On-line monitoring system for production process of milk powder strips

Abstract

The invention discloses an online monitoring system for a milk powder strip production process, and belongs to the field of online monitoring, and the online monitoring system comprises a raw material marking module, a metering and weighing module, a raw material ratio coordination processing module, an air tightness detection module and a visual processing module. The problems that in the prior art, control over the length and weight of milk powder strips is not accurate, packaging sealing performance detection is insufficient, production adjustment lags behind, and real-time data collection and tracing are lacked are solved. The invention aims to realize accurate monitoring and automatic adjustment of the milk powder strip production process, ensure the consistency of product quality and the improvement of production efficiency, and provide quick response and flexible customization capabilities for a production line.

Description

Technical Field

[0001] The present invention belongs to the field of on-line monitoring, and particularly relates to an on-line monitoring system for the production process of milk powder strips. Background Art

[0002] As a dairy product with high portability and standardized dosage, the market demand for milk powder strips is increasing day by day. The production and canning process of milk powder strips includes multiple links such as powder filling, strip cutting, sealing packaging and finished product inspection. The quality of these links directly determines the appearance, performance and shelf life of the product. However, there are many problems in the existing milk powder strip canning process, such as inconsistent length and weight of strip products, packaging airtightness problems, and insufficient real-time monitoring and data traceability capabilities.

[0003] Most of the existing technical solutions have optimized single process links and failed to comprehensively consider all key factors in the production process of milk powder strips, such as length, weight, packaging airtightness, and real-time monitoring and data traceability during the production process. Therefore, the existing technical solutions have obvious limitations in improving production efficiency and quality stability. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides an on-line monitoring system for the production process of milk powder strips, including:

[0005] A raw material marking module, used to mark the raw materials before canning the milk powder strips to obtain the marked raw materials;

[0006] A metering and weighing module, used to meter the marked raw materials to obtain the metered raw materials;

[0007] A raw material ratio coordination processing module, used to perform partitioned loading of raw materials on the metered raw materials to obtain the loaded raw materials;

[0008] An airtightness detection module, used to detect the airtightness of the loaded raw materials before and after roll film encapsulation;

[0009] A vision processing module, used to obtain image information of all processes based on a CCD industrial camera and construct a neural network model, input the image information into the neural network model for calculation, and obtain a monitoring result.

[0010] Preferably, the metering and weighing module is further used for: metering the raw materials, and if the weight of each raw material exceeds the set normal threshold, it is determined that there is a quality problem with the current milk powder strip ratio, and the current raw materials continue to enter the raw material ratio coordination processing module through the rotary table, and the raw material ratio is readjusted through the raw material ratio coordination processing module to enter the next round of raw material ratio.

[0011] Preferably, the airtightness detection module includes:

[0012] A pre-encapsulation detection unit for detecting the airtightness of the rolled film;

[0013] A post-encapsulation detection unit for monitoring the airtightness of the encapsulated raw materials.

[0014] Preferably, the pre-encapsulation detection unit includes: a ventilation hole and a container;

[0015] The operation process of the pre-encapsulation detection unit includes: blowing uniform gas through the ventilation hole towards the surface of the rolled film, and arranging a container filled with water on the other side of the surface of the rolled film. If bubbles appear in the water in the container, it is determined that the rolled film has leaked air.

[0016] Preferably, the operation process of the post-encapsulation detection unit includes: using a push rod to extrude the outer packaging of the milk powder strips after the rolled film is encapsulated. While extruding the outer packaging of the milk powder strips, using a CCD industrial camera to obtain the deformation situation of the strip-shaped outer packaging of the milk powder strips, generating a deformation image, calculating the deformation image through an image processing algorithm to obtain the process of edge deformation, and obtaining the rolled film detection result based on the process of edge deformation.

