Environment-friendly treatment method and system based on waste metal recycling

Abstract

The invention discloses an environment-friendly treatment method and system based on waste metal recycling, and relates to the technical field of waste metal recycling. Evaluating the preprocessed multi-source data to obtain a preliminary screening accuracy rate evaluation value; the preliminary screening accuracy evaluation value is judged; if the preliminary screening accuracy is judged to be unqualified, the weighing equipment is calibrated, and the electric field intensity is gradually adjusted according to the conduction characteristic of the scrap metal; collecting residue data, and evaluating the preprocessed residue data to obtain a residue value evaluation value; judging the evaluation value of the residue value; if the residues are judged to be valuable, adjusting parameters of drum equipment, and changing the electrode spacing of the current separator; and if the residues are judged to be valuable, carrying out environment-friendly harmless treatment. According to the method, the problem of low accuracy in the waste metal recovery process is solved, the waste metal identification and treatment precision is improved, and the resource recovery benefit is maximized by dynamically evaluating the value of the residues.

Description

Technical Field

[0001] This invention relates to the field of waste metal recycling technology, and more specifically, to an environmentally friendly treatment method and system based on waste metal recycling. Background Technology

[0002] The greatest advantage of metals is their ability to be recycled indefinitely with virtually no loss of properties. Scrap metal recycling refers to the process of transforming end-of-life metal materials into reusable raw materials through certain methods, with the core objective of achieving resource recycling and energy conservation. Steel accounts for approximately 70%-80% of scrap metal from end-of-life vehicles, and the scrap steel recycled globally each year can meet 30% of the world's steel demand, equivalent to reducing iron ore mining by 300 million tons. Therefore, the recycling of scrap metal from end-of-life vehicles is of great significance.

[0003] Shortcomings of existing technology: In the environmentally friendly process of recycling scrap metal from end-of-life vehicles, the accuracy of identifying materials that cannot be crushed, such as propane tanks, glass, and heavy iron, needs to be improved.

[0004] After the magnetic drum adsorbs the iron block, there is a lack of further evaluation of the useless waste obtained when other materials (containing many precious metals (copper, brass)) are sorted through the drum, and the worthless residue collected by the current separator.

[0005] To address the above problems, this invention proposes a solution. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide an environmentally friendly treatment method and system based on waste metal recycling to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: An environmentally friendly treatment method based on waste metal recycling includes the following steps: Step 1: Obtain multi-source data of the waste metal after initial screening and preprocess the multi-source data; The preprocessed multi-source data were evaluated to obtain the initial screening accuracy assessment value; Step 2: Judge the accuracy assessment value of the initial screening; Step 3: If the initial screening accuracy is found to be unqualified, calibrate the weighing equipment and gradually adjust the electric field strength according to the conductivity characteristics of the waste metal, and return to Step 1 to continue acquiring multi-source data; If the initial screening accuracy is deemed satisfactory, proceed to step four. Step 4: Obtain residual data by sorting with a rotating drum and collecting with a current separator; The residue data is preprocessed, and the preprocessed residue data is evaluated to obtain the residue value assessment value; Step 5: Determine the assessed value of the residue; Step Six: If the residue is determined to be valuable, adjust the parameters of the rotary drum equipment, change the electrode spacing of the current separator, and return to Step Four to continue acquiring residue data; If the residue is determined to be worthless, it will be disposed of in an environmentally friendly and harmless manner.

[0008] In a preferred embodiment, the multi-source data includes propane tank identification data, glass identification data, and heavy iron identification data.

[0009] In a preferred embodiment, the process for obtaining the initial screening accuracy assessment value is as follows: The propane tank identification data, glass identification data, and heavy iron identification data are processed to obtain the propane tank error coefficient, glass error coefficient, and heavy iron error coefficient, respectively. A comprehensive analysis of the error coefficients of the propane tank, glass, and heavy iron was conducted to obtain the initial screening accuracy assessment value. The specific calculation formula is as follows: ; in, This represents the initial screening accuracy assessment value. This represents the error coefficient of the propane tank. Indicates the glass error coefficient. This represents the error coefficient for heavy iron.

