An intelligent supervision system for plastic film production

Abstract

The present invention discloses an intelligent supervision system for plastic film production, which relates to the technical field of plastic film production monitoring and effectively improves the quality stability of plastic film production. The present invention collects various production materials and historical production data of various plastic films through a variety of sensors, as well as collects the real-time production status data of each plastic film, generates the production preconditions and production structure links of the corresponding types of plastic films according to the plastic film data set, integrates the production preconditions and production structure links to obtain a production monitoring network, inputs the real-time production status data into the production monitoring network to generate corresponding production adjustment decisions and execute them.

Description

Technical Field

[0001] The present invention relates to the technical field of plastic film production monitoring, and specifically to an intelligent supervision system for plastic film production. Background Art

[0002] Plastic film production refers to the production of various plastic film products for different purposes by using plastic raw materials such as polyethylene and polypropylene through processes such as extrusion film forming, stretching film forming, and calendering film forming. Plastic films are thin film-like plastic products widely used in agricultural production, mainly including covering films, fresh-keeping films, windbreak nets, greenhouse films, etc.

[0003] Due to the numerous steps in the plastic film production process, it is easy to have abnormalities during the plastic film production process, resulting in unqualified plastic film products. Moreover, some plastic film production technologies have insurmountable problems in controlling production parameters, raw material ratios, etc., leading to unstable quality of plastic film products. Therefore, how to improve the real-time supervision accuracy of each step in the plastic film production process while improving the quality stability of plastic film products is a difficulty in the existing technology. For this reason, an intelligent supervision system for plastic film production is provided. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide an intelligent supervision system for plastic film production.

[0005] In order to achieve the above purpose, the present invention provides the following technical solutions:

[0006] An intelligent supervision system for plastic film production, including a cloud computing platform, which is communicatively connected to an original data acquisition module, a production monitoring network module, a plastic film production device, and a quality inspection module;

[0007] The original data acquisition module is used to collect production materials and historical production data of various plastic films through multiple sensors, and collect real-time production status data of each plastic film, and then send the historical production data and real-time production status data to the production monitoring network module;

[0008] The production monitoring network module is used to analyze the production materials and historical production data to generate corresponding normal range intervals and multiple groups of multiple linear regression equations, and then establish corresponding production pre-attributes and production structure links. Integrate the production pre-attributes and production structure links to obtain a production monitoring network, input the real-time production status data into the production monitoring network, and then generate corresponding production adjustment decisions;

[0009] The quality inspection module is used to obtain various data of the plastic film products from the plastic film production device to judge the quality of the corresponding plastic films.

[0010] Further, inside the plastic film production device, the plastic film production device is provided with a conveyor belt, an extruder, a raw material processing device, a condensation device, a stretching machine, a quality inspection device, and a decision execution unit;

[0011] The decision execution unit is used to adjust the parameters of each device of the plastic film production device according to the production adjustment decision, and after the plastic film production is completed, retrieve the various data of the plastic film finished product from the quality inspection device and send them to the quality inspection module.

[0012] Further, the process of collecting the historical production data and production materials includes:

[0013] The staff sets the types and quantities of plastic films to be produced in the raw data production module, and pours the required production raw materials and corresponding quantities into the raw material processing device in sequence.

[0014] After all the production raw materials are loaded, the raw data acquisition module retrieves the sensors in each plastic film production device to collect various change curves during the production process.

[0015] The raw data acquisition module integrates the various change curves in the plastic film production to generate historical production data, obtains the corresponding loading quantity according to the pressure change value at the end of the loading of each production raw material, and classifies each production raw material according to the image data, thereby obtaining the production materials of the corresponding type of plastic film.

