standard circuit board temperature change diagram to obtain temperature rise distribution deviation data;

[0025] Step S25: performing coating uniformity anomaly analysis on the temperature rise distribution deviation data, thereby obtaining coating uniformity anomaly data;

[0026] Step S26: Based on the real-time aluminum circuit board thermal imaging view, the coating uniformity abnormal data is used to locate the abnormal area, thereby obtaining the coating abnormal area.

[0027] The present invention generates temperature change data of an aluminum circuit board by analyzing the temperature change of a time-series thermal imaging view sequence, and understands the temperature change of the aluminum circuit board in different time periods. The temperature change data of the aluminum circuit board is fitted with dynamic temperature rise changes, and a dynamic temperature rise curve is constructed to more clearly show the trend and law of temperature change. Based on the temperature distribution data of the aluminum circuit board, the dynamic temperature rise curve is reconstructed to perform a temperature rise distribution trend, and a temperature rise distribution trend chart is constructed to help understand the distribution of temperature rise in different areas. By comparing the temperature rise distribution trend chart with the preset standard circuit board temperature change chart, the deviation of the temperature rise distribution is detected, and potential problems and anomalies are discovered. The temperature rise distribution deviation data is analyzed to obtain coating uniformity abnormal data, which helps to identify the uniformity problem of the aluminum circuit board coating. Based on the real-time aluminum circuit board thermal imaging view, the coating uniformity abnormal data is located in the abnormal area, which helps to accurately locate the abnormal area of ​​the coating and provide guidance for subsequent repairs and improvements.

[0028] Preferably, step S3 specifically comprises the following steps:

[0029] Step S31: performing multi-directional ultrasonic pulse detection on the coating abnormal area according to the multi-array ultrasonic probe array to collect multi-array sound wave signals;

[0030] Step S32: low-pass filtering the multi-array sound wave signal to obtain a filtered noise-reduced sound wave signal;

[0031] Step S33: performing multi-scale wavelet decomposition on the filtered and noise-reduced wave signal to obtain frequency domain data of multiple scales;

[0032] Step S34: Calculate the frequency components of each scale on the frequency domain data of multiple scales, and extract the frequency components of each scale;

[0033] Step S35: optimizing the mean signal-to-noise ratio of the frequency domain data of multiple scales based on the frequency components of each scale, thereby obtaining multiple signal-to-noise ratio optimized frequency domain data;

[0034] Step S36: performing multi-sound wave fusion reconstruction on multiple signal-to-noise ratio optimized frequency domain data to construct a high signal-to-noise ratio sound wave signal;

[0035] Step S37: Perform three-dimensional projection reconstruction on the high signal-to-noise ratio sound wave signal to construct a three-dimensional sound wave projection image.

[0036] The present invention uses a multi-array ultrasonic probe array to perform multi-directional ultrasonic pulse detection on abnormal areas of the coating, and can obtain sound wave signals from different angles. This multi-directional detection obtains more comprehensive sound wave data to enable a more comprehensive evaluation and analysis of the coating. Low-pass filtering removes high-frequency noise in the sound wave signal, thereby improving the clarity and quality of the signal. This noise reduction process reduces interference, making subsequent signal processing more accurate. Multi-scale wavelet decomposition decomposes the sound wave signal into frequency domain data of different scales, allowing for a more detailed analysis of the frequency components in the sound wave signal, and a more in-depth study of the signal characteristics detected at different scales. By calculating the frequency components of each scale, the frequency information at each scale is extracted to understand the sound at different scales. The frequency characteristics of the wave signal provide a basis for subsequent analysis and processing. By optimizing the mean signal-to-noise ratio of multi-scale frequency domain data, the signal-to-noise ratio ratio is increased, thereby enhancing the clarity and recognizability of the signal. This optimization improves the accuracy of detection. Multi-acoustic wave fusion reconstruction integrates multiple signal-to-noise ratio optimized frequency domain data to construct a high signal-to-noise ratio acoustic wave signal. This reconstruction improves the quality and reliability of the acoustic wave signal and provides a better basis for subsequent analysis and processing. By reshaping the high signal-to-noise ratio acoustic wave signal through three-dimensional projection, a three-dimensional acoustic wave projection map is constructed to intuitively display the distribution of the acoustic wave signal in three-dimensional space. This projection map provides a more comprehensive understanding of the characteristics of the acoustic wave signal and the location of the abnormal area, providing an important reference for further analysis and processing.

[0037] Preferably, the specific steps of step S37 are:

[0038] Calculate the spatial position coordinates of each probe of the multi-array ultrasonic probe array;

[0039] Perform sound wave reflection delay calculation on the high signal-to-noise ratio sound wave signal to obtain the delay value of each sound wave signal;

[0040] Performing three-dimensional acoustic wave positioning on the delay value of each acoustic wave signal based on the spatial position coordinates of each probe, thereby obtaining multiple acoustic wave interaction points of the aluminum circuit board;

[0041] Calculate the spatial relative positions of multiple acoustic wave interaction points on aluminum circuit boards to obtain the spatial relative position coordinates of the interaction points;

[0042] The spatial relative position coordinates of the interaction points are reshaped by three-dimensional projection to construct a three-dimensional sound wave projection map.

[0043] The present invention accurately calculates the spatial position coordinates of each ultrasonic probe to ensure that the position of each probe is accurately understood during the sound wave signal acquisition process. The sound wave signal is subsequently processed and analyzed, and the propagation delay of the sound wave signal at different material interfaces or defects is accurately measured by calculating the reflection delay value of the sound wave signal, so as to detect the abnormality or defect on the surface of the aluminum circuit board. The spatial position coordinates of each probe are combined with the sound wave delay value to perform three-dimensional sound wave positioning, accurately determine the interaction point of the sound wave signal on the aluminum circuit board, and locate the potential problem area or abnormal point. The spatial relative position coordinates of the interaction point are calculated to understand the spatial relationship of the intersection of different sound wave signals, analyze the abnormal situation on the surface of the aluminum circuit board, such as uneven coating thickness or other defects, and construct a three-dimensional sound wave projection diagram by performing three-dimensional projection reconstruction on the spatial relative position of the interaction point, providing an intuitive visual display, helping engineers or operators to more intuitively understand the sound wave interaction situation on the surface of the aluminum circuit board, so as to better identify the potential problem area.

[0044] Preferably, the specific steps of step S4 are:

[0045] Step S41: performing acoustic wave semantic analysis on the three-dimensional acoustic wave projection image to generate acoustic wave semantic feature data;

[0046] Step S42: Calculate the signal amplitude of the acoustic wave semantic feature data to obtain the acoustic wave signal amplitude;

[0047] Step S43: performing sound wave frequency distribution analysis on the three-dimensional sound wave projection image to generate frequency distribution data;

[0048] Step S44: performing attenuation inversion of the sound wave penetrating the circuit board based on the sound wave signal amplitude and frequency distribution data, thereby obtaining sound wave attenuation characteristic data;

[0049] Step S45: performing attenuation reflection uniformity abnormality identification on the sound wave attenuation characteristic data, and extracting sound wave attenuation reflection abnormality data;

[0050] Step S46: performing surface defect analysis on the three-dimensional acoustic wave projection image based on the acoustic wave attenuation reflection anomaly data, thereby obtaining the surface defect area of ​​the aluminum circuit board.

