CN-122016767-A - High-purity quartz tube online detection method and system

CN122016767ACN 122016767 ACN122016767 ACN 122016767ACN-122016767-A

Abstract

The invention relates to the technical field of quartz material detection, in particular to a high-purity quartz tube online detection method and system, wherein the method comprises the steps of constructing an online detection platform, acquiring appearance images, internal defect signals and component content data of the high-purity quartz tube in real time through a multi-dimensional sensing unit, and sending the appearance images, the internal defect signals and the component content data to a data processing center; the data processing center preprocesses the acquired data to eliminate environmental interference and equipment errors, performs defect identification, purity analysis and dimensional accuracy detection on the preprocessed data based on a deep learning model and a feature fusion algorithm, generates a quartz tube quality assessment report, and performs real-time marking and sorting control on unqualified products. According to the invention, by integrating the multi-sensing detection technology, the intelligent data analysis algorithm and the online closed-loop control logic, the real-time monitoring, the accurate detection and the quick response in the production process of the high-purity quartz tube are realized, and the product quality qualification rate and the production efficiency of the high-purity quartz tube are obviously improved.

Inventors

SUN GUOFANG

Assignees

连云港市盛昌照明电器有限公司

Dates

Publication Date: 20260512
Application Date: 20260209

Claims (8)

