A method, system and device for monitoring the operating state of a material degradation granulator

Abstract

The application discloses a kind of degrading material granulator operating state monitoring method, system and equipment, comprising the following steps: in unit monitoring period, obtain the multiple monitoring images of target strip and multiple CT scanning images;With respectively obtain the surface defect trend value and porosity trend value of target strip;Obtain the running abnormal trend value of granulator;Surface defect trend value, porosity trend value and running abnormal trend value are input into the risk assessment model of pre-set, to obtain risk assessment value;Judge whether risk assessment value is greater than pre-set risk threshold value, if yes, then send first alarm information, if no, then obtain the current weight of granular material from the pelletizing mechanism of granulator;Judge whether current weight is in pre-set weight threshold interval, if no, then send second alarm information, the application has improved the advantages of abnormal monitoring accuracy of granulation process.

Description

Technical Field

[0001] This application relates to the field of equipment operation monitoring technology, and in particular to a method, system and equipment for monitoring the operating status of a biodegradable material granulator. Background Technology

[0002] Degradable materials are materials that can decompose into harmless substances under specific conditions. They have received widespread attention due to their environmental protection characteristics. In the production process of degradable materials, granulation is an important process step, which can process granular degradable materials into granules for easy transportation and application.

[0003] Currently, when producing granules from biodegradable materials using a granulator, the biodegradable materials are first stirred and mixed before being extruded into strips. These strips are then conveyed and subjected to processes such as air drying and cooling before finally being granulated. This process requires constant monitoring of the granulation production status. If any abnormalities are detected in the granulation process (including product abnormalities and equipment malfunctions), the granulator must be immediately shut down for maintenance. Currently, this is mainly done through manual visual monitoring. However, due to the large number of monitoring items, there is a high risk of missed or incorrect detections, leading to inaccurate monitoring. Summary of the Invention

[0004] The main objective of this application is to provide a method, system, and equipment for monitoring the operating status of a degradable material granulator, aiming to solve the technical problem that existing granulator monitoring methods have low accuracy in monitoring abnormalities during the granulation process.

[0005] To achieve the above objectives, this application provides a method for monitoring the operating status of a biodegradable material granulator, comprising the following steps:

[0006] Multiple monitoring images and multiple CT scan images of the target material strip are acquired within a unit monitoring cycle; the target material strip is the strip that comes out of the extrusion mechanism of the granulator, and the target material strips in the multiple monitoring images are spliced ​​together to form a strip with a preset length;

[0007] Based on multiple monitoring images, the surface defect trend value of the target material strip is obtained;

[0008] Based on multiple CT scan images, the porosity trend value of the target material strip is obtained;

[0009] Obtain abnormal operating trend values ​​of the granulator;

[0010] Input the surface defect trend value, porosity trend value, and operational anomaly trend value into the preset risk assessment model to obtain the risk assessment value;

[0011] Determine whether the risk assessment value is greater than the preset risk threshold. If so, send the first alarm message. If not, obtain the current weight of the granulated material coming out of the pelletizing mechanism of the pelletizer.

[0012] Determine whether the current weight is within the preset weight threshold range. If not, send a second alarm message. If yes, return to the multiple monitoring images and multiple CT scan images of the target material strip acquired within the unit monitoring cycle.

[0013] Optionally, based on multiple monitoring images, surface defect trend values ​​of the target strip are obtained, including:

[0014] Identify the fracture characteristics of the target material strip in each monitoring image to obtain the fracture defect trend value η1 of the target material strip;

[0015] Identify the abnormal diameter features of the target material strip in each monitoring image to obtain the abnormal diameter trend value η2 of the target material strip;

[0016] Identify the color dispersion of the target material strip in each monitoring image to obtain the color dispersion trend value η3;

[0017] The surface defect trend value η is obtained, and the expression for η is:

[0018] η=K1·η1+K2·η2+K3·η3;

[0019] In the formula, K1 is the first adjustment coefficient, K2 is the second adjustment coefficient, and K3 is the third adjustment coefficient.

[0020] Optionally, the fracture characteristics of the target strip in each monitoring image are identified to obtain the fracture defect trend value η1 of the target strip, including:

[0021] The monitoring images that do not have fracture characteristics are removed, and the remaining monitoring images are sorted according to time series.

[0022] Obtain the fracture characteristic value ΔL of the target material strip in each sorted monitoring image; where ΔL=N1·L, N1 is the number of fractures in a single monitoring image, and L is the longest fracture length in a single monitoring image;

[0023] Obtain the first ratio of the fracture feature value ΔL corresponding to adjacent monitoring images;

[0024] Get the number M1 of the first ratios that are greater than or equal to 1;

[0025] If the quantity M1 is greater than the preset first quantity threshold, the maximum value among the multiple first ratios will be output as the fracture defect trend value η1.

[0026] If the quantity M1 is less than or equal to the first quantity threshold, the output fracture defect trend value η1 is 0.

[0027] Optionally, the diameter anomaly features of the target material strip in each monitoring image are identified to obtain the diameter anomaly trend value η2 of the target material strip, including:

[0028] The monitoring images that do not have abnormal diameter characteristics are removed, and the remaining monitoring images are sorted according to time series.

