Spectral analysis monitoring method for aluminum alloy component detection

Abstract

The invention relates to the technical field of metal materials, and discloses a spectral analysis monitoring method for aluminum alloy component detection, and the method comprises the following steps: carrying out standardized surface pretreatment on an aluminum alloy sample; collecting spectral data of the sample by adopting a laser-induced breakdown spectroscopy technology; performing noise filtering and background correction on the spectral data through a wavelet transform algorithm; establishing a spectral feature database for element qualitative identification; carrying out element quantitative analysis by adopting an internal standard method, and calculating the content of each element through a calibration curve; comparing a detection result with a standard mark component range to generate component conformity evaluation; performing deviation analysis and performance prediction on samples which do not meet the standard; and finally, generating a quality grading report based on the multi-dimensional evaluation indexes. According to the method, the accuracy of component qualitative identification is ensured, the detection error is reduced, the repeatability and efficiency of detection are improved, and the effectiveness of risk management and control and the overall reliability of a detection result are improved.

Description

Technical Field

[0001] This invention relates to the field of metallic materials technology, specifically to a spectral analysis monitoring method for detecting the composition of aluminum alloys. Background Technology

[0002] Metallic materials refer to metallic elements and materials with metallic properties that are mainly composed of metallic elements. They include pure metals, alloys, intermetallic compounds, and special metallic materials. Common examples include iron, copper, aluminum, tin, nickel, gold, silver, lead, and zinc. Different metallic materials require performance testing after processing and synthesis. The testing scope of metallic materials covers mechanical property testing, chemical composition analysis, metallographic analysis, precision dimensional measurement, non-destructive testing, corrosion resistance testing, and environmental simulation testing of ferrous metals, non-ferrous metals, machinery and equipment, and parts.

[0003] Currently, due to the complex production process and diverse composition of aluminum alloys, the spectral analysis equipment used for rapid analysis of aluminum alloy composition excites the sample surface and cannot monitor in real time whether there are fluctuations in the plasma state during the spectral acquisition process. If the plasma excitation is unstable, it may cause a decrease in the signal-to-noise ratio of the spectral signal, making it impossible to guarantee the accuracy of composition quantification.

[0004] Therefore, a spectral analysis monitoring method for aluminum alloy composition detection is proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a spectral analysis and monitoring method for aluminum alloy composition detection, which solves the problems mentioned in the background art, such as the inability to monitor plasma state fluctuations during spectral acquisition in real time and the inability to guarantee the accuracy of composition quantification.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a spectral analysis monitoring method for aluminum alloy composition detection, the method comprising the following steps:

[0007] S1. Prepare aluminum alloy samples and perform surface pretreatment to obtain standardized test samples;

[0008] S2. The sample to be tested is subjected to spectral acquisition using a laser-induced breakdown spectroscopy device to obtain raw spectral data;

[0009] S3. Preprocess the original spectral data to reduce noise and background interference and obtain optimized spectral data;

[0010] S4. Based on the optimized spectral data, perform qualitative identification of aluminum alloy composition using a preset spectral analysis model to generate elemental composition data;

[0011] S5. Based on the elemental composition data, calculate the content of each element using a quantitative analysis algorithm to generate component content data;

[0012] S6. Compare the component content data with the component range of standard aluminum alloy grades to generate component compliance evaluation data;

[0013] S7. When the component conformity evaluation data does not meet the preset standard, perform component deviation analysis processing to generate component deviation report data.

[0014] S8. Based on the component content data and the component deviation report data, perform aluminum alloy performance prediction processing to generate performance prediction data;

[0015] S9. Based on the composition conformity evaluation data, the composition deviation report data, and the performance prediction data, perform aluminum alloy quality grading evaluation processing to generate quality grade data;

[0016] S10. Integrate component content, component compliance evaluation, component deviation report, performance prediction and quality grade data to generate aluminum alloy component test report data.

[0017] Preferably, the process of preparing an aluminum alloy sample and performing surface pretreatment in step S1 to obtain a standardized test sample includes the following steps:

[0018] S11. Take samples from aluminum alloy products and use cutting equipment to obtain sample blocks of uniform size;

[0019] S12. The sample block is subjected to surface polishing treatment by using sandpaper of different grit sizes to polish it step by step until the surface smoothness reaches the preset standard.