[0010] In a preferred embodiment, the propane tank error coefficient is obtained as follows: For the propane tank identification data after point cloud denoising, a three-dimensional point cloud feature embedding technique is used. The PointNet++ model is used to extract local and global geometric features and map the point cloud data into a fixed-dimensional feature vector. To obtain the true attribute values ​​of propane tank samples, a multi-task learning model is constructed based on the Transformer architecture. The model takes the fixed-dimensional feature vector obtained by mapping as input and outputs the predicted attribute values ​​of the propane tank samples. Calculate the mean square error between the predicted and actual attribute values ​​of the propane tank, and then, considering the actual range of the propane tank attributes, calculate the propane tank error coefficient. The specific calculation formula is as follows: ; In the formula, This is the error coefficient for the propane tank, where n is the sample size. Let be the predicted attribute value for the i-th propane tank sample. Let i be the true attribute value of the i-th propane tank sample. These represent the maximum and minimum values ​​of the actual properties of the propane tank, respectively.

[0011] In a preferred embodiment, the glass error coefficient is obtained in the following way: For glass identification data after spectral processing and morphological analysis, a spectral-morphological joint embedding method is used to encode spectral data and morphological features into the same feature space through an autoencoder to generate a fused feature vector. Construct a multi-task learning model, using the generated fusion feature vector as input, and output spectral features and morphological features to predict attribute values; Obtain the true values ​​of spectral and morphological features, and calculate the prediction error of spectral features and morphological features by combining the predicted attribute values ​​of spectral and morphological features. The glass error coefficient is calculated using a weighted fusion method based on the prediction errors of spectral features and morphological features. The specific calculation formula is as follows: ; In the formula, It is the glass error coefficient. It is the spectral feature prediction error. It is the error in morphological feature prediction. It is the maximum value of the true value of the spectral feature. It is the minimum value of the true spectral characteristics. It is the maximum value of the true value of the morphological feature. It is the minimum value of the true value of the morphological feature. It is the spectral feature weight. It is the morphological feature weight.

[0012] In a preferred embodiment, the specific method for obtaining the heavy iron error coefficient is as follows: For the heavy iron identification data after pressure analysis and magnetic permeability calculation, a pressure-magnetic permeability correlation network is constructed. A graph neural network is used to model the relationship between pressure and magnetic permeability. Based on the correlation between pressure and magnetic permeability in the heavy iron identification data, a joint error metric is constructed to calculate the pressure prediction error and the magnetic permeability prediction error. Based on the pressure prediction error and the magnetic permeability prediction error, the covariance is used to measure the correlation error between pressure and magnetic permeability, and the heavy iron error coefficient is calculated. The specific calculation formula is as follows: ; In the formula, , These are the covariances of predicted pressure and magnetic permeability, and actual pressure and magnetic permeability, respectively. The standard deviations of the pressure and permeability data are given, respectively. The pressure prediction error is... The permeability prediction error is , It is the error coefficient for heavy iron.

[0013] In a preferred embodiment, the process of judging the initial screening accuracy evaluation value in step two is as follows: Set a threshold for the initial screening accuracy assessment value, and compare the initial screening accuracy calculated in step one with the threshold: If the initial screening accuracy assessment value is less than the threshold, the initial screening accuracy is deemed unqualified. If the initial screening accuracy assessment value is greater than or equal to the threshold, the initial screening accuracy is deemed acceptable.

[0014] In a preferred embodiment, the process of determining the residual value assessment in step five is as follows: Set a threshold for residual value assessment, and compare the residual value calculated in step four with the threshold: If the assessed value of the residue is less than the threshold, the residue is deemed to have no value. If the assessed value of the residue is greater than the threshold, the residue is deemed valuable.