[0017] Preferably, the operation process of the post-encapsulation detection unit further includes: calculating the rate of change of the edge angle with time during the process of edge deformation, and setting a standard non-air leakage deformation rate. If the standard non-air leakage deformation rate is greater than the rate of change of the edge angle with time, it is determined that the milk powder strip packaging is not leaking air. If the standard non-air leakage deformation rate is less than or equal to the rate of change of the edge angle with time, it is determined that the milk powder strip packaging is leaking air.

[0018] Preferably, it further includes: a size measurement module for measuring the size of the rolled film.

[0019] Preferably, the size measurement module is further used for: obtaining the image information of the rolled film;

[0020] Performing grayscale processing and threshold segmentation on the image information to obtain the image information of the surface of the rolled film;

[0021] Splitting the image information of the surface of the rolled film into RGB channels and performing HSV color conversion to obtain new image information;

[0022] Calculating the thresholds of each channel in the new image information to obtain the standard deviation values of pixels in each channel;

[0023] Obtaining the Th value based on the standard deviation values of pixels in each channel and the standard deviation of pixels in the original image;

[0024] Obtain a preset threshold, compare the Th value with the preset threshold. If the Th value is greater than the preset threshold, use the Canny algorithm to extract the edge pixel information of the roll film, and perform a first-level wavelet transform to obtain the edge pixel information.

[0025] Statistically process the edge pixel information to obtain the current size of the roll film.

[0026] Compared with the prior art, the present invention has the following advantages and technical effects:

[0027] The online monitoring device system of the present invention adopts a modular design and integrates multiple functional modules such as a vision processing module, an intelligent metering and weighing module, a size measurement module, an airtightness detection module, a raw material marking module, and a raw material ratio coordination processing module. The functions of each module are clear and cooperate with each other, covering the entire process of milk powder strip canning, so as to achieve the purpose of real-time online monitoring device and process control. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The drawings constituting a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation to this application. In the drawings:

[0029] Figure 1 Schematic diagram of the milk powder strip canning device and online monitoring system for the embodiment of the present invention;

[0030] Figure 2 Schematic diagram of the airtightness detection device before encapsulation for the embodiment of the present invention;

[0031] Figure 3 Schematic diagram of the airtightness detection device after encapsulation for the embodiment of the present invention;

[0032] Figure 4 Schematic diagram of the image algorithm design process for the embodiment of the present invention;

[0033] Figure 5 Schematic diagram of the information storage and processing module for the embodiment of the present invention;

[0034] Figure 6 Schematic diagram of the neural network product quality evaluation model for the embodiment of the present invention;

[0035] Among them, 1. Horizontal transfer workbench; 2. Airtightness detection module; 3. Rotary workbench; 4. Metering and weighing module; 5. Raw material ratio coordination processing module; 6. CCD industrial camera; 7. Action mechanism; 8. Raw material marking module; 9. Vent hole; 10. Roll film surface; 11. Container; 12. Water; 13. Push rod; 14. Milk powder strip. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0037] It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0038] Embodiment 1

[0039] As Figure 1 shown, in this embodiment, an on-line monitoring system for the production process of milk powder strips is provided, including:

[0040] A raw material marking module 8, used to mark the raw materials before the milk powder strips are canned to obtain the marked raw materials;

[0041] A metering and weighing module 4, used to meter the marked raw materials to obtain the metered raw materials;

[0042] A raw material ratio coordination processing module 5, used to perform partitioned loading of raw materials on the metered raw materials to obtain the loaded raw materials;

[0043] An airtightness detection module 2, used to detect the airtightness of the loaded raw materials before and after roll film packaging;

[0044] A vision processing module, used to obtain the image information of all processes based on a CCD industrial camera and construct a neural network model, input the image information into the neural network model for calculation, and obtain the monitoring result.

[0045] The raw material marking module 8 is used for marking the raw materials before the milk powder strips are canned, realizing the rapid marking of the raw materials and the storage of the corresponding raw material information. After the raw materials are filled, the raw materials are placed in each fixed container, and each container has an independent two-dimensional code label. The specific information of the raw materials is stored in the two-dimensional code. The container contains raw materials. After being transported by the conveyor belt, the raw materials enter the rotary workbench. Further, the CCD industrial camera 6 collects the images of the work in real time, such as the images of the raw materials and the two-dimensional code on the container containing the raw materials. Further, the two-dimensional code can be decoded for information through the images collected by the CCD industrial camera 6, and the decoded information is transmitted to the computer to determine the source of the raw materials.