[0016] Further, the process of generating the normal range interval according to the production materials and historical production data includes:

[0017] Match the various data in the production materials generated at different times with each other. If the ratio between the types of plastic films, the names of production raw materials, and the expected production quantity contained in the two is greater than or equal to 95%, it is determined that the two match, otherwise it is determined that the two do not match;

[0018] Establish multiple two-dimensional rectangular coordinate systems, map the respective change curves in the historical production data corresponding to the mutually matching production materials onto the two-dimensional rectangular coordinate systems, divide several time nodes on the two-dimensional rectangular coordinate systems, and then divide each change curve into the same number of curve points according to the time nodes;

[0019] Set the Euclidean distance threshold, and then sequentially obtain the Euclidean distances between each curve node and other curve nodes at the same time node, compare the Euclidean distances between the curve nodes with the Euclidean distance threshold. If the Euclidean distance between the curve nodes is greater than the Euclidean distance threshold, it is determined that there is no correlation between the corresponding two curve nodes. If the Euclidean distance between the curve nodes is less than or equal to the Euclidean distance threshold, mark the corresponding two curve nodes as associated curve nodes;

[0020] Select the curve with the most associated curve nodes as the reference curve node at the corresponding time node, and eliminate the curve nodes of the associated curve nodes that are not reference curve nodes at this time node;

[0021] Obtain the reference curve nodes at each time node, and integrate their associated curve nodes to obtain the normal range interval;

[0022] Repeat the above operations to obtain the normal range intervals for each type of corresponding production materials.

[0023] Furthermore, the process of generating a multiple linear regression equation based on the production materials and historical production data includes:

[0024] Establish multiple three-dimensional space coordinate systems, map the same normal range intervals of the same type of plastic film under different expected production quantities onto the same three-dimensional space coordinate system, and set a three-dimensional straight line starting from the origin and passing through the central positions of each normal range interval;

[0025] Mark several line segment intervals on the three-dimensional straight line to obtain the multiple linear regression equations of each normal range interval within each line segment interval.

[0026] Select Num spatial coordinates within the normal range interval and substitute them into the multiple linear regression equation of its adjacent line segment interval. If more than 0.9Num spatial coordinate points conform to the multiple linear regression equation of the adjacent line segment interval, it is determined that the normal range intervals of two adjacent line segment intervals are coherent; otherwise, it is determined that the normal range intervals of two adjacent line segment intervals are not coherent, where Num is a natural number greater than 50;

[0027] Substitute the spatial coordinate points at the connection of the normal range intervals within two adjacent line segment intervals that are coherent into the two multiple linear regression equations respectively, and then correct the various parameters in the two multiple linear regression equations. Obtain a new multiple linear regression equation based on the corrected various parameters, and merge the two adjacent line segment intervals that are coherent;

[0028] Repeat the above operations until there are no adjacent line segment intervals that can be merged, and then obtain several multiple linear regression equations and mark the corresponding parameter interval ranges, where the parameter interval ranges include time, curve parameters, and expected production quantity.

[0029] Furthermore, the process of generating production preconditions and production structure links based on the normal range interval and multiple groups of multiple linear regression equations includes:

[0030] Establishing the same number of production pre-attributes according to the parameter range intervals corresponding to each multivariate linear regression equation, wherein the production pre-attributes include the type of mulch film, the name and quantity of each production raw material, and the estimated production quantity;

[0031] At the same time, a production structure link is established according to the number of devices in the mulch film production device. The production structure link is provided with 4 link nodes, and then the corresponding multivariate linear regression equation is input into the link node, each link node is connected in sequence, and the production structure link is bound to the corresponding production pre-attribute, and the production structure links and production pre-attributes of each type of mulch film are integrated to obtain a production monitoring network.

[0032] Further, the process of monitoring the mulch film production according to the production monitoring network includes:

[0033] Match the pre-production data attributes in the production monitoring network according to the real-time status data, and obtain the corresponding production structure link according to the matching result;

[0034] Input each change curve and the estimated production quantity in the production status data subsequently sent by the original data acquisition module into the corresponding link node, set the time node to convert each change curve and the estimated production quantity into a number of spatial coordinate points, obtain the spatial distance between the real-time spatial coordinate point and the multivariate linear regression equation, and set the spatial distance threshold;

[0035] If the spatial distance between the real-time spatial coordinate point and the multivariate linear regression equation is greater than or equal to the spatial distance threshold, it is determined that the corresponding device in the mulch film production device has an operating abnormality, and a corresponding production adjustment decision is generated, otherwise no operation is performed;

[0036] The production monitoring network module sends the production adjustment decision to the decision execution unit in the mulch film production device, and then the decision execution unit generates an abnormal adjustment instruction according to the production adjustment decision and sends it to the corresponding device for execution.