[0020] S13. Use an ultrasonic cleaner to clean the polished sample to remove residual abrasive and contaminants from the surface;

[0021] S14. The cleaned sample is dried using a drying device to obtain a test sample with a clean surface and no oxide layer.

[0022] S15. Perform surface roughness detection on the sample to be tested. When the roughness value is lower than the preset threshold, the sample is deemed qualified and proceeds to the spectral acquisition step.

[0023] Preferably, the process of acquiring the original spectral data of the sample under test using a laser-induced breakdown spectroscopy device in step S2 includes the following steps:

[0024] S21. Place the sample to be tested on the spectrometer sample stage and adjust the distance and angle between the laser and the sample;

[0025] S22. Set laser parameters, including laser energy, pulse width, and focusing position, to ensure consistency of excitation conditions;

[0026] S23. Laser excitation is performed in an inert gas protective environment to generate plasma and collect the emission spectrum;

[0027] S24. Use a high-resolution spectrometer to collect plasma emission spectra and obtain raw spectral data containing characteristic spectral lines of multiple elements;

[0028] S25. Record environmental parameters during the data collection process, including temperature, humidity, and air pressure data, for subsequent data correction.

[0029] Preferably, the process of preprocessing the original spectral data in step S3 to reduce noise and background interference and obtain optimized spectral data includes the following steps:

[0030] S31. Perform baseline correction on the original spectral data to reduce the influence of continuous background radiation;

[0031] S32. Wavelet transform algorithm is used to perform noise filtering on spectral data to improve the signal-to-noise ratio. The wavelet transform formula is as follows:

[0032] ;

[0033] in, These are wavelet coefficients, representing the quantized feature values ​​under the scale parameter a and the translation parameter b. The original spectral signal is discretized and normalized to the range [0, 1], where n is the index of the sampling point. For signal length, The wavelet function is generated based on the scale parameter a and the translation parameter b. It is used to extract local features. The preset noise threshold is 0.05. When the absolute value of the wavelet coefficient is less than 0.05, it is regarded as noise and set to zero. This formula achieves noise filtering through convolution operation and retains useful spectral information.

[0034] S33. Perform spectral intensity normalization to reduce the impact of laser energy fluctuations;

[0035] S34. Perform instrument response correction using the spectral data of standard samples to obtain accurate spectral intensity data;

[0036] S35. Perform peak identification and spectral line matching on the corrected spectral data to obtain optimized spectral data.

[0037] Preferably, the process of generating elemental composition data by performing qualitative identification of aluminum alloy composition based on the optimized spectral data and using a preset spectral analysis model in step S4 includes the following steps:

[0038] S41. Establish a spectral characteristic database of common elements in aluminum alloys, including the wavelength and intensity information of the main characteristic spectral lines of each element;

[0039] S42. Match the optimized spectral data with the spectral feature database to identify existing elements;

[0040] S43. Use principal component analysis to verify the identification results and reduce false identifications;

[0041] S44. Based on the element identification results, generate elemental composition data containing major elements and trace elements;

[0042] S45. Re-analyze elements whose confidence level is lower than the preset threshold to ensure the accuracy of element identification.

[0043] Preferably, the process of calculating the content of each element and generating component content data in step S5 based on the elemental composition data using a quantitative analysis algorithm includes the following steps:

[0044] S51. The elemental content is calculated using the internal standard method, and elements with high stability in aluminum alloys are selected as internal standard elements.

[0045] S52. Establish calibration curves and build a model of the relationship between element content and spectral intensity using spectral data from multiple standard samples.

[0046] S53. Based on the optimized spectral data and the calibration curve, calculate the content of each element, wherein the internal standard method calculation formula is:

[0047] ;

[0048] in, The concentration of the analyte element i is expressed as a percentage. The concentration of the internal standard element s is expressed as a percentage and is usually a known constant. The intensity of the characteristic spectral line of the element i to be measured is given, and the unit is arbitrary. The intensity of the characteristic spectral line of the internal standard element s is given, in arbitrary units. The calibration coefficient is obtained by fitting a standard sample. This formula reduces errors and improves quantitative accuracy through ratio calculation.