[0029] Obtain the abnormal diameter value ΔD of the target material strip in each sorted monitoring image; where ΔD=N2·Δd, N2 is the number of diameter anomalies in a single monitoring image, Δd is the maximum absolute value of the difference between the abnormal diameter and the standard diameter in a single monitoring image, and the abnormal diameter is the diameter of the part of the target material strip that is smaller or larger than the standard diameter.

[0030] Obtain the second ratio of the diameter anomaly value ΔD corresponding to adjacent monitoring images;

[0031] Obtain the number M2 of second ratios that are greater than or equal to 1;

[0032] If the quantity M2 is greater than the preset second quantity threshold, the maximum value among the multiple second ratios will be output as the diameter abnormal trend value η2.

[0033] If the quantity M2 is less than or equal to the second quantity threshold, the output diameter abnormal trend value η2 is 0.

[0034] Optionally, the color dispersion of the target material strip in each monitored image is identified to obtain the color dispersion trend value η3, including:

[0035] The monitoring images with color dispersion within the standard range are removed, and the remaining monitoring images are sorted according to the time series.

[0036] Obtain the third ratio of the color dispersion of adjacent monitoring images after sorting;

[0037] Obtain the number M3 of third ratios that are greater than or equal to 1;

[0038] If the quantity M3 is greater than the preset third quantity threshold, the maximum value among the multiple third ratios will be output as the color discrete trend value η3.

[0039] If the quantity M3 is less than or equal to the third quantity threshold, the output color discrete trend value η3 is 0.

[0040] Optionally, based on multiple CT scan images, the porosity trend value of the target material strip is obtained, including:

[0041] Obtain the average porosity of the target material strip in each CT scan image;

[0042] Obtain the fourth ratio of the average porosity corresponding to adjacent monitoring images;

[0043] Obtain the number M4 of fourth ratios that are greater than or equal to 1;

[0044] If the quantity M4 is greater than the preset fourth quantity threshold, the maximum value among the multiple fourth ratios will be output as the porosity trend value.

[0045] If the quantity M4 is less than or equal to the fourth quantity threshold, the output porosity trend value is 0.

[0046] Optionally, the average porosity of the target material strip in each CT scan image is obtained, including:

[0047] The target material strip in the CT scan image is divided into multiple sub-regions at equal intervals;

[0048] Obtain the measured porosity for each sub-region;

[0049] Sub-regions with measured porosity greater than theoretical porosity were selected;

[0050] The average value of the measured porosity of the selected sub-regions is output as the average porosity of the target material strip.

[0051] Optionally, the expression for the risk assessment model is:

[0052] F = (W1·η + W2·Q)·P;

[0053] In the formula, F is the risk assessment value, Q is the porosity trend value, P is the abnormal operation trend value, W1 is the first weighting coefficient, and W2 is the second weighting coefficient.

[0054] To achieve the above objectives, this application also provides a monitoring system for the operating status of a biodegradable material granulator, comprising:

[0055] The image acquisition module is used to acquire multiple monitoring images and multiple CT scan images of the target material strip within a unit monitoring cycle; wherein, the target material strip is the strip that comes out of the extrusion mechanism of the granulator, and the target material strips in multiple monitoring images are spliced ​​together to form a strip with a preset length;

[0056] The first trend prediction module is used to obtain the surface defect trend value of the target material strip based on multiple monitoring images;

[0057] The second trend prediction module is used to obtain the porosity trend value of the target material strip based on multiple CT scan images;

[0058] The third trend prediction module is used to obtain abnormal trend values ​​of the granulator's operation;

[0059] The risk assessment module is used to input surface defect trend values, porosity trend values, and operational anomaly trend values ​​into a preset risk assessment model to obtain risk assessment values.

[0060] The first data processing module is used to determine whether the risk assessment value is greater than the preset risk threshold. If it is, the first alarm information is sent. If not, the current weight of the granulated material coming out of the pelletizing mechanism of the pelletizer is obtained.

[0061] The second data processing module is used to determine whether the current weight is within the preset weight threshold range. If not, it sends a second alarm message. If yes, it returns to the multiple monitoring images and multiple CT scan images of the target material strip acquired within the unit monitoring cycle.

[0062] To achieve the above objectives, this application also provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0063] To achieve the above objectives, this application also provides a computer-readable storage medium storing a computer program, on which a processor executes the computer program to implement the above-described method.

[0064] The beneficial effects that this application can achieve are as follows:

