Intelligent weighing control system of powder automatic packaging machine

Abstract

The application discloses an intelligent weighing control system of a powder automatic packaging machine, relates to the technical field of weighing control, and solves the technical problem that material accumulation imbalance cannot be avoided, causing a great detection error of material weight; the system comprises a weighing data acquisition module, a discharging image acquisition module, a powder accumulation analysis module, a discharging efficiency analysis module, a discharging adjustment module and a data correction module; real-time video data of powder in the discharging process is acquired, and the top end accumulation image of powder material discharged into a material frame is acquired; the powder accumulation analysis module analyzes the distribution and accumulation of powder according to the top end accumulation image of powder material, and judges the uniformity of powder discharging; image data of powder accumulation is acquired in real time, the powder distribution in the discharging process is fed back in time, the discharging port position can be adjusted in time according to the analysis of the distribution and accumulation of powder, accumulation of material on one side of the material frame is avoided, and the detection accuracy is improved.

Description

Technical Field

[0001] This invention belongs to the field of weighing control, specifically an intelligent weighing control system for an automatic powder packaging machine. Background Technology

[0002] With the continuous development of industrial production and trade, the demand for accurate measurement and efficient packaging of powder materials is increasing. Especially in the food, chemical, and pharmaceutical industries, quantitative packaging of powder materials is crucial for ensuring product quality and improving production efficiency. Meanwhile, the continuous advancement of electronic technology provides technical support for the intelligent weighing control system of automatic powder packaging machines. In particular, the mature application of core technologies such as weighing sensors and PLCs has significantly improved the weighing accuracy and stability of automatic powder packaging machines. The intelligent weighing control system mainly relies on high-precision weighing sensors. When the weight of the material is applied to the sensor, it causes deformation of the strain gauge inside the sensor, thereby changing the sensor's resistance value and ultimately outputting an electrical signal proportional to the material weight. This technology enables accurate measurement of material weight and is mainly used in automatic powder packaging machines or can filling systems.

[0003] Powder weighing mechanisms typically have large material frames. If the material piles up too high on one side of the frame, it will affect the acquisition of weighing data. Most weighing control systems rely solely on a large number of weighing sensors to detect and calibrate the weight of powder materials, which makes it difficult to avoid unbalanced material accumulation and results in excessive errors in the detection of material weight. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent weighing control system for an automatic powder packaging machine to solve the technical problem that it is difficult to avoid unbalanced material accumulation, resulting in excessive error in the detection of material weight.

[0005] To address the above problems, a first aspect of the present invention provides an intelligent weighing control system for an automatic powder packaging machine, comprising:

[0006] Weighing data acquisition module: used to detect the weight of the powder material in the material frame through a gravity sensing unit arranged at the bottom of the powder weighing mechanism;

[0007] Material feeding image acquisition module: used to acquire real-time video data of powder during the feeding process, and to acquire images of the top accumulation of powder material in the feeding frame;

[0008] Powder Accumulation Analysis Module: By analyzing the top accumulation image of powder materials, the distribution and accumulation of powder are analyzed to determine the uniformity of powder feeding;

[0009] Material feeding efficiency analysis module: This module is used to detect the feeding efficiency of powder materials by analyzing the dynamic changes in video data during the feeding process.

[0010] Feeding adjustment module: used to set the target feeding weight, and adjust the feeding efficiency and feeding position of the feeding device according to the target feeding weight, the uniformity of powder feeding detected by the powder accumulation analysis module, and the feeding efficiency detected by the feeding efficiency analysis module.

[0011] Data correction module: Corrects the weighing data based on the detection values ​​from the weighing data acquisition module and the material feeding efficiency detected by the material feeding efficiency analysis module.

[0012] As a further aspect of the present invention: a plurality of gravity sensing units are provided, and the plurality of gravity sensing units are arranged in an array at the bottom end of the material frame of the powder weighing mechanism.

[0013] As a further aspect of the present invention: the feeding image acquisition module acquires real-time video data of the powder during the feeding process, and acquires images of the top accumulation of the powder material fed into the feeding frame, including the following steps:

[0014] During the material feeding process, 2 to 6 cameras are set up at the outlet of the powder material to collect real-time video data of the powder material feeding from different directions;

[0015] A camera is placed above the material frame of the powder weighing mechanism to capture images of the top accumulation of powder material as it is fed into the frame.