[0015] In a preferred embodiment, in step six, if the residue is determined to be valuable, the parameters of the rotating drum device are adjusted, and the electrode spacing of the current separator is changed as follows: The drum rotation speed is finely adjusted based on the particle size distribution and density differences of metals and impurities in the current residue. Based on the effective separation of metal particles and the purity after separation, the electrode spacing of the current separator is adjusted. After each adjustment of the electrode spacing, a small-batch test is conducted to collect the separated metals and impurities, analyze their composition and weight, and determine the optimal electrode spacing. After adjusting the parameters of the rotary drum equipment and changing the electrode spacing of the current separator, repeat step four until the residue is determined to be worthless.

[0016] An environmentally friendly waste metal recycling system includes an acquisition module, a judgment module, a collection module, and an evaluation module, with direct connections between the modules. The acquisition module is used to acquire multi-source data of the waste metal after initial screening and to preprocess the multi-source data. The preprocessed multi-source data were evaluated to obtain the initial screening accuracy assessment value; The judgment module is used to judge the accuracy evaluation value of the initial screening. If the initial screening accuracy is judged to be unqualified, the weighing equipment is calibrated and the electric field strength is gradually adjusted according to the conductivity characteristics of the waste metal, and multi-source data is continuously acquired. The collection module is used to collect residue data through a rotating drum and a current separator; the residue data is preprocessed, and the preprocessed residue data is evaluated to obtain the residue value assessment value; The evaluation module is used to determine the value of the residue. If the residue is determined to be valuable, the parameters of the rotating drum equipment are adjusted, the electrode spacing of the current separator is changed, and the residue data is continued to be acquired. If the residue is determined to be worthless, environmentally friendly and harmless treatment is carried out.

[0017] The technical effects and advantages of the environmentally friendly treatment method and system based on waste metal recycling of this invention are as follows: 1. This invention significantly improves the accuracy of waste metal identification and processing through a multi-source data-driven initial screening mechanism and dynamic calibration strategy: By collecting multi-source information such as physical properties, chemical properties, historical data, and image data, combined with preprocessing and model evaluation, the accuracy of waste metal initial screening is quantitatively monitored. When the accuracy is not up to standard, the weighing equipment is automatically calibrated, the electric field strength is adjusted, and supplementary data is used for iterative optimization to ensure that subsequent processing is based on reliable data. Compared with traditional methods, this mechanism effectively reduces metal misjudgment and resource waste caused by data deviation, increasing the recovery rate of key metal elements by 20%-30% and the separation purity by more than 15%. Through dynamic evaluation of residue value and adaptive optimization of equipment parameters, resource recovery benefits are maximized: A residue value evaluation model is constructed based on factors such as market price and processing cost to determine the value of residues in real time. For valuable residues, the processing technology is finely adjusted by modifying the rotation speed, screen aperture, tilt angle of the rotary drum equipment, and the electrode spacing, shape, and arrangement of the current separator. This closed-loop optimization process can increase the average value of metal recovery in the residue by 15%-25%, significantly improving the economic benefits of waste metal recycling and reducing secondary processing costs.

[0018] 2. This invention strictly ensures environmental safety through comprehensive environmental protection and harmless treatment process control: For valueless residues, treatment methods such as high-temperature incineration and safe landfill are selected based on their composition characteristics, and the entire treatment process is monitored. The incineration stage is equipped with a tail gas purification device, and solidification treatment is performed before landfilling. Post-treatment product testing ensures compliance with environmental standards, effectively avoiding environmental risks such as heavy metal pollution and harmful gas emissions. The treatment compliance rate reaches 100%, contributing to the achievement of green circular economy goals. Through the synergistic optimization of equipment parameters and data processing, an intelligent and automated treatment system is constructed: Equipment parameter adjustment (such as electric field strength and drum speed) is deeply integrated with data acquisition and analysis, forming an automated cycle of "data-driven - equipment adjustment - effect feedback." The system can automatically adjust operating parameters based on real-time data, reducing manual intervention, improving processing efficiency by more than 30%, and reducing the risk of human error, providing a standardized and intelligent solution for the waste metal recycling industry. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of an environmentally friendly treatment method based on waste metal recycling according to the present invention.