[0065] Since the forming quality of the target strip is crucial to the quality of the granulated material after pelletizing, this application first acquires multiple monitoring images and multiple CT scan images of the target strip within a unit monitoring cycle to comprehensively evaluate the forming quality of strips with a preset length. Here, the surface defect trend changes of the target strip can be predicted based on multiple monitoring images, and the trend change is quantified using the surface defect trend value. The internal structural defects of the target strip are mainly porosity; therefore, the internal porosity trend changes of the target strip can be predicted using multiple CT scan images, and the trend change is quantified using the porosity trend value. Simultaneously, the abnormal operation trend value of the pelletizer is also considered. Finally, the surface defect trend value, porosity trend value, and abnormal operation trend value are input into a preset risk assessment model to obtain a risk assessment value. This risk assessment value is determined by the trend changes of surface and internal structural defects in the target strip. The system simultaneously correlates the trend of abnormal changes in the granulator's operation with the trend of abnormal changes in the granulator's equipment. By predicting changes in abnormal trends, it assesses the risk of abnormalities, avoiding product quality defects or equipment malfunctions caused by accidental anomalies, and reducing the probability of false alarms. This approach is more reasonable and reliable than directly assessing risk based on abnormal characteristics. Then, it checks whether the risk assessment value exceeds a preset risk threshold. If so, a first alarm is sent. If not, considering monitoring errors, it further calculates the current weight of the granules from the pelletizing mechanism of the granulator and checks whether the current weight is within a preset weight threshold range. If not, it may be due to low granule density leading to a lower weight, or insufficient drying resulting in high moisture content leading to a higher weight. Therefore, a reasonable weight threshold range is set as the judgment benchmark, and a second alarm is sent. If so, the monitoring cycle continues. In summary, this application, based on the trend changes of defects in the front-end material strip forming and the abnormal trends of the granulator equipment, combined with the weight monitoring after pelletizing at the back end, effectively correlates various influencing factors. This allows for a comprehensive evaluation of abnormal situations in the granulation process, making it more reasonable, efficient, and improving monitoring accuracy. Attached Figure Description

[0066] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0067] Figure 1 This is a schematic diagram of the structure of a method, system, and equipment for monitoring the operating status of a degradable material granulator, as described in an embodiment of this application.

[0068] Figure 2This is a schematic diagram illustrating the removal of monitoring images lacking fracture characteristics in an embodiment of this application;

[0069] Figure 3 This is a schematic diagram illustrating the removal of monitoring images lacking diameter abnormalities in an embodiment of this application.

[0070] Figure label:

[0071] 110 - Target material strip, 120 - Fracture characteristics, 130 - Diameter anomaly characteristics.

[0072] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0073] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0074] It should be noted that if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0075] Example 1

[0076] Reference Figures 1-3 This embodiment provides a method for monitoring the operating status of a biodegradable material granulator, including the following steps:

[0077] Multiple monitoring images and multiple CT scan images of the target material strip 110 are acquired within a unit monitoring cycle; wherein, the target material strip 110 is a strip that comes out of the extrusion mechanism of the granulator, and the target material strip 110 in the multiple monitoring images is spliced ​​together to form a strip with a preset length;

[0078] Based on multiple monitoring images, the surface defect trend value of the target material strip 110 is obtained;

[0079] Based on multiple CT scan images, the porosity trend value of the target material strip 110 was obtained;

[0080] Obtain abnormal operating trend values ​​of the granulator;

[0081] Input the surface defect trend value, porosity trend value, and operational anomaly trend value into the preset risk assessment model to obtain the risk assessment value;

[0082] Determine whether the risk assessment value is greater than the preset risk threshold. If so, send the first alarm message. If not, obtain the current weight of the granulated material coming out of the pelletizing mechanism of the pelletizer.

[0083] Determine whether the current weight is within the preset weight threshold range. If not, send a second alarm message. If yes, return to the multiple monitoring images and multiple CT scan images of the target material bar 110 acquired within the unit monitoring cycle.

[0084] In this embodiment, since the forming quality of the target strip 110 is crucial to the quality of the granulated material after subsequent pelletizing, this embodiment first acquires multiple monitoring images and multiple CT scan images of the target strip 110 within a unit monitoring cycle to comprehensively evaluate the forming quality of the strip with a preset length. Here, the surface defect trend change of the target strip 110 can be predicted based on multiple monitoring images, and its trend change can be quantified and assessed through the surface defect trend value. The internal structural defects of the target strip 110 are mainly porosity, so the internal porosity trend change of the target strip 110 can be predicted through multiple CT scan images, and its trend change can be quantified and assessed through the porosity trend value. At the same time, the abnormal operation trend value of the pelletizer is also correlated and combined. Finally, the surface defect trend value, porosity trend value, and abnormal operation trend value are input into a preset risk assessment model to obtain a risk assessment value. This risk assessment value is obtained by simultaneously correlated calculation of the surface and internal structural defect change trends of the target strip 110 and the abnormal operation change trends of the pelletizer. Anomaly risk assessment is performed by analyzing abnormal trend changes, avoiding product quality defects or equipment malfunctions caused by accidental anomalies (such as misoperation), thus reducing the probability of false alarms. If the product defect or equipment malfunction is caused by an accidental anomaly, it will gradually return to normal (i.e., an anomaly mitigation trend). Therefore, this embodiment is more reasonable and reliable than directly assessing risk based on anomaly characteristics. Then, it is determined whether the risk assessment value is greater than a preset risk threshold. If so, it indicates an abnormal upward trend, and a first alarm message is sent to notify staff to check and repair in time. If not, considering the possibility of monitoring errors, the current weight of the granulated material from the pelletizing mechanism of the pelletizer is further calculated to determine whether the current weight is within a preset weight threshold range. If not, it may be due to low granule density leading to a low weight, or insufficient drying of the granule material resulting in high moisture content leading to a high weight. Therefore, a reasonable weight threshold range is set as the judgment benchmark. At this time, a second alarm message is sent. If so, the monitoring cycle continues in the next round. In summary, this embodiment effectively correlates various influencing factors based on the trend changes of defects in the front-end material strip forming and the abnormal trend changes of the granulator equipment, combined with the weight monitoring after pelletizing at the back end. This allows for a comprehensive evaluation of abnormal situations in the granulation process, making it more reasonable and efficient, and improving the accuracy of monitoring.