[0046] The described metering and weighing module 4 is used for metering the weight of the raw materials of the milk powder strips. Specifically, before filling, it is necessary to meter the raw materials to ensure the correctness of the raw material ratio. If the weight of each raw material exceeds the set normal threshold, the system determines that there is a quality problem with the current milk powder strip ratio, and the computer system issues an instruction to make the actuating mechanism 7 act, so that the current raw materials continue to enter the raw material ratio coordination processing module 5 through the rotary workbench 3. Further, the metering information is also stored in the information storage module.

[0047] Furthermore, if a quality problem is detected in the milk powder strips through the described metering and weighing module 4, it is necessary to readjust the ratio of the raw materials through the raw material ratio coordination processing module and enter the next round of raw material ratio to monitor the raw material ratio in real time and ensure that the prepared raw materials meet the requirements.

[0048] The described raw material ratio coordination processing module 5 is used for partitioning and filling the raw materials according to the set raw material ratio. Specifically, different portions of raw materials are pushed out by a push rod and mixed in a container to achieve the effect of the ratio relationship of different portions of raw materials. Further, if the ratio of the raw materials does not meet the standard all the time, the non-compliant raw materials continue to be proportioned through the described raw material ratio coordination processing module until the ratio meets the standard.

[0049] The described vision processing module mainly includes a high-end industrial intelligent CCD industrial camera 6 and a vision image algorithm processing framework, etc. The described CCD industrial camera 6 is used for on-line monitoring of the process of making milk powder strips and obtaining image information of each process. Further, monitoring the canning process is to ensure that every step can be traced, ensure safety, and at the same time facilitate the detection of process problems in each step, discover problems in time, and achieve the effect of on-line regulation, so as to improve the food safety and achieve the purpose of on-line regulation of quality and visualization of quality assessment. Further, the obtained image information of each process can be used as the source information for on-line monitoring device system training and regulation to enhance the real-time decision-making ability of the neural network.

[0050] The horizontal transfer workbench 1 is connected to the airtightness detection module 2. The described airtightness detection module is used for detecting the airtightness before and after the film is encapsulated. In order to avoid problems such as the film being damaged after leaving the factory or accidentally by workers during installation. Before the film enters the encapsulation, the film will pass through the airtightness detection module for airtightness detection. Specifically, uniform gas is blown through the air vent 9 onto the surface 10 of the film, and a container 11 filled with water is arranged on the other side of the surface 10 of the film. If there is a leak in the air film, the water 12 in the container will immediately show bubbles, so as to judge whether the film is leaking. Figure 2As shown. This device has the advantages of simple installation and low cost, and is suitable for the preliminary airtightness non-investigation of various devices in the early stage.

[0051] Further, the airtightness detection after the film roll packaging is automatically carried out by extrusion. Specifically, first, the outer package of the milk powder strip 14 after the film roll packaging is extruded using the push rod 13. Further, the extrusion process is a gradual process to prevent damage to the outer package of the milk powder strip 14. At the same time, while extruding the outer package of the milk powder strip 14, the CCD industrial camera 13 is used to obtain the deformation situation of the strip-shaped outer package of the milk powder strip 14. As Figure 3 shown, further, the obtained image is processed using an image processing algorithm to calculate the process of edge deformation, so as to judge whether there is air leakage in the film roll after packaging. Calculate the rate T1 of the change of the edge angle of the milk powder strip 14 with time, and at the same time set the standard non-air leakage deformation rate ST. Compare the two. If ST is greater than T1, it is judged as a non-air leakage state. If ST is less than or equal to T1, it is judged as an air leakage state. Because if there is air leakage in the milk powder strip packaging, under the same extrusion pressure, the larger the angle of the edges at both ends of the strip-shaped milk powder strip 14 warps, and the more serious the deformation situation. This simple airtightness detection equipment saves unnecessary expenses for enterprises and has high applicability.

