Automatic coarse crushing process for traditional Chinese medicinal materials with different material sizes
A dual-layer closed-loop control mechanism constructed using image analysis technology adaptively adjusts the pulverization parameters, solving the problems of uneven pulverization and clogging during the coarse pulverization of Chinese medicinal materials, and achieving stable and efficient pulverization of Chinese medicinal materials.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHAANXI SUN SIMIAO HIGH-TECH PHARM CO LTD
- Filing Date
- 2026-02-02
- Publication Date
- 2026-05-12
AI Technical Summary
In existing coarse crushing processes for Chinese medicinal materials, the crushing parameters rely on manual experience to set, which leads to problems such as insufficient crushing, uneven output, and equipment blockage when crushing different material sizes and fibrous medicinal materials, making it difficult to meet the requirements of automated and continuous production.
Image analysis technology is used to acquire image information of the feed and discharge ports in real time. Particle regions are divided by image difference, edge detection and clustering algorithms. Combined with the rotation speed and feed speed parameters of the crushing device, a two-layer closed-loop control mechanism is constructed to adaptively adjust the crushing parameters and correct crushing abnormalities in real time.
It improves the stability of crushing, reduces the risk of equipment blockage, and increases crushing efficiency. It is suitable for continuous coarse crushing of Chinese medicinal materials with different particle sizes and strong fibrous properties.
Smart Images
Figure CN122006883A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of general pulverization technology, and more specifically to an automated coarse pulverization process for Chinese medicinal materials of different sizes. Background Technology
[0002] Before being used in pharmaceutical preparations, traditional Chinese medicinal materials typically require coarse crushing to break down lumps or strips into granules suitable for subsequent grinding, extraction, or granulation. Due to the wide variety of sources and significant differences in morphology, size, and structure, raw materials of different sizes and forms often coexist within the same production batch. Currently, most pharmaceutical manufacturing enterprises rely on general-purpose coarse crushing equipment, which completes the crushing operation by setting fixed feeding methods and grinding speeds. This type of equipment is structurally mature and highly adaptable, and has been widely used in the processing of prepared slices of traditional Chinese medicine and the pretreatment stages of pharmaceutical preparations, playing a crucial role in ensuring basic production capacity and continuous operation.
[0003] Existing problems: In current coarse crushing processes for Chinese medicinal herbs of different sizes, crushing parameters are typically set manually based on experience and remain relatively constant throughout the process. However, because the initial particle size distribution of Chinese medicinal herbs is not constant, when the initial particles are large and the crushing speed or feed rate is not set appropriately, problems such as excessive instantaneous load, insufficient crushing, and material accumulation in the crushing chamber can easily occur, affecting crushing stability and making it difficult to balance processing efficiency or stability. Furthermore, in actual production, some types of Chinese medicinal herbs contain a lot of fiber. Materials with strong fibrous properties are prone to compression, adhesion, or clumping during coarse crushing, causing some strip-shaped medicinal material particles to fail to be effectively crushed and be discharged directly from the outlet, resulting in uneven particle size and fluctuating crushing effects. These problems are difficult to detect and effectively address in traditional coarse crushing processes that rely on manual parameter adjustments, especially when there are different initial particle sizes and strong fiber content, making it easier to cause unstable coarse crushing quality and failing to meet the requirements of automated and continuous production. Summary of the Invention
[0004] This invention provides an automated coarse crushing process for Chinese medicinal materials of different sizes to solve existing problems.
[0005] The automated coarse crushing process for Chinese medicinal materials of different sizes according to the present invention adopts the following technical solution: One embodiment of the present invention provides an automated coarse crushing process for Chinese medicinal materials of different sizes, the process comprising the following steps: After the coarse crushing device starts operating, acquire images of the feed medicinal materials, the discharge medicinal materials, and the feed speed at each sampling time; acquire the minimum rotation speed, maximum rotation speed, minimum feed speed, and maximum feed speed of the coarse crushing device. From the image of the feed medicinal materials at the current sampling time, several feed medicinal material particle regions are divided. Based on the area difference between the feed medicinal material particle regions, combined with the minimum speed, maximum speed, minimum feed speed and maximum feed speed of the coarse crushing device, the reference speed and reference feed speed for the next parameter adjustment after the current sampling time are obtained. The image of the discharged medicinal material at each sampling time is divided into the discharged medicinal material area and the discharged background area, and several discharged medicinal material particle areas are further divided from the discharged medicinal material area. Based on the area difference between the discharged medicinal material particle areas at the current sampling time, combined with the difference in the area ratio of the discharged medicinal material area and the discharged background area at different sampling times corresponding to the same feeding speed, the crushing abnormality factor at the current sampling time is obtained. Based on the magnitude of the crushing anomaly factor, the reference rotation speed and the reference feed rate are adjusted to obtain the target rotation speed and target feed rate for the next parameter adjustment after the current sampling time.
[0006] Furthermore, the specific steps for dividing the image of the fed medicinal materials at the current sampling time into several regions of fed medicinal material particles are as follows: Obtain a background image of the feed material when the coarse crushing device is not running and there are no medicinal materials being fed in; Based on the image of the feed medicinal material at the current sampling time and the image of the feed background, the feed medicinal material region in the image of the feed medicinal material is obtained using an image difference algorithm; The Canny edge detection algorithm is used to perform edge detection on the feed material area. Based on the edge detection results, the contour extraction method is used to obtain several edge connected components. Based on the differences between edge connected components, obtain the clustering distance between any two edge connected components; Based on the clustering distance between any two edge connected components, the DBSCAN clustering algorithm is used to perform clustering operations on all edge connected components, resulting in several clusters. All edge connected domains in each cluster are combined into a region of feed medicinal material particles.