[0016] As a further aspect of the present invention: the powder accumulation analysis module analyzes the distribution and accumulation of powder through the top accumulation image of the powder material, and determines the uniformity of powder feeding, including the following steps:

[0017] The top-piled image of the powder material is acquired, the image is converted to grayscale, and the top boundary of the powder is detected by edge detection. The image of the powder material piled in the frame is then labeled.

[0018] In the labeled powder image, the stacking height of the powder on the top surface is measured by calculating pixels or actual physical units;

[0019] Several detection points are set on the top surface of the powder. By obtaining the accumulation height of each detection point on the top surface of the powder, the distribution and accumulation of the powder are analyzed to determine the uniformity of the powder feeding.

[0020] As a further aspect of the present invention: by obtaining the accumulation height of each detection point on the top surface of the powder, analyzing the distribution and accumulation of the powder, and determining the uniformity of powder feeding, the following steps are included:

[0021] The accumulation height of the powder material at each detection point on the top surface is obtained, the powder distribution and accumulation are analyzed, and the uniformity of the powder distribution is calculated using the following formula:

[0022]

[0023] Where K is the evaluation value for the uniformity of powder distribution, and L max L represents the maximum accumulation height of the top surface detection points of the powder material. min S represents the minimum accumulation height of the top surface detection points of the powder material. L L0 is the standard deviation of the accumulation height of the top surface detection points of the powder material, and L0 is the average value of the accumulation height of the top surface detection points of the powder material.

[0024] If the uniformity evaluation value K of the powder feeding distribution is greater than the preset threshold a, 0.6, then the uniformity of the powder feeding is judged to be unqualified.

[0025] If the uniformity evaluation value K of the powder feeding distribution is less than or equal to the preset threshold a, 0.6, then the uniformity of the powder feeding is deemed to be qualified.

[0026] As a further aspect of the present invention: the feeding efficiency analysis module detects the feeding efficiency of powder materials by analyzing the dynamic changes in video data during the feeding process, including the following steps:

[0027] The feeding speed of powder materials is detected by analyzing real-time video data of powder material feeding.

[0028] Acquire historical video data of powder material feeding from different directions, and the corresponding feeding efficiency during the historical data collection.

[0029] Capture image frames from historical data during the powder feeding process and mark the outline of the powder material at different time points in the image frames of the powder material feeding process;

[0030] Based on the outline of the powder material at different time points in the labeled image frame, the powder material image is converted to grayscale, and the grayscale value data of each point is obtained.

[0031] A deep learning model is trained by detecting the feeding speed of the powder material, the contour data of the powder material at different time points, and the gray value data of the powder material.

[0032] Real-time video data of powder material feeding is collected from different directions, and the contour data of powder material at different time points is marked in the image frames of the video data, and the gray value data of powder material is calculated.

[0033] The feeding speed of the detected powder material, the contour data of the labeled powder material at different time points, and the gray value data of the powder material are input into the trained deep learning model to detect the feeding efficiency of the powder material.

[0034] As a further aspect of the present invention: the feeding speed of the powder material is detected by analyzing the collected video data of the powder material feeding, including the following steps:

[0035] Acquire historical data of feeding videos of different powder materials at different feeding speeds;

[0036] Without a preset time interval, it continuously captures image frames from historical data during the powder feeding process; these frames are usually acquired at certain time intervals.

[0037] Select corner points or edge points in the image frame as feature points, perform feature point detection, and then label the feature points;

[0038] For each feature point, the average speed of each feature point in consecutive adjacent image frames is calculated by removing bits of the feature point between adjacent image frames and taking the time interval between adjacent image frames, and is used as the feeding speed of the corresponding feature point.

[0039] The historical data obtained, including the feeding speed of the feature points corresponding to the feeding video image frames of different powder materials at different feeding speeds, the actual feeding speed of the materials, and the type of powder material, are fed into the machine learning model for training.

[0040] During the real-time feeding process, real-time video data of powder material feeding from different directions is acquired. Feature points in the video data image frames are selected, and the feeding speed of each feature point is calculated. The feeding speed of each feature point and the type of powder material are input into the trained machine learning model to detect the feeding speed of the powder material.

[0041] As a further aspect of the present invention: the feeding adjustment module sets a target feeding weight, and adjusts the feeding efficiency and feeding position of the feeding device based on the target feeding weight, the uniformity of powder feeding detected by the powder accumulation analysis module, and the feeding efficiency detected by the feeding efficiency analysis module, including the following steps:

[0042] Set the target feeding weight using the feeding adjustment module;

[0043] If the powder accumulation analysis module determines that the uniformity of the powder feeding is unqualified, the feeding position of the feeding device will be adjusted to above the detection point corresponding to the minimum accumulation height of the powder material on the top surface until the powder accumulation analysis module determines that the uniformity of the powder feeding is qualified, and then the movement of the feeding position of the feeding device will be stopped.