[0052] A size measurement module for measuring the size of the film roll. Before packaging, it is necessary to ensure the specific size of the film roll. Similarly, after packaging, it is also necessary to ensure the size after packaging. Specifically, the realization of size measurement depends on the acquisition of the CCD industrial camera image and the design of the image processing algorithm. In the production process of the assembly line, various harsh environments are inevitable, which will seriously affect the quality of the image obtained by the camera. While ensuring the efficient operation of the hardware, the design of the software algorithm is crucial.

[0053] Further, the algorithm mentioned above involves image processing algorithms. To enable the system to adapt to the influence of various complex ambient lights, it is necessary to preprocess the collected images to extract the required key information. First, the obtained images are grayscaled and threshold segmented. The noise on the surface of the roll film caused by ambient light is filtered to obtain the image information of the roll film surface. Further, the images are split into RGB channels and subjected to HSV color conversion to obtain new image information, namely the image information of six channels: RGB and HSV respectively. Further, the thresholds of each channel are calculated to obtain the standard deviation value Stdi of each channel pixel. The difference between the standard deviation value Stdi and the standard deviation value Stdi of the original image pixels is used to obtain the Th value. The size of the current Th value and the current Threshold value is judged. If the current Th value is greater than the current Threshold value, the next step will be continued. If the current Th value is less than the current Threshold value, the Threshold will be adjusted appropriately according to the real-time working conditions until the current Th value is greater than the current Threshold value. Then, the Canny algorithm is used to extract the edge pixel information of the roll film, and a first-level wavelet transform is performed to filter the high-frequency signals, store the low-frequency signals, and store the current information to establish the working condition correlation of the current process. Further, the edge pixel information is statistically analyzed to calculate the current size of the roll film, and the current size information is stored. The specific visual algorithm flowchart is as Figure 4 shown.

[0054] Further, the information of all modules will be stored in the information storage and processing module, as Figure 5 shown, to provide the original data for the training of the data model. Specifically, through the input of the original data, the neural network model can be trained in real time online to establish the correlation model between the milk powder strip filling and the final product quality, providing a guarantee for ensuring the filling quality. As Figure 6 shown, the neural network model includes three main parts: the input layer, the hidden layer, and the output layer.

[0055] Further, the neural network model is mainly used for the online evaluation of product quality, as well as for online adjustment of the process problems in the milk powder strip packaging process, to discover the quality problems of products in real time, which provides great help for reducing the losses of enterprises.

[0056] The beneficial effects of this embodiment are as follows:

[0057] (1) The online monitoring device system of this embodiment adopts a modular design and integrates multiple functional modules such as a vision processing module, an intelligent metering and weighing module, a size measurement module, an airtightness detection module, a raw material marking module, and a raw material ratio coordination and processing module. The functions of each module are clear and cooperate with each other, covering the whole process of milk powder strip canning, so as to achieve the purpose of real-time online monitoring device and process control.

[0058] (2) This embodiment uses a CCD industrial camera and vision image processing algorithms to collect image information of the milk powder strip canning process in real time. Through image processing techniques such as grayscale conversion, RGB / HSV channel segmentation, and wavelet transform, key features are accurately extracted to achieve real-time monitoring of the production process, problem traceability, and process optimization, improving the visual control level of food safety and product quality.

[0059] (3) This embodiment designs two airtightness detection methods before and after film packaging. Before packaging, a preliminary screening is carried out through the "gas-water container bubbling method" to detect whether there are air leakage points on the film. The method is simple, low-cost, and highly applicable; after packaging, through the method of "extrusion + image deformation analysis", the CCD camera is used to detect the change rate of the deformation angle of the milk powder strip package, so as to accurately judge whether there is an air leakage phenomenon, effectively improving the accuracy and reliability of the airtightness detection.