[0049] S54. Verify the calculation results using the standard addition method to ensure the accuracy of quantitative analysis;

[0050] S55. Generate component content data including the content of each element and the measurement uncertainty.

[0051] Preferably, the process of comparing the component content data with the standard aluminum alloy grade component range in step S6 to generate component conformity evaluation data includes the following steps:

[0052] S61. Establish a standard database of aluminum alloy grades and compositions, including the element content range requirements for each grade of aluminum alloy.

[0053] S62. Compare the component content data with the component range of the target brand one by one;

[0054] S63. Calculate the deviation of the content of each element from the standard range and generate deviation analysis data;

[0055] S64. Determine the component compliance based on the degree of deviation and generate component compliance evaluation data;

[0056] S65. Mark elements that exceed the allowed range and generate a list of elements to focus on.

[0057] Preferably, the process of performing component deviation analysis and generating component deviation report data when the component conformity evaluation data in step S7 does not meet the preset standard includes the following steps:

[0058] S71. When the component conformity evaluation data does not meet the preset standard, analyze the reasons for the component deviation.

[0059] S72. Distinguish between systematic bias and random bias, and determine the nature of the bias;

[0060] S73. Evaluate the extent to which compositional deviations affect the properties of aluminum alloys;

[0061] S74. Generate component deviation report data that includes the cause of the deviation, the degree of impact, and improvement suggestions;

[0062] S75. Depending on the severity of the deviation, it is recommended whether to remelt and adjust the production process.

[0063] Preferably, the process of generating performance prediction data for aluminum alloy based on the component content data and the component deviation report data in step S8 includes the following steps:

[0064] S81. Establish a model for the relationship between aluminum alloy composition and properties, including prediction models for mechanical properties, corrosion properties, and processing properties;

[0065] S82. Input the component content data into the performance relationship model to predict various performance indicators of the aluminum alloy;

[0066] S83. Consider the impact of component deviation on performance and correct the prediction results;

[0067] S84. Generate performance prediction data that includes mechanical properties, physical properties, and process performance indicators;

[0068] S85. Provide early warnings and suggestions regarding potential performance defects.

[0069] Preferably, the process of performing aluminum alloy quality grading evaluation processing and generating quality grade data based on the composition conformity evaluation data, the composition deviation report data, and the performance prediction data in step S9 includes the following steps:

[0070] S91. Establish aluminum alloy quality grading standards, including multiple dimensions such as composition accuracy, performance compliance rate, and degree of deviation;

[0071] S92. A comprehensive evaluation is performed based on the component compliance evaluation data, the component deviation report data, and the performance prediction data;

[0072] S93. The quality score is calculated using the fuzzy comprehensive evaluation method.

[0073] S94. Aluminum alloy products are classified into grades such as high-quality, qualified, and unqualified based on their quality scores.

[0074] S95. Generate quality grade data that includes quality grade and evaluation criteria.

[0075] Compared with existing technologies, this invention provides a spectral analysis monitoring method for aluminum alloy composition detection, which has the following beneficial effects:

[0076] 1. In this invention, by employing spectral acquisition technology, the uniform treatment of the sample surface and the consistency of high-precision excitation conditions are ensured during aluminum alloy composition detection. This enables real-time monitoring and timely adjustment of plasma state fluctuations, reducing signal interference and ensuring the stability and reliability of elemental characteristic spectral data. Consequently, the accuracy of quantitative composition analysis is improved, and detection errors are reduced.

[0077] 2. In this invention, by implementing a correction mechanism, when performing spectral data analysis and element identification, it is possible to determine in real time whether the characteristic spectral lines are subject to overlapping interference and automatically perform data compensation, thereby reducing the problem of element identification errors. Furthermore, it can quickly correct interference when it occurs, ensuring the accuracy of qualitative identification of components and further improving the repeatability and efficiency of detection.