[0044] If the powder accumulation analysis module determines that the uniformity of powder feeding is qualified, it acquires the feeding efficiency data detected by the feeding efficiency analysis module and the weight of the powder material in the material box detected by the weighing data acquisition module.

[0045] Calculate the difference between the target feeding weight and the detected weight of the powder material in the feed box, and use this as the weight difference. Divide the weight difference by the feeding efficiency to obtain the remaining feeding time.

[0046] When the remaining feeding time is greater than the preset threshold b, the feeding efficiency of the feeding device remains unchanged.

[0047] When the remaining feeding time is less than or equal to the preset threshold b, the feeding efficiency of the feeding device is adjusted to the minimum.

[0048] As a further aspect of the present invention: the data correction module corrects the weighing data based on the detection value of the weighing data acquisition module and the material feeding efficiency detected by the material feeding efficiency analysis module, including the following steps:

[0049] Acquire detection data from gravity sensing units arranged in an array at the bottom of the powder weighing mechanism's material frame;

[0050] The volatility of the detection data from the gravity sensing unit is evaluated based on the detection data from the gravity sensing unit.

[0051] If the fluctuation assessment value of the detection data of the gravity sensing unit is less than the preset threshold c, then there is no need to correct the weighing data of the weighing data acquisition module.

[0052] If the fluctuation evaluation value of the detection data of the gravity sensing unit is greater than or equal to the preset threshold c, the weighing data of the weighing data acquisition module will be corrected.

[0053] The weight of the powder material is calculated based on the feeding time and feeding efficiency detected by the feeding efficiency analysis module.

[0054] The average value of the detection values ​​from the gravity sensing unit is obtained. The average value of the detection values ​​from the gravity sensing unit and the average value of the material feeding weight are used as the standard value of the weighing data to correct the weighing data of the gravity sensing unit at the bottom of the powder weighing mechanism.

[0055] As a further aspect of the present invention: the fluctuation of the detection data of the gravity sensing unit is evaluated based on the detection data of the gravity sensing unit, using the following formula;

[0056]

[0057] Where R is the evaluation value of the fluctuation of the detection data of the gravity sensing unit, and G iLet G be the variance of the detection data of the i-th gravity sensing unit, i∈(1,…,n), n is the total number of gravity sensing units at the bottom of the powder weighing mechanism's material frame, and G is the average value of the detection data of the gravity sensing units at the bottom of the powder weighing mechanism's material frame.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] This invention acquires real-time video data of powder during the feeding process and captures images of the top accumulation of powder material in the feeding frame. The powder accumulation analysis module analyzes the distribution and accumulation of powder using these images to determine the uniformity of the powder feeding. Since the feeding frame of a powder weighing mechanism is typically large, excessive accumulation of material on one side can affect the acquisition of weighing data. Real-time acquisition of powder accumulation images provides timely feedback on the powder distribution during feeding, allowing the weighing data acquisition module to accurately detect the material weight. It also facilitates timely adjustment of the feeding port position based on the powder distribution and accumulation analysis, preventing excessive accumulation on one side of the feeding frame. Furthermore, image analysis provides precise powder accumulation information by calculating pixel values ​​or features in the image to quantify the uniformity of powder distribution, which is more accurate than traditional manual inspection or simple sensor measurements.

[0060] This invention sets the target feeding weight through a feeding adjustment module, and adjusts the feeding efficiency and position of the feeding device based on the target feeding weight, the uniformity of powder feeding detected by the powder accumulation analysis module, and the feeding efficiency detected by the feeding efficiency analysis module. By monitoring the feeding efficiency of powder materials during the feeding process, it provides real-time feedback on the adjustment needs of feeding speed and position, ensuring the stability and efficiency of the production process. It adjusts the position and speed of the feeding device based on real-time data, and evaluates the uniformity of feeding through powder accumulation analysis, helping to optimize the feeding process, reduce errors in material weight detection and material accumulation imbalance, and ensure that the target feeding weight is achieved. Attached Figure Description

[0061] 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.