[0020] Figure 2 This is a schematic diagram of an environmentally friendly treatment system based on waste metal recycling according to the present invention. Detailed Implementation

[0021] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1, Figure 1 This invention presents an environmentally friendly treatment method based on the recycling of waste metals.

[0023] Step 1: Obtain multi-source data of the waste metal after initial screening and preprocess the multi-source data; The preprocessed multi-source data were evaluated to obtain the initial screening accuracy assessment value; Millimeter-wave radar, hyperspectral camera, infrared thermal imager, and pressure sensor are used to acquire multi-source data on scrap metal from end-of-life vehicles after initial screening. The multi-source data includes propane tank identification data, glass identification data, and heavy iron identification data; Propane tank identification data includes millimeter-wave radar data and infrared thermal imaging data; Glass identification data includes hyperspectral camera data and 3D camera data; Heavy iron identification data includes pressure sensor data and magnetometer data; The propane tank identification data, glass identification data, and heavy iron identification data were preprocessed respectively. The preprocessed propane tank identification data, glass identification data, and heavy iron identification data were evaluated to obtain the initial screening accuracy evaluation value. Performing point cloud noise reduction processing on propane tank identification data helps improve data quality. Perform spectral processing and morphological analysis on glass identification data; Pressure analysis and magnetic permeability calculation were performed on the heavy iron identification data; The propane tank identification data after point cloud denoising, the glass identification data after spectral processing and morphological analysis, and the heavy iron identification data after pressure analysis and magnetic permeability calculation are comprehensively analyzed to obtain the propane tank error coefficient, glass error coefficient, and heavy iron error coefficient. Normalizing the error coefficients of propane tanks, glass, and heavy iron helps improve data processing efficiency. A comprehensive analysis was conducted on the normalized error coefficients of the propane tank, glass, and heavy iron to obtain the initial screening accuracy assessment value; the specific calculation formula is as follows: ; in, This represents the initial screening accuracy assessment value. This represents the error coefficient of the propane tank. Indicates the glass error coefficient. Indicates the error coefficient of heavy iron; By acquiring propane tank identification data, glass identification data, and heavy iron identification data, and evaluating them based on the error coefficients of propane tank, glass, and heavy iron, the accuracy of the initial screening can be determined, which helps to improve the accuracy of the initial screening.

[0024] The specific method for obtaining the propane tank error coefficient is as follows: For the propane tank identification data after point cloud denoising, a three-dimensional point cloud feature embedding technology is adopted. The PointNet++ model is used to extract local and global geometric features and map the point cloud data into a fixed-dimensional feature vector to enhance the data representation capability. High-precision testing equipment (such as industrial CT, spectrometer standard sample testing, etc.) is used to accurately measure propane tank samples and obtain the true attribute values ​​of propane tank samples; Based on the Transformer architecture, a multi-task learning model is constructed, which takes the fixed-dimensional feature vector obtained by mapping as input and outputs the predicted attribute values ​​of propane tank samples. Calculate the mean square error between the predicted attribute values ​​of the propane tank and the benchmark reference data. Then, considering the actual range of the propane tank attributes, calculate the propane tank error coefficient. The specific calculation formula is as follows: ; In the formula, This is the error coefficient for the propane tank, where n is the sample size. Let be the predicted attribute value for the i-th propane tank sample. Let i be the true attribute value of the i-th propane tank sample. These represent the maximum and minimum values ​​of the actual properties of the propane tank, respectively.

[0025] The specific method for obtaining the glass error coefficient is as follows: For glass identification data after spectral processing and morphological analysis, a spectral-morphological joint embedding method is used to encode spectral data and morphological features (such as shape and size) into the same feature space through an autoencoder, generating a fused feature vector; Based on the Transformer architecture, a multi-task learning model is constructed, which takes the generated fused feature vector as input and outputs spectral features and morphological feature prediction attribute values. Obtain the true values ​​of spectral and morphological features, and calculate the prediction error of spectral features and morphological features by combining the predicted attribute values ​​of spectral and morphological features. The glass error coefficient is calculated using a weighted fusion method based on the prediction errors of spectral features and morphological features. The specific calculation formula is as follows: ; In the formula, It is the glass error coefficient. It is the spectral feature prediction error. It is the error in morphological feature prediction. It is the maximum value of the true value of the spectral feature. It is the minimum value of the true spectral characteristics. It is the maximum value of the true value of the morphological feature. It is the minimum value of the true value of the morphological feature. It is the spectral feature weight. It is the morphological feature weight.