[0007] Furthermore, the specific steps for obtaining the clustering distance between any two edge connected components are as follows: The mean curvature of all pixels on the boundary of each edge connected region is obtained and denoted as the average curvature of each edge connected region. For any two edge connected regions A and B, obtain the Euclidean distance between the centroids of edge connected regions A and B, denoted as the first correlation value; obtain the absolute value of the difference in the average curvature of edge connected regions A and B, denoted as the second correlation value; obtain the ratio of the maximum area to the minimum area of edge connected regions A and B, denoted as the third correlation value; and use the sum of the normalized values of the first, second, and third correlation values as the clustering distance between edge connected regions A and B.
[0008] Furthermore, the specific steps for obtaining the reference rotational speed and reference feed rate at the next parameter adjustment after the current sampling time are as follows: Based on the area difference of the feed herb particles at the current sampling time, obtain the pulverization parameter adjustment coefficient; The product of the crushing parameter adjustment coefficient and the minimum speed of the coarse crushing device is recorded as the third product. The product of the inverse proportional value of the crushing parameter adjustment coefficient and the maximum speed of the coarse crushing device is recorded as the fourth product. The sum of the third product and the fourth product is recorded as the reference speed for the next parameter adjustment after the current sampling time. The product of the crushing parameter adjustment coefficient and the minimum feed rate of the coarse crushing device is recorded as the fifth product. The product of the inverse proportional value of the crushing parameter adjustment coefficient and the maximum feed rate of the coarse crushing device is recorded as the sixth product. The sum of the fifth and sixth products is recorded as the reference feed rate for the next parameter adjustment after the current sampling time.
[0009] Furthermore, the specific steps for obtaining the crushing parameter adjustment coefficient are as follows: At the current sampling time, obtain the standard deviation of the area of all feed herb granule regions, and record it as the area reference coefficient; obtain the maximum value of the area of all feed herb granule regions, and record it as the maximum feed area; obtain the mean of the area of all feed herb granule regions, and record it as the average feed area. The product of the normalized value of the area reference coefficient and the maximum feed area is denoted as the first product. The product of the inversely proportional normalized value of the area reference coefficient and the average feed area is denoted as the second product. The normalized value of the sum of the first product and the second product is denoted as the crushing parameter adjustment coefficient.
[0010] Furthermore, the specific steps for obtaining the pulverization anomaly factor at the current sampling time are as follows: Based on the area difference of the discharged medicinal material particles at the current sampling time, the degree of abnormality of the crushed particles at the current sampling time is obtained; Based on the difference in the area ratio of the discharged medicinal material area to the discharged background area at different sampling times corresponding to the same feeding speed, the degree of abnormality of the discharge speed at the current sampling time can be obtained. Based on the degree of abnormality of the crushed particles and the degree of abnormality of the discharge rate at the current sampling time, the crushing abnormality factor at the current sampling time is obtained.
[0011] Furthermore, the specific steps for obtaining the degree of abnormality of the crushed particles at the current sampling time are as follows: At the current sampling time, the product of the number of all discharged medicinal material granule areas and the preset quantity coefficient is rounded up and recorded as the target quantity. ; Among all the areas of the discharged medicinal material granules, obtain the largest front The average of the areas is denoted as the maximum discharge area; The average area of all discharged medicinal material granule areas is recorded as the average discharge area. The difference between the maximum discharge area and the average discharge area is recorded as the first difference. The ratio of the first difference to the maximum discharge area is recorded as the degree of abnormality of the crushed particles at the current sampling time.
[0012] Furthermore, the specific steps for obtaining the degree of abnormality in the discharge rate at the current sampling time are as follows: In the image of the discharged medicinal material at each sampling time, the ratio of the area of the discharged medicinal material region to the area of the discharged background region is obtained and recorded as the discharge speed at each sampling time. The feed rate at the current sampling moment is denoted as the target feed rate; Obtain all sampling times at which the feed rate is the target feed rate, and record them as reference sampling times. Obtain the maximum discharge rate among all discharge rates at all reference sampling times, and record it as the maximum standard discharge rate. The difference between the maximum standard discharge speed and the discharge speed at the current sampling time is recorded as the second difference. The ratio of the second difference to the maximum standard discharge speed is recorded as the degree of abnormality of the discharge speed at the current sampling time.
[0013] Furthermore, the specific steps for obtaining the crushing anomaly factor at the current sampling time based on the degree of anomaly in the crushed particles and the degree of anomaly in the discharge rate at the current sampling time are as follows: At the current sampling time, the product of the abnormality degree of crushed particles and the preset first weighting coefficient is recorded as the seventh product, and the product of the abnormality degree of discharge speed and the preset second weighting coefficient is recorded as the eighth product. The sum of the seventh product and the eighth product is recorded as the crushing abnormality factor at the current sampling time.