[0062] Figure 1 This is a schematic diagram of the system framework of the present invention;

[0063] Figure 2 This is a schematic diagram showing the camera arrangement position at the outlet of the powder material in one embodiment of the present invention. Detailed Implementation

[0064] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] Please see Figures 1-2 The first aspect of this invention provides an intelligent weighing control system for an automatic powder packaging machine, comprising:

[0066] Weighing data acquisition module: used to detect the weight of the powder material in the material frame through a gravity sensing unit arranged at the bottom of the powder weighing mechanism;

[0067] Material feeding image acquisition module: used to acquire real-time video data of powder during the feeding process, and to acquire images of the top accumulation of powder material in the feeding frame;

[0068] Powder Accumulation Analysis Module: By analyzing the top accumulation image of powder materials, the distribution and accumulation of powder are analyzed to determine the uniformity of powder feeding;

[0069] Material feeding efficiency analysis module: This module is used to detect the feeding efficiency of powder materials by analyzing the dynamic changes in video data during the feeding process.

[0070] Feeding adjustment module: used to set the target feeding weight, and adjust the feeding efficiency and feeding position of the feeding device according to the target feeding weight, the uniformity of powder feeding detected by the powder accumulation analysis module, and the feeding efficiency detected by the feeding efficiency analysis module.

[0071] Data correction module: Corrects the weighing data based on the detection values ​​from the weighing data acquisition module and the material feeding efficiency detected by the material feeding efficiency analysis module.

[0072] Specifically, in this embodiment, the powder feeding image acquisition module acquires real-time video data of the powder during the feeding process and captures images of the top accumulation of powder material in the feeding frame. The powder accumulation analysis module analyzes the distribution and accumulation of powder material in the top accumulation image to determine the uniformity of powder feeding. The powder weighing mechanism's feeding frame is usually large, and excessive accumulation of material on one side of the frame can affect the acquisition of weighing data. Real-time acquisition of powder accumulation image data allows for timely feedback on the powder distribution during the feeding process. This facilitates the assessment of whether the weighing data acquisition module can accurately detect the material weight based on the powder distribution, and also allows for timely adjustment of the feeding port position based on the analysis of powder distribution and accumulation to prevent excessive accumulation of material on one side of the frame. Simultaneously, image analysis can provide precise powder accumulation information by calculating pixel values ​​or features in the image to quantify the uniformity of powder distribution, which is more accurate than traditional manual inspection or simple sensor measurements.

[0073] The material feeding efficiency analysis module analyzes the dynamic changes in video data during the powder material feeding process to detect the feeding efficiency of the powder material; by monitoring the feeding efficiency of the powder material in real time, it provides immediate feedback on the adjustment needs of the feeding speed and position.

[0074] The target feeding weight is set through the feeding adjustment module. Based on the target feeding weight, the uniformity of powder feeding detected by the powder accumulation analysis module, and the feeding efficiency detected by the feeding efficiency analysis module, the feeding efficiency and feeding position of the feeding device are adjusted. By monitoring the feeding efficiency of powder materials during the feeding process, the adjustment requirements for feeding speed and position are fed back in real time to ensure the stability and efficiency of the production process. The position and speed of the feeding device are adjusted according to real-time data, and the uniformity of feeding is evaluated through powder accumulation analysis to help optimize the feeding process, reduce errors in material weight detection and material accumulation imbalance, and ensure that the target feeding weight is achieved.

[0075] The data correction module corrects the weighing data based on the detection values ​​from the weighing data acquisition module and the material feeding efficiency detected by the material feeding efficiency analysis module. By comparing the data from the two modules, potential deviations or anomalies are identified and promptly corrected and confirmed, ensuring the consistency and accuracy of production data. Combining the data from the weighing data acquisition module and the material feeding efficiency analysis module allows for a more accurate assessment of the weight and efficiency of each feeding operation, improving the reliability and precision of the data.

[0076] In one embodiment of the present invention, a plurality of gravity sensing units are provided, and the plurality of gravity sensing units are arranged in an array at the bottom end of the material frame of the powder weighing mechanism.

[0077] In one embodiment of the present invention, the feeding image acquisition module acquires real-time video data of the powder during the feeding process and acquires images of the top accumulation of the powder material in the feeding frame, including the following steps:

[0078] During the material feeding process, 2 to 6 cameras are positioned at the powder material outlet to collect real-time video data of the powder material feeding from different directions; in this embodiment, 4 cameras are arranged at the powder material outlet, such as... Figure 2 As shown.

[0079] A camera is placed above the material frame of the powder weighing mechanism to capture images of the top accumulation of powder material as it is fed into the frame.