[0078] 3. In this invention, through graded evaluation processing, simultaneous unbiased analysis of major and trace elements is achieved when predicting the performance and comprehensively evaluating the quality of aluminum alloys. Accurate risk quantification and grading are performed based on multidimensional indicators, and comprehensive evaluation reports are generated for different aluminum alloy grades, reducing systematic bias and improving the effectiveness of risk control and the overall reliability of test results. Attached Figure Description

[0079] Figure 1 This is a flowchart of the steps of the spectral analysis monitoring method for detecting aluminum alloy composition according to the present invention. Detailed Implementation

[0080] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0081] Please see Figure 1 The specific implementation of the spectral analysis monitoring method for aluminum alloy composition detection is as follows, and the method includes the following steps:

[0082] S1. Prepare aluminum alloy samples and perform surface pretreatment to obtain standardized test samples;

[0083] S2. Use a laser-induced breakdown spectroscopy device to acquire the spectrum of the sample under test and obtain the raw spectral data;

[0084] S3. Preprocess the raw spectral data to reduce noise and background interference and obtain optimized spectral data;

[0085] S4. Based on optimized spectral data, perform qualitative identification of aluminum alloy composition using a preset spectral analysis model to generate elemental composition data;

[0086] S5. Based on the elemental composition data, use quantitative analysis algorithms to calculate the content of each element and generate component content data;

[0087] S6. Compare the component content data with the component range of standard aluminum alloy grades to generate component compliance evaluation data.

[0088] S7. When the component conformity evaluation data does not meet the preset standard, perform component deviation analysis and generate component deviation report data.

[0089] S8. Based on the component content data and component deviation report data, perform aluminum alloy performance prediction processing to generate performance prediction data;

[0090] S9. Based on the composition conformity evaluation data, composition deviation report data, and performance prediction data, perform aluminum alloy quality grading evaluation processing to generate quality grade data.

[0091] S10. Integrate component content, component compliance evaluation, component deviation report, performance prediction and quality grade data to generate aluminum alloy component test report data.

[0092] In S1, the process of preparing an aluminum alloy sample and performing surface pretreatment to obtain a standardized sample to be tested includes the following steps:

[0093] S11. Cut a sample from the aluminum alloy product and obtain a sample block with uniform size using cutting equipment;

[0094] S12. Perform surface grinding on the sample block and gradually grind it with sandpapers of different grits until the surface finish reaches the preset standard;

[0095] S13. Clean the polished sample using an ultrasonic cleaner to remove residual abrasives and contaminants on the surface;

[0096] S14. Dry the cleaned sample using a drying device to obtain a sample to be tested with a clean surface and no oxide layer;

[0097] S15. Detect the surface roughness of the sample to be tested. Among them, the roughness formula is normalized to a dimensionless form:

[0098] ;

[0099] Among them, is the normalized roughness index, with a range of 0 - 1, is the evaluation length, with the unit of millimeter, is the profile deviation function, is the maximum reference roughness, and the preset roughness threshold is 0.8. When , the sample is determined to be qualified;

[0100] When the roughness value is lower than the preset threshold, the sample is determined to be qualified and enters the spectral acquisition step.

[0101] In S2, the process of using a laser-induced breakdown spectroscopy device to perform spectral acquisition on the sample to be tested and obtain the original spectral data includes the following steps:

[0102] S21. Place the sample to be tested on the sample stage of the spectrometer and adjust the distance and angle between the laser and the sample;

[0103] S22. Set the laser parameters, including laser energy, pulse width, and focusing position, to ensure the consistency of the excitation conditions;

[0104] S23. Perform laser excitation under an inert gas protection environment to generate plasma and collect the emission spectrum;

[0105] S24. Use a high-resolution spectrometer to collect the plasma emission spectrum. Among them, the plasma stability formula is:

[0106] ;

[0107] in, The plasma stability index, This represents the standard deviation of spectral intensity, in au, normalized to relative intensity. The spectral intensity is the mean value in au, normalized to relative intensity, with a preset stability threshold of 0.15. The data is then determined to be valid.

[0108] Obtain raw spectral data containing characteristic spectral lines of multiple elements;

[0109] S25. Record environmental parameters during the data collection process, including temperature, humidity, and air pressure data, for subsequent data correction.