[0014] Furthermore, the specific steps for obtaining the target rotational speed and target feed rate at the next parameter adjustment after the current sampling time are as follows: The product of the inverse proportional value of the crushing anomaly factor at the current sampling time and the reference speed at the next parameter adjustment after the current sampling time is recorded as the target speed at the next parameter adjustment after the current sampling time. The product of the inverse proportional value of the crushing anomaly factor at the current sampling time and the reference feed rate at the next parameter adjustment after the current sampling time is recorded as the target feed rate at the next parameter adjustment after the current sampling time.
[0015] The beneficial effects of the technical solution of the present invention are: In this embodiment of the invention, after the coarse crushing device starts operating, the reference rotation speed and reference feed rate for the next parameter adjustment after the current sampling time are obtained based on the image information of the fed medicinal materials. The crushing anomaly factor at the current sampling time is obtained based on the image information of the discharged medicinal materials. The reference rotation speed and reference feed rate are adjusted according to the magnitude of the crushing anomaly factor. The target rotation speed and target feed rate for the next parameter adjustment after the current sampling time are obtained. Thus, by simultaneously introducing the image information of the feed inlet and discharge outlet of the coarse crushing device, a dual-layer closed-loop control of the crushing process is achieved. This not only enables adaptive setting of crushing parameters according to the size of the medicinal material particles, but also allows for real-time correction when the medicinal material is highly fibrous or crushing anomalies occur, thereby improving crushing stability, reducing the risk of clogging, and improving overall crushing efficiency. This invention, by simultaneously acquiring image information from the inlet and outlet during the crushing process, constructs a dual-layer closed-loop crushing control mechanism based on image analysis. This mechanism adaptively sets the crusher speed and feed rate according to the initial particle size and distribution of the feed herbs, and continuously senses the crushing results and discharge status in real time. It dynamically corrects crushing anomalies caused by factors such as the high fibrousness and large particle size differences of the herbs, thereby avoiding problems such as insufficient crushing, particle compression during discharge, or equipment blockage. This ensures both crushing quality and efficiency, improves the stability and reliability of the continuous coarse crushing process, and reduces the risk of human intervention and equipment failure. It is suitable for continuous coarse crushing of Chinese medicinal herbs with different particle sizes and high fibrousness, and has high industrial application value. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 This is a flowchart of the steps of the automated coarse crushing process for Chinese medicinal materials of different sizes according to the present invention; Figure 2 A schematic diagram of the control coefficients for the coarse crushing device; Figure 3 This is a schematic diagram of the medicinal materials being fed into the plant. Detailed Implementation
[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation methods, structure, features, and effects of the automated coarse crushing process for Chinese medicinal materials of different sizes proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of the automated coarse crushing process for Chinese medicinal materials of different sizes provided by the present invention.
[0021] Please see Figure 1 The diagram illustrates a process flow chart of an automated coarse crushing process for Chinese medicinal materials of different sizes provided by an embodiment of the present invention. The process includes the following steps: Step S001: After the coarse crushing device starts operating, acquire the image of the feed medicinal material, the image of the discharge medicinal material, and the feed speed at each sampling time; acquire the minimum rotation speed, maximum rotation speed, minimum feed speed, and maximum feed speed of the coarse crushing device.
[0022] In this embodiment, by acquiring image information at the inlet and outlet of the coarse crushing device in real time, the material particles and feeding / discharging speed at the inlet and outlet are estimated in real time, and abnormal crushing conditions are analyzed in real time, so as to achieve intelligent control of crushing parameters in real time.
[0023] Specifically, the process is as follows: First, an industrial camera is installed above the feed inlet of the coarse crushing device and on the side of the discharge outlet to collect real-time images of the incoming and outgoing medicinal materials, thus obtaining the distribution of particles. The cameras are kept at a fixed distance from the crushing device and are equipped with appropriate lighting to ensure clear image acquisition of different batches of medicinal materials under different lighting conditions. The image frame rate is 30 frames per second, which can be adjusted according to the material flow rate to ensure real-time capture of rapidly changing states. The image resolution is 1920×1080 for accurate analysis of particle morphology and size. Then, the image data from the feed inlet and discharge outlet are transmitted to the control system (e.g., PLC or industrial computer) after acquisition. The system calculates the average particle size and distribution statistics, and generates real-time feed speed adjustment signals and crusher speed signals, which are transmitted to the crusher actuator through the control interface to achieve closed-loop control and stable coarse crushing for different medicinal materials. A schematic diagram of the control coefficients for the coarse crushing device is shown below. Figure 2 As shown. Figure 2 The system uses industrial cameras at the feed inlet and discharge outlet of the crusher to collect data and send it to the control system, which then sends control commands to the crusher.
[0024] It should be noted that in this embodiment, the parameters of the coarse crushing device are adjusted every 10 seconds. Taking this as an example, specifically, after the coarse crushing device starts running, the initial operating parameters are used for the first 10 seconds. Then, based on the image information analysis results at the 5th second (after the device starts running and is relatively stable), the operating parameters for the 11th to 20th seconds are determined. Then, based on the image information analysis results at the 15th second (after updating the operating parameters and the device is relatively stable), the operating parameters for the 21st to 30th seconds are determined, and so on, completing the real-time adjustment of the coarse crushing device's operating parameters. This avoids frequent changes in crushing parameters due to image detection noise or instantaneous material fluctuations. The time interval can be set according to the structural characteristics of the crusher and the response time of the crushing process, so that the crusher can complete a stable crushing process and form an observable output result within each time interval. The operating parameters are rotational speed and feed rate. Therefore, the image sampling times are: 5 seconds, 15 seconds, 25 seconds, 35 seconds, ... after the coarse crushing device starts running.
