Quality detection and analysis system for polypropylene plastic particles
The quality inspection of polypropylene plastic particles is solved through the multi-layer convolutional network model, and the problem of insufficient detection accuracy in the existing technology is solved, and the accuracy of multi-dimensional quality evaluation of polypropylene plastic particles is improved and the safety of food contact is improved.
Patent Information
- Application Number
- CN202411029821.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-15
AI Technical Summary
The prior art has insufficient detection accuracy and misjudgment in the quality detection of polypropylene plastic particles. It has failed to fully consider the quality detection of polypropylene plastic particles in multiple dimensions, which limits the accuracy and reliability of the detection.
A multi-layer convolutional network is used to construct a polypropylene particle quality detection model. Through sample image preprocessing, feature extraction and fusion, particle performance feature vectors are obtained, and multiple quality detection values are calculated, so as to comprehensively determine whether the mass of polypropylene plastic particles is qualified.
It significantly improves the accuracy and reliability of the quality evaluation of polypropylene plastic particles, comprehensively evaluates multi-dimensional physical characteristics, ensures that product quality meets high standards, and enhances the overall effect of food contact safety performance detection.
Smart Images

Figure CN118960840A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of polypropylene plastic particles, in particular to a quality detection and analysis system for polypropylene plastic particles. Background Art
[0002] Polypropylene plastic pellets are a thermoplastic polymer used in a wide variety of industrial and consumer products. This material is favored for its excellent chemical resistance, insulation properties, low water absorption and high melting point. Polypropylene pellets are easy to melt and shape during processing, making them ideal for manufacturing products of various shapes and sizes through injection molding, blow molding and other thermoplastic processing techniques.
[0003] Polypropylene plastic granules have applications in many industries, especially in the field of food packaging. Due to its non-toxicity and ability to create products that meet food contact safety standards, polypropylene is widely used to make various food containers, beverage bottles, plastic wrap and tableware. In addition, polypropylene's heat resistance makes it an ideal material for making microwave-safe food containers. Its light weight and durability also make polypropylene a preferred material for making packaging films and reusable packaging boxes.
[0004] It is crucial to test the quality of polypropylene plastic pellets, as this directly affects the performance and safety of the finished product. In the field of food packaging, the use of poor quality polypropylene may lead to structural imbalance or more serious health risks during the use of the product, such as the migration of chemical substances. Therefore, ensuring that polypropylene pellets meet strict quality standards can ensure consumer safety while also complying with regulatory requirements and avoiding possible legal liabilities and loss of brand reputation.
[0005] Traditional techniques commonly used in quality testing of polypropylene plastic pellets include UV-visible spectroscopy, Fourier transform infrared spectroscopy, thermogravimetric analysis, mechanical properties testing, and X-ray diffraction analysis. These tests provide important information about the quality of polypropylene plastic pellets, helping manufacturers ensure that each batch of materials meets the required industrial standards and performance specifications.
[0006] In the prior art, there are still certain defects in the quality detection of polypropylene plastic particles. For example, a Chinese patent with publication number CN113191616A provides a polypropylene product quality anomaly detection method based on double-layer correlation feature analysis. This method obtains characteristic variables and components related to the quality of polypropylene products through double-layer correlation feature analysis technology, and indirectly implements anomaly detection. When implementing polypropylene product quality anomaly detection, this method does not rely on direct measurement of polypropylene melt index, which has a long measurement cycle and serious lag. Instead, it uses double-layer correlation feature analysis such as nearest component analysis algorithm and typical correlation analysis algorithm to obtain the most relevant characteristic variables or components of polypropylene melt index to indirectly detect anomalies, overcoming the lag problem of traditional methods; however, reliance on indirect measurement may lead to insufficient detection accuracy, especially when the correlation between the characteristic variables and the actual melt index is not completely consistent, misjudgment or missed judgment may occur. The Chinese patent with publication number CN113673915B provides a polypropylene infusion bag quality detection and analysis system based on full-cycle production tracking, including a polypropylene infusion bag process step marking module, a production cycle modeling detection module, a production database and a detection result analysis terminal. The system performs pre-production equipment raw material matching detection, process operation precision detection and molded product quality detection on the upright polypropylene infusion bag, and then comprehensively evaluates the comprehensive production quality coefficient corresponding to the upright polypropylene infusion bag based on the above detection results. The quality detection process integrates the detection of the entire process before, during and after the production of the infusion bag, and to a certain extent broadens the detection direction of the corresponding production quality of the upright polypropylene infusion bag, thereby realizing the full-cycle tracking detection of the corresponding production quality of the upright polypropylene infusion bag; although the method provides a full-cycle quality detection system for polypropylene infusion bags, it fails to cover the potential impact of the variability of raw material quality on the quality of the final product, which may limit the comprehensiveness of the detection system in preventing production problems. The existing technology fails to fully consider the quality detection of polypropylene plastic particles from multiple dimensions, thereby limiting the accuracy and reliability of the detection.
[0007] Therefore, a quality detection and analysis system for polypropylene plastic particles is proposed. Summary of the invention
[0008] The purpose of the present invention is to provide a quality detection and analysis system for polypropylene plastic particles. The present invention relates to the technical field of polypropylene plastic particles, and specifically to a quality detection and analysis system for polypropylene plastic particles. First, a plurality of polypropylene plastic particles produced in the same batch are collected to construct a polypropylene plastic particle sample set; then, the polypropylene plastic particles in the polypropylene plastic particle sample set are respectively subjected to polypropylene particle quality detection, first processing quality detection, second processing quality detection, third processing quality detection, and finished product contact safety detection to obtain a first quality detection value, a second quality detection value, a third quality detection value, a fourth quality detection value, and a fifth quality detection value; finally, a comprehensive quality detection value of the polypropylene plastic particles is obtained according to the five quality detection values, and the quality of the polypropylene plastic particles produced in the same batch is judged to be qualified by the comprehensive quality detection value.
[0009] To achieve the above object, the present invention provides the following technical solutions:
[0010] A quality detection and analysis system for polypropylene plastic particles, comprising:
[0011] A sample set building unit, used for collecting a plurality of polypropylene plastic particles produced in the same batch to build a polypropylene plastic particle sample set;
[0012] The quality detection unit includes a polypropylene particle quality detection module, a first processing quality detection module, a second processing quality detection module, a third processing quality detection module and a finished product contact safety detection module, and is used to perform different tests on the polypropylene plastic particles in the polypropylene plastic particle sample set to obtain a first quality detection value, a second quality detection value, a third quality detection value, a fourth quality detection value and a fifth quality detection value;
[0013] A comprehensive evaluation unit, used to obtain a comprehensive quality test value; the comprehensive quality test value is used to determine whether the quality of the polypropylene plastic particles produced in the same batch is qualified;
[0014] The process of obtaining the first quality detection value includes:
[0015] Acquire a sample image of each polypropylene plastic particle in the polypropylene plastic particle sample set;
[0016] The sample image is subjected to a polypropylene particle quality detection model to obtain a particle performance feature vector;
[0017] Acquire the first quality detection value according to the particle performance characteristic vector;
[0018] The particle performance feature vector includes size feature, shape feature and color feature.
[0019] The polypropylene particle quality detection model includes a preprocessing layer, a feature map extraction layer, a feature fusion layer and an output layer;
[0020] The preprocessing layer is used to perform a preprocessing operation on the sample image;
[0021] The preprocessing operation includes uniformly cropping the sizes of the sample images to obtain sample images of the same size;
[0022] Removing noise from the sample image of the same size to obtain a denoised sample image; enhancing the denoised sample image using an image enhancement algorithm to obtain an enhanced sample image;
[0023] The feature map extraction layer is used to obtain a plurality of polypropylene plastic particle feature maps according to the enhanced sample image;
[0024] The feature fusion layer is used to perform feature fusion on a plurality of feature maps of the polypropylene plastic particles to obtain a particle performance feature vector;
[0025] The output layer is used to output the particle performance feature vector.