[0110] The process of preprocessing raw spectral data in S3 to reduce noise and background interference and obtain optimized spectral data includes the following steps:

[0111] S31. Perform baseline correction on the original spectral data to reduce the influence of continuous background radiation;

[0112] S32. Wavelet transform algorithm is used to perform noise filtering on spectral data to improve the signal-to-noise ratio. The wavelet transform formula is as follows:

[0113] ;

[0114] in, These are wavelet coefficients, representing the quantized feature values ​​under the scale parameter a and the translation parameter b. The original spectral signal is discretized and normalized to the range [0, 1], where n is the index of the sampling point. For signal length, The wavelet function is generated based on the scale parameter a and the translation parameter b. It is used to extract local features. The preset noise threshold is 0.05. When the absolute value of the wavelet coefficient is less than 0.05, it is regarded as noise and set to zero. This formula achieves noise filtering through convolution operation and retains useful spectral information.

[0115] S33. Perform spectral intensity normalization to reduce the impact of laser energy fluctuations;

[0116] S34. Perform instrument response correction using the spectral data of standard samples to obtain accurate spectral intensity data;

[0117] S35. Perform peak identification and spectral line matching on the corrected spectral data to obtain optimized spectral data.

[0118] In S4, the process of qualitatively identifying the composition of aluminum alloys and generating elemental composition data based on optimized spectral data and a preset spectral analysis model includes the following steps:

[0119] S41. Establish a spectral characteristic database of common elements in aluminum alloys, including the wavelength and intensity information of the main characteristic spectral lines of each element;

[0120] S42. Match the optimized spectral data with the spectral feature database to identify the existing elements;

[0121] S43. Use principal component analysis to verify the identification results and reduce false identifications;

[0122] S44. Based on the element identification results, generate elemental composition data containing major and trace elements, wherein the element confidence formula is normalized to the range [0, 1].

[0123] ;

[0124] in, Confidence level for element identification, ranging from 0 to 1. The measured spectral line wavelengths are in nm. The standard wavelength for the database is in nm. The wavelength normalization factor is set to a preset signal threshold of 0.95. Determine if the element exists;

[0125] S45. Re-analyze elements whose confidence level is lower than the preset threshold to ensure the accuracy of element identification.

[0126] The process of calculating the content of each element and generating component content data in S5 based on elemental composition data using quantitative analysis algorithms includes the following steps:

[0127] S51. The elemental content is calculated using the internal standard method, and elements with high stability in aluminum alloys are selected as internal standard elements.

[0128] S52. Establish calibration curves and build a model of the relationship between element content and spectral intensity using spectral data from multiple standard samples.

[0129] S53. Based on the optimized spectral data and calibration curves, calculate the content of each element. The formula for the internal standard method is as follows:

[0130] ;

[0131] in, The concentration of the analyte element i is expressed as a percentage. The concentration of the internal standard element s is expressed as a percentage and is usually a known constant. The intensity of the characteristic spectral line of the element i to be measured is given, and the unit is arbitrary. The intensity of the characteristic spectral line of the internal standard element s is given, in arbitrary units. The calibration coefficient is obtained by fitting a standard sample. This formula reduces errors and improves quantitative accuracy through ratio calculation.

[0132] S54. Verify the calculation results using the standard addition method to ensure the accuracy of quantitative analysis;

[0133] S55. Generate component content data including the content of each element and the measurement uncertainty.

[0134] The process of comparing the composition content data with the composition range of standard aluminum alloy grades in S6 to generate composition conformity evaluation data includes the following steps:

[0135] S61. Establish a standard database of aluminum alloy grades and compositions, including the element content range requirements for each grade of aluminum alloy.

[0136] S62. Compare the component content data with the component range of the target brand one by one;

[0137] S63. Calculate the deviation of the content of each element from the standard range and generate deviation analysis data;

[0138] S64. Determine the component compliance based on the degree of deviation and generate component compliance evaluation data;

[0139] S65. Mark elements that exceed the allowed range and generate a list of elements to focus on.