1. The online detection method of the high-purity quartz tube is characterized by comprising the following steps of: step S10, an online detection platform is built, a multi-dimensional sensing unit is deployed at a key detection node of a quartz tube production line, appearance image data, an internal defect sensing signal and chemical component content data of a high-purity quartz tube are collected in real time, and the collected data are synchronously transmitted to a data processing center through a communication link; Step S20, after the data processing center receives the data, respectively performing targeted preprocessing on the detection data of different types, eliminating environmental interference, equipment system errors and data redundancy, and obtaining standardized detection data; S30, carrying out fusion analysis on appearance image data and internal defect signals in standardized detection data based on a deep learning model, identifying surface defects, internal cracks and bubbles of a quartz tube, detecting and judging the purity of impurity element content in the quartz tube by combining a laser-induced breakdown spectroscopy technology with characteristic spectrum analysis, and detecting the sizes of the inner diameter, the outer diameter, the wall thickness and the straightness of the quartz tube based on a machine vision measurement algorithm; And S40, integrating the defect identification result, the purity detection data and the dimensional accuracy data by the data processing center, generating a real-time quality evaluation report according to a preset high-purity quartz tube quality standard, triggering an online sorting signal for the unqualified products, controlling a sorting device to finish automatic separation of the unqualified products, and synchronizing the detection data and the quality report to a production management and control platform.
2. The method for on-line detection of a high purity quartz tube according to claim 1, wherein the multi-dimensional sensing unit in the step S10 comprises an appearance detection component, an internal defect detection component and a component detection component, wherein the appearance detection component comprises an industrial camera, an annular LED light source and a polarized light filter, the internal defect detection component comprises an ultrasonic flaw detection sensor, an infrared thermal imager and a laser interferometer, the ultrasonic flaw detection sensor is used for detecting positions, sizes and forms of cracks and bubbles in the quartz tube by transmitting and receiving ultrasonic signals, the infrared thermal imager is used for detecting abnormal heat conduction caused by defects in the quartz tube based on amplitude and phase changes of ultrasonic reflection signals, the laser interferometer is used for forming interference fringes by transmitting laser beams, the internal refractive index nonuniform region of the quartz tube is analyzed based on the degree of fringe distortion, the internal bubbles, impurity agglomeration and hidden crack defects are identified, and the component detection component is a laser induced breakdown spectrum detector used for exciting plasma in a micro-area of the surface of the quartz tube, and characteristic spectrum data are acquired.
3. The method for online detection of high purity quartz tubes according to claim 1, wherein the step of performing targeted pretreatment on different types of detection data in step S20 includes: appearance image preprocessing, namely denoising, image enhancement and interested region extraction are carried out on appearance images acquired by an industrial camera; the sensor signal preprocessing, namely filtering, trending and peak value extraction are carried out on the signals acquired by the ultrasonic flaw detection sensor; and (3) preprocessing spectrum data, namely performing background subtraction, spectral line smoothing and normalization processing on the laser-induced breakdown spectrum data.
4. The method according to claim 1, wherein the step of performing fusion analysis on the appearance image data and the internal defect signal in the standardized detection data based on the deep learning model in the step S30 comprises: Extracting defect characteristics, namely extracting characteristics of the preprocessed appearance image based on an improved YOLOv deep learning model, identifying scratches, pits and dirt defects on the surface of the quartz tube, extracting the position, size, shape and gray scale characteristics of the defects, fusing the preprocessed ultrasonic signals and infrared thermal imaging data, and extracting ultrasonic reflection intensity, depth and heat conduction abnormal region area characteristic parameters of the internal defects; Constructing a multi-feature fusion classifier, fusing surface defect features and internal defect features, classifying defect types through a support vector machine algorithm, setting a defect judgment threshold, and judging that the defect exceeds the threshold as exceeding the standard; the purity detection, namely extracting the characteristic spectral line intensity of impurity elements in a quartz tube based on laser-induced breakdown spectroscopy data, and calculating the content of each impurity element by a calibration curve method; And the dimensional accuracy detection is to calculate the inner diameter, the outer diameter, the wall thickness and the straightness of the quartz tube by combining distance data acquired by a laser interferometer through the outline edge of the quartz tube in the appearance image based on a machine vision measurement algorithm.
5. The method for online detection of the high-purity quartz tube according to claim 4, wherein the establishment process of the calibration curve method comprises the steps of preparing a series of quartz standard samples with known impurity element contents, collecting characteristic spectral line intensities of the standard samples through a laser-induced breakdown spectroscopy detector, taking the impurity element content as an abscissa and the characteristic spectral line intensities as an ordinate, and obtaining the calibration curve by adopting a least square fitting method.
6. An on-line detection system for a high purity quartz tube, the system performing the on-line detection method for a high purity quartz tube according to claim 1, comprising: The online detection platform construction module is used for constructing an online detection platform, deploying a multi-dimensional sensing unit at a key detection node of a quartz tube production line, acquiring appearance image data, an internal defect sensing signal and chemical component content data of the high-purity quartz tube in real time, and synchronously transmitting the acquired data to the data processing center through a communication link; the data preprocessing module is used for respectively carrying out targeted preprocessing on the detection data of different types after the data processing center receives the data, eliminating environmental interference, equipment system errors and data redundancy and obtaining standardized detection data; The multi-dimensional detection analysis module is used for carrying out fusion analysis on appearance image data and internal defect signals in the standardized detection data based on a deep learning model, identifying surface defects, internal cracks and bubbles of the quartz tube, carrying out detection and purity judgment on the content of impurity elements in the quartz tube by combining a laser-induced breakdown spectroscopy technology with characteristic spectrum analysis, and detecting the sizes of the inner diameter, the outer diameter, the wall thickness and the straightness of the quartz tube based on a machine vision measurement algorithm; The quality evaluation and sorting control module is used for integrating defect identification results, purity detection data and dimensional accuracy data in the data processing center, generating a real-time quality evaluation report according to a preset high-purity quartz tube quality standard, triggering an online sorting signal for unqualified products, controlling the sorting device to complete automatic separation of the unqualified products, and synchronizing the detection data and the quality report to the production management and control platform.
7. The utility model provides a high purity quartz capsule on-line measuring equipment which characterized in that includes: A memory, a processor and a high purity quartz tube online detection program stored on the memory and executable on the processor, which when executed by the processor, implements a high purity quartz tube online detection method as claimed in any one of claims 1 to 5.
8. A computer program product, characterized in that it comprises a high purity quartz tube online detection program which, when executed by a processor, implements a high purity quartz tube online detection method according to any of claims 1 to 5.