[0026] The process of obtaining the particle performance characteristic vector includes:
[0027] Input the sample image into a convolution layer with a size of 1×1 and 8 convolution kernels to obtain a first physical feature map;
[0028] Input the first physical feature map into a convolutional layer with a size of 3×3 and 12 convolution kernels, and add a spatial attention mechanism after the convolutional layer to obtain a second physical feature map;
[0029] The second physical feature map is input into another convolutional layer with a size of 3×3 and 12 convolution kernels, and a nonlinear transformation is performed through a ReLU activation function to obtain a third physical feature map;
[0030] Performing feature fusion on the first physical feature map, the second physical feature map, and the third physical feature map to obtain a particle performance feature vector;
[0031] The size characteristics include the length, width and height of the polypropylene plastic pellets;
[0032] The shape characteristics include surface roughness, shape factor and overall geometric symmetry of the polypropylene particles;
[0033] The color characteristics include color consistency and color distribution of the polypropylene plastic particles.
[0034] The calculation formula of the first quality detection value includes:
[0035]
[0036] Among them, Q1 is the first quality detection value, N is the number of samples in the polypropylene plastic particle sample set, S i is the size characteristic value of the i-th sample, C i is the shape feature value of the i-th sample, F i is the color feature value of the i-th sample; α, β, γ represent the size feature coefficient, color feature coefficient, and shape feature coefficient, respectively.
[0037] The calculation formula of the size characteristic value is:
[0038]
[0039] Among them, S i is the size characteristic value of the i-th sample, L i ,W i ,H i represents the length, width and height of the i-th sample, L thresh ,W thresh ,H thresh Respectively represent the set thresholds for length, width and height, ω l ,ω w ,ω h are the coefficients of length, width and height respectively;
[0040] The calculation formula of the shape characteristic value is:
[0041]
[0042] Among them, F i is the shape characteristic value of the i-th sample; R i Indicates the surface roughness of the i-th sample; Area i ,Perimeter i represent the area and perimeter of the particle in the i-th sample image, respectively, and are used to calculate the shape factor; Gyr ideal Gyr represents the gyro radius in the standard state, which represents the relationship between the moment of inertia of the polypropylene particles and the mass of the polypropylene particles, and reflects the shape and mass distribution characteristics of the polypropylene particles; i represents the gyro radius of the i-th sample shape; σ r ,σ s ,σ g are the weight coefficients of surface roughness, shape factor and overall geometric symmetry, respectively;
[0043] The calculation formula of the color characteristic value is:
[0044]
[0045] Among them, C i is the color feature value of the i-th sample, p i,c represents the actual distribution probability of the i-th sample on color c, q i,c represents the expected color distribution probability of the i-th sample on color c, where C represents the color class; The sum of the absolute values of the differences between the actual distribution probability of each color and the uniform distribution is calculated, η u ,η d A weight coefficient representing color consistency and color distribution.
[0046] The calculation formula of the second quality detection value is:
[0047]
[0048] Among them, Q2 is the second detection mass value; N1 and N2 are the number of samples selected from the polypropylene plastic particle sample set for density detection and heat resistance detection, respectively, and N1 and N2 meet N is the number of samples in the polypropylene plastic pellet sample set; D jm is the density value detected by the jmth sample; HD jr is the thermal deformation temperature value of the jrth sample tested; D std ,HD std are the standard values of density and heat deformation temperature respectively; d1 and d2 are the weight coefficients of density and heat deformation temperature respectively.
[0049] The calculation formula of the third quality detection value is:
[0050]
[0051] Among them, Q3 is the third quality test value; N3 and N4 are the number of samples selected from the polypropylene plastic particle sample set for purity test and isotacticity test, respectively, and N3 and N4 meet N is the number of samples in the polypropylene plastic particle sample set; P jc is the purity value of the jcth sample detected; I jd is the isotacticity value of the jdth sample detection; P std ,I std are the standard values of purity and isotacticity respectively; d3 and d4 are the weight coefficients of purity and isotacticity respectively;
[0052] Classifying the polypropylene plastic particle sample to be tested for purity as a first sample to be tested;
[0053] Dissolving the first sample to be tested in a dichloromethane solvent to obtain a polypropylene solution;
[0054] The purity value is obtained by gas chromatography-mass spectrometry, and the specific calculation formula is:
[0055]
[0056] Where P is the purity value, T is the total concentration of the polypropylene solution, and ∑T1 represents the total concentration of all impurities in the polypropylene solution;
[0057] The isotacticity value is obtained by a nuclear magnetic resonance spectrometer, and the specific calculation formula is:
[0058]
[0059] Among them, I is the isotacticity value, A ij is the area of the peak corresponding to the isotactic arrangement of methyl side chains in the NMR spectra, A t It is the total area of all peaks in the NMR spectra.
[0060] The calculation formula of the fourth quality detection value is:
[0061]
[0062] Among them, Q4 is the fourth quality detection value; N5 and N6 are the number of samples in the polypropylene plastic particle sample set selected for the first flow rate detection and the second flow rate detection, and N5 and N6 meet N is the number of samples in the polypropylene plastic pellet sample set; MFR jz The first flow rate detected for the jzth sample; MVR jt The second flow rate detected for the jtth sample; MFR std ,MVR std are standard values of the first flow rate and the second flow rate, respectively; d5 and d6 are weight coefficients of the first flow rate and the second flow rate, respectively;
[0063] The calculation formula of the first flow rate is:
[0064]
[0065] Wherein, MFR is the first flow rate, W represents the mass of the polypropylene melt flowing out of the flow rate measuring instrument within a specified time, and T represents the specified time;
[0066] The calculation formula of the second flow rate is:
[0067]
[0068] Wherein, MVR is the second flow rate, and ρ represents the density of the second sample to be detected.
[0069] The calculation formula of the fifth quality detection value is:
[0070]
[0071] Among them, Q5 is the fifth quality test value; N7 and N8 are the number of samples selected from the polypropylene plastic particle sample set for volatile gas detection and food contact safety detection, and N7 and N8 meet N is the number of samples in the polypropylene plastic pellet sample set; VL jq is the volatile gas detection value of the jqth sample; SL ja VL is the environmental protection test value of the jath sample; std ,SL std are the standard values of the volatile gas detection value and the environmental protection detection value respectively; d7 and d8 are the weight coefficients of the volatile gas detection value and the environmental protection detection value respectively, which are used to adjust the influence weights of the volatile gas detection value and the environmental protection detection value in the fifth quality detection value;
[0072] The process of obtaining the volatile gas detection value includes:
[0073] Placing the third sample to be tested in a sealed container and heating it at a temperature of 120° C.;
[0074] Use gas chromatography-mass spectrometry to analyze the air in the heated sealed container to obtain the types and concentrations of various volatile gases;
[0075] Comparing the types and concentrations of the various volatile gases with environmental safety standards to obtain volatile gas detection values;
[0076] The environmental protection detection value is obtained by immersing the fourth detection sample in a food simulant, collecting the corresponding extract and performing chemical analysis using a liquid chromatograph;
[0077] The food simulant is ethanol; the calculation formula of the environmental protection detection value is:
[0078]
[0079] Among them, SL is the environmental protection test value, CN h is the concentration of the detected harmful chemical substances, and SF is the food safety threshold. The calculation formula of the comprehensive quality test value is:
[0080] Q z =φ1*Q1+φ2*Q2+φ3*Q3+φ4*Q4+φ5*Q5;
[0081] Among them, Q zis the comprehensive quality detection value, Q1, Q2, Q3, Q4, Q5 are the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value respectively, φ1, φ2, φ3, φ4, φ5 are the correlation coefficients of the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value respectively.
[0082] Compared with the prior art, the present invention has the following beneficial effects:
[0083] 1. The present invention constructs a polypropylene particle quality detection model to obtain particle performance characteristic vectors. The model is used to obtain the size characteristics, shape characteristics and color characteristics of polypropylene plastic particles, and then calculate the first quality detection value of polypropylene plastic particles. This method not only comprehensively evaluates the multi-dimensional physical properties of polypropylene plastic particles, but also accurately reflects its overall quality level. By comprehensively analyzing these key physical properties, the present invention greatly improves the accuracy and reliability of polypropylene plastic particle quality assessment.