[0085] It should be noted that an industrial camera and a CT scanner can be installed at the material discharge position of the extrusion mechanism to continuously acquire monitoring images and CT scan images. The CT scan images can use X-rays to penetrate the target material strip 110 and reconstruct its three-dimensional internal structure to quantify porosity. Since many factors affect the forming quality of the target material strip 110, such as the uniformity of raw material mixing, temperature, humidity, and extrusion pressure within the granulator, and when the target material strip 110 exhibits an abnormal quality trend, it is generally due to corresponding abnormal operating parameters of the granulator. Therefore, when calculating the abnormal operating trend value of the granulator, the type of abnormal operating parameter is first detected by the corresponding sensor. For example, if the temperature is abnormal, the temperature change value is output as the abnormal operating trend value. If there are multiple abnormal operating parameters, the weighted sum of the changes of the multiple abnormal operating parameters is output as the abnormal operating trend value. The first alarm message and the second alarm message can each contain the corresponding parameter values ​​to assist personnel in quickly screening for anomalies.

[0086] As an optional implementation, the surface defect trend value of the target strip 110 is obtained based on multiple monitoring images, including:

[0087] Identify the fracture features 120 of the target material strip 110 in each monitoring image to obtain the fracture defect trend value η1 of the target material strip 110;

[0088] Identify the diameter anomaly features 130 of the target material strip 110 in each monitoring image to obtain the diameter anomaly trend value η2 of the target material strip 110;

[0089] Identify the color dispersion of the target material strip 110 in each monitoring image to obtain the color dispersion trend value η3;

[0090] The surface defect trend value η is obtained, and the expression for η is:

[0091] η=K1·η1+K2·η2+K3·η3;

[0092] In the formula, K1 is the first adjustment coefficient, K2 is the second adjustment coefficient, and K3 is the third adjustment coefficient.

[0093] In this embodiment, since the surface defects of the target material strip 110 are mainly poor continuity (i.e., having fracture feature 120), poor diameter uniformity (i.e., having diameter anomaly feature 130), and uneven surface color distribution (which can be characterized by color dispersion), the defect features can be identified separately after image processing of the monitoring images. This allows for the identification of the corresponding fracture feature 120, diameter anomaly feature 130, and color dispersion in each monitoring image, thereby quantifying and calculating the fracture defect trend value η1, diameter anomaly trend value η2, and color dispersion trend value η3. Based on these three key parameters, the overall trend of surface defect changes of the target material strip 110 can be characterized. Inputting the above formula allows for the quantification and calculation of the surface defect trend value η. Since the three parameters have different unit attributes, adjustment coefficients are used to convert and adjust them so that each parameter can be quantized and superimposed to calculate an accurate and reliable surface defect trend value η.

[0094] It should be noted that since the extrusion structure generally has multiple discharge holes, the acquired monitoring image may contain multiple target material strips 110. When calculating the above trend values, it is necessary to identify the defect features (i.e., fracture feature 120, diameter anomaly feature 130 and color dispersion) of all target material strips 110 in each monitoring image and superimpose them for calculation, so as to make an overall defect trend prediction.

[0095] As an optional implementation, the fracture features 120 of the target material strip 110 in each monitoring image are identified to obtain the fracture defect trend value η1 of the target material strip 110, including:

[0096] The monitoring images that do not have the fracture feature 120 are removed, and the remaining monitoring images are sorted according to the time series.

[0097] Obtain the fracture characteristic value ΔL of the target material strip 110 in each sorted monitoring image; where ΔL=N1·L, N1 is the number of fractures in a single monitoring image, and L is the longest fracture length in a single monitoring image;

[0098] Obtain the first ratio of the fracture feature value ΔL corresponding to adjacent monitoring images;

[0099] Get the number M1 of the first ratios that are greater than or equal to 1;

[0100] If the quantity M1 is greater than the preset first quantity threshold, the maximum value among the multiple first ratios will be output as the fracture defect trend value η1.

[0101] If the quantity M1 is less than or equal to the first quantity threshold, the output fracture defect trend value η1 is 0.