[0037] Furthermore, when the conveyor belt sends the finished mulch film to the quality inspection device, the quality inspection device calls the laser sensors on both sides of the conveyor belt to send laser signals to the finished mulch film, and then generates a laser reflection signal spectrum according to the laser sensors on both sides, and calls the ultraviolet light sending device to randomly select N detection points on the finished mulch film, and then sends ultraviolet light beams to the detection points, and at the same time captures the detection images of the detection points through the camera, where N is a natural number greater than 0;

[0038] Then the quality detection module determines whether the finished mulch film is qualified according to the laser reflection signal spectrum and the detection image.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. The present invention establishes a corresponding production monitoring network by collecting historical production data of various types of plastic films under different quantities of production raw materials, and then inputs the real-time status data in the plastic film production process into the production monitoring network. According to the evaluation results of the production monitoring network, the production device is adjusted in real time, effectively improving the real-time supervision accuracy rate of each step in the plastic film production process.

[0041] 2. The present invention collects and analyzes historical production data under different quantities of raw materials, generates corresponding production pre-attributes and multiple linear regression equations according to the analysis results, and then conducts specific production supervision through the multiple linear regression equations, improving the quality stability of plastic film products to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention.

[0043] Figure 1 It is the schematic diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0045] As Figure 1 shown, an intelligent supervision system for plastic film production includes a cloud computing platform, and the cloud computing platform is communicatively connected to a raw data collection module, a production monitoring network module, a plastic film production device, and a quality detection module;

[0046] The raw data collection module is used to collect production materials and historical production data of various plastic films through a variety of sensors, as well as collect real-time production status data of each plastic film, and then send the historical production data and real-time production status data to the production monitoring network module;

[0047] The production monitoring network module is used to analyze production materials and historical production data to generate corresponding normal range intervals and multiple groups of multiple linear regression equations, and then establish corresponding production pre-attributes and production structure links. Integrate the production pre-attributes and production structure links to obtain a production monitoring network, input the real-time production status data into the production monitoring network, and then generate corresponding production adjustment decisions;

[0048] The mulch film production device is provided with a conveyor belt, an extruder, a raw material processing device, a condensing device, a stretching machine, a quality detection device and a decision execution unit;

[0049] The decision execution unit is used to adjust the parameters of each device of the ground film production device according to the production adjustment decision, and after the ground film production is completed, the quality inspection device is called to obtain various data of the finished ground film, and the data is sent to the quality inspection module;

[0050] The quality detection module is used to obtain various data of finished ground film from the ground film production device to determine the quality of the corresponding ground film.

[0051] Further, the working principle of the present invention is described below by way of examples:

[0052] The raw data acquisition module sets a variety of sensors on each device in the ground film production device, and then when the ground film production device produces various ground films, the sensors are used to collect real-time status data of each device and the ground film during the processing of each device, wherein the types of sensors include temperature sensors, laser sensors, cameras, etc.

[0053] In the mulch film production device, the raw material processing device, the extruder, the condensing device, the stretching machine and the quality inspection device are connected in sequence through a conveyor belt, wherein the raw material processing device is provided with a temperature sensor, a pressure sensor and a camera, the extruder is provided with a laser sensor and a pressure sensor, the condensing device is provided with a temperature sensor, the stretching machine is provided with a stretching robot arm and a pressure sensor, and the quality inspection device is provided at the conveying end, which has a built-in laser sensor, a camera and an ultraviolet emitting device.

[0054] The process of producing mulch film by the mulch film production device includes:

[0055] The staff sets the expected production type and quantity of mulch in the raw data production module, where the types of mulch include transparent mulch, black mulch, silver-gray mulch, etc.