[0051] The present invention performs semantic analysis on the acoustic wave projection image, identifies and extracts semantic features in the acoustic wave signal, understands the acoustic wave interaction on the surface of the aluminum circuit board, and generates acoustic wave semantic feature data to help identify the features of different acoustic wave signals, providing an important basis for subsequent signal analysis and processing. By calculating the amplitude of the acoustic wave signal, the intensity and energy distribution of the acoustic wave signal are evaluated, and the signal intensity change in the abnormal or defective area is detected. The amplitude data can be used to identify signal intensity anomalies and indicate the problem area. The frequency distribution analysis understands the frequency components of the acoustic wave signal and identifies the acoustic wave feature changes at different frequencies, providing important data support for subsequent analysis. The frequency distribution data detects abnormal frequency components and indicates the presence of special features on the surface of the aluminum circuit board. Identify problems or defects, obtain the attenuation characteristics of sound waves during propagation inside the aluminum circuit board through attenuation inversion, analyze the stability and quality of signal transmission, and the attenuation characteristic data helps understand the sound wave propagation of the internal structure of the aluminum circuit board, revealing the existing internal problems or defects. By identifying the abnormal attenuation reflection data, locate the abnormal reflection points inside the aluminum circuit board, indicate potential problem areas, and extract abnormal data to accurately locate the defective areas on the surface of the aluminum circuit board, providing guidance for subsequent repairs and improvements. By analyzing the abnormal sound wave attenuation reflection data, accurately identify the defective areas on the surface of the aluminum circuit board, provide detailed information for quality inspection, determine the surface defective areas and formulate corresponding repair plans to improve the quality and performance of the aluminum circuit board.

[0052] Preferably, the specific steps of step S5 are:

[0053] Step S51: performing high-precision repeated acoustic wave scanning on the surface defect area of ​​the aluminum circuit board to extract high-precision acoustic wave scanning signals;

[0054] Step S52: performing an acoustic wave propagation path distribution analysis on the high-precision acoustic wave scanning signal to generate surface defect propagation path distribution data;

[0055] Step S53: performing defect morphological feature analysis on the surface defect propagation path distribution data to generate surface defect morphological feature data;

[0056] Step S54: performing deep mining of defect diffusion on the high-precision acoustic wave scanning signal to obtain the defect depth diffusion law;

[0057] Step S55: Based on the defect depth diffusion law, the surface defect morphological feature data is subjected to three-dimensional defect structure evolution to construct a three-dimensional defect structure model.

[0058] The present invention obtains high-precision acoustic wave signals by repeated acoustic wave scanning, captures small changes and details of surface defects, and improves detection accuracy. Repeated scanning can improve signal stability, reduce errors and interference, and ensure data reliability and consistency. By analyzing the distribution of acoustic wave propagation paths, the propagation mode and path of acoustic waves in the surface defect area are understood, and the defects are helped to be located and analyzed. The propagation path distribution data detects abnormal propagation paths and indicates the existing surface defect problems. The defect propagation path distribution data is analyzed to extract the morphological characteristics of the surface defects, including size, shape, distribution and other information. The morphological characteristic data quantitatively describes the morphological characteristics of the surface defects, and provides a basis for subsequent structural analysis and comparison. By deeply mining the acoustic wave scanning signal, the deep diffusion law of the defects is understood, and the diffusion of surface defects inside the aluminum circuit board is revealed. The deep diffusion law of the defects is obtained to understand the internal structure and development trend of the defects, and a basis is provided for further analysis. Based on the deep diffusion law and the surface morphological characteristic data, a three-dimensional defect structure model is constructed to intuitively display the structure and distribution of the defects inside the aluminum circuit board. The three-dimensional defect structure model analyzes the spatial layout and morphological evolution of the defects, and provides an intuitive reference for in-depth understanding of the surface treatment quality.

[0059] Preferably, the specific steps of step S6 are:

[0060] Step S61: performing defect type cluster analysis on the three-dimensional defect structure model to obtain surface defect types;

[0061] Step S62: performing defect propagation risk analysis on the surface defect type, thereby obtaining surface defect propagation risk data;

[0062] Step S63: performing a deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data;

[0063] Step S64: quantifying the repair difficulty of the deep defect repair simulation response data to generate a defect repair difficulty value;

[0064] Step S65: Perform a comprehensive surface quality inspection evaluation on the surface defect propagation risk data and the defect repair difficulty value to generate a quality inspection evaluation result of the aluminum circuit board to be inspected.

[0065] The present invention identifies different types of surface defects through cluster analysis, classifies and marks the defects, and facilitates subsequent processing and analysis. The obtained surface defect type information helps to understand the characteristics and influences of different defect types, guides the selection of subsequent processing methods, evaluates the propagation risk of surface defects through risk analysis, identifies potential propagation paths and affected areas, and prevents and controls the spread of problems in advance. The obtained propagation risk data provides a basis for formulating defect repair plans and quality control strategies. By simulating the repair process of deep defects, the effects and costs of different repair strategies are evaluated, and the actual repair operations are guided. The generated repair simulation response data provides a quantitative evaluation of the effects of different repair schemes, providing a basis for decision-making. By quantifying the repair difficulty, the repair workload and difficulty of defects of different depths are evaluated, which helps to reasonably arrange repair plans and resources. The obtained repair difficulty value helps to determine the repair priority and improve the repair efficiency and quality. The propagation risk data and the repair difficulty value are combined for comprehensive evaluation, and the surface quality detection of aluminum circuit boards is fully understood. The generated quality detection evaluation results provide decision support for formulating improvement measures and maintenance strategies, and improve the surface treatment quality.

[0066] In this specification, a system for detecting the surface treatment quality of an aluminum circuit board is provided, which is used to perform the above-mentioned method for detecting the surface treatment quality of an aluminum circuit board, comprising:

[0067] The temperature distribution module obtains the real-time thermal imaging view of the aluminum circuit board; performs multi-point sampling calculation on the real-time thermal imaging view of the aluminum circuit board, and performs temperature distribution analysis to generate the temperature distribution data of the aluminum circuit board;

[0068] The temperature rise trend module reconstructs the temperature rise distribution trend based on the temperature distribution data of the aluminum circuit board and locates the abnormal area to obtain the coating abnormal area;

[0069] The three-dimensional projection module performs multi-directional ultrasonic pulse detection on the abnormal area of ​​the coating and collects multi-array sound wave signals; the multi-array sound wave signals are low-pass filtered and then reshaped by three-dimensional projection to construct a three-dimensional sound wave projection map.

[0070] The surface defect analysis module performs attenuation inversion of the sound wave penetrating the circuit board on the three-dimensional sound wave projection image, and then performs surface defect analysis to obtain the surface defect area of ​​the aluminum circuit board;

[0071] The three-dimensional defect structure module conducts deep mining of defect diffusion in the defect area on the surface of aluminum circuit boards, performs three-dimensional defect structure evolution, and constructs a three-dimensional defect structure model;

[0072] The defect repair simulation module performs deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data; and generates quality inspection and evaluation results of the aluminum circuit board to be inspected based on the deep defect repair simulation response data.

[0073] The present invention obtains real-time thermal imaging views of aluminum circuit boards and performs multi-point sampling calculations to monitor the temperature distribution of aluminum circuit boards in real time, performs temperature distribution analysis to generate temperature distribution data of aluminum circuit boards, provides basic data support for subsequent temperature rise trend analysis, reconstructs temperature rise distribution trends based on the temperature distribution data of aluminum circuit boards, effectively detects abnormal areas and locates them, such as coating abnormal areas, locates abnormal areas to find problems in advance and take measures to improve production quality and efficiency, performs multi-directional ultrasonic pulse detection on coating abnormal areas, obtains more comprehensive sound wave signal data, constructs a clearer three-dimensional sound wave projection image after low-pass filtering and three-dimensional projection reconstruction, improves the accuracy of defect analysis, and attenuates reflections through sound waves penetrating the circuit board The system can accurately identify the defective areas on the surface of aluminum circuit boards, and use the obtained surface defect area information to evaluate the surface treatment quality, guide the subsequent defect repair and quality improvement work, conduct deep defect diffusion mining and three-dimensional defect structure evolution on the surface defective areas of aluminum circuit boards, understand the formation and development process of defects, and build a three-dimensional defect structure model. It can perform deep defect repair simulation on the three-dimensional defect structure model, generate deep defect repair simulation response data, and help evaluate the effects of different repair schemes. The quality inspection and evaluation results of the aluminum circuit boards to be inspected generated based on the repair simulation response data can comprehensively evaluate the quality status of the aluminum circuit boards, and guide the subsequent quality improvement work to provide more intuitive defect information and provide a basis for subsequent repair simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 A schematic diagram of the steps of a method for detecting the surface treatment quality of an aluminum circuit board according to the present invention;