Description

High-purity quartz tube online detection method and system Technical Field The invention relates to the technical field of quartz material detection, in particular to a high-purity quartz tube online detection method and system. Background The high-purity quartz tube has excellent performances of high temperature resistance, corrosion resistance, good light transmittance, strong chemical stability and the like, and is widely applied to high-end fields such as semiconductor chip manufacture, optical instrument processing, photovoltaic cell production and the like. The application scene has extremely high requirements on the purity (the total impurity content is generally less than or equal to 5 ppm), the appearance quality (no obvious scratches and pits), the internal defects (no cracks and bubbles) and the dimensional accuracy (the uniformity of the inner diameter, the outer diameter and the wall thickness, etc.) of the quartz tube. At present, the detection method of the high-purity quartz tube has a plurality of defects: The detection mode is mainly off-line sampling detection, so that real-time monitoring in the production process cannot be realized, a large number of unqualified products flow into subsequent procedures, and production loss and cost are increased; The appearance detection mainly depends on manual visual inspection, has strong subjectivity and low efficiency, is difficult to find micro scratches (less than or equal to 0.1 mm) and fine stains, and has high omission rate; the internal defect detection mostly adopts a single ultrasonic flaw detection technology, has insufficient recognition precision on micro bubbles (less than or equal to 0.3 mm) and shallow surface cracks, and is easy to be interfered by environmental noise; the purity detection mainly adopts an inductively coupled plasma mass spectrometry (ICP-MS), the detection period is long (more than 30 minutes are needed for single detection), and the requirement of on-line rapid detection cannot be met; the size precision detection adopts contact measuring tool more, easily causes the damage to quartz tube surface, and measurement efficiency is low, can't match with production line speed. Therefore, a technical scheme for realizing real-time, accurate, multidimensional and non-contact online detection of the high-purity quartz tube is needed, so as to solve the problems of lag off-line detection, low detection precision, poor efficiency, easy damage to products and the like in the existing detection method. Disclosure of Invention Aiming at the technical defects, the invention provides a high-purity quartz tube online detection method and a high-purity quartz tube online detection system, which realize real-time, accurate and nondestructive detection in the production process of the high-purity quartz tube by integrating a multi-sensor detection technology, an intelligent data analysis algorithm and online closed-loop control logic. The invention is realized by the following technical scheme: the method for detecting the high-purity quartz tube on line comprises the following steps: Step S10, an online detection platform is built, a multi-dimensional sensing unit is deployed at a key detection node of a quartz tube production line, appearance image data, an internal defect sensing signal and chemical component content data of a high-purity quartz tube are collected in real time, and the collected data are synchronously transmitted to a data processing center through a high-speed communication link; Step S20, after the data processing center receives the data, respectively performing targeted preprocessing on the detection data of different types, eliminating environmental interference, equipment system errors and data redundancy, and obtaining standardized detection data; S30, carrying out fusion analysis on appearance image data and internal defect signals in standardized detection data based on a deep learning model, identifying surface defects, internal cracks and bubbles of a quartz tube, detecting and judging the purity of impurity element content in the quartz tube by combining a laser-induced breakdown spectroscopy technology with characteristic spectrum analysis, and detecting the sizes of the inner diameter, the outer diameter, the wall thickness and the straightness of the quartz tube based on a machine vision measurement algorithm; And S40, integrating the defect identification result, the purity detection data and the dimensional accuracy data by the data processing center, generating a real-time quality evaluation report according to a preset high-purity quartz tube quality standard, triggering an online sorting signal for the unqualified products, controlling a sorting device to finish automatic separation of the unqualified products, and synchronizing the detection data and the quality report to a production management and control platform. Preferably, the multi-dimensional sensing unit in the step S10 comprises an ap