[0140] In S7, when the component conformity evaluation data does not meet the preset standard, the process of performing component deviation analysis and generating component deviation report data includes the following steps:

[0141] S71. When the component conformity evaluation data shows that it does not meet the preset standard, analyze the reasons for the component deviation;

[0142] S72. Distinguish between systematic bias and random bias, and determine the nature of the bias;

[0143] S73. Evaluate the impact of compositional deviations on the properties of aluminum alloys, where the performance impact coefficient formula includes the normalized deviation term:

[0144] ;

[0145] in, The comprehensive performance impact coefficient, Let i be the weight factor of element i. The concentration deviation of element i, in %. The standard concentration of element i is expressed as a percentage. As a concentration normalization benchmark, when High-risk warnings are generated in real time;

[0146] S74. Generate component deviation report data that includes the cause of the deviation, the degree of impact, and improvement suggestions;

[0147] S75. Depending on the severity of the deviation, it is recommended whether to remelt and adjust the production process.

[0148] In S8, the process of generating performance prediction data for aluminum alloys based on composition content data and composition deviation report data includes the following steps:

[0149] S81. Establish a model for the relationship between aluminum alloy composition and properties, including prediction models for mechanical properties, corrosion properties, and processing properties;

[0150] S82. Input the component content data into the performance relationship model to predict various performance indicators of aluminum alloys;

[0151] S83. Consider the impact of component deviation on performance and correct the prediction results;

[0152] S84. Generate performance prediction data that includes mechanical properties, physical properties, and process performance indicators;

[0153] S85. Provide early warnings and suggestions regarding potential performance defects.

[0154] The process of generating quality grade data for aluminum alloys in S9, based on composition conformity evaluation data, composition deviation report data, and performance prediction data, includes the following steps:

[0155] S91. Establish aluminum alloy quality grading standards, including multiple dimensions such as composition accuracy, performance compliance rate, and degree of deviation;

[0156] S92. Conduct a comprehensive evaluation based on the component compliance evaluation data, component deviation report data, and performance prediction data;

[0157] S93. The quality score is calculated using the fuzzy comprehensive evaluation method, wherein the quality score formula is normalized to the range [0, 100].

[0158] ;

[0159] in, The final quality score ranges from 0 to 100. To evaluate the membership degree of index j, the range is 0-1. The raw score for evaluation index j, ranging from 0 to 1. The maximum score reference value, preset to 1.0, is used for normalization and is the grading threshold. It is a high-quality product. It is a qualified product. These are substandard products.

[0160] S94. Aluminum alloy products are classified into grades such as high-quality, qualified, and unqualified based on their quality scores.

[0161] S95. Generate quality grade data that includes quality grade and evaluation criteria.

[0162] The operational steps of the spectral analysis monitoring method for aluminum alloy composition detection are as follows:

[0163] Step 1: Sample Standardization Preparation

[0164] Aluminum alloy samples of uniform size are obtained by mechanical cutting, and the surface is polished with multi-level sandpaper, followed by ultrasonic cleaning to remove residual contaminants. Finally, the sample surface is dried to ensure that it is clean and free of oxide layer, providing a homogenized basis for spectral detection.

[0165] Step 2: Plasma Excitation and Spectral Acquisition

[0166] Under inert gas protection, the sample surface is irradiated with a high-energy pulsed laser. The instantaneous high temperature causes the material to vaporize and form plasma. The elemental characteristic spectral lines radiated by the plasma are captured in real time by a high-resolution spectrometer, and the ambient temperature and humidity parameters are recorded simultaneously to correct for environmental interference.

[0167] Step 3: Spectral Intelligent Noise Reduction Processing

[0168] Baseline correction is performed on the original spectral signal to reduce background radiation, and random noise is filtered out by combining wavelet transform algorithm. Then, the instrument deviation is corrected by the spectral response curve of the standard sample to extract pure elemental characteristic spectral line data.

[0169] Step 4: Qualitative and Quantitative Elemental Analysis

[0170] The optimized spectral data were compared with the elemental characteristic wavelength database to identify the main and trace elemental composition of the aluminum alloy. Based on the internal standard method, stable elements were selected as references, and the precise content of each element was calculated through calibration curves to verify the reliability of the measurement results.

[0171] Step 5: Dynamic Verification of Ingredient Compliance

[0172] The system automatically compares the element content data with the standard composition range of the target grade aluminum alloy, generates a deviation analysis report, systematically traces the source of elements exceeding the standard, distinguishes between process deviations and detection errors, and predicts the impact of compositional anomalies on the mechanical properties of the material.