[0075] Figure 2 Detailed implementation flow chart of step S1;

[0076] Figure 3 Detailed implementation flow chart of step S2. DETAILED DESCRIPTION

[0077] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0078] The present application example provides a system and method for detecting the surface treatment quality of aluminum circuit boards. The execution subject of the system and method for detecting the surface treatment quality of aluminum circuit boards includes but is not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0079] See also Figures 1 to 3 The present invention provides a method for detecting the surface treatment quality of an aluminum circuit board, and the method comprises the following steps:

[0080] Step S1: obtaining a real-time thermal imaging view of an aluminum circuit board; performing multi-point sampling calculation on the real-time thermal imaging view of the aluminum circuit board, and performing temperature distribution analysis, thereby generating temperature distribution data of the aluminum circuit board;

[0081] Step S2: reconstructing the temperature rise distribution trend based on the temperature distribution data of the aluminum circuit board, and locating the abnormal area, thereby obtaining the coating abnormal area;

[0082] Step S3: performing multi-directional ultrasonic pulse detection on the coating abnormal area to collect multi-array sound wave signals; performing low-pass filtering on the multi-array sound wave signals, and then performing three-dimensional projection reconstruction to construct a three-dimensional sound wave projection map;

[0083] Step S4: performing attenuation inversion of the sound wave penetrating the circuit board on the three-dimensional sound wave projection image, and then performing surface defect analysis to obtain the surface defect area of ​​the aluminum circuit board;

[0084] Step S5: performing deep mining of defect diffusion on the surface defect area of ​​the aluminum circuit board, and performing three-dimensional defect structure evolution to construct a three-dimensional defect structure model;

[0085] Step S6: Perform a deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data; and generate a quality inspection and evaluation result of the aluminum circuit board to be inspected based on the deep defect repair simulation response data.

[0086] The present invention acquires real-time thermal imaging views and temperature distribution data of aluminum circuit boards, monitors the temperature changes of aluminum circuit boards in real time, analyzes the temperature distribution of aluminum circuit boards, discovers potential abnormal areas and problems of uneven heat distribution, and accurately locates abnormal coating areas through temperature rise distribution trend reconstruction and abnormal area positioning, providing accurate information for subsequent detection and repair. After locating the abnormal coating area, subsequent analysis of defects on the surface of the aluminum circuit board is performed, and detection of abnormal coating areas is achieved through multi-directional ultrasonic pulse detection and three-dimensional acoustic wave projection map construction, while maintaining the integrity of the aluminum circuit board. A more accurate three-dimensional acoustic wave projection map is obtained through the processing and reshaping of acoustic wave signals, thereby improving the accuracy of detection, and attenuation of acoustic waves penetrating the circuit board is reduced. Inversion and surface defect analysis can accurately identify the surface defect area of ​​aluminum circuit boards, and further process and repair them. After locating the surface defects, the nature and extent of the defects are deeply analyzed to provide guidance for subsequent repair work. Deep mining and three-dimensional defect structure evolution of the defect area are carried out to establish a comprehensive three-dimensional defect structure model to help understand the formation and evolution process of defects. After understanding the depth and structure of the defects, targeted repair plans are planned to improve repair efficiency and quality. Through deep defect repair simulation and quality inspection and evaluation, comprehensive quality assessment results of aluminum circuit boards are generated to help formulate subsequent processing and improvement plans. Based on the results generated by the repair simulation response data, the quality status of the aluminum circuit board to be inspected is predicted to provide guidance for subsequent inspection and repair.

[0087] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a method for detecting the surface treatment quality of an aluminum circuit board according to the present invention. In this example, the steps of the method include:

[0088] Step S1: obtaining a real-time thermal imaging view of an aluminum circuit board; performing multi-point sampling calculation on the real-time thermal imaging view of the aluminum circuit board, and performing temperature distribution analysis, thereby generating temperature distribution data of the aluminum circuit board;

[0089] In this embodiment, a high-resolution infrared thermal imaging camera is equipped to ensure that it is suitable for the measurement range of the aluminum circuit board. Before the actual measurement, the infrared camera is calibrated to ensure the temperature measurement accuracy. The camera parameters, including the measurement mode, exposure time and focal length, are configured to ensure that the real-time thermal imaging view of the aluminum circuit board can be captured. The thermal imaging camera is started to continuously collect thermal imaging images of the aluminum circuit board to ensure that the image is clear and reflects the real-time temperature change. The acquired thermal imaging view is imported using image processing software (such as MATLAB, OpenCV or dedicated thermal imaging analysis software), and the thermal imaging image is divided into multiple sampling regions (ROIs) according to the design structure and focus of the circuit board. Each divided ROI is analyzed. Temperature data extraction, record the temperature value of each area, use image processing algorithms (such as threshold processing) to identify the area of ​​interest, extract the average temperature, maximum temperature and minimum temperature of each area, organize the extracted temperature data into a data frame, including the temperature value, location and other relevant information of each area, calculate the temperature distribution of the entire aluminum circuit board, use interpolation methods (such as Kriging interpolation or spline interpolation) to generate a temperature distribution map, use visualization tools (such as Matplotlib, Seaborn or heat map function) to visualize the temperature distribution data, generate a temperature distribution map of the aluminum circuit board, store the generated aluminum circuit board temperature distribution data and visualization results in the data report, and ensure that the analysis results are clear and easy to understand.

[0090] Step S2: reconstructing the temperature rise distribution trend based on the temperature distribution data of the aluminum circuit board, and locating the abnormal area, thereby obtaining the coating abnormal area;

[0091] In this embodiment, the temperature distribution data obtained from the thermal imaging analysis of the aluminum circuit board is organized into a structured format to ensure that the temperature value and the corresponding spatial coordinates of each sampling point are included. A suitable mathematical model or algorithm is selected to reconstruct the temperature rise distribution trend, such as polynomial fitting, interpolation method (such as spline interpolation) or machine learning regression model. According to the temperature distribution data, a temperature rise trend model is constructed, and data smoothing is performed to reduce the impact of noise on the model. The temperature rise trend is determined by the fitting method, and the expected temperature of each point is calculated. The reconstructed temperature rise trend data is visualized to generate a corresponding temperature rise distribution diagram to show the temperature rise of the aluminum circuit board in different areas. An abnormality detection algorithm is selected. Commonly used methods include threshold-based methods, Z-s core method or cluster analysis in machine learning (such as K-means clustering), set thresholds or standards, identify areas in the temperature rise distribution map that are significantly higher than the average level, mark them as abnormal areas, set the temperature threshold as the average temperature plus a certain multiple of the standard deviation, and any area exceeding the threshold is considered abnormal. The coordinates and temperature information of the identified coating abnormal area are recorded in the data structure to ensure the traceability of subsequent analysis and processing, analyze the identified abnormal area, evaluate its cause and impact, consider the uniformity of the coating, material defects or electrical failures, etc., write an analysis report, and record in detail the process of reconstructing the temperature rise distribution trend, the specific location of the abnormal area and its analysis results, so as to provide a basis for subsequent troubleshooting and maintenance decisions.