[0025] Therefore, after the coarse crushing device starts operating, images of the feed medicinal materials, images of the discharge medicinal materials, and the feed speed are acquired at each sampling time.
[0026] Then obtain the minimum speed, maximum speed, minimum feed rate, and maximum feed rate of the coarse crushing device.
[0027] It should be noted that in this embodiment, the minimum rotational speed, maximum rotational speed, minimum feed rate, and maximum feed rate of the coarse crushing device within its safe operating range are obtained from the technical manual of the coarse crushing device. For example, the minimum and maximum rotational speeds are 200 and 1000 revolutions per minute, respectively, and the minimum and maximum feed rates are 0.1 and 0.8 meters per second, respectively.
[0028] Step S002: Divide the image of the feed medicinal materials at the current sampling time into several feed medicinal material particle regions. Based on the area difference between the feed medicinal material particle regions, and combined with the minimum speed, maximum speed, minimum feed speed and maximum feed speed of the coarse crushing device, obtain the reference speed and reference feed speed for the next parameter adjustment after the current sampling time.
[0029] It should be noted that in existing coarse crushing processes for Chinese medicinal herbs, the operating parameters of the pulverizer are usually set manually by the operator based on experience. The main parameters adjusted are the pulverizer speed and feed rate. However, in actual production, these parameters are often set once at the initial stage and are not adjusted in real time during the coarse crushing process. This method ignores the fluctuations in the size of the feed particles. When the particles at the feed inlet are too large, it can lead to material accumulation or an increase in load in a short period. Conversely, when the particles at the feed inlet are too small, the speed and feed rate settings will be too low, resulting in a decrease in crushing efficiency. Therefore, an industrial camera is added at the feed inlet to perform real-time detection of the particle size of the medicinal herbs at the feed inlet. A schematic diagram of the feed medicinal herbs is shown below. Figure 3 As shown.
[0030] Preferably, in one embodiment of the present invention, the method for obtaining the reference rotational speed and reference feed rate at the next parameter adjustment after the current sampling time includes: In this embodiment, the region of the feed medicinal material particles in the image information is first obtained, specifically: Industrial cameras at the inlet and outlet were used to acquire background images of the feed and discharge processes when the coarse crushing device was not running and there were no medicinal materials.
[0031] First, based on the image of the feed medicinal material and the image of the feed background at the current sampling time, the image difference algorithm is used to obtain the feed medicinal material region and the feed background region in the image of the feed medicinal material.
[0032] Then, the Canny edge detection algorithm is used to perform edge detection on the feed material area, and based on the edge detection results, the contour extraction method (findContours) is used to obtain several edge connected components.
[0033] It should be noted that image differencing, Canny edge detection, and findContours are all well-known techniques, and their specific methods will not be described here. Image differencing is used to remove irrelevant information such as background. Then, based on traditional edge detection results, the edge positions of different medicinal materials are obtained to extract several edge connected regions. Considering that the edges of individual medicinal particles may not be closed in the edge detection results, leading to the generation of multiple fragmented contour regions in subsequent edge connected region extraction, when multiple edge connected regions are detected to be spatially adjacent and their edge curvature distribution characteristics and area size are strongly correlated, these multiple edge connected regions are merged into the same medicinal particle region to characterize the equivalent particle scale characteristics of a single medicinal material.
[0034] Obtain the curvature of each pixel on the boundary of each edge connected region, and obtain the mean curvature of all pixels on the boundary of each edge connected region, which is denoted as the average curvature of each edge connected region.
[0035] For any two edge connected regions A and B, obtain the Euclidean distance between the centroids of edge connected regions A and B, denoted as the first correlation value; obtain the absolute value of the difference in the average curvature of edge connected regions A and B, denoted as the second correlation value; obtain the ratio of the maximum area to the minimum area of edge connected regions A and B, denoted as the third correlation value; and use the sum of the normalized values of the first, second, and third correlation values as the clustering distance between edge connected regions A and B.
[0036] Specifically, the min-max normalization method is used to normalize the first, second, and third association values corresponding to any two edge connected components to a range of 0 to 1. The min-max normalization method is a well-known technique, and its specific method will not be described here.
[0037] Based on the clustering distance between any two edge connected components, the DBSCAN clustering algorithm is used to cluster all edge connected components, resulting in several clusters. All edge connected components in each cluster are combined into a feed herb particle region, thus obtaining several feed herb particle regions.
[0038] It should be noted that the DBSCAN clustering algorithm (Density-Based Spatial Clustering of Applications with Noise) is a well-known technique, and its specific method will not be described here. The above method determines the correlation between edge-connected domains by calculating the spatial distance, curvature difference, and area ratio between different edge-connected regions. Therefore, multiple edge-connected domains with small spatial distances, curvature differences, and area ratios are merged into the same feed herb particle region.
[0039] Based on the method of obtaining the feeding medicinal material area and feeding background area in the feeding medicinal material image at the current sampling time, as well as several feeding medicinal material particle areas in the feeding medicinal material area, the discharging medicinal material area and discharging background area in the discharging medicinal material image at the current sampling time, as well as several discharging medicinal material particle areas in the discharging medicinal material area, are obtained.