[0173] Step Six: Quality Grading and Decision Output

[0174] Based on the comprehensive composition compliance, deviation risk coefficient, and performance prediction results, a multi-dimensional evaluation model is used to generate a quality grade score. According to the preset threshold, the product is divided into high-quality, qualified, and unqualified grades, and an inspection report containing improvement measures is output to guide production optimization.

[0175] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0176] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A spectral analysis monitoring method for aluminum alloy composition detection, characterized in that: The method includes the following steps: S1. Prepare aluminum alloy samples and perform surface pretreatment to obtain standardized test samples; S2. The sample to be tested is subjected to spectral acquisition using a laser-induced breakdown spectroscopy device to obtain raw spectral data; S3. Preprocess the original spectral data to reduce noise and background interference and obtain optimized spectral data; S4. Based on the optimized spectral data, perform qualitative identification of aluminum alloy composition using a preset spectral analysis model to generate elemental composition data; S5. Based on the elemental composition data, calculate the content of each element using a quantitative analysis algorithm to generate component content data; S6. Compare the component content data with the component range of standard aluminum alloy grades to generate component compliance evaluation data; S7. When the component conformity evaluation data does not meet the preset standard, perform component deviation analysis processing to generate component deviation report data. S8. Based on the component content data and the component deviation report data, perform aluminum alloy performance prediction processing to generate performance prediction data; S9. Based on the composition conformity evaluation data, the composition deviation report data, and the performance prediction data, perform aluminum alloy quality grading evaluation processing to generate quality grade data; S10. Integrate component content, component compliance evaluation, component deviation report, performance prediction and quality grade data to generate aluminum alloy component test report data.

2. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of preparing an aluminum alloy sample and performing surface pretreatment in S1 to obtain a standardized test sample includes the following steps: S11. Take samples from aluminum alloy products and use cutting equipment to obtain sample blocks of uniform size; S12. The sample block is subjected to surface polishing treatment by using sandpaper of different grit sizes to polish it step by step until the surface smoothness reaches the preset standard. S13. Use an ultrasonic cleaner to clean the polished sample to remove residual abrasive and contaminants from the surface; S14. The cleaned sample is dried using a drying device to obtain a test sample with a clean surface and no oxide layer. S15. Perform surface roughness detection on the sample to be tested. When the roughness value is lower than the preset threshold, the sample is deemed qualified and proceeds to the spectral acquisition step.

3. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of acquiring raw spectral data by using a laser-induced breakdown spectroscopy device to perform spectral acquisition on the sample under test in step S2 includes the following steps: S21. Place the sample to be tested on the spectrometer sample stage and adjust the distance and angle between the laser and the sample; S22. Set laser parameters, including laser energy, pulse width, and focusing position, to ensure consistency of excitation conditions; S23. Laser excitation is performed in an inert gas protective environment to generate plasma and collect the emission spectrum; S24. Use a high-resolution spectrometer to collect plasma emission spectra and obtain raw spectral data containing characteristic spectral lines of multiple elements; S25. Record environmental parameters during the data collection process, including temperature, humidity, and air pressure data, for subsequent data correction.

4. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of preprocessing the original spectral data in step S3 to reduce noise and background interference and obtain optimized spectral data includes the following steps: S31. Perform baseline correction on the original spectral data to reduce the influence of continuous background radiation; S32. Wavelet transform algorithm is used to perform noise filtering on spectral data to improve the signal-to-noise ratio. The wavelet transform formula is as follows: ； in, The wavelet coefficients represent the quantized feature values ​​under the scaling parameter a and the translation parameter b. The original spectral signal is discretized and normalized to the range [0, 1], where n is the index of the sampling point. For signal length, The wavelet function is generated based on the scale parameter a and the translation parameter b. It is used to extract local features. The preset noise threshold is 0.

05. When the absolute value of the wavelet coefficient is less than 0.05, it is regarded as noise and set to zero. This formula achieves noise filtering through convolution operation and retains useful spectral information. S33. Perform spectral intensity normalization to reduce the impact of laser energy fluctuations; S34. Perform instrument response correction using the spectral data of standard samples to obtain accurate spectral intensity data; S35. Perform peak identification and spectral line matching on the corrected spectral data to obtain optimized spectral data.

5. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of generating elemental composition data by qualitatively identifying the aluminum alloy composition based on the optimized spectral data and using a preset spectral analysis model in step S4 includes the following steps: S41. Establish a spectral characteristic database of common elements in aluminum alloys, including the wavelength and intensity information of the main characteristic spectral lines of each element; S42. Match the optimized spectral data with the spectral feature database to identify existing elements; S43. Use principal component analysis to verify the identification results and reduce false identifications; S44. Based on the element identification results, generate elemental composition data containing major elements and trace elements; S45. Re-analyze elements whose confidence level is lower than the preset threshold to ensure the accuracy of element identification.

6. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of calculating the content of each element and generating component content data in step S5 based on the elemental composition data using a quantitative analysis algorithm includes the following steps: S51. The elemental content is calculated using the internal standard method, and elements with high stability in aluminum alloys are selected as internal standard elements. S52. Establish calibration curves and build a model of the relationship between element content and spectral intensity using spectral data from multiple standard samples. S53. Based on the optimized spectral data and the calibration curve, calculate the content of each element, wherein the internal standard method calculation formula is: ； in, The concentration of the analyte i is expressed as a percentage. The concentration of the internal standard element s is expressed as a percentage and is usually a known constant. The intensity of the characteristic spectral line of the element i to be measured is given, and the unit is arbitrary. The intensity of the characteristic spectral line of the internal standard element s is given, in any unit. The calibration coefficient is obtained by fitting a standard sample. This formula reduces errors and improves quantitative accuracy through ratio calculation. S54. Verify the calculation results using the standard addition method to ensure the accuracy of quantitative analysis; S55. Generate component content data including the content of each element and the measurement uncertainty.

7. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of comparing the component content data with the standard aluminum alloy grade component range in step S6 to generate component conformity evaluation data includes the following steps: S61. Establish a standard database of aluminum alloy grades and compositions, including the element content range requirements for each grade of aluminum alloy. S62. Compare the component content data with the component range of the target brand one by one; S63. Calculate the deviation of the content of each element from the standard range and generate deviation analysis data; S64. Determine the component compliance based on the degree of deviation and generate component compliance evaluation data; S65. Mark elements that exceed the allowed range and generate a list of elements to focus on.

8. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of performing component deviation analysis and generating component deviation report data when the component conformity evaluation data in step S7 does not meet the preset standard includes the following steps: S71. When the component conformity evaluation data does not meet the preset standard, analyze the reasons for the component deviation. S72. Distinguish between systematic bias and random bias, and determine the nature of the bias; S73. Evaluate the extent to which compositional deviations affect the properties of aluminum alloys; S74. Generate component deviation report data that includes the cause of the deviation, the degree of impact, and improvement suggestions; S75. Depending on the severity of the deviation, it is recommended whether to remelt and adjust the production process.

9. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of generating performance prediction data for aluminum alloy based on the component content data and the component deviation report data in step S8 includes the following steps: S81. Establish a model for the relationship between aluminum alloy composition and properties, including prediction models for mechanical properties, corrosion properties, and processing properties; S82. Input the component content data into the performance relationship model to predict various performance indicators of the aluminum alloy; S83. Consider the impact of component deviation on performance and correct the prediction results; S84. Generate performance prediction data that includes mechanical properties, physical properties, and process performance indicators; S85. Provide early warnings and suggestions regarding potential performance defects.

10. The spectral analysis monitoring method for aluminum alloy composition detection according to claim 1, characterized in that: The process of performing aluminum alloy quality grading evaluation processing and generating quality grade data based on the composition conformity evaluation data, the composition deviation report data, and the performance prediction data in step S9 includes the following steps: S91. Establish aluminum alloy quality grading standards, including multiple dimensions such as composition accuracy, performance compliance rate, and degree of deviation; S92. A comprehensive evaluation is performed based on the component compliance evaluation data, the component deviation report data, and the performance prediction data; S93. The quality score is calculated using the fuzzy comprehensive evaluation method. S94. Aluminum alloy products are classified into grades such as high-quality, qualified, and unqualified based on their quality scores. S95. Generate quality grade data that includes quality grade and evaluation criteria.

Images