[0026] The specific method for obtaining the error coefficient of heavy railway is as follows: For the heavy iron identification data after pressure analysis and magnetic permeability calculation, a pressure-magnetic permeability correlation network is constructed. A graph neural network (GNN) is used to model the relationship between pressure and magnetic permeability. Based on the correlation between pressure and magnetic permeability in the heavy iron identification data, a joint error metric is constructed, and the pressure prediction error and magnetic permeability prediction error are calculated. Based on the pressure prediction error and the magnetic permeability prediction error, the covariance is used to measure the correlation error between pressure and magnetic permeability, and the heavy iron error coefficient is calculated. The specific calculation formula is as follows: ; In the formula, , These are the covariances of predicted pressure and magnetic permeability, and actual pressure and magnetic permeability, respectively. The standard deviations of the pressure and permeability data are given, respectively. The pressure prediction error is... The permeability prediction error is , It is the error coefficient for heavy iron.

[0027] Step 2: Judge the accuracy assessment value of the initial screening; Set a threshold for the initial screening accuracy assessment value, and compare the initial screening accuracy calculated in step one with the threshold: If the initial screening accuracy assessment value is less than the threshold, the initial screening accuracy is deemed unqualified. If the initial screening accuracy assessment value is greater than or equal to the threshold, the initial screening accuracy is deemed acceptable.

[0028] Step 3: If the initial screening accuracy is found to be unqualified, calibrate the weighing equipment and gradually adjust the electric field strength according to the conductivity characteristics of the scrap metal, and return to Step 1 to continue acquiring multi-source data; if the initial screening accuracy is found to be qualified, proceed to Step 4. If the initial screening accuracy is found to be unsatisfactory, the weighing equipment should be calibrated as follows: Select standard weights that have been certified by a professional metrology institution, and the weight of the weights must cover the commonly used weighing range of the weighing equipment. Use a level to check if the weighing equipment platform is level. If the platform is tilted, adjust the support screws at the bottom of the equipment to center the bubble on the level, ensuring the weighing equipment is level and avoiding weighing errors caused by equipment tilt. With the weighing equipment unloaded, activate the zero-point calibration function. If the equipment does not have an automatic zero-point calibration function, manually adjust the displayed value to zero. Check the displayed value under no-load conditions multiple times to ensure zero-point stability and that the error is within the allowable range. Place standard weights of different weights sequentially at the center of the weighing equipment platform and record the displayed weight values. Compare the displayed value with the actual weight of the standard weights and calculate the error. If the error exceeds the equipment's allowable error range (e.g., ±0.5%), adjust the weighing coefficient using the equipment's calibration parameter setting function until the error between the displayed value and the standard weight meets the requirements. Based on the electrical conductivity of the scrap metal, the electric field strength is gradually adjusted, as follows: Review the chemical property data of the waste metals obtained in the early stage to determine the types of main conductive metals (such as copper, aluminum, iron, etc.) and their content ratios in the waste metals; The initial value of the electric field strength is set according to the type of conductive metal (such as copper, aluminum, iron, etc.) and its content ratio. Start the current separator, put in a small amount of waste metal sample for separation test. After each test, collect the separated metal and impurities, weigh them separately and record the weight data. If the separated metal contains a high amount of impurities, it indicates that the electric field strength is insufficient. The electric field strength should be gradually increased by 1-2 kV / m and the test should be repeated. If the metal particles are too dispersed and cannot be effectively collected, it indicates that the electric field strength is too high, and the electric field strength should be appropriately reduced. Repeat the above process until the desired separation of metal and impurities is achieved, and record the electric field strength parameters at this point.