[0056] According to the type and quantity of the selected mulch film, the staff will pour the required production raw materials and the corresponding quantities into the raw material processing device in sequence. It should be noted that in the process of pouring the production raw materials into the raw material processing device, the raw data acquisition module calls the camera and the pressure sensor to capture the image data of each production raw material and obtain the pressure change value at the end of each production raw material loading;

[0057] After all the production raw materials are loaded, the raw material processing device mixes, heats, and plasticizes each production raw material to obtain a secondary processed product. At the same time, the original data acquisition module retrieves the temperature sensor and pressure sensor to obtain the temperature value and pressure value changes in the raw material processing device, and generates corresponding temperature change curves and pressure change curves;

[0058] The raw material processing device transfers the secondary processed product to the extruder through a conveyor belt. Then, the extruder extrudes the secondary processed product into a film shape. It should be noted that during the process of the extruder extruding the secondary processed product into a film shape, the original data acquisition module retrieves the pressure sensor to obtain the working pressure change curve of the extruder during this process, and scans the secondary processed product in the shape of a film from the side through a laser sensor, thereby obtaining the thickness change curve of the secondary processed product;

[0059] The conveyor belt passes the secondary processed product in the shape of a film through the condensation device and the stretching machine in sequence. The condensation device cools the secondary processed product in the shape, and then the stretching machine stretches the secondary processed product into a finished plastic film through the robotic arm device. At the same time, the original data acquisition module retrieves the temperature sensor to obtain the temperature change curve of the finished plastic film, and obtains the stretching force change curve of the robotic arm device through the pressure sensor;

[0060] The quality inspection device sets multiple laser sensors at the port position of the conveyor belt, and a UV transmitter and a camera at the upper position. When the finished plastic film reaches the quality inspection device through the conveyor belt, the laser sensors at the port position of the conveyor belt collect the thickness change curve of the finished plastic film. At the same time, the UV transmitter at the upper position of the conveyor belt randomly selects several detection points on the finished plastic film, and then sends UV light beams to the detection points;

[0061] After the plastic film production is completed, the plastic film production device sends a production prompt to the original data acquisition module.

[0062] Furthermore, the original data acquisition module integrates the various change curves in the plastic film production to generate historical production data, obtains the corresponding loading quantity according to the pressure change value at the end of loading each production raw material, and classifies each production raw material according to the image data, thereby obtaining the production data of the corresponding type of plastic film;

[0063] The original data acquisition module binds each historical production data with the corresponding production data and sends it to the production monitoring network module;

[0064] Then, the production monitoring network module establishes the production pre - attributes and production structure links of the corresponding type of plastic film according to the production data and historical production data. The specific process includes:

[0065] Match the data in the production materials generated at different times. If the ratio of the types of mulch films, the names of production raw materials and the estimated production quantities contained in the two is greater than or equal to 95%, then the two are considered to be matched; otherwise, the two are considered to be mismatched.

[0066] Establish multiple two-dimensional rectangular coordinate systems, map each change curve in the historical production data corresponding to the matching production materials onto the two-dimensional rectangular coordinate system, divide a number of time nodes on the two-dimensional rectangular coordinate system, and then divide each change curve into the same number of curve points according to the time nodes;

[0067] Set the Euclidean distance threshold, and then obtain the Euclidean distance between each curve node and other curve nodes at the same time node in turn, compare the Euclidean distance between each curve node with the Euclidean distance threshold, if the Euclidean distance between the curve nodes is greater than the Euclidean distance threshold, it is judged that there is no correlation between the corresponding two curve nodes, if the Euclidean distance between the curve nodes is less than or equal to the Euclidean distance threshold, then the corresponding two curve nodes are marked as mutually correlated curve nodes;

[0068] Select the curve with the most associated curve nodes as the reference curve node at the corresponding time node, and remove the curve nodes of the associated curve nodes that are not reference curve nodes at the time node;

[0069] Obtain reference curve nodes at each time node, and integrate the associated curve nodes to obtain the normal range interval;

[0070] Repeat the above operation to obtain the normal range interval of each type of corresponding production materials;