[0080] In one embodiment of the present invention, the powder accumulation analysis module analyzes the distribution and accumulation of powder by examining the top accumulation image of the powder material, and determines the uniformity of powder feeding, including the following steps:

[0081] The top-piled image of the powder material is acquired, the image is converted to grayscale, and the top boundary of the powder is detected by edge detection. The image of the powder material piled in the frame is then labeled.

[0082] In the labeled powder image, the stacking height of the powder on the top surface is measured by calculating pixels or actual physical units;

[0083] Several detection points are set on the top surface of the powder. By obtaining the accumulation height of each detection point on the top surface of the powder, the distribution and accumulation of the powder are analyzed to determine the uniformity of the powder feeding.

[0084] In one embodiment of the present invention, by obtaining the accumulation height of each detection point on the top surface of the powder, analyzing the distribution and accumulation of the powder, and determining the uniformity of powder feeding, the following steps are included:

[0085] The accumulation height of the powder material at each detection point on the top surface is obtained, the powder distribution and accumulation are analyzed, and the uniformity of the powder distribution is calculated using the following formula:

[0086]

[0087] Where K is the evaluation value for the uniformity of powder distribution, and L max L represents the maximum accumulation height of the top surface detection points of the powder material. min S represents the minimum accumulation height of the top surface detection points of the powder material. L L0 is the standard deviation of the accumulation height of the top surface detection points of the powder material, and L0 is the average value of the accumulation height of the top surface detection points of the powder material.

[0088] If the uniformity evaluation value K of the powder feeding distribution is greater than the preset threshold a, 0.6, then the uniformity of the powder feeding is judged to be unqualified.

[0089] If the uniformity evaluation value K of the powder feeding distribution is less than or equal to the preset threshold a, 0.6, then the uniformity of the powder feeding is deemed to be qualified.

[0090] Specifically, in this embodiment, based on the statistical analysis of a large amount of experimental data, when the uniformity evaluation value K is greater than 0.6, the uniformity of powder feeding is poor, and the accuracy of the weight data detected by the weighing data acquisition module is poor. In this embodiment, the preset threshold a is 0.6. If the uniformity evaluation value K of the powder feeding distribution is greater than the preset threshold 0.6, then the uniformity of the powder feeding is judged to be unqualified.

[0091] If the uniformity evaluation value K of the powder feeding distribution is less than or equal to the preset threshold of 0.6, then the uniformity of the powder feeding is deemed to be qualified.

[0092] In one embodiment of the present invention, the feeding efficiency analysis module detects the feeding efficiency of powder materials by analyzing the dynamic changes in video data during the feeding process, including the following steps:

[0093] The feeding speed of powder materials is detected by analyzing real-time video data of powder material feeding.

[0094] Acquire historical video data of powder material feeding from different directions, and the corresponding feeding efficiency during the historical data collection.

[0095] Capture image frames from historical data during the powder feeding process and mark the outline of the powder material at different time points in the image frames of the powder material feeding process;

[0096] Based on the outline of the powder material at different time points in the labeled image frame, the powder material image is converted to grayscale, and the grayscale value data of each point is obtained.

[0097] A deep learning model is trained by detecting the feeding speed of the powder material, the contour data of the powder material at different time points, and the gray value data of the powder material.

[0098] Real-time video data of powder material feeding is collected from different directions, and the contour data of powder material at different time points is marked in the image frames of the video data, and the gray value data of powder material is calculated.

[0099] The feeding speed of the detected powder material, the contour data of the labeled powder material at different time points, and the gray value data of the powder material are input into the trained deep learning model to detect the feeding efficiency of the powder material.

[0100] In one embodiment of the present invention, the feeding speed of the powder material is detected by analyzing the collected video data of the powder material feeding, including the following steps:

[0101] Acquire historical data of feeding videos of different powder materials at different feeding speeds;

[0102] Without a preset time interval, it continuously captures image frames from historical data during the powder feeding process; these frames are usually acquired at certain time intervals.

[0103] Select corner points or edge points in the image frame as feature points, perform feature point detection, and then label the feature points;

[0104] For each feature point, its movement vector in adjacent image frames is calculated using an optical flow algorithm. The average velocity of each feature point in consecutive adjacent image frames is calculated using the bit removal of the feature point between adjacent image frames and the time interval between adjacent image frames, and is taken as the feeding speed of the corresponding feature point.