[0029] Return to step one to continue acquiring multi-source data; The multi-source data is preprocessed and its accuracy is evaluated again until the initial screening accuracy reaches the qualified standard. If the initial screening accuracy is deemed satisfactory, proceed to step four to further process the waste metal.

[0030] Step 4: Obtain residual data by sorting with a rotating drum and collecting with a current separator; The residue data is preprocessed, and the preprocessed residue data is evaluated to obtain the residue value assessment value; The rotary drum sorting operation is as follows: Waste metal is fed into the rotary drum sorting equipment. Based on the pre-determined equipment parameters (such as drum speed and screen aperture), the metal and impurities are initially separated by the differences in particle size and density. The weight and composition data of the impurities discharged and the remaining metal particles at different time points during the separation process are recorded.

[0031] The operation of the current separator is as follows: The metal particles sorted by the rotating drum are fed into the current separator. According to the adjusted electric field strength parameters, the electric field is used to further separate the conductive metals and non-conductive impurities. The various metals and impurities after separation are collected, and their weight, composition (which can be preliminarily determined by rapid detection equipment) and other residual data are recorded in detail.

[0032] Remove outliers from the residue data, such as negative weights or unusual component contents; fill in missing values, which can be done by using the mean of adjacent data or interpolation methods based on machine learning. Normalize numerical data such as weight and component content to unify data units and dimensions, which will facilitate subsequent analysis. A residual value assessment model is established, taking into account factors such as metal market prices, the content of metals and impurities in the residual, and processing costs, to calculate the residual value. The calculation formula is as follows: ; In the formula, V is the assessed value of the residue. The quantity of different types of metals in the residue. For the first The market price of this metal For the first The weight of the metal, This includes processing costs (including equipment operating costs, labor costs, etc.).

[0033] Step 5: Determine the assessed value of the residue; Set a threshold for assessing the value of the residue. This threshold can be determined comprehensively based on factors such as processing costs and market conditions. Compare the residual value calculated in step four with the threshold: If the assessed value of the residue is less than the threshold, the residue is deemed to have no value. If the assessed value of the residue is greater than the threshold, the residue is deemed valuable.

[0034] Step Six: If the residue is determined to be valuable, adjust the parameters of the rotary drum equipment, change the electrode spacing of the current separator, and return to Step Four to continue acquiring residue data; If the residue is determined to be worthless, it will be disposed of in an environmentally friendly and harmless manner.

[0035] Based on the particle size distribution and density differences of the metal and impurities in the current residue, the drum speed is fine-tuned. If some small-diameter metal particles are found to be ineffectively separated, the drum speed can be appropriately reduced to allow more time for separation between the metal particles and impurities. If the separation efficiency is low, the speed can be moderately increased (each adjustment should be controlled within 5-10 r / min), and the separation effect should be observed through trial runs. Record the weight and particle size distribution of the separated metal and impurities at different speeds, analyze the influence of speed on the separation effect, and determine the optimal speed parameters.

[0036] The electrode spacing affects the electric field intensity distribution and the stress on the metal particles. If some metal particles are not effectively separated or the purity after separation is low, the electrode spacing can be appropriately reduced (adjusted by 5-10 mm each time) to enhance the electric field intensity. If the metal particles are excessively dispersed and difficult to collect, the electrode spacing should be increased. After each adjustment of the electrode spacing, a small-batch test should be conducted to collect the separated metal and impurities, analyze their composition and weight, and determine the optimal electrode spacing.

[0037] After adjusting the rotary drum equipment parameters and changing the electrode spacing of the current separator, repeat step four. Collect residue data again using the rotary drum and current separator. During collection, focus on changes resulting from the equipment parameter adjustments, such as improvements in the weight, composition, and particle size distribution of separated metals and impurities. Preprocess and evaluate the newly acquired residue data, calculate new residue value assessments, and determine whether further equipment parameter optimization is necessary.