[0071] Establish multiple three-dimensional space coordinate systems, map the same normal range interval of the same type of mulch film under different expected production quantities onto the same three-dimensional space coordinate system, and set a three-dimensional straight line from the origin and passing through the center position of each normal range interval;

[0072] Marking several line segment intervals of different sizes on the three-dimensional straight line, and then obtaining the multiple linear regression equations of each normal range interval in each line segment interval, wherein the multiple linear regression equation can be expressed as f=α+φ1x+φ2y+φ3z, wherein α, φ1, φ2, φ3 represent correction constants and relationship constants respectively, and x, y, z represent time, curve parameters and expected production quantity respectively, and the curve parameters can be temperature values ​​and pressure values;

[0073] Select Num spatial coordinates within the normal range interval and substitute them into the multiple linear regression equation of the adjacent line segment interval. If more than 0.9Num spatial coordinate points conform to the multiple linear regression equation of the adjacent line segment interval, it is determined that there is coherence in the normal range intervals of two adjacent line segment intervals; otherwise, it is determined that there is no coherence in the normal range intervals of two adjacent line segment intervals, where Num is a natural number greater than 50;

[0074] Substitute the spatial coordinate points at the connection of the normal range intervals within two adjacent line segment intervals with coherence into the two multiple linear regression equations respectively, and then correct α, φ1, φ2, and φ3 in the two multiple linear regression equations. Obtain a new multiple linear regression equation based on the corrected α, φ1, φ2, and φ3, and merge the two adjacent line segment intervals with coherence;

[0075] Repeat the above operations until there are no adjacent line segment intervals that can be merged, and then obtain several multiple linear regression equations and mark the corresponding parameter interval ranges, where the parameter interval ranges include time, curve parameters, and the expected production quantity;

[0076] Establish the same number of production precondition attributes according to the parameter range intervals corresponding to each multiple linear regression equation. The production precondition attributes include the type of plastic film, the names and quantities of various production raw materials, and the expected production quantity;

[0077] At the same time, establish a production structure link according to the number of devices in the plastic film production device. The production structure link has 4 link nodes. Then input the corresponding multiple linear regression equations to the link nodes, connect each link node in sequence, and bind the production structure link to the corresponding production precondition attributes;

[0078] Integrate the production structure links and production precondition attributes of each type of plastic film to obtain a production monitoring network.

[0079] Furthermore, the raw data acquisition module acquires the real-time status data of each device in the plastic film production device for producing plastic film and sends the real-time status data to the production monitoring network module;

[0080] The production monitoring network module matches the pre-production data attributes in the production monitoring network according to the real-time status data, and obtains the corresponding production structure link according to the matching result;

[0081] Input each change curve and the expected production quantity in the production status data subsequently sent by the raw data acquisition module into the corresponding link nodes. Set time nodes to convert each change curve and the expected production quantity into several spatial coordinate points, obtain the spatial distance between the real-time spatial coordinate points and the multiple linear regression equation, and set a spatial distance threshold;

[0082] If the spatial distance between the real-time spatial coordinate point and the multivariate linear regression equation is greater than or equal to the spatial distance threshold, it is determined that the corresponding device in the mulch film production device has an operating abnormality, and a corresponding production adjustment decision is generated, otherwise no operation is performed;

[0083] It should be noted that the production adjustment decision includes the name of the device with abnormal operation and abnormal data.

[0084] The production monitoring network module sends the production adjustment decision to the decision execution unit in the mulch film production device, and then the decision execution unit generates an abnormal adjustment instruction according to the production adjustment decision and sends it to the corresponding device for execution.