[0102] In this embodiment, when calculating the fracture defect trend value η1, monitoring images without fracture feature 120 are first removed, and the remaining monitoring images are sorted according to time sequence (i.e., the order in which the monitoring images were acquired). This facilitates the overall prediction of the continuity trend of the target material strip 110, ensuring the accuracy and reliability of the prediction results. Then, the fracture feature value ΔL of the target material strip 110 in each sorted monitoring image is calculated. Here, ΔL = N1·L, that is, the fracture feature value ΔL is characterized by two key factors: the number of fractures and the longest fracture length. Whether the number of fractures is too large or the fracture length is too long, it can represent the continuity situation (when there are multiple fracture locations, the longest fracture length is taken here, which is more representative). Therefore, when the fracture feature value ΔL is represented by the product of the two, if either parameter is too large or the two parameters are in the middle, the fracture feature value ΔL will be large, thus accurately reflecting the influence of the number and length of fractures on the continuity of the target material strip 110. The influence of the first ratio is then calculated, and the first ratio of the fracture feature value ΔL corresponding to adjacent monitoring images is calculated. The first ratio can reflect the trend of fracture defect change. If the first ratio is greater than 1, it indicates an upward trend. If the first ratio is equal to 1, it indicates a flat trend, but it also indicates that there is a certain degree of continuous fracture defect. If the first ratio is less than 1, it indicates a downward trend. Therefore, the number of first ratios greater than or equal to 1, M1, is taken as the judgment benchmark, which can represent the upward or flat trend as a whole. For example, taking 10 monitoring images with fracture features as an example, 9 sets of first ratios can be calculated. Assuming the first quantity threshold is 3, as long as at least 4 sets of first ratios are greater than or equal to 1, it can be characterized that the fracture defect is on an overall upward trend. Then, the maximum value among the multiple first ratios is output as the fracture defect trend value η1, which is more targeted and representative. Conversely, if it indicates an overall downward trend, the fracture defect trend value η1 is output as 0. Therefore, this quantitative assessment method can accurately and reliably judge the fracture defect trend.

[0103] It should be noted that, considering the possibility of significant fracture defects, a fracture characteristic threshold can be set here. After calculating multiple sets of fracture characteristic values ​​ΔL, it is determined whether there is a fracture characteristic value ΔL greater than the fracture characteristic threshold. If so, the fracture defect trend value η1 is directly output as the default value (which can be set according to an empirical model, for example, 10). This eliminates the need to calculate the first ratio to determine the defect trend, thereby amplifying the final risk assessment value and increasing the alarm probability to characterize emergency handling in the event of a significant defect. Here, the first quantity threshold should be a floating value, related to the remaining monitoring images. That is, the smaller the number of remaining monitoring images (i.e., the number of monitoring images with fracture characteristics), the smaller the first quantity threshold. This can be determined based on the correlation formula, E' = ε1E, where E' is the first quantity threshold, E is the number of remaining monitoring images, and ε1 is the first proportional coefficient, making this more reasonable. To further improve data reliability, a basic image number threshold Z' can be set, Z' = ε2Z, where Z is the number of acquired monitoring images and ε2 is the second proportional coefficient. For example, if the initial number of acquired monitoring images is 10 and ε2 = 0.3, then the basic image number threshold Z' = 3. When the remaining number of monitoring images E is less than this basic image number threshold Z', for example, if the remaining number of monitoring images E is only 2, it means that only 2 of the 10 initially acquired monitoring images have fracture characteristics, which can be judged as accidental fracture. At this time, the fracture defect trend value η1 can be directly output as 0, which optimizes the judgment principle and makes it more reasonable and reliable.

[0104] As an optional implementation, the diameter anomaly feature 130 of the target material strip 110 in each monitoring image is identified to obtain the diameter anomaly trend value η2 of the target material strip 110, including:

[0105] The monitoring images that do not have the diameter abnormality feature 130 are removed, and the remaining monitoring images are sorted according to the time series.

[0106] Obtain the abnormal diameter value ΔD of the target material strip 110 in each sorted monitoring image; where ΔD=N2·Δd, N2 is the number of diameter abnormalities in a single monitoring image, Δd is the maximum absolute value of the difference between the abnormal diameter and the standard diameter in a single monitoring image, and the abnormal diameter is the diameter of the part of the target material strip 110 that is smaller or larger than the standard diameter.

[0107] Obtain the second ratio of the diameter anomaly value ΔD corresponding to adjacent monitoring images;

[0108] Obtain the number M2 of second ratios that are greater than or equal to 1;

[0109] If the quantity M2 is greater than the preset second quantity threshold, the maximum value among the multiple second ratios will be output as the diameter abnormal trend value η2.

[0110] If the quantity M2 is less than or equal to the second quantity threshold, the output diameter abnormal trend value η2 is 0.

[0111] In this embodiment, similarly, when calculating the diameter anomaly trend value η2, it is necessary to first remove the monitoring images that do not have the diameter anomaly feature 130, and sort the remaining monitoring images according to the time series in order to predict the diameter uniformity trend of the target material strip 110 as a whole. Then, the diameter anomaly value ΔD of the target material strip 110 in each monitoring image is calculated. Here, ΔD = N2·Δd, which is characterized by the number of diameter anomalies and the anomaly difference. When diameter non-uniformity occurs, the diameter of some parts of the target material strip 110 is greater than or less than the standard diameter. Therefore, the maximum value of the absolute value of the difference between the abnormal diameter and the standard diameter is taken as one of the key factors, which is more targeted and prominent, so as to amplify the diameter anomaly value ΔD. Then, the second ratio of the diameter anomaly value ΔD corresponding to adjacent monitoring images is calculated. Similarly, the number M2 of the second ratio greater than or equal to 1 is used as the judgment benchmark. When the number M2 is greater than the second number threshold, the maximum value among the multiple second ratios can be output as the diameter anomaly trend value η2. Otherwise, the diameter anomaly trend value η2 is output as 0, so as to accurately and reliably judge the diameter uniformity trend.