[0040] It should be noted that in the feeding area of medicinal materials, when the area of large-particle medicinal materials is relatively high, the average particle area may be small due to the large number of small-particle medicinal materials. Therefore, it is necessary to consider the distribution dispersion of different medicinal material particles at the current crushing inlet. When the dispersion is high, the area of large-particle medicinal materials should be taken into account when adjusting the crushing parameters. Conversely, the crushing parameters should be adjusted based on the average particle area.
[0041] It should be further noted that if the number of all feed herb particle areas in the feed herb image at the current sampling time is less than 5, then the feeding operation is considered to have a problem, and the machine will be stopped and an alarm will be triggered. If the number of all feed herb particle areas in the feed herb image at the current sampling time is greater than or equal to 5, then the following analysis will be performed. This is described using this as an example. In other embodiments, other thresholds for the number of feed herb particle areas can be set, and this embodiment does not limit this.
[0042] Within the feed herb region of the image at the current sampling time, obtain the standard deviation of the area of all feed herb granule regions, denoted as the area reference coefficient. Obtain the maximum area among all feed herb granule regions, denoted as the maximum feed area. Obtain the mean area of all feed herb granule regions, denoted as the average feed area.
[0043] Obtain the normalized value of the area reference factor The product of the maximum feed area and the maximum feed area is denoted as the first product, which yields the inversely proportional normalized value of the area reference coefficient. The product of the first product and the average feed area is denoted as the second product. The normalized value of the sum of the first product and the second product is denoted as the crushing parameter adjustment coefficient. .
[0044] It should be noted that the above area reference coefficient corresponds to the current sampling time. In this embodiment, the minimum-maximum normalization method is used to normalize the area reference coefficient to between 0 and 1 for all sampling times. When the distribution of the area of the feed herb particles is more discrete (the standard deviation of the area is larger), the crushing situation of large-area herb particles is considered more. Conversely, the crushing parameters can be adjusted according to the area mean. Therefore, the larger the area reference coefficient, the greater the weight of the largest feed area and the smaller the weight of the average feed area. Conversely, the smaller the area reference coefficient, the smaller the weight of the largest feed area and the larger the weight of the average feed area, thus obtaining the crushing parameter adjustment coefficient. The maximum value among the areas of all feed herb particle regions at all sampling times is obtained and recorded as the maximum standard area. The ratio of the sum of the first product and the second product to the maximum standard area is recorded as the normalized value of the sum of the first product and the second product.
[0045] It should be further noted that when the pulverization parameter adjustment coefficient is large, meaning the area of the feed herb particles at the current sampling moment is closer to the maximum standard area, it indicates that there are many large particles or a relatively dispersed particle distribution at the feed inlet. In this case, a smaller feed speed and rotation speed should be adopted to ensure that the large particles are pulverized better. Conversely, a smaller value indicates that the feed inlet mainly contains small particles, and a relatively faster feed speed and rotation speed can be adopted to improve the pulverization effect.
[0046] Obtain the adjustment coefficient of the crushing parameters The product of this product and the minimum rotational speed of the coarse crushing device is denoted as the third product, which yields the inverse proportional value of the crushing parameter adjustment coefficient. The product of the third product and the maximum rotational speed of the coarse crushing device is denoted as the fourth product. The sum of the third product and the fourth product is denoted as the reference rotational speed at the next parameter adjustment after the current sampling time.
[0047] Obtain the adjustment coefficient of the crushing parameters The product of this product and the minimum feed rate of the coarse crushing device is denoted as the fifth product, which yields the inverse proportional value of the crushing parameter adjustment coefficient. The product of the fifth product and the maximum feed rate of the coarse crushing device is denoted as the sixth product. The sum of the fifth product and the sixth product is denoted as the reference feed rate for the next parameter adjustment after the current sampling time.
[0048] Step S003: Divide the image of the discharged medicinal material at each sampling time into the discharged medicinal material area and the discharged background area, and divide the discharged medicinal material area into several discharged medicinal material particle areas; based on the area difference between the discharged medicinal material particle areas at the current sampling time, and combined with the difference in the area ratio of the discharged medicinal material area to the discharged background area at different sampling times corresponding to the same feeding speed, obtain the pulverization abnormality factor at the current sampling time.
[0049] It should be noted that in the above analysis, the real-time image information acquired at the feed inlet enabled preliminary real-time control of the pulverizer's rotation speed and feed rate, which can adapt to the pulverization needs of different medicinal material particles to a certain extent. However, adjusting the pulverization parameters solely based on the image information at the feed inlet cannot directly reflect the actual particle size after pulverization, especially when the medicinal material has strong fibrous properties. The image information at the feed inlet cannot detect this type of problem in a timely manner, thus hindering the adjustment of pulverization parameters. When the medicinal material has strong fibrous properties, the fibrous material often bends, stretches, or peels upon impact, rather than instantly breaking into small pieces. Fiber strip-shaped particles can pass through the sieve holes with the smallest cross-section, either "sideways" or "longitudinally." They have a long residence time in the crushing chamber, and geometrically, although one cross-section exceeds the aperture, another cross-section is small enough to pass through. Therefore, the following consideration is to further combine the image information at the discharge outlet to make further adjustments to the pulverization parameters. The main consideration at the discharge outlet is whether there are a large number of large, coarsely crushed medicinal material particles.