[0085] Furthermore, when the conveyor belt sends the finished ground film to the quality inspection device, the quality inspection device calls the laser sensors on both sides of the conveyor belt to send laser signals to the finished ground film, and then generates a laser reflection signal spectrum according to the laser sensors on both sides;

[0086] At the same time, the quality inspection device calls the ultraviolet light sending device to randomly select N inspection points on the finished mulch film, and then sends ultraviolet light beams to the inspection points, and at the same time, the inspection images of the inspection points are captured by the camera, where N is a natural number greater than 0;

[0087] The decision execution unit sends the laser reflection signal spectrum and the detection image to the quality detection module. The quality detection module obtains the actual thickness of the ground film product according to the laser reflection signal spectrum generated by the laser sensors on both sides, and obtains the corresponding standard thickness according to the type of ground film. If the ratio between the actual thickness and the standard thickness is between (0.99, 1), the thickness of the corresponding ground film product is judged to be qualified, otherwise it is judged to be unqualified.

[0088] At the same time, the quality inspection module grayscales the inspection image and obtains the pixel value of each grayscale pixel in the inspection image, and overlaps and maps the inspection point positions in different inspection images. Since the various materials in the ground film are evenly distributed, the light absorption of the ground film at the inspection point position is the same under the ultraviolet irradiation of the same frequency;

[0089] Furthermore, if the ratios of the pixel values ​​of the grayscale pixels at the detection point positions in different detection images are all between (0.95, 1), the corresponding finished mulch film is judged to be qualified, otherwise it is judged to be unqualified.