[0112] As an optional implementation, the color dispersion of the target strip 110 in each monitored image is identified to obtain the color dispersion trend value η3, including:

[0113] The monitoring images with color dispersion within the standard range are removed, and the remaining monitoring images are sorted according to the time series.

[0114] Obtain the third ratio of the color dispersion of adjacent monitoring images after sorting;

[0115] Obtain the number M3 of third ratios that are greater than or equal to 1;

[0116] If the quantity M3 is greater than the preset third quantity threshold, the maximum value among the multiple third ratios will be output as the color discrete trend value η3.

[0117] If the quantity M3 is less than or equal to the third quantity threshold, the output color discrete trend value η3 is 0.

[0118] In this embodiment, color dispersion is defined as the degree to which the numerical distribution of a specific color channel (such as the H channel in HSV space) deviates from its mean. It is used to quantify the consistency of the color distribution on the surface of the target material strip 110. For example, when the surface color of the target material strip 110 is yellow, the color dispersion is high. Similarly, when calculating the color dispersion trend value η3, it is necessary to first remove the monitoring images with color dispersion within the standard range, sort the remaining monitoring images according to the time series, and then calculate the third ratio of the color dispersion corresponding to adjacent monitoring images. Thus, the color dispersion trend is predicted as a whole through multiple sets of third ratios. Here, the number of third ratios greater than or equal to 1, M3, is used as the judgment benchmark. If it is greater than the third number threshold, the maximum value among the multiple third ratios can be output as the color dispersion trend value η3. Otherwise, the color dispersion trend value η3 is output as 0, thereby realizing the effective prediction of the color dispersion trend of the target material strip 110 as a whole.

[0119] It should be noted that, let the color dispersion be Dc, the expression for Dc is:

[0120]

[0121] In the formula, H(x,y) represents the hue value of pixel (x,y). MN represents the mean color tone of the region, and MN represents the image resolution.

[0122] As an optional implementation, the porosity trend value of the target material strip 110 is obtained based on multiple CT scan images, including:

[0123] Obtain the average porosity of the target material strip 110 in each CT scan image;

[0124] Obtain the fourth ratio of the average porosity corresponding to adjacent monitoring images;

[0125] Obtain the number M4 of fourth ratios that are greater than or equal to 1;

[0126] If the quantity M4 is greater than the preset fourth quantity threshold, the maximum value among the multiple fourth ratios will be output as the porosity trend value.

[0127] If the quantity M4 is less than or equal to the fourth quantity threshold, the output porosity trend value is 0.

[0128] In this embodiment, the porosity of the target material strip 110 can be quantitatively calculated based on CT scan images. Since the target material strip 110 has a certain length, the average porosity is used as a key parameter, and then the fourth ratio of the average porosity corresponding to adjacent monitoring images is calculated to predict the porosity change trend. Similarly, the number M4 of the fourth ratio greater than or equal to 1 is used as the judgment benchmark. When the number M4 is greater than the fourth number threshold, the maximum value among the multiple fourth ratios is output as the porosity trend value; otherwise, the porosity trend value is output as 0, thereby predicting the porosity change trend as a whole.

[0129] As an optional implementation, obtaining the average porosity of the target material strip 110 in each CT scan image includes:

[0130] The target material strip 110 in the CT scan image is divided into multiple sub-regions at equal intervals;

[0131] Obtain the measured porosity for each sub-region;

[0132] Sub-regions with measured porosity greater than theoretical porosity were selected;

[0133] The average value of the measured porosity of the selected sub-regions is output as the average porosity of the target material strip 110.

[0134] In this embodiment, since the target material strip 110 has a certain length, in order to improve the calculation accuracy, the target material strip 110 in the CT scan image is first divided into multiple sub-regions at equal intervals. Then, the measured porosity corresponding to each sub-region is calculated, thereby selecting the sub-regions with measured porosity greater than theoretical porosity as defect areas for consideration, which is more targeted. Then, the average value of the measured porosity of the selected sub-regions is calculated, which can be output as the average porosity of the target material strip 110, which can better characterize the non-standard porosity variation of the target material strip 110.

[0135] As an optional implementation method, the expression for the risk assessment model is:

[0136] F = (W1·η + W2·Q)·P;

[0137] In the formula, F is the risk assessment value, Q is the porosity trend value, P is the abnormal operation trend value, W1 is the first weighting coefficient, and W2 is the second weighting coefficient.

[0138] In this embodiment, (W1·η+W2·Q) represents the weighted sum of the surface defect trend value η and the porosity trend value Q, thereby characterizing the trend changes of external and internal defects of the target material strip 110. At the same time, it is correlated with the abnormal operation trend value P of the granulator equipment. The product of the two is output as the risk assessment value F. Whether either parameter (W1·η+W2·Q) or P is too large or the parameter values ​​of the two are in the middle, the overall risk assessment value F will be too large. Thus, hidden risks can be predicted from two dimensions: the target material strip 110 product and the granulator equipment, which improves the reliability and accuracy of the data and has strong reference and guidance value.