[0092] Step S3: performing multi-directional ultrasonic pulse detection on the coating abnormal area to collect multi-array sound wave signals; performing low-pass filtering on the multi-array sound wave signals, and then performing three-dimensional projection reconstruction to construct a three-dimensional sound wave projection map;

[0093] In this embodiment, an appropriate ultrasonic detection device is selected to ensure that it has a multi-array probe configuration for multi-directional detection, the arrangement of the probes (such as linear, matrix or stereo array) is determined, and optimized according to the specific shape and size of the coating abnormal area, and the necessary parameters of ultrasonic detection are set, including the transmission frequency, pulse width, gain and sampling rate, etc., to ensure the quality of data acquisition, and the ultrasonic detection device is started to transmit sound wave pulses to the coating abnormal area from different directions and angles to ensure that the entire abnormal area is covered, and the sound wave signals received by each probe are collected, and the timestamp, signal amplitude and frequency information of each probe are recorded to form a multi-array sound wave signal data set, and a suitable low-pass filtering algorithm is selected, such as a Butterworth filter or a FIR filter, to remove high-frequency noise, and a low-pass filter is applied to the collected multi-array sound wave signals. The following steps can be adopted during the processing: a cutoff frequency is set to ensure that the frequencies above the cutoff frequency are filtered out. The signal components are suppressed, the signal of each probe is filtered to obtain smooth acoustic signal data, the influence of noise on subsequent analysis is reduced, the filtered acoustic signal data is integrated into a three-dimensional data structure, and the spatial position and corresponding signal strength of each probe are ensured to be included. A suitable three-dimensional visualization tool or software is selected, such as MATLAB, ParaView or Mayavi in ​​Python, and a three-dimensional acoustic projection diagram is generated according to the integrated three-dimensional data. The specific steps include: using interpolation technology to map the probe signal data into three-dimensional space to form a continuous acoustic wave intensity distribution, applying appropriate coloring and rendering technology so that the projection diagram can clearly show the acoustic signal intensity in different areas. The generated three-dimensional acoustic projection diagram can be used to analyze the characteristics of abnormal coating areas, identify potential defects and problems, record the three-dimensional acoustic projection diagram and its analysis results, and generate a detailed report to provide a basis for subsequent troubleshooting and maintenance decisions.

[0094] Step S4: performing attenuation inversion of the sound wave penetrating the circuit board on the three-dimensional sound wave projection image, and then performing surface defect analysis to obtain the surface defect area of ​​the aluminum circuit board;

[0095] In this embodiment, a suitable attenuation inversion model is selected, for example, a sound wave propagation model (such as a wave equation or a Rayleigh scattering model) is used to describe the propagation behavior of sound waves in an aluminum circuit board. The intensity and propagation path information of the sound wave signal are extracted from the three-dimensional sound wave projection diagram to form the correlation data of the sound wave intensity and position. The extracted data is preprocessed to remove outliers and noise to ensure the accuracy and availability of the data. The attenuation data is processed using an inversion algorithm (such as the least squares method or Bayesian inversion) to reconstruct the attenuation characteristics of the sound wave in the material and generate an attenuation distribution map. A suitable defect detection algorithm is selected, and commonly used methods include image processing techniques (such as edge detection, region segmentation) or machine learning methods (such as support vector machines, convolutional neural networks). Use the attenuation distribution map to analyze surface defects and identify areas where the sound wave attenuation is significant, which correspond to the surface defects of the aluminum circuit board. Determine the characteristics of the defect, such as position, size and shape. Visualize the analysis results, generate a heat map or annotation map of the surface defect area, and clearly show the location and characteristics of the defect. The analysis results of the surface defect area are recorded and a detailed report is generated, including the specific description, location and impact of the defect, providing a basis for subsequent repairs and improvements.

[0096] Step S5: performing deep mining of defect diffusion on the surface defect area of ​​the aluminum circuit board, and performing three-dimensional defect structure evolution to construct a three-dimensional defect structure model;

[0097] In this embodiment, relevant data is extracted from the defect area on the surface of the aluminum circuit board, including the location, size and shape of the defect, and a suitable deep mining model is selected, such as a diffusion model or a thermal diffusion model based on acoustic wave attenuation, to evaluate the diffusion characteristics of the defect inside the material, and a mathematical model (such as a diffusion equation) is used to model the depth diffusion of the defect, and the model parameters (such as the diffusion coefficient) are estimated based on the actual data, and the model is verified, and the defect diffusion data is integrated with the surface defect data to form a complete three-dimensional defect data set, including the surface characteristics and internal diffusion characteristics of the defect, and a suitable three-dimensional modeling software (such as SolidWorks, MATLAB, AutoCAD or Blender) is selected to construct a three-dimensional defect structure model. Based on the integrated data, a three-dimensional defect structure model is constructed: the outer boundary of the defect is drawn using modeling tools, and the internal structure is generated according to the deep diffusion law. Appropriate rendering technology is applied to display the geometric shape and internal structural characteristics of the defect. The constructed three-dimensional defect structure model is verified to ensure that the model can accurately reflect the distribution and shape of the actual defects. The model is verified by comparing with the actual detection data. According to the verification results, necessary optimization and adjustments are made to the model to ensure its accuracy and usability. The three-dimensional defect structure model is exported to a suitable file format for subsequent analysis and sharing. A detailed analysis report is generated to record the process of deep mining of defect diffusion, the characteristics of the three-dimensional defect structure model and its application value, providing a basis for subsequent troubleshooting, maintenance and improvement.

[0098] Step S6: Perform a deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data; and generate a quality inspection and evaluation result of the aluminum circuit board to be inspected based on the deep defect repair simulation response data.

[0099] In this embodiment, a suitable finite element analysis (FEA) software (such as ANSYS, COMSOL) is selected to perform repair simulation, and a three-dimensional model including defects and repair materials is established in the simulation software to ensure that the model accurately reflects the actual situation. The simulation parameters, including material properties, boundary conditions and loading conditions, are set to perform deep defect repair simulation, and the response data during the repair process is observed. The response data of the deep defect repair simulation (such as stress, strain, and temperature change) is stored in a data frame for subsequent analysis. According to the requirements of the repair process, relevant parameters are set, such as the coating thickness, curing temperature and time of the repair material, and the boundary conditions of the model are set to define the fixedness and freedom of the repair area; at the same time, external load conditions, such as temperature changes, mechanical stress, etc., are applied to simulate the actual working environment, and the repair simulation program is run. The sequence is to calculate the stress, strain and displacement response data of the material during the deep defect repair process, extract key information from the simulation results, including stress distribution, strain distribution and displacement field data, to analyze the repair effect, and use visualization tools to display the simulation results in graphical form, including stress intensity diagrams, deformation diagrams, etc., to facilitate understanding of the repair effect, determine the quality inspection and evaluation standards, including the material properties after repair, the probability of recurrence of defects and the stability of the overall structure, etc., calculate the quality inspection and evaluation results of the aluminum circuit board to be inspected based on the extracted response data and the set evaluation standards, use weighted scoring method or multi-attribute decision analysis method for comprehensive evaluation, record the quality inspection and evaluation results, and generate a detailed report, including the repair simulation process, response data analysis and final evaluation results, to provide a basis for subsequent decision-making and improvement.

[0100] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0101] Step S11: performing global heating treatment on the aluminum circuit board to be inspected, and continuously performing infrared thermal imaging scanning to generate a real-time thermal imaging view of the aluminum circuit board;

[0102] Step S12: performing time-series segmentation processing on the real-time aluminum circuit board thermal imaging view to obtain a time-series thermal imaging view sequence;

[0103] Step S13: extracting the first thermal imaging view based on the time-series thermal imaging view sequence;

[0104] Step S14: performing multi-point sampling calculation on the first thermal imaging view, thereby obtaining multi-point initial temperature values ​​of the aluminum circuit board;

[0105] Step S15: performing temperature distribution analysis on the initial temperature values ​​at multiple points of the aluminum circuit board, thereby generating temperature distribution data of the aluminum circuit board.