[0090] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent supervision system for plastic film production, including a cloud computing platform, characterized in that, The cloud computing platform is communicatively connected to a raw data acquisition module, a production monitoring network module, a plastic film production device, and a quality inspection module; The raw data acquisition module is used to collect production materials and historical production data of various plastic films through a variety of sensors, and collect real-time production status data of each plastic film, and then send the historical production data and real-time production status data to the production monitoring network module; Inside the plastic film production device, the plastic film production device is provided with a conveyor belt, an extruder, a raw material processing device, a condensation device, a stretching machine, a quality inspection device, and a decision execution unit; The decision execution unit is used to adjust the parameters of each device of the plastic film production device according to the production adjustment decision, and retrieve the data of the plastic film finished product from the quality inspection device after the plastic film production is completed, and send it to the quality inspection module; The process of collecting the historical production data and production materials includes: The staff sets the types and quantities of plastic films to be produced in the raw data production module, and pours the required production raw materials and corresponding quantities into the raw material processing device in sequence; When all the production raw materials are loaded, the raw data acquisition module retrieves the sensors in each plastic film production device to collect various change curves during the production process, the pressure change value at the end of the production raw material loading, and image data; The raw data acquisition module integrates the various change curves in the plastic film production to generate historical production data, obtains the corresponding loading quantity according to the pressure change value at the end of each production raw material loading, and classifies each production raw material according to the image data, so as to obtain the production materials of the corresponding type of plastic film; The production monitoring network module is used to analyze the production materials and historical production data, and generate corresponding normal range intervals and multiple groups of multiple linear regression equations, and then establish production pre-set attributes and production structure links, integrate the production pre-set attributes and production structure links to obtain a production monitoring network, input the real-time production status data into the production monitoring network, and then generate corresponding production adjustment decisions; The process of generating the normal range interval according to the production materials and historical production data includes: Match the data in the production materials generated at different times with each other. If the ratio between the types of plastic films, the names of production raw materials, and the expected production quantity contained in the two is greater than or equal to 95%, it is judged that the two match, otherwise it is judged that the two do not match; Establish multiple two-dimensional rectangular coordinate systems, map the respective change curves in the historical production data corresponding to the production materials that match each other onto the two-dimensional rectangular coordinate systems, divide several time nodes on the two-dimensional rectangular coordinate systems, and then divide each change curve into the same number of curve points according to the time nodes; Set the Euclidean distance threshold, and then obtain the Euclidean distance between each curve node and other curve nodes at the same time node in turn, compare the Euclidean distance between each curve node with the Euclidean distance threshold, if the Euclidean distance between the curve nodes is greater than the Euclidean distance threshold, it is judged that there is no correlation between the corresponding two curve nodes, if the Euclidean distance between the curve nodes is less than or equal to the Euclidean distance threshold, then the corresponding two curve nodes are marked as mutually correlated curve nodes; Select the curve with the most associated curve nodes as the reference curve node at the corresponding time node, and remove the curve nodes of the associated curve nodes that are not reference curve nodes at the time node; Obtain reference curve nodes at each time node, and integrate the associated curve nodes to obtain the normal range interval; The process of generating a multiple linear regression equation based on the production data and historical production data includes: Establish multiple three-dimensional space coordinate systems, map the same normal range interval of the same type of mulch film under different expected production quantities onto the same three-dimensional space coordinate system, and set a three-dimensional straight line from the origin and passing through the center position of each normal range interval; Mark several line segment intervals of different sizes on the three-dimensional straight line, and then obtain the multivariate linear regression equations of each normal range interval within each line segment interval. Select Num spatial coordinates within the normal range interval and substitute them into the multivariate linear regression equation of its adjacent line segment interval. If more than 0.9Num spatial coordinate points meet the multivariate linear regression equation of the adjacent line segment interval, it is judged that the normal range intervals of the two adjacent line segment intervals are continuous. Otherwise, it is judged that the normal range intervals of the two adjacent line segment intervals are not continuous, where Num is a natural number greater than 50. Substitute the spatial coordinate points of the connection of the normal range intervals in the two adjacent line segment intervals with coherence into the two multiple linear regression equations respectively, and then correct the parameters in the two multiple linear regression equations, obtain new multiple linear regression equations according to the corrected parameters, and merge the two adjacent line segment intervals with coherence; Repeat the above adjacent line segment interval merging operation until there are no adjacent line segment intervals that can be merged, and then obtain several multivariate linear regression equations and mark the corresponding parameter interval ranges, where the parameter interval ranges include time, curve parameters and expected production quantity; The process of generating production pre-attributes and production structure links based on normal range intervals and multiple sets of multivariate linear regression equations includes: Establishing the same number of production pre-attributes according to the parameter range intervals corresponding to each multivariate linear regression equation, wherein the production pre-attributes include the type of mulch film, the name and quantity of each production raw material, and the estimated production quantity; At the same time, a production structure link is established according to the number of devices in the mulch film production device, and the production structure link is provided with 4 link nodes, and then the multivariate linear regression equation is input into the corresponding link node, each link node is connected in sequence, and the production structure link is bound to the corresponding production pre-attribute, and the production structure link and production pre-attribute of each type of mulch film are integrated to obtain a production monitoring network; The quality detection module is used to obtain various data of finished mulch films from the mulch film production device to determine the quality of the corresponding mulch films; When the conveyor belt sends the finished mulch film to the quality inspection device, the quality inspection device calls the laser sensors on both sides of the conveyor belt to send laser signals to the finished mulch film, and then generates a laser reflection signal spectrum according to the laser sensors on both sides, and calls the ultraviolet sending device to randomly select N detection points on the finished mulch film, and then sends ultraviolet beams to the detection points, and at the same time uses the camera to capture the detection image of the detection point, where N is a natural number greater than 0; Then the quality detection module determines whether the finished mulch film is qualified according to the laser reflection signal spectrum and the detection image.

2. The intelligent supervision system for plastic film production according to claim 1, characterized in that The process of monitoring mulch film production according to the production monitoring network includes: According to the real-time status data, the pre-production data attributes and production structure links in the production monitoring network are matched, and each change curve and the estimated production quantity in the production status data subsequently sent by the original data acquisition module are input into the corresponding link node. The time node is set to convert each change curve and the estimated production quantity into a number of spatial coordinate points, and the spatial distance between the real-time spatial coordinate point and the multivariate linear regression equation is obtained, and the spatial distance threshold is set; If the spatial distance between the real-time spatial coordinate point and the multivariate linear regression equation is greater than or equal to the spatial distance threshold, it is determined that the corresponding device in the mulch film production device has an operating abnormality, and a corresponding production adjustment decision is generated, otherwise no operation is performed; The production monitoring network module sends the production adjustment decision to the decision execution unit in the mulch film production device, and then the decision execution unit generates an abnormal adjustment instruction according to the production adjustment decision and sends it to the corresponding device for execution.

Images