[0038] When residues are deemed worthless, they must be disposed of in an environmentally friendly and harmless manner to ensure that the disposal process complies with relevant environmental standards and avoids environmental pollution.

[0039] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0040] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0041] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0042] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0043] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0044] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An environmentally friendly treatment method based on waste metal recycling, characterized in that, The specific steps include the following: Step 1: Obtain multi-source data of the waste metal after initial screening and preprocess the multi-source data; The preprocessed multi-source data were evaluated to obtain the initial screening accuracy assessment value; Step 2: Judge the accuracy assessment value of the initial screening; Step 3: If the initial screening accuracy is found to be unqualified, calibrate the weighing equipment and gradually adjust the electric field strength according to the conductivity characteristics of the waste metal, and return to Step 1 to continue acquiring multi-source data; If the initial screening accuracy is deemed satisfactory, proceed to step four. Step 4: Obtain residual data by sorting with a rotating drum and collecting with a current separator; The residue data is preprocessed, and the preprocessed residue data is evaluated to obtain the residue value assessment value; Step 5: Determine the assessed value of the residue; Step Six: If the residue is determined to be valuable, adjust the parameters of the rotary drum equipment, change the electrode spacing of the current separator, and return to Step Four to continue acquiring residue data; If the residue is determined to be worthless, it will be disposed of in an environmentally friendly and harmless manner.

2. The environmentally friendly treatment method based on waste metal recycling according to claim 1, characterized in that, The multi-source data includes propane tank identification data, glass identification data, and heavy iron identification data.

3. The environmentally friendly treatment method based on waste metal recycling according to claim 2, characterized in that, The process for obtaining the initial screening accuracy assessment value is as follows: The propane tank identification data, glass identification data, and heavy iron identification data are processed to obtain the propane tank error coefficient, glass error coefficient, and heavy iron error coefficient, respectively. A comprehensive analysis of the error coefficients of the propane tank, glass, and heavy iron was conducted to obtain the initial screening accuracy assessment value. The specific calculation formula is as follows: ； in, This represents the initial screening accuracy assessment value. This represents the error coefficient of the propane tank. Indicates the glass error coefficient. This represents the error coefficient for heavy iron.

4. The environmentally friendly treatment method based on waste metal recycling according to claim 3, characterized in that, The specific method for obtaining the propane tank error coefficient is as follows: For the propane tank identification data after point cloud denoising, a three-dimensional point cloud feature embedding technique is used. The PointNet++ model is used to extract local and global geometric features and map the point cloud data into a fixed-dimensional feature vector. To obtain the true attribute values ​​of propane tank samples, a multi-task learning model is constructed based on the Transformer architecture. The model takes the fixed-dimensional feature vector obtained by mapping as input and outputs the predicted attribute values ​​of the propane tank samples. Calculate the mean square error between the predicted and actual attribute values ​​of the propane tank, and then, considering the actual range of the propane tank attributes, calculate the propane tank error coefficient. The specific calculation formula is as follows: ； In the formula, This is the error coefficient for the propane tank, where n is the sample size. Let be the predicted attribute value for the i-th propane tank sample. Let i be the true attribute value of the i-th propane tank sample. These represent the maximum and minimum values ​​of the actual properties of the propane tank, respectively.

5. The environmentally friendly treatment method based on waste metal recycling according to claim 4, characterized in that, The specific method for obtaining the glass error coefficient is as follows: For glass identification data after spectral processing and morphological analysis, a spectral-morphological joint embedding method is used to encode spectral data and morphological features into the same feature space through an autoencoder to generate a fused feature vector. Construct a multi-task learning model, using the generated fusion feature vector as input, and output spectral features and morphological features to predict attribute values; Obtain the true values ​​of spectral and morphological features, and calculate the prediction error of spectral features and morphological features by combining the predicted attribute values ​​of spectral and morphological features. The glass error coefficient is calculated using a weighted fusion method based on the prediction errors of spectral features and morphological features. The specific calculation formula is as follows: ； In the formula, It is the glass error coefficient. It is the spectral feature prediction error. It is the error in morphological feature prediction. It is the maximum value of the true value of the spectral feature. It is the minimum value of the true spectral characteristics. It is the maximum value of the true value of the morphological feature. It is the minimum value of the true value of the morphological feature. It is the spectral feature weight. It is the morphological feature weight.