[0060] (4) The raw material marking module of this embodiment marks the raw materials with two-dimensional codes before the milk powder strips are canned. The detailed information of the raw materials is recorded in the two-dimensional codes, which is convenient for subsequent traceability. The two-dimensional code information is read in real time through a high-end CCD camera to ensure the controllability of the raw material source and provide an effective basis for the assignment of responsibilities for quality problems.

[0061] (5) All the detection data generated by the modules are stored in the information storage and processing module for training the neural network model. This model can learn production data in real time online and dynamically adjust the production process according to the detection results, so as to establish a correlation model between the milk powder strip filling and the final product quality, optimize the process flow, and improve the consistency and qualification rate of the products.

[0062] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An online monitoring system for the production process of milk powder strips, characterized in that: include: The raw material marking module is used to mark the raw materials before the milk powder strips are canned to obtain the marked raw materials; A weighing module is used to weigh the marked raw materials to obtain the weighed raw materials; A raw material ratio coordination processing module is used to perform partition loading of the measured raw materials to obtain loaded raw materials; An airtightness detection module, used for performing airtightness detection on the loaded raw materials before and after film packaging; The visual processing module is used to obtain image information of all processes based on a CCD industrial camera and build a neural network model, and input the image information into the neural network model for calculation to obtain monitoring results.

2. The system according to claim 1, characterized in that The metering and weighing module is also used to: measure the raw materials. If the weight of each raw material exceeds the set normal threshold, it is determined that there is a quality problem with the current milk powder bar ratio, so that the current raw materials continue to pass through the rotating workbench into the raw material ratio coordination processing module, and the raw material ratio is readjusted through the raw material ratio coordination processing module to enter the next round of raw material ratio.

3. The system according to claim 1, characterized in that The airtightness detection module comprises: Pre-packaging inspection unit, used to inspect the air tightness of the roll film; The post-packaging detection unit is used to monitor the air tightness of the raw materials after packaging.

4. The system according to claim 3, characterized in that The pre-packaging detection unit comprises: a vent and a container; The operation process of the pre-packaging detection unit includes: blowing uniform gas to the surface of the roll film through the vent hole, and setting a container filled with water on the other side of the roll film surface. If bubbles appear in the water in the container, it is determined that the roll film has leaked.

5. The system according to claim 3, characterized in that The operation process of the post-packaging detection unit includes: using a push rod to squeeze the outer packaging of the milk powder bar after the roll film is packaged, and while squeezing the outer packaging of the milk powder bar, using a CCD industrial camera to obtain the deformation of the outer packaging of the milk powder bar, generating a deformed image, calculating the deformed image through an image processing algorithm, obtaining the edge deformation process, and obtaining the roll film detection result based on the edge deformation process.

6. The system according to claim 5, characterized in that The operation process of the post-packaging detection unit also includes: calculating the rate at which the edge angle changes over time during the edge deformation process, and setting a standard airtight deformation rate. If the standard airtight deformation rate is greater than the rate at which the edge angle changes over time, the milk powder bar package is judged to be airtight. If the standard airtight deformation rate is less than or equal to the rate at which the edge angle changes over time, the milk powder bar package is judged to be leaking.

7. The system according to claim 1, characterized in that Also includes: Size measurement module, used for measuring the size of roll film.

8. The system according to claim 7, characterized in that The size measurement module is also used to: obtain roll film image information; grayscale processing and threshold segmentation are performed on the image information to obtain image information of the film roll surface; The image information on the surface of the roll film is split into RGB channels, and HSV color conversion is performed to obtain new image information; Calculating the threshold of each channel in the new image information to obtain the standard deviation value of each channel pixel; Obtaining a Th value based on the standard deviation of pixels in each channel and the standard deviation of pixels in the original image; Obtain a preset threshold, compare the Th value with the preset threshold, and if the Th value is greater than the preset threshold, use the Canny algorithm to extract edge pixel information of the roll film, and perform a first-level wavelet transform to obtain edge pixel information; The edge pixel information is counted to obtain the current size of the roll film.

Images