[0050] It should be further noted that if the number of all outgoing medicinal material particles in the outgoing medicinal material image at the current sampling time is less than 10, the outlet is considered blocked, and the machine is immediately stopped and an alarm is triggered. If the number of all outgoing medicinal material particles in the outgoing medicinal material image at the current sampling time is greater than or equal to 10, the following analysis is performed. This is described using this as an example. In other embodiments, other thresholds for the number of outgoing medicinal material particles can be set, and this embodiment does not limit this.
[0051] Preferably, in one embodiment of the present invention, the method for obtaining the pulverization anomaly factor at the current sampling time includes: The preset quantity coefficient is 10%, and this will be used as an example for explanation.
[0052] Within the discharged medicinal material region of the image at the current sampling time, obtain the rounded-up value of the product of the number of all discharged medicinal material particle regions and a preset quantity coefficient, and denote it as the target quantity. .
[0053] Among all the areas of the discharged medicinal material granules, obtain the largest front The average of the areas is denoted as the maximum discharge area.
[0054] The average area of all discharged medicinal material granule areas is obtained and recorded as the average discharge area.
[0055] The difference between the maximum discharge area and the average discharge area is recorded as the first difference. The ratio of the first difference to the maximum discharge area is recorded as the degree of abnormality of the crushed particles at the current sampling time.
[0056] It should be noted that when the maximum discharge area is much larger than the average discharge area, it indicates that the size of the pulverized medicinal materials is more uneven, meaning that there is a higher probability that there are strong fibrous medicinal materials inside the pulverizer. The value for the degree of abnormality in pulverized particles ranges from 0 to 1.
[0057] It should be further noted that when the medicinal materials have a strong fibrous structure, high toughness, and are difficult to be quickly sheared or crushed before pulverization, not only will large-sized pulverized medicinal material particles appear at the discharge port, but the fibrous materials are also prone to entanglement, bending, or clustering during pulverization, leading to a decrease in the effective pulverization efficiency per unit time. This causes some pulverized medicinal materials to remain in the screening area, resulting in localized clogging of the filter screen. In other words, the amount of medicinal materials entering the pulverizer continuously increases, but the amount that can be pulverized in time and discharged through the filter screen decreases relatively. Therefore, based on the above-mentioned abnormality in pulverized particles, the relationship between the feed rate at the inlet and the discharge rate at the outlet at the current sampling time is considered to further determine the pulverization anomaly factor at the current sampling time.
[0058] In the image of the discharged medicinal material at each sampling time, the ratio of the area of the discharged medicinal material region to the area of the discharged background region is obtained and recorded as the discharge speed at each sampling time.
[0059] The larger the area of the herb discharge zone, the faster the discharge speed.
[0060] The feed rate at the current sampling moment is denoted as the target feed rate.
[0061] Obtain all sampling times at which the feed rate is the target feed rate, and record them as reference sampling times. Obtain the maximum discharge rate among all discharge rates at all reference sampling times, and record it as the maximum standard discharge rate.
[0062] The difference between the maximum standard discharge speed and the discharge speed at the current sampling time is recorded as the second difference. The ratio of the second difference to the maximum standard discharge speed is recorded as the degree of abnormality of the discharge speed at the current sampling time.
[0063] It should be noted that when the feed rate is high, the discharge rate should also be high. Therefore, the maximum discharge rate among all discharge rates at the same sampling time as the current sampling time is taken as the maximum standard discharge rate. If the discharge rate at the current sampling time is close to the maximum standard discharge rate, it indicates that the discharge rate at the current sampling time is normal. Therefore, the larger the second difference, the more abnormal the discharge rate at the current sampling time. The value of the abnormality level of the discharge rate is between 0 and 1.
[0064] The first weight coefficient is preset to 0.4, the second weight coefficient is preset to 0.6, and the sum of the first and second weight coefficients is set to 1. This will be used as an example for explanation.
[0065] At the current sampling time, the product of the abnormality degree of crushed particles and the preset first weighting coefficient is recorded as the seventh product, and the product of the abnormality degree of discharge speed and the preset second weighting coefficient is recorded as the eighth product. The sum of the seventh product and the eighth product is recorded as the crushing abnormality factor at the current sampling time. .
[0066] It should be noted that in this embodiment, first and second weighting coefficients are preset to balance the influence of abnormal crushing structure and abnormal efficiency. The crushing abnormality factor takes a value between 0 and 1. If the crushing abnormality factor at the current sampling time is greater than or equal to 0.8, the coarse crushing device is considered to have malfunctioned, and the machine is directly stopped and an alarm is triggered. If the crushing abnormality factor at the current sampling time is less than 0.8, the following analysis is performed. This is described using this as an example. In other embodiments, it can be set as the judgment threshold for malfunctions of other coarse crushing devices. This embodiment does not limit this.
[0067] Step S004: Adjust the reference rotation speed and the reference feed rate according to the magnitude of the crushing anomaly factor, and obtain the target rotation speed and target feed rate at the next parameter adjustment after the current sampling time.
[0068] It should be noted that in this embodiment, the pulverization anomaly factor at the current sampling time is used as the correction basis for the reference rotation speed and reference feed rate when adjusting the parameters next time after the current sampling time. When the pulverization anomaly factor is large, it indicates that there are abnormalities such as insufficient pulverization, particle extrusion, or material retention during the current pulverization process. Due to the presence of strong fibers in the medicinal materials, the reference rotation speed and reference feed rate may not achieve a good pulverization effect, so further correction is required.