[0084] 2. The present invention calculates the second quality detection value, the third quality detection value and the fourth quality detection value, detects the quality of the polypropylene particles in the production process, and reflects the structural stability of the polypropylene plastic particles by detecting the density and heat resistance of the samples; reflects the chemical composition accuracy of the polypropylene plastic particles by detecting the purity and isotacticity of the samples; reflects the processing applicability of the polypropylene plastic particles by evaluating the first flow rate and the second flow rate of the polypropylene plastic particles; this detection method ensures the evaluation of the material properties of the polypropylene plastic particles in the production process, and provides a solid quality assurance for high-standard applications.
[0085] 3. The present invention detects the food contact safety performance of polypropylene plastic particles by calculating the fifth quality detection value to ensure that they meet relevant safety standards when manufacturing food packaging and containers. The calculation of the fifth quality detection formula covers volatile gas detection and the release of environmentally friendly chemicals, providing a comprehensive safety analysis for polypropylene plastic particles and improving the overall effect of quality detection of polypropylene plastic particles. BRIEF DESCRIPTION OF THE DRAWINGS
[0086] Figure 1 A flow chart of an implementation method of a quality detection and analysis system for polypropylene plastic particles provided by an embodiment of the present invention;
[0087] Figure 2 A structural diagram of a quality detection and analysis system for polypropylene plastic particles provided in an embodiment of the present invention;
[0088] Figure 3 A schematic diagram of the structure of a quality detection unit provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0089] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0090] Polypropylene plastic pellets are a thermoplastic polymer used in a wide variety of industrial and consumer products. This material is favored for its excellent chemical resistance, insulation properties, low water absorption and high melting point. Polypropylene pellets are easy to melt and shape during processing, making them ideal for manufacturing products of various shapes and sizes through injection molding, blow molding and other thermoplastic processing techniques.
[0091] Polypropylene plastic granules have applications in many industries, especially in the field of food packaging. Due to its non-toxicity and ability to create products that meet food contact safety standards, polypropylene is widely used to make various food containers, beverage bottles, plastic wrap and tableware. In addition, polypropylene's heat resistance makes it an ideal material for making microwave-safe food containers. Its light weight and durability also make polypropylene a preferred material for making packaging films and reusable packaging boxes.
[0092] It is crucial to test the quality of polypropylene plastic pellets, as this directly affects the performance and safety of the finished product. In the field of food packaging, the use of poor quality polypropylene may lead to structural imbalance or more serious health risks during the use of the product, such as the migration of chemical substances. Therefore, ensuring that polypropylene pellets meet strict quality standards can ensure consumer safety while also complying with regulatory requirements and avoiding possible legal liabilities and loss of brand reputation.
[0093] There are many studies on quality inspection of polypropylene plastic particles in the prior art, but there are some defects. In order to solve the defects in the prior art, the purpose of the present invention is to provide a quality inspection and analysis system for polypropylene plastic particles. The specific implementation method flow chart and system structure diagram of the system are shown in FIG. Figure 1 and Figure 2 .
[0094] Embodiment 1
[0095] This example tests polypropylene plastic particles produced by Factory A;
[0096] Reference Figure 1S10 in the embodiment of the present invention is applied to a sample set construction unit of a quality inspection and analysis system for polypropylene plastic particles, and is used to collect a plurality of polypropylene plastic particles produced in the same batch to construct a sample set of polypropylene plastic particles;
[0097] In this embodiment, a comprehensive polypropylene plastic particle sample set is constructed by collecting multiple polypropylene plastic particle samples produced in the same batch. This systematic sample set construction method not only ensures the representativeness and consistency of the samples, but also provides a solid foundation for subsequent multi-dimensional quality analysis, thereby making the entire quality inspection process more accurate and reliable.
[0098] Reference Figure 1 S20 in the figure is applied to a quality detection unit of a quality detection and analysis system for polypropylene plastic particles, including a polypropylene particle quality detection module, a first processing quality detection module, a second processing quality detection module, a third processing quality detection module and a finished product contact safety detection module, which are used to perform different detections on the polypropylene plastic particles in the polypropylene plastic particle sample set to obtain a first quality detection value, a second quality detection value, a third quality detection value, a fourth quality detection value and a fifth quality detection value; the structural schematic diagram of the quality detection unit is shown in Figure 3 shown.
[0099] When making polypropylene plastic boxes for storing food, the quality inspection of polypropylene plastic particles needs to be considered from multiple key dimensions to ensure that the products made of polypropylene plastic particles meet the strict regulations of food contact materials and the final products can be used safely. In this embodiment, the quality inspection of polypropylene plastic particles is divided into five core modules: physical property inspection, durability inspection, chemical property inspection, processing performance inspection and food contact safety performance inspection. This comprehensive inspection method not only ensures that all performance indicators of the product meet industry standards, but also strengthens the competitiveness of the product in the market and the trust of consumers.
[0100] Furthermore, the process of obtaining the first quality detection value includes:
[0101] Acquire a sample image of each polypropylene plastic particle in the polypropylene plastic particle sample set;
[0102] The sample image is subjected to a polypropylene particle quality detection model to obtain a particle performance feature vector;
[0103] Acquire the first quality detection value according to the particle performance characteristic vector;
[0104] The particle performance feature vector includes size feature, shape feature and color feature.
[0105] Furthermore, the polypropylene particle quality detection model includes a preprocessing layer, a feature map extraction layer, a feature fusion layer and an output layer;
[0106] The preprocessing layer is used to perform a preprocessing operation on the sample image;
[0107] The preprocessing operation includes uniformly cropping the sizes of the sample images to obtain sample images of the same size;
[0108] Removing noise from the sample image of the same size to obtain a denoised sample image; enhancing the denoised sample image using an image enhancement algorithm to obtain an enhanced sample image;
[0109] The feature map extraction layer is used to obtain a plurality of polypropylene plastic particle feature maps according to the enhanced sample image;
[0110] The feature fusion layer is used to perform feature fusion on a plurality of feature maps of the polypropylene plastic particles to obtain a particle performance feature vector;
[0111] The output layer is used to output the particle performance feature vector.
[0112] In this embodiment, the polypropylene particle quality detection model significantly improves the quality analysis accuracy and efficiency of polypropylene plastic particles through its refined hierarchical structure. The preprocessing layer of the model performs a series of preprocessing operations on the sample images, including uniform cropping size, noise removal and image enhancement, to ensure the quality of the obtained images is optimized, laying a solid foundation for feature extraction. Next, the feature map extraction layer extracts key polypropylene plastic particle feature maps from these enhanced images, effectively capturing detailed information related to physical properties. In the feature fusion layer, by synthesizing these feature maps, a comprehensive particle performance feature vector is generated, which is finally provided by the output layer. This method not only enhances the representativeness and discrimination of the features, but also optimizes the response speed and accuracy of the entire quality detection process.
[0113] Furthermore, the process of acquiring the particle performance characteristic vector includes:
[0114] Input the sample image into a convolution layer with a size of 1×1 and 8 convolution kernels to obtain a first physical feature map;
[0115] Input the first physical feature map into a convolutional layer with a size of 3×3 and 12 convolution kernels, and add a spatial attention mechanism after the convolutional layer to obtain a second physical feature map;
[0116] The second physical feature map is input into another convolutional layer with a size of 3×3 and 12 convolution kernels, and a nonlinear transformation is performed through a ReLU activation function to obtain a third physical feature map;
[0117] Performing feature fusion on the first physical feature map, the second physical feature map, and the third physical feature map to obtain a particle performance feature vector;
[0118] In this embodiment, the particle performance feature vector acquisition process significantly improves the depth and accuracy of polypropylene plastic particle quality analysis by using a multi-layer convolutional network structure. In the initial stage, a small convolution kernel (1x1) is used to capture subtle local features, and the physical characteristics of the particles are further extracted by the continuous application of a larger convolution kernel (3x3). The added spatial attention mechanism ensures the outstanding performance of key features, and the application of the ReLU activation function enhances the nonlinear expression ability of the features, thereby obtaining a richer and more effective feature representation. Finally, by fusing feature maps at different levels, the obtained particle performance feature vector not only fully reflects the physical properties of polypropylene plastic particles, but also greatly improves the prediction accuracy and reliability of the quality detection model, providing strong data support for further quality control and classification.