[0106] In this embodiment, a suitable heating device (such as a hot air gun, an infrared heater or a heating plate) is selected to ensure that it can evenly heat the aluminum circuit board, the heating temperature and time are determined, and suitable heating conditions are usually set according to the characteristics of the aluminum material and the detection requirements. The aluminum circuit board is placed under the heating device, and the heating is started to ensure that the entire surface is evenly heated. An infrared thermal imaging camera is used to ensure that it is correctly set and can capture the temperature information of the aluminum circuit board in real time. During the heating process, infrared thermal imaging scanning is continuously performed to obtain a temperature distribution image of the surface of the aluminum circuit board, and the thermal imaging view generated by each scan is recorded in real time for subsequent processing. The thermal imaging view data obtained from the infrared thermal imaging camera is sorted and stored in a data structure (such as an array or a list) in chronological order. A time series segmentation algorithm (such as a method based on a time window or a threshold) is applied to obtain the real-time thermal image distribution image. Extract a continuous image sequence from the image view, select the first thermal imaging view from the extracted time-series thermal imaging view sequence, usually the first frame in the sequence, save the selected first thermal imaging view to an independent variable or file, and prepare for further analysis. In the first thermal imaging view, select several representative sampling points, usually in different areas to obtain comprehensive temperature data. Use thermal imaging software or algorithms to extract the corresponding temperature value from each sampling point, record these initial temperature data, store the extracted multi-point temperature values ​​in an array or data table, perform statistical analysis on the collected multi-point initial temperature values, calculate the mean, maximum, minimum and standard deviation of the temperature and other indicators, use visualization tools (such as Matplotlib, Excel, etc.) to draw a temperature distribution diagram, and intuitively display the temperature distribution of the aluminum circuit board.

[0107] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0108] Step S21: performing temperature change analysis on the time-series thermal imaging view sequence to generate temperature change data of the aluminum circuit board;

[0109] Step S22: Perform dynamic temperature rise fitting on the temperature change data of the aluminum circuit board to construct a dynamic temperature rise curve;

[0110] Step S23: reconstructing the temperature rise distribution trend of the dynamic temperature rise curve based on the temperature distribution data of the aluminum circuit board, and constructing a temperature rise distribution trend diagram;

[0111] Step S24: performing temperature rise distribution deviation detection on the temperature rise distribution trend diagram based on a preset standard circuit board temperature change diagram to obtain temperature rise distribution deviation data;

[0112] Step S25: performing coating uniformity anomaly analysis on the temperature rise distribution deviation data, thereby obtaining coating uniformity anomaly data;

[0113] Step S26: Based on the real-time aluminum circuit board thermal imaging view, the coating uniformity abnormal data is used to locate the abnormal area, thereby obtaining the coating abnormal area.

[0114] In this embodiment, the temperature data of each view is extracted from the time-series thermal imaging view sequence. The temperature value of each point is usually obtained in pixels. The temperature change of each sampling point at different time points is calculated to generate temperature change data. The formula is: = , where T represents the temperature at a certain moment. Select an appropriate fitting method, such as linear fitting, polynomial fitting or exponential fitting, which is usually selected according to the characteristics of temperature change. Use the least squares method or other optimization algorithms to fit the temperature rise data to obtain a dynamic temperature rise curve. Use the SciPy library in Python for fitting and evaluate the accuracy of the fitting curve. According to the temperature distribution data of the aluminum circuit board, use methods such as moving average or weighted average to reconstruct the trend. Combine the reconstructed dynamic temperature rise curve with the temperature distribution data of the aluminum circuit board to generate a temperature rise distribution trend chart. Use visualization tools (such as Matplotlib or Tableau) to draw a temperature rise distribution trend chart to intuitively display the trend of temperature change. Obtain a preset standard circuit board temperature change chart to ensure that it is a valid reference. Calculate the deviation between the temperature rise distribution trend chart and the standard chart. Differentially analyze the temperature rise distribution deviation of each area, generate temperature rise distribution deviation data, store the deviation data in a data frame, ensure that the deviation value and position of each area are clear and available, analyze the temperature rise distribution deviation data, and find out whether there are obvious outliers. This is usually achieved by setting thresholds or using statistical methods (such as Z-score). Record the detected coating uniformity anomaly data in a data structure, mark the specific location of the anomaly and its temperature deviation, use real-time aluminum circuit board thermal imaging view to visualize the coating uniformity anomaly data, mark the coating abnormal area on the thermal imaging map, use image processing technology (such as contour detection) to identify, and combine the coating uniformity anomaly data to generate the positioning information of the coating abnormal area. Store the location information of the coating abnormal area and related image data in the data frame for visualization and subsequent processing.

[0115] In this embodiment, the detailed implementation steps of step S3 include:

[0116] Step S31: performing multi-directional ultrasonic pulse detection on the coating abnormal area according to the multi-array ultrasonic probe array to collect multi-array sound wave signals;

[0117] Step S32: low-pass filtering the multi-array sound wave signal to obtain a filtered noise-reduced sound wave signal;

[0118] Step S33: performing multi-scale wavelet decomposition on the filtered and noise-reduced wave signal to obtain frequency domain data of multiple scales;

[0119] Step S34: Calculate the frequency components of each scale on the frequency domain data of multiple scales, and extract the frequency components of each scale;

[0120] Step S35: optimizing the mean signal-to-noise ratio of the frequency domain data of multiple scales based on the frequency components of each scale, thereby obtaining multiple signal-to-noise ratio optimized frequency domain data;

[0121] Step S36: performing multi-sound wave fusion reconstruction on multiple signal-to-noise ratio optimized frequency domain data to construct a high signal-to-noise ratio sound wave signal;

[0122] Step S37: Perform three-dimensional projection reconstruction on the high signal-to-noise ratio sound wave signal to construct a three-dimensional sound wave projection image.

[0123] In this embodiment, according to the shape and size of the coating abnormal area, a multi-array ultrasonic probe is configured to ensure that all key areas are covered, the ultrasonic detection equipment is started, a pulse ultrasonic signal is emitted to the coating abnormal area, the sound wave signal reflected from the coating is received through the probe array, the signal data of each probe is recorded, the collected multi-array sound wave signal is stored as a data matrix, it is ensured that the signal of each probe corresponds to its position information, an appropriate low-pass filter (such as Butterworth or Chebyshev filter) is designed, the cutoff frequency is determined to remove high-frequency noise, the multi-array sound wave signal is low-pass filtered to eliminate the high-frequency noise component, and the filtered The noise-reduced acoustic wave signal is stored in a data frame, ensuring that the filtered signal of each probe is available for subsequent analysis, selecting a suitable wavelet basis (such as Daubechies wavelet or Haar wavelet) for signal decomposition, performing multi-scale wavelet decomposition on the filtered acoustic wave signal, obtaining frequency domain data of different scales, and storing the obtained frequency domain data of multiple scales in a data frame, ensuring that the decomposition information of each scale is clear and available, performing Fourier transform on the frequency domain data of each scale, calculating its frequency components, and storing the extracted frequency components in a data frame, ensuring that the frequency information of each scale is clear and available, and calculating the mean signal-to-noise ratio (SNR) for the frequency components of each scale using the following formula: ,in, is the signal power, is the noise standard deviation. Based on the signal-to-noise ratio result of each scale, the frequency domain data of multiple scales are optimized to improve the signal-to-noise ratio. The optimized frequency domain data is stored in the data frame to ensure that the optimization information of each scale is clear and available. A suitable fusion algorithm is selected. Weighted average method: weighted synthesis of different frequency domain signals according to the signal-to-noise ratio. Wavelet reconstruction: reconstruct high signal-to-noise ratio acoustic wave signals based on wavelet transform, fuse multiple signal-to-noise ratio optimized frequency domain data, construct high signal-to-noise ratio acoustic wave signals, store the reconstructed high signal-to-noise ratio acoustic wave signals in the data frame to ensure that the signal information is complete. A three-dimensional projection algorithm (such as volume rendering or surface rendering) is selected to convert the acoustic wave signal into a three-dimensional image. The high signal-to-noise ratio acoustic wave signal is reshaped by three-dimensional projection to generate a three-dimensional acoustic wave projection map. Visualization tools (such as Matplotlib, ParaView, etc.) are used to display the three-dimensional acoustic wave projection map and analyze the abnormal areas of the coating.