[0069] Preferably, in one embodiment of the present invention, the method for obtaining the target rotational speed and target feed rate at the next parameter adjustment after the current sampling time includes: The inverse proportional value of the shattering anomaly factor at the current sampling time. The product of the current sampling time and the reference speed at the next parameter adjustment is denoted as the target speed at the next parameter adjustment after the current sampling time.
[0070] The inverse proportional value of the shattering anomaly factor at the current sampling time. The product of the current sampling time and the reference feed rate at the next parameter adjustment is denoted as the target feed rate at the next parameter adjustment.
[0071] It should be noted that if the crushing anomaly factor is higher at the current sampling time, the feed rate and rotation speed need to be appropriately reduced. Therefore, the target rotation speed and target feed rate are lower than the reference rotation speed and reference feed rate. When the target rotation speed is lower than the minimum rotation speed of the coarse crushing device, the target rotation speed is set to the minimum rotation speed of the coarse crushing device. When the target feed rate is lower than the minimum feed rate of the coarse crushing device, the target feed rate is set to the minimum feed rate of the coarse crushing device.
[0072] It should be further explained that if the current sampling time is 25 seconds after the start of the coarse crushing device, then the target rotational speed and target feed rate at the next parameter adjustment after the current sampling time will be used as the rotational speed and feed rate between the 31st and 40th seconds after the start of the coarse crushing device. This achieves real-time control of the rotational speed and feed rate of the coarse crushing device. Compared to the traditional method of relying on manual experience to set fixed crushing parameters, this embodiment simultaneously introduces image information from the feed inlet and discharge outlet of the coarse crushing device to achieve dual-layer closed-loop control of the crushing process. This not only allows for adaptive setting of crushing parameters based on the size of the medicinal material particles, but also enables real-time correction when the medicinal material is highly fibrous or crushing abnormalities occur, thereby improving crushing stability, reducing the risk of clogging, and increasing overall crushing efficiency.
[0073] This invention is now complete.
[0074] In summary, in this embodiment of the invention, after the coarse crushing device starts operating, images of the feed medicinal materials, images of the discharge medicinal materials, and the feed speed are acquired at each sampling time. Based on the area difference between the particle areas of the feed medicinal materials in the image of the feed medicinal materials at the current sampling time, combined with the minimum rotational speed, maximum rotational speed, minimum feed speed, and maximum feed speed of the coarse crushing device, reference rotational speed and reference feed speed are obtained for the next parameter adjustment after the current sampling time. Based on the area difference between the particle areas of the discharge medicinal materials at the current sampling time, combined with the difference in the area ratio of the discharge medicinal material area to the discharge background area at different sampling times corresponding to the same feed speed, a crushing anomaly factor is obtained for the current sampling time. Based on the magnitude of the crushing anomaly factor, the reference rotational speed and the reference feed speed are adjusted to obtain the target rotational speed and target feed speed for the next parameter adjustment after the current sampling time. This invention can improve crushing stability, reduce the risk of clogging, and improve overall crushing efficiency.
[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.
Claims
1. An automated coarse crushing process for Chinese medicinal herbs of different sizes, characterized in that, The process includes the following steps: After the coarse crushing device starts operating, acquire images of the feed medicinal materials, the discharge medicinal materials, and the feed speed at each sampling time; acquire the minimum rotation speed, maximum rotation speed, minimum feed speed, and maximum feed speed of the coarse crushing device. From the image of the feed medicinal materials at the current sampling time, several feed medicinal material particle regions are divided. Based on the area difference between the feed medicinal material particle regions, combined with the minimum speed, maximum speed, minimum feed speed and maximum feed speed of the coarse crushing device, the reference speed and reference feed speed for the next parameter adjustment after the current sampling time are obtained. The image of the discharged medicinal material at each sampling time is divided into the discharged medicinal material area and the discharged background area, and several discharged medicinal material particle areas are further divided from the discharged medicinal material area. Based on the area difference between the discharged medicinal material particle areas at the current sampling time, combined with the difference in the area ratio of the discharged medicinal material area and the discharged background area at different sampling times corresponding to the same feeding speed, the crushing abnormality factor at the current sampling time is obtained. Based on the magnitude of the crushing anomaly factor, the reference rotation speed and the reference feed rate are adjusted to obtain the target rotation speed and target feed rate for the next parameter adjustment after the current sampling time.
2. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 1, characterized in that, The specific steps for dividing the image of the fed medicinal materials at the current sampling time into several regions of fed medicinal material particles are as follows: Obtain a background image of the feed material when the coarse crushing device is not running and there are no medicinal materials being fed in; Based on the image of the feed medicinal material at the current sampling time and the image of the feed background, the feed medicinal material region in the image of the feed medicinal material is obtained using an image difference algorithm; The Canny edge detection algorithm is used to perform edge detection on the feed material area. Based on the edge detection results, the contour extraction method is used to obtain several edge connected components. Based on the differences between edge connected components, obtain the clustering distance between any two edge connected components; Based on the clustering distance between any two edge connected components, the DBSCAN clustering algorithm is used to perform clustering operations on all edge connected components, resulting in several clusters. All edge connected domains in each cluster are combined into a region of feed medicinal material particles.
3. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 2, characterized in that, The specific steps for obtaining the clustering distance between any two edge connected components are as follows: The mean curvature of all pixels on the boundary of each edge connected region is obtained and denoted as the average curvature of each edge connected region. For any two edge connected regions A and B, obtain the Euclidean distance between the centroids of edge connected regions A and B, denoted as the first correlation value; obtain the absolute value of the difference in the average curvature of edge connected regions A and B, denoted as the second correlation value; obtain the ratio of the maximum area to the minimum area of edge connected regions A and B, denoted as the third correlation value; and use the sum of the normalized values of the first, second, and third correlation values as the clustering distance between edge connected regions A and B.
4. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 1, characterized in that, The specific steps for obtaining the reference rotation speed and reference feed rate for the next parameter adjustment after the current sampling time are as follows: Based on the area difference of the feed herb particles at the current sampling time, obtain the pulverization parameter adjustment coefficient; The product of the crushing parameter adjustment coefficient and the minimum speed of the coarse crushing device is recorded as the third product. The product of the inverse proportional value of the crushing parameter adjustment coefficient and the maximum speed of the coarse crushing device is recorded as the fourth product. The sum of the third product and the fourth product is recorded as the reference speed for the next parameter adjustment after the current sampling time. The product of the crushing parameter adjustment coefficient and the minimum feed rate of the coarse crushing device is recorded as the fifth product. The product of the inverse proportional value of the crushing parameter adjustment coefficient and the maximum feed rate of the coarse crushing device is recorded as the sixth product. The sum of the fifth and sixth products is recorded as the reference feed rate for the next parameter adjustment after the current sampling time.
5. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 4, characterized in that, The specific steps for obtaining the crushing parameter adjustment coefficient are as follows: At the current sampling time, obtain the standard deviation of the area of all feed herb granule regions, and record it as the area reference coefficient; obtain the maximum value of the area of all feed herb granule regions, and record it as the maximum feed area; obtain the mean of the area of all feed herb granule regions, and record it as the average feed area. The product of the normalized value of the area reference coefficient and the maximum feed area is denoted as the first product. The product of the inversely proportional normalized value of the area reference coefficient and the average feed area is denoted as the second product. The normalized value of the sum of the first product and the second product is denoted as the crushing parameter adjustment coefficient.
6. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 1, characterized in that, The specific steps for obtaining the pulverization anomaly factor at the current sampling time are as follows: Based on the area difference of the discharged medicinal material particles at the current sampling time, the degree of abnormality of the crushed particles at the current sampling time is obtained; Based on the difference in the area ratio of the discharged medicinal material area to the discharged background area at different sampling times corresponding to the same feeding speed, the degree of abnormality of the discharge speed at the current sampling time can be obtained. Based on the degree of abnormality of the crushed particles and the degree of abnormality of the discharge rate at the current sampling time, the crushing abnormality factor at the current sampling time is obtained.
7. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 6, characterized in that, The specific steps for obtaining the degree of abnormality of the crushed particles at the current sampling time are as follows: At the current sampling time, the product of the number of all discharged medicinal material granule areas and the preset quantity coefficient is rounded up and recorded as the target quantity. ; Among all the areas of the discharged medicinal material granules, obtain the largest front The average of the areas is denoted as the maximum discharge area; The average area of all discharged medicinal material granule areas is recorded as the average discharge area. The difference between the maximum discharge area and the average discharge area is recorded as the first difference. The ratio of the first difference to the maximum discharge area is recorded as the degree of abnormality of the crushed particles at the current sampling time.
8. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 6, characterized in that, The specific steps for obtaining the degree of anomaly in the discharge rate at the current sampling time are as follows: In the image of the discharged medicinal material at each sampling time, the ratio of the area of the discharged medicinal material region to the area of the discharged background region is obtained and recorded as the discharge speed at each sampling time. The feed rate at the current sampling moment is denoted as the target feed rate; Obtain all sampling times at which the feed rate is the target feed rate, and record them as reference sampling times. Obtain the maximum discharge rate among all discharge rates at all reference sampling times, and record it as the maximum standard discharge rate. The difference between the maximum standard discharge speed and the discharge speed at the current sampling time is recorded as the second difference. The ratio of the second difference to the maximum standard discharge speed is recorded as the degree of abnormality of the discharge speed at the current sampling time.
9. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 6, characterized in that, The specific steps for obtaining the crushing anomaly factor at the current sampling time based on the degree of anomaly in the crushed particles and the degree of anomaly in the discharge rate at the current sampling time are as follows: At the current sampling time, the product of the abnormality degree of crushed particles and the preset first weighting coefficient is recorded as the seventh product, and the product of the abnormality degree of discharge speed and the preset second weighting coefficient is recorded as the eighth product. The sum of the seventh product and the eighth product is recorded as the crushing abnormality factor at the current sampling time.
10. The automated coarse crushing process for Chinese medicinal materials of different sizes according to claim 1, characterized in that, The specific steps for obtaining the target rotational speed and target feed rate at the next parameter adjustment after the current sampling time are as follows: The product of the inverse proportional value of the crushing anomaly factor at the current sampling time and the reference speed at the next parameter adjustment after the current sampling time is recorded as the target speed at the next parameter adjustment after the current sampling time. The product of the inverse proportional value of the crushing anomaly factor at the current sampling time and the reference feed rate at the next parameter adjustment after the current sampling time is recorded as the target feed rate at the next parameter adjustment after the current sampling time.