[0119] Furthermore, the calculation formula of the first quality detection value includes:
[0120]
[0121] Among them, Q1 is the first quality detection value, N is the number of samples in the polypropylene plastic particle sample set, S i is the size characteristic value of the i-th sample, C i is the shape feature value of the i-th sample, F i is the color feature value of the i-th sample; α, β, γ represent the size feature coefficient, color feature coefficient and shape feature coefficient respectively, which are used to adjust the influence weight of each feature in the first quality detection value. The correlation coefficient is set according to professionals in this field.
[0122] The calculation of the first mass detection value corresponds to the detection of physical properties of the polypropylene plastic particles;
[0123] The first quality test value calculation formula provides a method to accurately evaluate the physical properties of polypropylene plastic particles by combining the size, shape and color characteristics of polypropylene plastic particles. The formula uses a weight coefficient to adjust the contribution of each feature to the total quality score, ensuring the flexibility and accuracy of the evaluation process.
[0124] Furthermore, the calculation formula of the size characteristic value is:
[0125]
[0126] Among them, S i is the size characteristic value of the i-th sample, L i ,Wi ,H i represents the length, width and height of the i-th sample, L thresh ,W thresh ,H thresh Respectively represent the set thresholds for length, width and height, ω l ,ω w ,ω h are the coefficients of length, width and height respectively, and the correlation coefficients are set according to professionals in this field;
[0127] The calculation formula of the shape characteristic value is:
[0128]
[0129] Among them, F i is the shape characteristic value of the i-th sample; R i Indicates the surface roughness of the i-th sample; Area i ,Perimeter i represent the area and perimeter of the particle in the i-th sample image, respectively, and are used to calculate the shape factor; Gyr ideal Gyr represents the gyro radius in the standard state, which represents the relationship between the moment of inertia of the polypropylene particles and the mass of the polypropylene particles, and reflects the shape and mass distribution characteristics of the polypropylene particles; i represents the gyro radius of the i-th sample shape; σ r ,σ s ,σ g are weight coefficients of surface roughness, shape factor and overall geometric symmetry, respectively, and the correlation coefficients are set by professionals in this field;
[0130] In this embodiment, examples of some physical property characteristic values of samples in the polypropylene plastic particle sample set subjected to physical property testing are shown in Table 1:
[0131] Table 1 Example of some physical properties characteristic values
[0132] serial number Length(mm) Width(mm) Height(mm) Surface roughness Form Factor Overall geometric symmetry 1 15 10 5 0.002 0.8 0.9 2 15.2 10.1 5.1 0.003 0.81 0.85 3 14.9 9.8 4.9 0.002 0.79 0.88 ... ... ... ... ... ... ...
[0133] The calculation formula of the color characteristic value is:
[0134]
[0135] Among them, C i is the color feature value of the i-th sample, p i,c represents the actual distribution probability of the i-th sample on color c, q i,crepresents the expected color distribution probability of the ith sample on color c, where C represents the color class, including transparent, white, orange, red, blue, green, yellow, and purple; The sum of the absolute values of the differences between the actual distribution probability of each color and the uniform distribution is calculated, η u ,η d The weight coefficient representing color consistency and color distribution, and the correlation coefficient are set according to professionals in this field.
[0136] In this embodiment, the quality assessment accuracy of polypropylene plastic particles is significantly improved by accurately calculating the size, shape and color characteristic values. The size characteristic value effectively monitors the size consistency of polypropylene particles by analyzing the length, width and height of each sample and comparing it with the set threshold value. The shape characteristic value combines the surface roughness, the area and perimeter of the particles and the gyro radius, which together evaluate the morphological complexity and geometric symmetry of the particles. The color characteristic value accurately reflects the color consistency and distribution range by comparing the actual and expected color distribution. These multi-dimensional feature evaluations make the quality control process more comprehensive and reliable, ensuring that the performance of polypropylene products in actual applications meets high standards, thereby enhancing the market competitiveness and consumer satisfaction of the products.
[0137] Furthermore, the calculation formula of the second quality detection value is:
[0138]
[0139] Among them, Q2 is the second detection mass value; N1 and N2 are the number of samples selected from the polypropylene plastic particle sample set for density detection and heat resistance detection, respectively, and N1 and N2 meet N is the number of samples in the polypropylene plastic pellet sample set; D jm is the density value detected by the jmth sample; HD jr is the thermal deformation temperature value of the jrth sample tested; D std ,HD std are the standard values of density and heat deformation temperature respectively; d1 and d2 are the weight coefficients of density value and heat deformation temperature value respectively, which are used to adjust the influence weight of density and heat resistance in the second quality detection value, and the correlation coefficient is set according to professionals in this field.
[0140] The calculation of the second mass detection value corresponds to the detection of the durability of the polypropylene plastic particles;
[0141] In the quality inspection of polypropylene plastic particles, density and heat deformation temperature are selected as inspection indicators, mainly because these two parameters critically reflect the consistency and structural stability of the material. Density not only reveals the tightness and uniformity of the material, but is also crucial for predicting the mechanical strength and durability of the product. The heat deformation temperature provides the material's ability to maintain performance under heating conditions. It is an important indicator for evaluating the feasibility of using polypropylene plastic particles in high temperature environments, especially in application scenarios that need to withstand certain temperature conditions, such as automotive parts and food packaging materials. Through the detection of these two parameters, the safety and functionality of polypropylene products in actual use can be ensured, and product quality can be guaranteed.
[0142] The calculation formula of the second quality test value provides an efficient method to comprehensively evaluate the density and heat resistance of polypropylene plastic particles, two key physical properties. By accurately measuring the density and heat deformation temperature of selected samples and comparing these values with preset standard values, this method can not only accurately monitor the consistency and quality of the material, but also optimize the emphasis on these characteristics by adjusting the weight coefficient.
[0143] Furthermore, the calculation formula of the third quality detection value is:
[0144]
[0145] Among them, Q3 is the third quality test value; N3 and N4 are the number of samples selected from the polypropylene plastic particle sample set for purity test and isotacticity test, respectively, and N3 and N4 meet N is the number of samples in the polypropylene plastic particle sample set; P jc is the purity value of the jcth sample detected; I jd is the isotacticity value of the jdth sample detection; P std ,I std are the standard values of purity and isotacticity, respectively; d3 and d4 are the weight coefficients of purity value and isotacticity value, respectively, which are used to adjust the influence weight of purity and isotacticity in the third quality detection value, and the correlation coefficient is set according to professionals in this field;
[0146] The calculation of the third mass detection value corresponds to the detection of chemical properties of the polypropylene plastic particles;
[0147] The process of obtaining the purity value includes:
[0148] Dissolving the first sample to be detected in a dichloromethane solvent to obtain a polypropylene solution; analyzing the polypropylene solution using a gas chromatography-mass spectrometer to obtain the concentrations of all compounds in the first sample to be detected;
[0149] The gas chromatograph-mass spectrometer is provided with chromatographic conditions and mass spectrometric conditions; the chromatographic conditions include using a 5% phenylmethylsiloxane column, setting a temperature program from 40°C for 1 minute, heating at 10°C / min to 280°C and maintaining for 10 minutes; the mass spectrometric conditions include setting an ion source temperature of 250°C, a scanning range of 50 to 600 m / z, and using an electron impact (EI) mode;
[0150] In order to verify the effectiveness of the polypropylene particle quality detection model proposed in this embodiment, a comparative experiment was carried out in this embodiment. The comparative models include Model 2 and Model 3. The characteristic detection accuracy of the other two models was obtained by referring to the calculation process of the polypropylene particle quality detection model. The comparative experiment was carried out five times. The results of the comparative experiment are shown in Table 2.