[0139] Example 2

[0140] Based on the same inventive concept as the foregoing embodiments, this embodiment also provides a monitoring system for the operating status of a degradable material granulator, comprising:

[0141] The image acquisition module is used to acquire multiple monitoring images and multiple CT scan images of the target material strip 110 within a unit monitoring cycle; wherein, the target material strip 110 is a strip that comes out of the extrusion mechanism of the granulator, and the target material strip 110 in multiple monitoring images is spliced ​​together to form a strip with a preset length;

[0142] The first trend prediction module is used to obtain the surface defect trend value of the target material strip 110 based on multiple monitoring images;

[0143] The second trend prediction module is used to obtain the porosity trend value of the target material strip 110 based on multiple CT scan images.

[0144] The third trend prediction module is used to obtain abnormal trend values ​​of the granulator's operation;

[0145] The risk assessment module is used to input surface defect trend values, porosity trend values, and operational anomaly trend values ​​into a preset risk assessment model to obtain risk assessment values.

[0146] The first data processing module is used to determine whether the risk assessment value is greater than the preset risk threshold. If it is, the first alarm information is sent. If not, the current weight of the granulated material coming out of the pelletizing mechanism of the pelletizer is obtained.

[0147] The second data processing module is used to determine whether the current weight is within the preset weight threshold range. If not, it sends a second alarm message. If yes, it returns to the multiple monitoring images and multiple CT scan images of the target material bar 110 acquired within the unit monitoring cycle.

[0148] The explanations and examples of the modules in this embodiment can be found in the methods of the foregoing embodiments, and will not be repeated here.

[0149] Example 3

[0150] Based on the same inventive concept as the foregoing embodiments, this embodiment provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0151] Example 4

[0152] Based on the same inventive concept as the foregoing embodiments, this embodiment provides a computer-readable storage medium storing a computer program, and a processor executes the computer program to implement the above-described method.

[0153] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method of monitoring the operating state of a material degradation granulator, characterized in that Includes the following steps: Multiple monitoring images and multiple CT scan images of the target material strip are acquired within a unit monitoring cycle; the target material strip is the strip that comes out of the extrusion mechanism of the granulator, and the target material strips in the multiple monitoring images are spliced ​​together to form a strip with a preset length; Based on multiple monitoring images, the surface defect trend value of the target material strip is obtained; this includes: identifying the fracture characteristics of the target material strip in each monitoring image to obtain the fracture defect trend value η1 of the target material strip; identifying the diameter anomaly characteristics of the target material strip in each monitoring image to obtain the diameter anomaly trend value η2 of the target material strip; identifying the color dispersion of the target material strip in each monitoring image to obtain the color dispersion trend value η3; obtaining the surface defect trend value η, the expression of η is: η=K1·η1+K2·η2+K3·η3; where K1 is the first adjustment coefficient, K2 is the second adjustment coefficient, and K3 is the third adjustment coefficient; identifying the fracture characteristics of the target material strip in each monitoring image to obtain the fracture defect trend value η1 of the target material strip includes: removing monitoring images that do not have fracture characteristics. The remaining monitoring images are sorted according to time series. The fracture feature value ΔL of the target material strip in each sorted monitoring image is obtained. Where ΔL=N1·L, N1 is the number of fractures in a single monitoring image, and L is the longest fracture length in a single monitoring image. It is determined whether there is a fracture feature value ΔL greater than the fracture feature threshold. If it exists, the fracture defect trend value η1 is directly output as the default value. If it does not exist, the first ratio of the fracture feature values ​​ΔL corresponding to adjacent monitoring images is obtained. The number M1 of the first ratios greater than or equal to 1 is obtained. If the number M1 is greater than the preset first number threshold, the maximum value among the multiple first ratios is output as the fracture defect trend value η1. If the number M1 is less than or equal to the first number threshold, the fracture defect trend value η1 is output as 0. Based on multiple CT scan images, the porosity trend value of the target material strip is obtained; Obtain abnormal operating trend values ​​of the granulator; Input the surface defect trend value, porosity trend value, and operational anomaly trend value into the preset risk assessment model to obtain the risk assessment value; Determine whether the risk assessment value is greater than the preset risk threshold. If so, send the first alarm message. If not, obtain the current weight of the granulated material coming out of the pelletizing mechanism of the pelletizer. Determine whether the current weight is within the preset weight threshold range. If not, send a second alarm message. If yes, return to the multiple monitoring images and multiple CT scan images of the target material strip acquired within the unit monitoring cycle.

2. The method of claim 1, wherein the method further comprises: Identify the diameter anomaly features of the target material strip in each monitoring image to obtain the diameter anomaly trend value η2 of the target material strip, including: The monitoring images that do not have abnormal diameter characteristics are removed, and the remaining monitoring images are sorted according to time series. Obtain the abnormal diameter value ΔD of the target material strip in each sorted monitoring image; where ΔD=N2·Δd, N2 is the number of diameter abnormalities in a single monitoring image, Δd is the maximum absolute value of the difference between the abnormal diameter and the standard diameter in a single monitoring image, and the abnormal diameter is the diameter of the part of the target material strip that is smaller or larger than the standard diameter. Obtain the second ratio of the diameter anomaly value ΔD corresponding to adjacent monitoring images; Obtain the number M2 of second ratios that are greater than or equal to 1; If the quantity M2 is greater than the preset second quantity threshold, the maximum value among the multiple second ratios will be output as the diameter abnormal trend value η2. If the quantity M2 is less than or equal to the second quantity threshold, the output diameter abnormal trend value η2 is 0.