[0105] The historical data obtained, including the feeding speed of the feature points corresponding to the feeding video image frames of different powder materials at different feeding speeds, the actual feeding speed of the materials, and the type of powder material, are fed into the machine learning model for training.

[0106] During the real-time feeding process, real-time video data of powder material feeding from different directions is acquired. Feature points in the video data image frames are selected, and the feeding speed of each feature point is calculated. The feeding speed of each feature point and the type of powder material are input into the trained machine learning model to detect the feeding speed of the powder material.

[0107] In one embodiment of the present invention, the feeding adjustment module sets a target feeding weight, and adjusts the feeding efficiency and feeding position of the feeding device according to the target feeding weight, the uniformity of powder feeding detected by the powder accumulation analysis module, and the feeding efficiency detected by the feeding efficiency analysis module, including the following steps:

[0108] Set the target feeding weight using the feeding adjustment module;

[0109] If the powder accumulation analysis module determines that the uniformity of the powder feeding is unqualified, the feeding position of the feeding device will be adjusted to above the detection point corresponding to the minimum accumulation height of the powder material on the top surface until the powder accumulation analysis module determines that the uniformity of the powder feeding is qualified, and then the movement of the feeding position of the feeding device will be stopped.

[0110] If the powder accumulation analysis module determines that the uniformity of powder feeding is qualified, it acquires the feeding efficiency data detected by the feeding efficiency analysis module and the weight of the powder material in the material box detected by the weighing data acquisition module.

[0111] Calculate the difference between the target feeding weight and the detected weight of the powder material in the feed box, and use this as the weight difference. Divide the weight difference by the feeding efficiency to obtain the remaining feeding time.

[0112] When the remaining feeding time is greater than the preset threshold b, the feeding efficiency of the feeding device remains unchanged.

[0113] When the remaining feeding time is less than or equal to the preset threshold b, the feeding efficiency of the feeding device is adjusted to the minimum.

[0114] Specifically, in this embodiment, the preset threshold b is set to 5 seconds. When the remaining feeding time is greater than 5 seconds, the feeding efficiency of the feeding device remains unchanged.

[0115] When the remaining feeding time is less than or equal to 5 seconds, the feeding efficiency of the feeding device is adjusted to the minimum.

[0116] In one embodiment of the present invention, the data correction module corrects the weighing data based on the detection value of the weighing data acquisition module and the feeding efficiency detected by the feeding efficiency analysis module, including the following steps:

[0117] Acquire detection data from gravity sensing units arranged in an array at the bottom of the powder weighing mechanism's material frame;

[0118] The volatility of the detection data from the gravity sensing unit is evaluated based on the detection data from the gravity sensing unit.

[0119] If the fluctuation assessment value of the detection data of the gravity sensing unit is less than the preset threshold c, then there is no need to correct the weighing data of the weighing data acquisition module.

[0120] If the fluctuation evaluation value of the detection data of the gravity sensing unit is greater than or equal to the preset threshold c, the weighing data of the weighing data acquisition module will be corrected.

[0121] The weight of the powder material is calculated based on the feeding time and feeding efficiency detected by the feeding efficiency analysis module.

[0122] The average value of the detection values ​​from the gravity sensing unit is obtained. The average value of the detection values ​​from the gravity sensing unit and the average value of the material feeding weight are used as the standard value of the weighing data to correct the weighing data of the gravity sensing unit at the bottom of the powder weighing mechanism.

[0123] Specifically, in this embodiment, after statistical analysis of a large amount of experimental data, when the fluctuation evaluation value of the detection data of the gravity sensing unit is less than 0.1, the variation value of the weighing data of the weighing data acquisition module usually does not exceed the maximum allowable variation range in this embodiment. Therefore, in this embodiment, the preset threshold c is 0.1. If the fluctuation evaluation value of the detection data of the gravity sensing unit is less than 0.1, there is no need to correct the weighing data of the weighing data acquisition module.

[0124] If the fluctuation assessment value of the detection data of the gravity sensing unit is greater than or equal to 0.1, the weighing data of the weighing data acquisition module will be corrected.

[0125] In one embodiment of the present invention, the fluctuation of the detection data of the gravity sensing unit is evaluated based on the detection data of the gravity sensing unit by the following formula;

[0126]

[0127] Where R is the evaluation value of the fluctuation of the detection data of the gravity sensing unit, and G i Let G be the variance of the detection data of the i-th gravity sensing unit, i∈(1,…,n), n is the total number of gravity sensing units at the bottom of the powder weighing mechanism's material frame, and G is the average value of the detection data of the gravity sensing units at the bottom of the powder weighing mechanism's material frame.