[0151] Table 2 Comparative experimental results
[0152] Number of experiments Model 1 Model 2 Model 3 1 95.77% 93.43% 90.65% 2 93.22% 93.14% 90.86% 3 95.52% 93.62% 90.47% 4 95.73% 93.02% 90.45% 5 94.35% 92.84% 89.51%
[0153] Referring to Table 2, Model 1 is the polypropylene particle quality detection model proposed by the present invention, Model 2 is based on Model 1 without adding the spatial attention mechanism, and Model 3 is based on Model 2 without adding the ReLU activation function. It can be seen from the results in Table 2 that the polypropylene particle quality detection model proposed in this embodiment can accurately obtain the particle performance characteristic vector of the polypropylene plastic particles.
[0154] The calculation formula of the purity value is:
[0155]
[0156] Wherein, P is the purity value, T is the total concentration of the sample, and ∑T1 represents the total concentration of all impurities in the first sample to be tested;
[0157] The isotacticity value is obtained by a nuclear magnetic resonance spectrometer;
[0158] The calculation formula of the isotacticity value is:
[0159]
[0160] Among them, I is the isotacticity value, A ij is the area of the peak corresponding to the isotactic arrangement of methyl side chains in the NMR spectra, A t It is the total area of all peaks in the NMR spectra.
[0161] In the quality inspection of polypropylene plastic granules, purity and isotacticity are selected as the detection indicators of key chemical properties because these two parameters profoundly affect the chemical structure and performance of the material. Purity determines the content of impurities in polypropylene. High purity ensures the consistency of plastic performance and the predictability of the processing process, which is related to the strength and transparency of the final product. Isotacticity describes the regular arrangement of polymer chains and directly affects the crystallinity and mechanical strength of the material. By accurately measuring these two parameters, the performance standards of polypropylene plastic granules in various applications, such as packaging materials and other industrial applications, can be ensured, thereby improving product quality and market competitiveness.
[0162] The calculation of the third quality test value provides an accurate evaluation method for the chemical properties of polypropylene plastic particles by combining the test results of two key chemical properties, purity and isotacticity. Purity testing uses a gas chromatograph-mass spectrometer to analyze a polypropylene solution dissolved in dichloromethane to accurately determine the total concentration of impurities in the sample, while the isotacticity value is obtained by a nuclear magnetic resonance spectrometer to analyze the molecular structure arrangement of polypropylene in detail. This comprehensive chemical performance testing method not only ensures the high quality of polypropylene products at the molecular level, but also accurately reflects the actual impact of these chemical properties on the performance of polypropylene products by adjusting the weight coefficients of purity and isotacticity.
[0163] Furthermore, the calculation formula of the fourth quality detection value is:
[0164]
[0165] Among them, Q4 is the fourth quality detection value; N5 and N6 are the number of samples in the polypropylene plastic particle sample set selected for the first flow rate detection and the second flow rate detection, and N5 and N6 meet N is the number of samples in the polypropylene plastic pellet sample set; MFR jz The first flow rate detected for the jzth sample; MVR jt The second flow rate detected for the jtth sample; MFR std ,MVR std are standard values of the first flow rate and the second flow rate, respectively; d5 and d6 are weight coefficients of the first flow rate and the second flow rate, respectively, used to adjust the influence weights of the first flow rate and the second flow rate in the fourth quality detection value, and the correlation coefficients are set according to professionals in the field;
[0166] The calculation of the fourth quality detection value corresponds to the detection of the processing performance of the polypropylene plastic particles;
[0167] The process of obtaining the first flow rate and the second flow rate includes:
[0168] Using a weighing device to obtain the weight of the second sample to be tested;
[0169] Preheating the second rate measuring instrument to a standard test temperature for polypropylene; the standard test temperature is 230°C;
[0170] Loading the second sample to be tested into the barrel of the flow rate measuring instrument, and ensuring that the barrel is completely filled;
[0171] Apply a prescribed load according to ISO standards; the prescribed load is 2.16 kg;
[0172] The piston of the barrel is opened, and the polypropylene melt flows out through the mold under the load;
[0173] Record the mass of the melt within a specified time;
[0174] The calculation formula of the first flow rate is:
[0175]
[0176] Wherein, MFR is the first flow rate, W represents the mass of the polypropylene melt flowing out within a specified time, and T represents the specified time;
[0177] The calculation formula of the second flow rate is:
[0178]
[0179] Wherein, MVR is the second flow rate, and ρ represents the density of the second sample to be detected.
[0180] The first flow rate (MFR) and the second flow rate are key parameters for evaluating the fluidity of polypropylene during processing. The fourth quality test value is calculated by measuring the first flow rate (MFR) and the second flow rate (MVR) of the sample to be tested and comparing them with the preset standard values, and further applying the weight coefficient to adjust the impact of these two parameters in the final quality score, which can accurately reflect the processing quality of polypropylene particles. This method not only improves the evaluation accuracy of polypropylene processing performance, but also helps manufacturers optimize production processes and improve product quality, ensuring that polypropylene products exhibit excellent performance in actual applications.
[0181] Furthermore, the calculation formula of the fifth quality detection value is:
[0182]
[0183] Among them, Q5 is the fifth quality test value; N7 and N8 are the number of samples selected from the polypropylene plastic particle sample set for volatile gas detection and food contact safety detection, and N7 and N8 meet N is the number of samples in the polypropylene plastic pellet sample set; VL jq is the volatile gas detection value of the jqth sample; SL ja VL is the environmental protection test value of the jath sample; std ,SL std They are the standard values of the volatile gas detection value and the environmental protection detection value respectively; d7 and d8 are the weight coefficients of the volatile gas detection value and the environmental protection detection value respectively, which are used to adjust the influence weights of the volatile gas detection value and the environmental protection detection value in the fifth quality detection value, and the correlation coefficients are set according to professionals in this field.
[0184] The calculation of the fifth mass test value corresponds to the testing of food contact safety performance of polypropylene plastic particles;
[0185] The process of obtaining the volatile gas detection value includes:
[0186] Placing the third sample to be tested in a sealed container and heating it at a temperature of 120° C.;
[0187] Using a gas chromatograph-mass spectrometer to analyze the air in the sealed container after heating to obtain the types and concentrations of various volatile gases; comparing the types and concentrations of the various volatile gases with environmental safety standards to obtain volatile gas detection values;
[0188] The process of obtaining the environmental detection value includes:
[0189] soaking the fourth test sample in a food simulant; the food simulant is ethanol;
[0190] The extracts were collected and chemically analyzed using liquid chromatography;
[0191] The calculation formula of the environmental protection detection value is:
[0192]
[0193] Among them, SL is the environmental protection test value, CN h is the concentration of harmful chemicals detected, and SF is the food safety threshold.
[0194] By comprehensively considering the results of volatile gas detection and food contact safety testing, the calculation of the fifth quality test value not only includes the type and concentration of volatile gases released by the sample under heating conditions, but also evaluates the migration of chemical substances after the sample comes into contact with food simulants. The analysis performed using gas chromatography-mass spectrometry and liquid chromatography ensures the accuracy and reliability of the test. By weighting these key safety indicators, the fifth quality test value can accurately reflect the safety of the material in actual use, thereby ensuring the compliance of polypropylene plastic particles in applications such as food packaging.
[0195] Furthermore, the calculation formula of the comprehensive quality detection value is:
[0196] Q z =φ1*Q1+φ2*Q2+φ3*Q3+φ4*Q4+φ5*Q5;
[0197] Among them, Q z is the comprehensive quality detection value, Q1, Q2, Q3, Q4, Q5 are the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value respectively; φ1, φ2, φ3, φ4, φ5 are the correlation coefficients of the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value respectively.