3. The method of claim 1, wherein the method further comprises: determining the operating state of the granulator based on the monitored parameter. Identify the color dispersion of the target material strip in each monitoring image to obtain the color dispersion trend value η3, including: The monitoring images with color dispersion within the standard range are removed, and the remaining monitoring images are sorted according to the time series. Obtain the third ratio of the color dispersion of adjacent monitoring images after sorting; Obtain the number M3 of third ratios that are greater than or equal to 1; If the quantity M3 is greater than the preset third quantity threshold, the maximum value among the multiple third ratios will be output as the color discrete trend value η3. If the quantity M3 is less than or equal to the third quantity threshold, the output color discrete trend value η3 is 0.

4. The method of claim 1, wherein the method further comprises: determining a number of the material degradation granulators in operation; and determining a number of the material degradation granulators in standby mode. Based on multiple CT scan images, the porosity trend value of the target material strip was obtained, including: Obtain the average porosity of the target material strip in each CT scan image; Obtain the fourth ratio of the average porosity corresponding to adjacent monitoring images; Obtain the number M4 of fourth ratios that are greater than or equal to 1; If the quantity M4 is greater than the preset fourth quantity threshold, the maximum value among the multiple fourth ratios will be output as the porosity trend value. If the quantity M4 is less than or equal to the fourth quantity threshold, the output porosity trend value is 0.

5. The method of claim 4, wherein the method further comprises: determining the number of revolutions of the rotor; and determining the number of revolutions of the rotor per minute. Obtain the average porosity of the target material strip in each CT scan image, including: The target material strip in the CT scan image is divided into multiple sub-regions at equal intervals; Obtain the measured porosity for each sub-region; Sub-regions with measured porosity greater than theoretical porosity were selected; The average value of the measured porosity of the selected sub-regions is output as the average porosity of the target material strip.

6. A method of monitoring the operating state of a granulator for degrading material according to any one of claims 1 to 5, characterized in that, The expression for the risk assessment model is: F = (W1·η + W2·Q)·P; In the formula, F is the risk assessment value, Q is the porosity trend value, P is the abnormal operation trend value, W1 is the first weighting coefficient, and W2 is the second weighting coefficient.

7. A system for monitoring the operating state of a material degradation granulator, characterized in that include: The image acquisition module is used to acquire multiple monitoring images and multiple CT scan images of the target material strip within a unit monitoring cycle; wherein, the target material strip is the strip that comes out of the extrusion mechanism of the granulator, and the target material strips in multiple monitoring images are spliced ​​together to form a strip with a preset length; The first trend prediction module is used to obtain the surface defect trend value of the target material strip based on multiple monitoring images; including: identifying the fracture characteristics of the target material strip in each monitoring image to obtain the fracture defect trend value η1 of the target material strip; identifying the diameter anomaly characteristics of the target material strip in each monitoring image to obtain the diameter anomaly trend value η2 of the target material strip; identifying the color dispersion of the target material strip in each monitoring image to obtain the color dispersion trend value η3; obtaining the surface defect trend value η, the expression of η is: η=K1·η1+K2·η2+K3·η3; where K1 is the first adjustment coefficient, K2 is the second adjustment coefficient, and K3 is the third adjustment coefficient; identifying the fracture characteristics of the target material strip in each monitoring image to obtain the fracture defect trend value η1 of the target material strip includes: eliminating the target material strip with no fracture characteristics. The system collects monitoring images of the target material and sorts the remaining monitoring images according to time series. It then obtains the fracture characteristic value ΔL of the target material strip in each sorted monitoring image; where ΔL = N1·L, N1 is the number of fractures in a single monitoring image, and L is the longest fracture length in a single monitoring image. The system determines whether there is a fracture characteristic value ΔL greater than a fracture characteristic threshold. If it exists, the system directly outputs the fracture defect trend value η1 as the default value. If it does not exist, it obtains the first ratio of the fracture characteristic values ​​ΔL corresponding to adjacent monitoring images. It then obtains the number M1 of first ratios greater than or equal to 1. If the number M1 is greater than a preset first quantity threshold, the maximum value among the multiple first ratios is output as the fracture defect trend value η1. If the number M1 is less than or equal to the first quantity threshold, the fracture defect trend value η1 is output as 0. The second trend prediction module is used to obtain the porosity trend value of the target material strip based on multiple CT scan images; The third trend prediction module is used to obtain abnormal trend values ​​of the granulator's operation; The risk assessment module is used to input surface defect trend values, porosity trend values, and operational anomaly trend values ​​into a preset risk assessment model to obtain risk assessment values. The first data processing module is used to determine whether the risk assessment value is greater than the preset risk threshold. If it is, the first alarm information is sent. If not, the current weight of the granulated material coming out of the pelletizing mechanism of the pelletizer is obtained. The second data processing module is used to determine whether the current weight is within the preset weight threshold range. If not, it sends a second alarm message. If yes, it returns to the multiple monitoring images and multiple CT scan images of the target material strip acquired within the unit monitoring cycle.

8. A computer device, comprising: The computer device includes a memory and a processor, wherein the memory stores a computer program and the processor executes the computer program to implement the method as described in any one of claims 1-6.

Images