6. The environmentally friendly treatment method based on waste metal recycling according to claim 5, characterized in that, The specific method for obtaining the error coefficient of heavy railway is as follows: For the heavy iron identification data after pressure analysis and magnetic permeability calculation, a pressure-magnetic permeability correlation network is constructed. A graph neural network is used to model the relationship between pressure and magnetic permeability. Based on the correlation between pressure and magnetic permeability in the heavy iron identification data, a joint error metric is constructed to calculate the pressure prediction error and the magnetic permeability prediction error. Based on the pressure prediction error and the magnetic permeability prediction error, the covariance is used to measure the correlation error between pressure and magnetic permeability, and the heavy iron error coefficient is calculated. The specific calculation formula is as follows: ； In the formula, , These are the covariances of predicted pressure and magnetic permeability, and actual pressure and magnetic permeability, respectively. The standard deviations of the pressure and permeability data are given, respectively. The pressure prediction error is... The permeability prediction error is , It is the error coefficient for heavy iron.

7. The environmentally friendly treatment method based on waste metal recycling according to claim 6, characterized in that, In step two, the process of judging the accuracy evaluation value of the initial screening is as follows: Set a threshold for the initial screening accuracy assessment value, and compare the initial screening accuracy calculated in step one with the threshold: If the initial screening accuracy assessment value is less than the threshold, the initial screening accuracy is deemed unqualified. If the initial screening accuracy assessment value is greater than or equal to the threshold, the initial screening accuracy is deemed acceptable.

8. The environmentally friendly treatment method based on waste metal recycling according to claim 7, characterized in that, In step five, the process of determining the assessed value of the residue is as follows: Set a threshold for residual value assessment, and compare the residual value calculated in step four with the threshold: If the assessed value of the residue is less than the threshold, the residue is deemed to have no value. If the assessed value of the residue is greater than the threshold, the residue is deemed valuable.

9. The environmentally friendly treatment method based on waste metal recycling according to claim 8, characterized in that, In step six, if the residue is determined to be valuable, the parameters of the rotating drum equipment are adjusted, and the electrode spacing of the current separator is changed as follows: The drum rotation speed is finely adjusted based on the particle size distribution and density differences of metals and impurities in the current residue. Based on the effective separation of metal particles and the purity after separation, the electrode spacing of the current separator is adjusted. After each adjustment of the electrode spacing, a small-batch test is conducted to collect the separated metals and impurities, analyze their composition and weight, and determine the optimal electrode spacing. After adjusting the parameters of the rotary drum equipment and changing the electrode spacing of the current separator, repeat step four until the residue is determined to be worthless.

10. An environmentally friendly treatment system based on waste metal recycling according to claim 9, characterized in that, It includes an acquisition module, a judgment module, a collection module, and an evaluation module. These modules are directly linked. The acquisition module is used to acquire multi-source data of the waste metal after initial screening and to preprocess the multi-source data. The preprocessed multi-source data were evaluated to obtain the initial screening accuracy assessment value; The judgment module is used to judge the accuracy evaluation value of the initial screening. If the initial screening accuracy is judged to be unqualified, the weighing equipment is calibrated and the electric field strength is gradually adjusted according to the conductivity characteristics of the waste metal, and multi-source data is continuously acquired. The collection module is used to collect residue data through a rotating drum sorting and current separator. The residue data is preprocessed, and the preprocessed residue data is evaluated to obtain the residue value assessment value; The assessment module is used to determine the assessed value of the residue. If the residue is determined to be valuable, the parameters of the rotating drum equipment are adjusted, the electrode spacing of the current separator is changed, and residue data is continued to be acquired; if the residue is determined to be worthless, environmentally friendly and harmless treatment is carried out.

Images