[0198] The calculation of the comprehensive quality test value comprehensively considers the impact of the first quality test value, the second quality test value, the third quality test value, the fourth quality test value and the fifth quality test value on the comprehensive quality assessment, and provides a comprehensive assessment of the overall quality of polypropylene plastic particles. This method can not only systematically evaluate the quality of polypropylene plastic particles from multiple key dimensions, including physical properties, durability, chemical properties, processing properties and food contact safety performance, but also make appropriate weight adjustments based on the importance of each dimension in the overall performance. Through this comprehensive and quantitative evaluation method, it ensures that each performance is properly reflected in the overall quality judgment, significantly improves the accuracy of product quality control and the efficiency of operation, and helps manufacturers optimize product quality and meet higher safety and performance standards.
[0199] In this embodiment, a quality inspection and analysis system for polypropylene plastic particles is proposed. First, the sample set construction unit ensures the consistency and representativeness of the sample by collecting multiple polypropylene plastic particles produced in the same batch, laying the foundation for subsequent quality analysis. Then, the quality inspection unit conducts a detailed analysis of the sample set through the polypropylene particle quality inspection module, the first processing quality inspection module, the second processing quality inspection module, the third processing quality inspection module and the finished product contact safety inspection module. The test results of each module correspond to a specific quality inspection value, ensuring the overall quality of the product from multiple dimensions. Finally, the comprehensive evaluation unit calculates a comprehensive quality inspection value based on all the individual test results to determine whether the batch of polypropylene plastic particles meets the quality standards. This comprehensive evaluation method significantly improves the transparency and traceability of product quality, and provides a reliable guarantee for the production of high-quality polypropylene plastic particles.
[0200] Embodiment 2
[0201] This example tests the polypropylene plastic particles produced by Factory B;
[0202] Reference Figure 1 S10 in the embodiment of the present invention is applied to a sample set construction unit of a quality inspection and analysis system for polypropylene plastic particles, and is used to collect a plurality of polypropylene plastic particles produced in the same batch to construct a sample set of polypropylene plastic particles;
[0203] Reference Figure 1 S20 in which S20 is applied to a quality inspection unit of a quality inspection and analysis system for polypropylene plastic particles, including a polypropylene particle quality inspection module, a first processing quality inspection module, a second processing quality inspection module, a third processing quality inspection module and a finished product contact safety inspection module, and is used to perform different inspections on the polypropylene plastic particles in the polypropylene plastic particle sample set, respectively, to obtain a first quality inspection value, a second quality inspection value, a third quality inspection value, a fourth quality inspection value and a fifth quality inspection value;
[0204] Furthermore, the process of obtaining the first quality detection value includes:
[0205] Acquire a sample image of each polypropylene plastic particle in the polypropylene plastic particle sample set;
[0206] The sample image is subjected to a polypropylene particle quality detection model to obtain a particle performance feature vector;
[0207] Acquire the first quality detection value according to the particle performance characteristic vector;
[0208] The particle performance feature vector includes size feature, shape feature and color feature.
[0209] Furthermore, the polypropylene particle quality detection model includes a preprocessing layer, a feature map extraction layer, a feature fusion layer and an output layer;
[0210] The preprocessing layer is used to perform a preprocessing operation on the sample image; the preprocessing operation includes uniformly cropping the size of the sample image to obtain sample images of the same size; removing noise from the sample images of the same size to obtain a denoised sample image; and enhancing the denoised sample image using an image enhancement algorithm to obtain an enhanced sample image;
[0211] The feature map extraction layer is used to obtain a plurality of polypropylene plastic particle feature maps according to the enhanced sample image;
[0212] The feature fusion layer is used to perform feature fusion on a plurality of feature maps of the polypropylene plastic particles to obtain a particle performance feature vector;
[0213] The output layer is used to output the particle performance feature vector.
[0214] Furthermore, the process of acquiring the particle performance characteristic vector includes:
[0215] Input the sample image into a convolution layer with a size of 1×1 and 8 convolution kernels to obtain a first physical feature map;
[0216] Input the first physical feature map into a convolutional layer with a size of 3×3 and 12 convolution kernels, and add a spatial attention mechanism after the convolutional layer to obtain a second physical feature map;
[0217] The second physical feature map is input into another convolutional layer with a size of 3×3 and 12 convolution kernels, and a nonlinear transformation is performed through a ReLU activation function to obtain a third physical feature map;
[0218] Performing feature fusion on the first physical feature map, the second physical feature map, and the third physical feature map to obtain a particle performance feature vector;
[0219] The size characteristics include the length, width and height of the polypropylene plastic particles; the shape characteristics include the surface roughness, shape factor and overall geometric symmetry of the polypropylene particles; the color characteristics include the color consistency and color distribution of the polypropylene plastic particles. Furthermore, the calculation formula of the first quality detection value includes:
[0220]
[0221] Among them, Q1 is the first quality detection value, N is the number of samples in the polypropylene plastic particle sample set, S i is the size characteristic value of the i-th sample, Ci is the shape feature value of the i-th sample, F i is the color feature value of the i-th sample; α, β, γ represent the size feature coefficient, color feature coefficient and shape feature coefficient respectively, which are used to adjust the influence weight of each feature in the first quality detection value, and the correlation coefficient is set according to professionals in this field.
[0222] Furthermore, the calculation formula of the size characteristic value is:
[0223]
[0224] Among them, S i is the size characteristic value of the i-th sample, L i ,W i ,H i represents the length, width and height of the i-th sample, L thresh ,W thresh ,H thresh Respectively represent the set thresholds for length, width and height, ω l ,ω w ,ω h are the coefficients of length, width and height respectively, and the correlation coefficients are set according to professionals in this field;
[0225] The calculation formula of the shape characteristic value is:
[0226]
[0227] Among them, F i is the shape characteristic value of the i-th sample; R i Indicates the surface roughness of the i-th sample; Area i ,Perimeter i represent the area and perimeter of the particle in the i-th sample image, respectively, and are used to calculate the shape factor; Gyr ideal Gyr represents the gyro radius in the standard state, which represents the relationship between the moment of inertia of the polypropylene particles and the mass of the polypropylene particles, and reflects the shape and mass distribution characteristics of the polypropylene particles; i represents the gyro radius of the i-th sample shape; σ r ,σ s ,σ g are the weight coefficients of surface roughness, shape factor and overall geometric symmetry, respectively;
[0228] The calculation formula of the color characteristic value is:
[0229]
[0230] Among them, C iis the color feature value of the i-th sample, p i,c represents the actual distribution probability of the i-th sample on color c, q i,c represents the expected color distribution probability of the ith sample on color c, where C represents the color class, including transparent, white, orange, red, blue, green, yellow, and purple; The sum of the absolute values of the differences between the actual distribution probability of each color and the uniform distribution is calculated, η u ,η d The weight coefficient representing color consistency and color distribution, and the correlation coefficient are set according to professionals in this field.
[0231] Furthermore, the calculation formula of the second quality detection value is:
[0232]
[0233] Among them, Q2 is the second detection mass value; N1 and N2 are the number of samples selected from the polypropylene plastic particle sample set for density detection and heat resistance detection, respectively, and N1 and N2 meet N is the number of samples in the polypropylene plastic pellet sample set; D jm is the density value detected by the jmth sample; HD jr is the thermal deformation temperature value of the jrth sample tested; D std ,HD std are the standard values of density and heat deformation temperature respectively; d1 and d2 are the weight coefficients of density value and heat deformation temperature value respectively, which are used to adjust the influence weight of density and heat resistance in the second quality detection value, and the correlation coefficient is set according to professionals in this field.
[0234] Furthermore, the calculation formula of the third quality detection value is:
[0235]
[0236] Among them, Q3 is the third quality test value; N3 and N4 are the number of samples selected from the polypropylene plastic particle sample set for purity test and isotacticity test, respectively, and N3 and N4 meet N is the number of samples in the polypropylene plastic particle sample set; P jc is the purity value of the jcth sample detected; I jd is the isotacticity value of the jdth sample detection; P std ,I std are the standard values of purity and isotacticity, respectively; d3 and d4 are the weight coefficients of purity and isotacticity, respectively, which are used to adjust the influence weights of purity and isotacticity in the third quality test value, and the correlation coefficients are set according to professionals in this field;
[0237] In order to verify the effectiveness of the polypropylene particle quality detection model proposed in this embodiment, a comparative experiment was carried out in this embodiment. The comparative models include model four and model five. Referring to the calculation process of the polypropylene particle quality detection model, the characteristic detection accuracy of the other two models was obtained. The comparative experiment was carried out five times. The results of the comparative experiment are shown in Table 3.