[0128] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. An intelligent weighing control system for an automatic powder packaging machine, characterized in that, include: Weighing data acquisition module: used to detect the weight of the powder material in the material frame through a gravity sensing unit arranged at the bottom of the powder weighing mechanism; Material feeding image acquisition module: used to acquire real-time video data of powder during the feeding process, and to acquire images of the top accumulation of powder material in the feeding frame; Powder Accumulation Analysis Module: By analyzing the top accumulation image of powder materials, the distribution and accumulation of powder are analyzed to determine the uniformity of powder feeding; Material feeding efficiency analysis module: This module is used to detect the feeding efficiency of powder materials by analyzing the dynamic changes in video data during the feeding process. Feeding adjustment module: used to set the target feeding weight, and adjust the feeding efficiency and feeding position of the feeding device according to the target feeding weight, the uniformity of powder feeding detected by the powder accumulation analysis module, and the feeding efficiency detected by the feeding efficiency analysis module. Data correction module: Corrects the weighing data based on the detection values ​​from the weighing data acquisition module and the material feeding efficiency detected by the material feeding efficiency analysis module; The feeding efficiency analysis module detects the feeding efficiency of powder materials by analyzing the dynamic changes in video data during the feeding process, including the following steps: The feeding speed of powder materials is detected by analyzing real-time video data of powder material feeding. Acquire historical video data of powder material feeding from different directions, and the corresponding feeding efficiency during the historical data collection. Capture image frames from historical data during the powder feeding process and mark the outline of the powder material at different time points in the image frames of the powder material feeding process; Based on the outline of the powder material at different time points in the labeled image frame, the powder material image is converted to grayscale, and the grayscale value data of each point is obtained. A deep learning model is trained by detecting the feeding speed of the powder material, the contour data of the powder material at different time points, and the gray value data of the powder material. Real-time video data of powder material feeding is collected from different directions, and the contour data of powder material at different time points is marked in the image frames of the video data, and the gray value data of powder material is calculated. The feeding speed of the detected powder material, the outline data of the labeled powder material at different time points, and the gray value data of the powder material are input into the trained deep learning model to detect the feeding efficiency of the powder material. By analyzing the collected video data of powder material feeding, the feeding speed of the powder material is detected, including the following steps: Acquire historical data of feeding videos of different powder materials at different feeding speeds; Without a preset time interval, it continuously captures image frames from historical data during the powder feeding process; these frames are usually acquired at certain time intervals. Select corner points or edge points in the image frame as feature points, perform feature point detection, and then label the feature points; For each feature point, the average speed of each feature point in consecutive adjacent image frames is calculated by removing bits of the feature point between adjacent image frames and taking the time interval between adjacent image frames, and is used as the feeding speed of the corresponding feature point. The historical data obtained, including the feeding speed of the feature points corresponding to the feeding video image frames of different powder materials at different feeding speeds, the actual feeding speed of the materials, and the type of powder material, are fed into the machine learning model for training. During the real-time feeding process, real-time video data of powder material feeding from different directions is acquired. Feature points in the video data image frames are selected, and the feeding speed of each feature point is calculated. The feeding speed of each feature point and the type of powder material are input into the trained machine learning model to detect the feeding speed of the powder material.

2. The intelligent weighing control system for an automatic powder packaging machine according to claim 1, characterized in that, The gravity sensing unit is provided in several units, and the several gravity sensing units are arranged in an array at the bottom of the material frame of the powder weighing mechanism.

3. The intelligent weighing control system for an automatic powder packaging machine according to claim 1, characterized in that, The feeding image acquisition module acquires real-time video data of the powder during the feeding process and captures images of the top accumulation of powder material in the feeding frame, including the following steps: During the material feeding process, 2 to 6 cameras are set up at the outlet of the powder material to collect real-time video data of the powder material feeding from different directions; A camera is placed above the material frame of the powder weighing mechanism to capture images of the top accumulation of powder material as it is fed into the frame.

4. The intelligent weighing control system for an automatic powder packaging machine according to claim 3, characterized in that, The powder accumulation analysis module analyzes the distribution and accumulation of powder materials through the top accumulation image of the powder material, and determines the uniformity of powder feeding, including the following steps: The top-piled image of the powder material is acquired, the image is converted to grayscale, and the top boundary of the powder is detected by edge detection. The image of the powder material piled in the frame is then labeled. In the labeled powder image, the stacking height of the powder on the top surface is measured by calculating pixels or actual physical units; Several detection points are set on the top surface of the powder. By obtaining the accumulation height of each detection point on the top surface of the powder, the distribution and accumulation of the powder are analyzed to determine the uniformity of the powder feeding.