[0124] In this embodiment, the specific steps of step S37 are:

[0125] Calculate the spatial position coordinates of each probe of the multi-array ultrasonic probe array;

[0126] Perform sound wave reflection delay calculation on the high signal-to-noise ratio sound wave signal to obtain the delay value of each sound wave signal;

[0127] Performing three-dimensional acoustic wave positioning on the delay value of each acoustic wave signal based on the spatial position coordinates of each probe, thereby obtaining multiple acoustic wave interaction points of the aluminum circuit board;

[0128] Calculate the spatial relative positions of multiple acoustic wave interaction points on aluminum circuit boards to obtain the spatial relative position coordinates of the interaction points;

[0129] The spatial relative position coordinates of the interaction points are reshaped by three-dimensional projection to construct a three-dimensional sound wave projection map.

[0130] In this embodiment, the relative position and quantity of each probe are determined according to the installation method of the probe (such as a planar array or a three-dimensional array). Select a suitable coordinate system (such as a Cartesian coordinate system) to define the origin and axis. According to the geometric layout of the probe, calculate the spatial position coordinates of each probe. Store the spatial position coordinates of each probe in a data frame to ensure that the coordinate information is clear and available. Determine the propagation speed of the sound wave in the aluminum circuit board (usually about 5000 m / s) and use the following formula to calculate the delay value of the sound wave signal, delay time = , for each sound wave signal, calculate the total delay time from the probe to the target reflection surface and then back to the probe. Store the delay value of each sound wave signal in a data frame to ensure that each signal corresponds to its delay value. Select a suitable positioning algorithm, for example: calculate the interaction point by knowing the probe coordinates and delay value. Calculate the three-dimensional coordinates of the sound wave interaction point based on the spatial position coordinates of each probe and the corresponding delay value. The specific steps include: using the triangulation method to solve the coordinates of the interaction point using three or more sets of probe coordinates and delay values. Store the calculated coordinates of multiple aluminum circuit board sound wave interaction points in a data frame. Select a three-dimensional sound wave positioning algorithm, such as triangulation or least squares method, to locate the sound wave signal. Calculate the position of the sound wave interaction point based on the spatial position coordinates of the probe and the sound wave delay value. , where c is the propagation speed of the sound wave in the material. Store the calculated position coordinates of multiple aluminum circuit board sound wave interaction points in a data structure. Select a suitable three-dimensional projection reconstruction algorithm, such as: Volume rendering: map the relative position coordinates to three-dimensional space. Slicing method: display data at different depths through multiple slices. Use the relative position coordinates to perform three-dimensional projection reconstruction of the interaction points to generate a three-dimensional sound wave projection map. Visualization tools (such as Matplotlib, Mayavi, etc.) can be used for graphical display. Store the generated three-dimensional sound wave projection map as an image file to ensure that the data can be used for subsequent analysis and visualization.

[0131] In this embodiment, step S4 includes the following steps:

[0132] Step S41: performing acoustic wave semantic analysis on the three-dimensional acoustic wave projection image to generate acoustic wave semantic feature data;

[0133] Step S42: Calculate the signal amplitude of the acoustic wave semantic feature data to obtain the acoustic wave signal amplitude;

[0134] Step S43: performing sound wave frequency distribution analysis on the three-dimensional sound wave projection image to generate frequency distribution data;

[0135] Step S44: performing attenuation inversion of the sound wave penetrating the circuit board based on the sound wave signal amplitude and frequency distribution data, thereby obtaining sound wave attenuation characteristic data;

[0136] Step S45: performing attenuation reflection uniformity abnormality identification on the sound wave attenuation characteristic data, and extracting sound wave attenuation reflection abnormality data;

[0137] Step S46: performing surface defect analysis on the three-dimensional acoustic wave projection image based on the acoustic wave attenuation reflection anomaly data, thereby obtaining the surface defect area of ​​the aluminum circuit board.

[0138] In this embodiment, a suitable semantic parsing method is selected, for example: feature extraction algorithm: use feature selection methods in machine learning (such as principal component analysis PCA) or deep learning models (such as convolutional neural networks CNN) to parse the three-dimensional sound wave projection image, extract sound wave features (such as reflectivity, propagation path, etc.), generate sound wave semantic feature data, and store the generated sound wave semantic feature data in a data frame to ensure that the data structure is clear and can be used for subsequent processing. Use a peak detection method to calculate the amplitude of the sound wave signal. The specific method includes: peak detection algorithm: for example, use the find_peaks function in a signal processing library (such as SciPy) to identify the peak in the signal, perform amplitude calculation on the sound wave signal data, obtain the amplitude value of the sound wave signal collected by each probe, store the calculated sound wave signal amplitude in the data frame, ensure that the amplitude information can be used for subsequent analysis, select a suitable frequency analysis method, such as fast Fourier transform (FFT), to analyze the frequency distribution of the sound wave signal, perform frequency analysis on the sound wave signal in the three-dimensional sound wave projection image, record the intensity values ​​of different frequencies, generate frequency distribution data, and establish a sound wave attenuation model, usually using an exponential attenuation model: ,in, is the amplitude at distance d, Initial amplitude, The attenuation coefficient is obtained by performing attenuation inversion based on the amplitude and frequency distribution data of the acoustic wave signal, and the attenuation characteristic data of the acoustic wave when penetrating the circuit board is calculated. The attenuation characteristic data is stored in a data frame to ensure that each data point and its corresponding attenuation characteristic information are clear and available. An anomaly detection algorithm (such as a threshold-based method, a statistical method, or a machine learning algorithm) is used to identify the uniformity anomaly of the attenuation reflection, analyze the attenuation characteristic data, extract the reflection data with obvious anomalies, record the specific location and characteristics of the anomalies, and store the extracted acoustic wave attenuation reflection anomaly data in a data structure for subsequent analysis. Appropriate defect analysis technology (such as pattern recognition, image processing, etc.) is selected to analyze and identify surface defects of the aluminum circuit board. The acoustic wave attenuation reflection anomaly data is used to analyze the three-dimensional acoustic wave projection image, identify the surface defect area, mark the detected surface defect area on the three-dimensional acoustic wave projection image, and generate a report for subsequent processing or decision-making.

[0139] In this embodiment, the specific steps of step S5 are:

[0140] Step S51: performing high-precision repeated acoustic wave scanning on the surface defect area of ​​the aluminum circuit board to extract high-precision acoustic wave scanning signals;

[0141] Step S52: performing an acoustic wave propagation path distribution analysis on the high-precision acoustic wave scanning signal to generate surface defect propagation path distribution data;

[0142] Step S53: performing defect morphological feature analysis on the surface defect propagation path distribution data to generate surface defect morphological feature data;

[0143] Step S54: performing deep mining of defect diffusion on the high-precision acoustic wave scanning signal to obtain the defect depth diffusion law;

[0144] Step S55: Based on the defect depth diffusion law, the surface defect morphological feature data is subjected to three-dimensional defect structure evolution to construct a three-dimensional defect structure model.