[0238] Table 3 Comparative experimental results
[0239] Number of experiments Model 1 Model 4 Model 5 1 95.77% 93.65% 92.53% 2 93.22% 92.57% 92.46% 3 95.52% 94.37% 91.95% 4 95.73% 94.86% 92.68% 5 94.35% 93.67% 91.27%
[0240] Referring to Table 3, Model 1 is the polypropylene particle quality detection model proposed by the present invention, Model 4 is the LSTM model, and Model 5 is the GRU model. From the results in Table 3, it can be seen that the polypropylene particle quality detection model proposed in this embodiment can accurately obtain the particle performance characteristic vector of the polypropylene plastic particles.
[0241] The process of obtaining the purity value includes:
[0242] Dissolving the first sample to be tested in a dichloromethane solvent to obtain a polypropylene solution;
[0243] Analyzing the polypropylene solution using a gas chromatography-mass spectrometer to obtain the concentrations of all compounds in the first sample to be detected;
[0244] The gas chromatograph-mass spectrometer is provided with chromatographic conditions and mass spectrometric conditions; the chromatographic conditions include using a 5% phenylmethylsiloxane column, setting a temperature program from 40°C for 1 minute, heating at 10°C / min to 280°C and maintaining for 10 minutes; the mass spectrometric conditions include setting an ion source temperature of 250°C, a scanning range of 50 to 600 m / z, and using an electron impact (EI) mode;
[0245] The calculation formula of the purity value is:
[0246]
[0247] Where P is the purity value, T is the total concentration of the polypropylene solution, and ∑T1 represents the total concentration of all impurities in the polypropylene solution;
[0248] The isotacticity value is obtained by a nuclear magnetic resonance spectrometer;
[0249] The calculation formula of the isotacticity value is:
[0250]
[0251] Among them, I is the isotacticity value, A ij is the area of the peak corresponding to the isotactic arrangement of methyl side chains in the NMR spectra, A tIt is the total area of all peaks in the NMR spectra.
[0252] Furthermore, the calculation formula of the fourth quality detection value is:
[0253]
[0254] Among them, Q4 is the fourth quality detection value; N5 and N6 are the number of samples in the polypropylene plastic particle sample set selected for the first flow rate detection and the second flow rate detection, and N5 and N6 meet N is the number of samples in the polypropylene plastic pellet sample set; MFR jz The first flow rate detected for the jzth sample; MVR jt The second flow rate detected for the jtth sample; MFR std ,MVR std are standard values of the first flow rate and the second flow rate, respectively; d5 and d6 are weight coefficients of the first flow rate and the second flow rate, respectively, used to adjust the influence weights of the first flow rate and the second flow rate in the fourth quality detection value, and the correlation coefficients are set according to professionals in the field;
[0255] The process of obtaining the first flow rate and the second flow rate includes:
[0256] Using a weighing device to obtain the weight of the second sample to be tested;
[0257] Preheat the melt flow rate measuring instrument to the standard test temperature of polypropylene; the standard test temperature is 230°C;
[0258] Loading the second sample to be tested into the barrel of the flow rate measuring instrument, and ensuring that the barrel is completely filled;
[0259] Apply a prescribed load according to ISO standards; the prescribed load is 2.16 kg;
[0260] The piston of the barrel is opened, and the polypropylene melt flows out through the mold under the load;
[0261] Record the mass of the melt within a specified time;
[0262] The calculation formula of the first flow rate is:
[0263]
[0264] Wherein, MFR is the first flow rate, W represents the mass of the polypropylene melt flowing out within a specified time, and T represents the specified time;
[0265] The calculation formula of the second flow rate is:
[0266]
[0267] Wherein, MVR is the second flow rate, and ρ represents the density of the second sample to be detected.
[0268] Furthermore, the calculation formula of the fifth quality detection value is:
[0269]
[0270] Among them, Q5 is the fifth quality test value; N7 and N8 are the number of samples selected from the polypropylene plastic particle sample set for volatile gas detection and food contact safety detection, and N7 and N8 meet N is the number of samples in the polypropylene plastic pellet sample set; VL jq is the volatile gas detection value of the jqth sample; SL ja VL is the environmental protection test value of the jath sample; std ,SL std are the standard values of the volatile gas detection value and the environmental protection detection value respectively; d7 and d8 are the weight coefficients of the volatile gas detection value and the environmental protection detection value respectively, which are used to adjust the influence weights of the volatile gas detection value and the environmental protection detection value in the fifth quality detection value, and the correlation coefficients are set according to professionals in this field;
[0271] The process of obtaining the volatile gas detection value includes:
[0272] Placing the third sample to be tested in a sealed container and heating it at a temperature of 120° C.;
[0273] Using a gas chromatograph-mass spectrometer to analyze the air in the sealed container after heating to obtain the types and concentrations of various volatile gases; comparing the types and concentrations of the various volatile gases with environmental safety standards to obtain volatile gas detection values;
[0274] The process of obtaining the environmental detection value includes:
[0275] The fourth test sample is immersed in a food simulant; the food simulant is ethanol; the extract is collected and chemically analyzed using a liquid chromatograph; the calculation formula of the environmental protection test value is:
[0276]
[0277] Among them, SL is the environmental protection test value, CN h is the concentration of the detected harmful chemical substances, and SF is the food safety threshold. Furthermore, the calculation formula of the comprehensive quality detection value is:
[0278] Qz =φ1*Q1+φ2*Q2+φ3*Q3+φ4*Q4+φ5*Q5;
[0279] Among them, Q z is the comprehensive quality detection value, Q1, Q2, Q3, Q4, Q5 are the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value respectively, φ1, φ2, φ3, φ4, φ5 are the correlation coefficients of the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value respectively, and the correlation coefficients are set according to professionals in this field.
[0280] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A quality detection and analysis system for polypropylene plastic particles, characterized in that: include: A sample set building unit, used for collecting a plurality of polypropylene plastic particles produced in the same batch to build a polypropylene plastic particle sample set; The quality detection unit includes a polypropylene particle quality detection module, a first processing quality detection module, a second processing quality detection module, a third processing quality detection module and a finished product contact safety detection module, and is used to perform different tests on the polypropylene plastic particles in the polypropylene plastic particle sample set to obtain a first quality detection value, a second quality detection value, a third quality detection value, a fourth quality detection value and a fifth quality detection value; The comprehensive evaluation unit is used to obtain a comprehensive quality detection value according to the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value; the comprehensive quality detection value is used to judge whether the quality of the polypropylene plastic particles produced in the same batch is qualified.
2. A quality detection and analysis system for polypropylene plastic particles according to claim 1, characterized in that: The process of obtaining the first quality detection value includes: Acquire a sample image of each polypropylene plastic particle in the polypropylene plastic particle sample set; The sample image is subjected to a polypropylene particle quality detection model to obtain a particle performance feature vector; Acquire the first quality detection value according to the particle performance characteristic vector; The particle performance feature vector includes size feature, shape feature and color feature; The polypropylene particle quality detection model includes a preprocessing layer, a feature map extraction layer, a feature fusion layer and an output layer; The preprocessing layer is used to perform a preprocessing operation on the sample image; the preprocessing operation includes uniformly cropping the size of the sample image to obtain sample images of the same size; removing noise from the sample images of the same size to obtain a denoised sample image; and enhancing the denoised sample image using an image enhancement algorithm to obtain an enhanced sample image; The feature map extraction layer is used to obtain a plurality of polypropylene plastic particle feature maps according to the enhanced sample image; The feature fusion layer is used to perform feature fusion on a plurality of feature maps of the polypropylene plastic particles to obtain a particle performance feature vector; The output layer is used to output the particle performance feature vector.