5. The intelligent weighing control system for an automatic powder packaging machine according to claim 4, characterized in that, By acquiring the accumulation height of the powder at each detection point on the top surface, analyzing the powder distribution and accumulation, and determining the uniformity of powder feeding, the following steps are included: The accumulation height of the powder material at each detection point on the top surface is obtained, the powder distribution and accumulation are analyzed, and the uniformity of the powder distribution is calculated using the following formula: Where K is the evaluation value for the uniformity of powder distribution, and L max L represents the maximum accumulation height of the top surface detection points of the powder material. min S represents the minimum accumulation height of the top surface detection points of the powder material. L L0 is the standard deviation of the accumulation height of the top surface detection points of the powder material, and L0 is the average value of the accumulation height of the top surface detection points of the powder material. If the uniformity evaluation value K of the powder feeding distribution is greater than the preset threshold a, then the uniformity of the powder feeding is judged to be unqualified. If the uniformity evaluation value K of the powder feeding distribution is less than or equal to the preset threshold a, then the uniformity of the powder feeding is deemed to be qualified.

6. The intelligent weighing control system for an automatic powder packaging machine according to claim 5, characterized in that, The feeding adjustment module sets the target feeding weight, and adjusts the feeding efficiency and feeding position of the feeding device based on the target feeding weight, the uniformity of powder feeding detected by the powder accumulation analysis module, and the feeding efficiency detected by the feeding efficiency analysis module, including the following steps: Set the target feeding weight using the feeding adjustment module; If the powder accumulation analysis module determines that the uniformity of the powder feeding is unqualified, the feeding position of the feeding device will be adjusted to above the detection point corresponding to the minimum accumulation height of the powder material on the top surface until the powder accumulation analysis module determines that the uniformity of the powder feeding is qualified, and then the movement of the feeding position of the feeding device will be stopped. If the powder accumulation analysis module determines that the uniformity of powder feeding is qualified, it acquires the feeding efficiency data detected by the feeding efficiency analysis module and the weight of the powder material in the material box detected by the weighing data acquisition module. Calculate the difference between the target feeding weight and the detected weight of the powder material in the feed box, and use this as the weight difference. Divide the weight difference by the feeding efficiency to obtain the remaining feeding time. When the remaining feeding time is greater than the preset threshold b, the feeding efficiency of the feeding device remains unchanged. When the remaining feeding time is less than or equal to the preset threshold b, the feeding efficiency of the feeding device is adjusted to the minimum.

7. The intelligent weighing control system for an automatic powder packaging machine according to claim 2, characterized in that, The data correction module corrects the weighing data based on the detection values ​​from the weighing data acquisition module and the feeding efficiency detected by the feeding efficiency analysis module, including the following steps: Acquire detection data from gravity sensing units arranged in an array at the bottom of the powder weighing mechanism's material frame; The volatility of the detection data from the gravity sensing unit is evaluated based on the detection data from the gravity sensing unit. If the fluctuation assessment value of the detection data of the gravity sensing unit is less than the preset threshold c, then there is no need to correct the weighing data of the weighing data acquisition module. If the fluctuation evaluation value of the detection data of the gravity sensing unit is greater than or equal to the preset threshold c, the weighing data of the weighing data acquisition module will be corrected. The weight of the powder material is calculated based on the feeding time and feeding efficiency detected by the feeding efficiency analysis module. The average value of the detection values ​​from the gravity sensing unit is obtained. The average value of the detection values ​​from the gravity sensing unit and the average value of the material feeding weight are used as the standard value of the weighing data to correct the weighing data of the gravity sensing unit at the bottom of the powder weighing mechanism.

8. The intelligent weighing control system for an automatic powder packaging machine according to claim 7, characterized in that, The volatility of the gravity sensing unit's detection data is evaluated based on the following formula. Where R is the evaluation value of the fluctuation of the detection data of the gravity sensing unit, Gi is the variance of the detection data of the i-th gravity sensing unit, i∈(1,…,n), n is the total number of gravity sensing units at the bottom of the powder weighing mechanism, and G is the average value of the detection data of the gravity sensing units at the bottom of the powder weighing mechanism.

Images