[0145] In this embodiment, high-precision ultrasonic detection equipment is configured to ensure that the sensitivity and frequency of the probe are suitable for the aluminum circuit board material, and scanning parameters are set, including scanning frequency, probe spacing and scanning path, to ensure that all surface defect areas are covered. High-precision repeated acoustic wave scanning is performed on the surface defect area of ​​the aluminum circuit board, and the acoustic wave signal data of each scan is recorded. According to the acoustic wave propagation theory, a propagation path model of the acoustic wave through the aluminum circuit board is established, and multi-path propagation and reflection are considered. The high-precision acoustic wave scanning signal is analyzed to identify the acoustic wave propagation path, and a signal processing algorithm (such as time delay estimation) is used to obtain the acoustic wave propagation path distribution data. The generated surface defect propagation path distribution data is stored in a data frame to ensure that the feature information of each path is clear and available. The acoustic wave propagation path distribution data is analyzed using a feature extraction algorithm (such as morphological processing or image segmentation), and the propagation characteristics of the acoustic wave in the defect area (such as reflection, scattering, etc.) are analyzed to generate a surface defect propagation path distribution data. The morphological characteristic data of the defects are stored in a data frame to ensure that the morphological characteristic information of each defect is available for subsequent analysis. A mathematical model (such as an exponential decay model) is used to describe the depth attenuation and diffusion law of the acoustic wave signal. The deep diffusion law of the high-precision acoustic wave scanning signal is analyzed to identify the diffusion characteristics of the defects in the material. The defect depth data is extracted using an inversion technique (such as back propagation). The obtained defect depth diffusion law is stored in a data frame to ensure that the diffusion characteristic information is clear and available. Based on the defect depth diffusion law and the defect morphological characteristic data, a three-dimensional defect structure model is constructed using three-dimensional modeling software (such as MATLAB, SolidWorks, etc.). The defect characteristics obtained by the analysis are combined with the depth information to generate a three-dimensional structural evolution diagram of the defect in the aluminum circuit board. The constructed three-dimensional defect structure model is stored in a 3D file format (such as STL or OBJ) for subsequent visualization and analysis.

[0146] In this embodiment, the specific steps of step S6 are:

[0147] Step S61: performing defect type cluster analysis on the three-dimensional defect structure model to obtain surface defect types;

[0148] Step S62: performing defect propagation risk analysis on the surface defect type, thereby obtaining surface defect propagation risk data;

[0149] Step S63: performing a deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data;

[0150] Step S64: quantifying the repair difficulty of the deep defect repair simulation response data to generate a defect repair difficulty value;

[0151] Step S65: Perform a comprehensive surface quality inspection evaluation on the surface defect propagation risk data and the defect repair difficulty value to generate a quality inspection evaluation result of the aluminum circuit board to be inspected.

[0152] In this embodiment, defect feature data in the three-dimensional defect structure model are collected, including parameters such as shape, size, and depth, and a suitable clustering algorithm is selected, such as: K-means clustering: suitable for dividing defect data into a preset number of categories, DBSCAN: used to discover defect types of any shape, the defect feature data is input into the selected clustering algorithm, the defect type is classified, the identified surface defect type and its corresponding feature data are stored in a data frame for subsequent analysis, a risk assessment model is established, and factors such as defect type, location, and depth are considered. A suitable risk assessment method (such as hierarchical analysis method or fuzzy logic) is selected to conduct a propagation risk assessment on each defect type, calculate its propagation risk value, and establish a mathematical model for deep defect repair, considering the material Characteristics and repair methods (such as filling, welding, coating, etc.), use finite element analysis software (such as ANSYS or COMSOL) to simulate the repair of deep defects, generate response data after repair, determine the evaluation indicators of repair difficulty, such as repair material cost, operation complexity, technical requirements, etc., analyze the simulated response data of deep defect repair, and calculate the repair difficulty value in combination with the selected evaluation indicators. Use the following formula: Repair difficulty value = material cost + complexity + technical requirements. Combine the defect propagation risk data and the repair difficulty value to establish a comprehensive evaluation model, for example: Weighted scoring method: Determine the weight according to the importance of each indicator, calculate the comprehensive score, conduct a comprehensive evaluation of the quality inspection of the aluminum circuit board to be inspected, and obtain the final quality evaluation result. Use the following formula: .

[0153] In this embodiment, a system for detecting the surface treatment quality of an aluminum circuit board is provided, which is used to perform the above-mentioned method for detecting the surface treatment quality of an aluminum circuit board, including:

[0154] The temperature distribution module obtains the real-time thermal imaging view of the aluminum circuit board; performs multi-point sampling calculation on the real-time thermal imaging view of the aluminum circuit board, and performs temperature distribution analysis to generate the temperature distribution data of the aluminum circuit board;

[0155] The temperature rise trend module reconstructs the temperature rise distribution trend based on the temperature distribution data of the aluminum circuit board and locates the abnormal area to obtain the coating abnormal area;

[0156] The three-dimensional projection module performs multi-directional ultrasonic pulse detection on the abnormal area of ​​the coating and collects multi-array sound wave signals; the multi-array sound wave signals are low-pass filtered and then reshaped by three-dimensional projection to construct a three-dimensional sound wave projection map.

[0157] The surface defect analysis module performs attenuation inversion of the sound wave penetrating the circuit board on the three-dimensional sound wave projection image, and then performs surface defect analysis to obtain the surface defect area of ​​the aluminum circuit board;

[0158] The three-dimensional defect structure module conducts deep mining of defect diffusion in the defect area on the surface of aluminum circuit boards, performs three-dimensional defect structure evolution, and constructs a three-dimensional defect structure model;

[0159] The defect repair simulation module performs deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data; and generates quality inspection and evaluation results of the aluminum circuit board to be inspected based on the deep defect repair simulation response data.

[0160] The present invention obtains real-time thermal imaging views of aluminum circuit boards and performs multi-point sampling calculations to monitor the temperature distribution of aluminum circuit boards in real time, performs temperature distribution analysis to generate temperature distribution data of aluminum circuit boards, provides basic data support for subsequent temperature rise trend analysis, reconstructs temperature rise distribution trends based on the temperature distribution data of aluminum circuit boards, effectively detects abnormal areas and locates them, such as coating abnormal areas, locates abnormal areas to find problems in advance and take measures to improve production quality and efficiency, performs multi-directional ultrasonic pulse detection on coating abnormal areas, obtains more comprehensive sound wave signal data, constructs a clearer three-dimensional sound wave projection image after low-pass filtering and three-dimensional projection reconstruction, improves the accuracy of defect analysis, and attenuates reflections through sound waves penetrating the circuit board The system can accurately identify the defective areas on the surface of aluminum circuit boards, and use the obtained surface defect area information to evaluate the surface treatment quality, guide the subsequent defect repair and quality improvement work, conduct deep defect diffusion mining and three-dimensional defect structure evolution on the surface defective areas of aluminum circuit boards, understand the formation and development process of defects, and build a three-dimensional defect structure model. It can perform deep defect repair simulation on the three-dimensional defect structure model, generate deep defect repair simulation response data, and help evaluate the effects of different repair schemes. The quality inspection and evaluation results of the aluminum circuit boards to be inspected generated based on the repair simulation response data can comprehensively evaluate the quality status of the aluminum circuit boards, and guide the subsequent quality improvement work to provide more intuitive defect information and provide a basis for subsequent repair simulation.