3. A quality detection and analysis system for polypropylene plastic particles according to claim 2, characterized in that: The process of obtaining the particle performance characteristic vector includes: Input the sample image into a convolution layer with a size of 1×1 and 8 convolution kernels to obtain a first physical feature map; Input the first physical feature map into a convolutional layer with a size of 3×3 and 12 convolution kernels, and add a spatial attention mechanism after the convolutional layer to obtain a second physical feature map; The second physical feature map is input into another convolutional layer with a size of 3×3 and 12 convolution kernels, and a nonlinear transformation is performed through a ReLU activation function to obtain a third physical feature map; Performing feature fusion on the first physical feature map, the second physical feature map, and the third physical feature map to obtain a particle performance feature vector; The size characteristics include the length, width and height of the polypropylene plastic particles; the shape characteristics include the surface roughness, shape factor and overall geometric symmetry of the polypropylene particles; and the color characteristics include the color consistency and color distribution of the polypropylene plastic particles.
4. A quality detection and analysis system for polypropylene plastic particles according to claim 1, characterized in that: The calculation formula of the first quality detection value includes: Among them, Q1 is the first quality detection value, N is the number of samples in the polypropylene plastic particle sample set, S i is the size characteristic value of the i-th sample, C i is the shape feature value of the i-th sample, F i is the color feature value of the i-th sample; α, β, γ represent the size feature coefficient, color feature coefficient, and shape feature coefficient, respectively.
5. A quality detection and analysis system for polypropylene plastic particles according to claim 4, characterized in that: The calculation formula of the size characteristic value is: Among them, S i is the size characteristic value of the i-th sample, L i ,W i ,H i represents the length, width and height of the i-th sample, L thresh ,W thresh ,H thresh Respectively represent the set thresholds for length, width and height, ω l ,ω w ,ω h are the coefficients of length, width and height respectively; The calculation formula of the shape characteristic value is: Among them, F i is the shape characteristic value of the i-th sample; R i Indicates the surface roughness of the i-th sample; Area i ,Perimeter i represent the area and perimeter of the particle in the i-th sample image, respectively, and are used to calculate the shape factor; Gyr ideal Gyr represents the gyro radius in the standard state, which represents the relationship between the moment of inertia of the polypropylene plastic particles and the mass of the polypropylene plastic particles, and reflects the shape and mass distribution characteristics of the polypropylene plastic particles; Gyr i represents the gyro radius of the i-th sample shape; σ r ,σ s ,σ g are the weight coefficients of surface roughness, shape factor and overall geometric symmetry, respectively; The calculation formula of the color characteristic value is: Among them, C i is the color feature value of the i-th sample, p i,c represents the actual distribution probability of the i-th sample on color c, q i,c represents the expected color distribution probability of the i-th sample on color c, C represents the color class; η u ,η d A weight coefficient representing color consistency and color distribution.
6. A quality detection and analysis system for polypropylene plastic particles according to claim 1, characterized in that: The calculation formula of the second quality detection value is: Among them, Q2 is the second detection mass value; N1 and N2 are the number of samples selected from the polypropylene plastic particle sample set for density detection and heat resistance detection, respectively, and N1 and N2 meet N is the number of samples in the polypropylene plastic pellet sample set; D jm is the density value detected by the jmth sample; HD jr is the thermal deformation temperature value of the jrth sample tested; D std ,HD std are the standard values of density and heat deformation temperature respectively; d1 and d2 are the weight coefficients of density and heat deformation temperature respectively.
7. A quality detection and analysis system for polypropylene plastic particles according to claim 1, characterized in that: The calculation formula of the third quality detection value is: Among them, Q3 is the third quality test value; N3 and N4 are the number of samples selected from the polypropylene plastic particle sample set for purity test and isotacticity test, respectively, and N3 and N4 meet N is the number of samples in the polypropylene plastic particle sample set; P jc is the purity value of the jcth sample detected; I jd is the isotacticity value of the jdth sample detection; P std ,I std are the standard values of purity and isotacticity respectively; d3 and d4 are the weight coefficients of purity and isotacticity respectively; The polypropylene plastic particle sample to be tested for purity value is divided into a first sample to be tested; the first sample to be tested is dissolved in a dichloromethane solvent to obtain a polypropylene solution; the purity value is obtained by a gas chromatography-mass spectrometer, and the specific calculation formula is: The purity value is obtained by gas chromatography-mass spectrometry, and the specific calculation formula is: Where P is the purity value, T is the total concentration of the polypropylene solution, and ∑T1 represents the total concentration of all impurities in the polypropylene solution; The isotacticity value is obtained by a nuclear magnetic resonance spectrometer, and the specific calculation formula is: Among them, I is the isotacticity value, A ij is the area of the peak corresponding to the isotactic arrangement of methyl side chains in the NMR spectra, A t It is the total area of all peaks in the NMR spectra.
8. A quality detection and analysis system for polypropylene plastic particles according to claim 1, characterized in that: The calculation formula of the fourth quality detection value is: Among them, Q4 is the fourth quality detection value; N5 and N6 are the number of samples in the polypropylene plastic particle sample set selected for the first flow rate detection and the second flow rate detection, and N5 and N6 meet N is the number of samples in the polypropylene plastic pellet sample set; MFR jz The first flow rate detected for the jzth sample; MVR jt The second flow rate detected for the jtth sample; MFR std ,MVR std are standard values of the first flow rate and the second flow rate, respectively; d5 and d6 are weight coefficients of the first flow rate and the second flow rate, respectively; The calculation formula of the first flow rate is: Wherein, MFR is the first flow rate, W represents the mass of the polypropylene melt flowing out of the flow rate measuring instrument within a specified time, and T represents the specified time; The calculation formula of the second flow rate is: Wherein, MVR is the second flow rate, and ρ represents the density of the second sample to be detected.
9. A quality detection and analysis system for polypropylene plastic particles according to claim 1, characterized in that: The calculation formula of the fifth quality detection value is: Among them, Q5 is the fifth quality test value; N7 and N8 are the number of samples selected from the polypropylene plastic particle sample set for volatile gas detection and food contact safety detection, and N7 and N8 meet N is the number of samples in the polypropylene plastic pellet sample set; VL jq is the volatile gas detection value of the jqth sample; SL ja VL is the environmental protection test value of the jath sample; std ,SL std are the standard values of the volatile gas detection value and the environmental protection detection value respectively; d7 and d8 are the weight coefficients of the volatile gas detection value and the environmental protection detection value respectively, which are used to adjust the influence weights of the volatile gas detection value and the environmental protection detection value in the fifth quality detection value; The process of obtaining the volatile gas detection value includes: placing the third sample to be tested in a sealed container and heating it; using a gas chromatograph-mass spectrometer to analyze the air in the sealed container after heating to obtain the types and concentrations of various volatile gases; comparing the types and concentrations of the various volatile gases with environmental safety standards to obtain volatile gas detection values; The environmental protection detection value is obtained by immersing the fourth test sample in a food simulant, collecting the corresponding extract and performing chemical analysis using a liquid chromatograph; the food simulant is ethanol; the calculation formula of the environmental protection detection value is: Among them, SL is the environmental protection test value, CN h is the concentration of harmful chemicals detected, and SF is the food safety threshold.
10. A quality detection and analysis system for polypropylene plastic particles according to claim 1, characterized in that: The calculation formula of the comprehensive quality detection value is: Q z =φ1*Q1+φ2*Q2+φ3*Q3+φ4*Q4+φ5*Q5; Among them, Q z is the comprehensive quality detection value, Q1, Q2, Q3, Q4, Q5 are the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value respectively, φ1, φ2, φ3, φ4, φ5 are the correlation coefficients of the first quality detection value, the second quality detection value, the third quality detection value, the fourth quality detection value and the fifth quality detection value respectively.
Citation Information
Cited By
Plastic recycled material purity monitoring method based on artificial intelligence
CN120581087A