[0161] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0162] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for detecting the surface treatment quality of an aluminum circuit board, characterized in that: The following steps are involved: Step S1: Obtaining a real-time thermal imaging view of an aluminum circuit board; Perform multi-point sampling calculation on the real-time aluminum circuit board thermal imaging view and perform temperature distribution analysis to generate aluminum circuit board temperature distribution data; the specific steps of step S1 are: Step S11: performing global heating treatment on the aluminum circuit board to be inspected, and continuously performing infrared thermal imaging scanning to generate a real-time thermal imaging view of the aluminum circuit board; Step S12: performing time-series segmentation processing on the real-time aluminum circuit board thermal imaging view to obtain a time-series thermal imaging view sequence; Step S13: extracting the first thermal imaging view based on the time-series thermal imaging view sequence; Step S14: performing multi-point sampling calculation on the first thermal imaging view, thereby obtaining multi-point initial temperature values ​​of the aluminum circuit board; Step S15: performing temperature distribution analysis on the initial temperature values ​​of multiple points of the aluminum circuit board, thereby generating temperature distribution data of the aluminum circuit board; Step S2: reconstruct the temperature rise distribution trend based on the temperature distribution data of the aluminum circuit board, and locate the abnormal area, so as to obtain the coating abnormal area; the specific steps of step S2 are: Step S21: performing temperature change analysis on the time-series thermal imaging view sequence to generate temperature change data of the aluminum circuit board; Step S22: Perform dynamic temperature rise fitting on the temperature change data of the aluminum circuit board to construct a dynamic temperature rise curve; Step S23: reconstructing the temperature rise distribution trend of the dynamic temperature rise curve based on the temperature distribution data of the aluminum circuit board, and constructing a temperature rise distribution trend diagram; Step S24: performing temperature rise distribution deviation detection on the temperature rise distribution trend diagram based on a preset standard circuit board temperature change diagram to obtain temperature rise distribution deviation data; Step S25: performing coating uniformity anomaly analysis on the temperature rise distribution deviation data, thereby obtaining coating uniformity anomaly data; Step S26: locating the abnormal area of ​​the coating uniformity abnormal data based on the real-time aluminum circuit board thermal imaging view, thereby obtaining the coating abnormal area; Step S3: Perform multi-directional ultrasonic pulse detection on the coating abnormal area to collect multi-array sound wave signals; perform low-pass filtering on the multi-array sound wave signals, and then perform three-dimensional projection reconstruction to construct a three-dimensional sound wave projection map; the specific steps of step S3 are: Step S31: performing multi-directional ultrasonic pulse detection on the coating abnormal area according to the multi-array ultrasonic probe array to collect multi-array sound wave signals; Step S32: low-pass filtering the multi-array sound wave signal to obtain a filtered noise-reduced sound wave signal; Step S33: performing multi-scale wavelet decomposition on the filtered and noise-reduced wave signal to obtain frequency domain data of multiple scales; Step S34: Calculate the frequency components of each scale on the frequency domain data of multiple scales, and extract the frequency components of each scale; Step S35: optimizing the mean signal-to-noise ratio of the frequency domain data of multiple scales based on the frequency components of each scale, thereby obtaining multiple signal-to-noise ratio optimized frequency domain data; Step S36: performing multi-sound wave fusion reconstruction on multiple signal-to-noise ratio optimized frequency domain data to construct a high signal-to-noise ratio sound wave signal; Step S37: reshape the high signal-to-noise ratio sound wave signal by three-dimensional projection to construct a three-dimensional sound wave projection image; The specific steps of step S37 are: Calculate the spatial position coordinates of each probe of the multi-array ultrasonic probe array; Perform sound wave reflection delay calculation on the high signal-to-noise ratio sound wave signal to obtain the delay value of each sound wave signal; Perform three-dimensional acoustic wave positioning on the delay value of each acoustic wave signal based on the spatial position coordinates of each probe, thereby obtaining multiple acoustic wave interaction points of the aluminum circuit board; Calculate the spatial relative positions of multiple acoustic wave interaction points on aluminum circuit boards to obtain the spatial relative position coordinates of the interaction points; Reshape the spatial relative position coordinates of the interaction points by three-dimensional projection to construct a three-dimensional sound wave projection map; Step S4: Perform attenuation inversion of the sound wave penetrating the circuit board on the three-dimensional sound wave projection image, and then perform surface defect analysis to obtain the surface defect area of ​​the aluminum circuit board; the specific steps of step S4 are: Step S41: performing acoustic wave semantic analysis on the three-dimensional acoustic wave projection image to generate acoustic wave semantic feature data; Step S42: Calculate the signal amplitude of the acoustic wave semantic feature data to obtain the acoustic wave signal amplitude; Step S43: performing sound wave frequency distribution analysis on the three-dimensional sound wave projection image to generate frequency distribution data; Step S44: performing attenuation inversion of the sound wave penetrating the circuit board based on the sound wave signal amplitude and frequency distribution data, thereby obtaining sound wave attenuation characteristic data; Step S45: performing attenuation reflection uniformity abnormality identification on the sound wave attenuation characteristic data, and extracting sound wave attenuation reflection abnormality data; Step S46: performing surface defect analysis on the three-dimensional acoustic wave projection image based on the acoustic wave attenuation reflection abnormality data, thereby obtaining the surface defect area of ​​the aluminum circuit board; Step S5: Deeply explore the defect diffusion of the defect area on the surface of the aluminum circuit board, and perform three-dimensional defect structure evolution to build a three-dimensional defect structure model; the specific steps of step S5 are: Step S51: performing high-precision repeated acoustic wave scanning on the surface defect area of ​​the aluminum circuit board to extract high-precision acoustic wave scanning signals; Step S52: performing an acoustic wave propagation path distribution analysis on the high-precision acoustic wave scanning signal to generate surface defect propagation path distribution data; Step S53: performing defect morphological feature analysis on the surface defect propagation path distribution data to generate surface defect morphological feature data; Step S54: performing deep mining of defect diffusion on the high-precision acoustic wave scanning signal to obtain the defect depth diffusion law; Step S55: performing three-dimensional defect structure evolution on the surface defect morphological feature data based on the defect depth diffusion law, and constructing a three-dimensional defect structure model; Step S6: Perform a deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data; and finally generate the quality inspection and evaluation results of the aluminum circuit board to be inspected. The specific steps of step S6 are: Step S61: performing defect type cluster analysis on the three-dimensional defect structure model to obtain surface defect types; Step S62: performing defect propagation risk analysis on the surface defect type, thereby obtaining surface defect propagation risk data; Step S63: performing a deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data; Step S64: quantifying the repair difficulty of the deep defect repair simulation response data to generate a defect repair difficulty value; Step S65: Perform a comprehensive surface quality inspection evaluation on the surface defect propagation risk data and the defect repair difficulty value to generate a quality inspection evaluation result of the aluminum circuit board to be inspected.

2. A surface treatment quality inspection system for aluminum circuit boards, characterized in that: Used to perform the aluminum circuit board surface treatment quality detection method as claimed in claim 1, comprising: The temperature distribution module obtains the real-time thermal imaging view of the aluminum circuit board; performs multi-point sampling calculation on the real-time thermal imaging view of the aluminum circuit board, and performs temperature distribution analysis to generate the temperature distribution data of the aluminum circuit board; The temperature rise trend module reconstructs the temperature rise distribution trend based on the temperature distribution data of the aluminum circuit board and locates the abnormal area to obtain the coating abnormal area; The three-dimensional projection module performs multi-directional ultrasonic pulse detection on the coating abnormal area and collects multi-array sound wave signals; performs low-pass filtering on the multi-array sound wave signals, and then performs three-dimensional projection reconstruction to construct a three-dimensional sound wave projection map. The surface defect analysis module performs attenuation inversion of the sound wave penetrating the circuit board on the three-dimensional sound wave projection image, and then performs surface defect analysis to obtain the surface defect area of ​​the aluminum circuit board; The three-dimensional defect structure module conducts deep mining of defect diffusion in the defect area on the surface of aluminum circuit boards, performs three-dimensional defect structure evolution, and constructs a three-dimensional defect structure model; The defect repair simulation module performs deep defect repair simulation on the three-dimensional defect structure model to generate deep defect repair simulation response data; and finally generates the quality inspection and evaluation results of the aluminum circuit board to be inspected.

Images