A data interaction management system for the entire manufacturing cycle of smart factory products

Abstract

The present invention relates to the technical field of smart factories, and discloses a full-cycle data interaction management system for product manufacturing in a smart factory, the system comprising a raw material procurement management module, a raw material storage management module, a product production management module, a product quality inspection management module, a product storage management module, a product after-sales management module, and a product management cloud platform; the raw material procurement management module, the raw material storage management module, the product production management module, the product quality inspection management module, the product storage management module, and the product after-sales management module collect corresponding module feature data for identification and matching, and the product management cloud platform monitors and manages the product manufacturing process in an orderly manner through the matching results, realizing full-cycle quality monitoring of raw material procurement, raw material storage, product production, product quality inspection, product storage, and product after-sales in the product manufacturing process under data interaction, realizing autonomous monitoring of the entire product life cycle, and reducing product quality defects.

Description

Technical Field

[0001] The present invention relates to the technical field of smart factories, and specifically to a data interaction management system for the entire manufacturing cycle of smart factory products. Background Art

[0002] Smart factories are a new stage in the development of modern factory informatization. Based on digital factories, they use the Internet of Things technology and equipment monitoring technology to strengthen information management and services, and build an efficient, energy-saving, green, environmentally friendly, and comfortable humanized factory. However, existing smart factories mainly focus on the collection of production process information and automatic control of the product production process for product manufacturing. The full-cycle quality management of raw material procurement, raw material storage, product production, product quality inspection, product storage, and product after-sales stages involved in product manufacturing mainly relies on product information collection. The various links of the product full-cycle quality management are too scattered and simple, lacking independent supervision, and unable to conduct online monitoring to ensure that product manufacturing meets manufacturing standards, which increases the probability of product quality defects.

[0003] China Patent Publication No. CN112132526A discloses a smart factory management system, in which the warehouse management module includes an inbound and outbound management unit, a replenishment management unit and a data statistics unit. The inbound and outbound management unit is used to record the name, quantity and inbound and outbound time of the products in and out of the warehouse, and the replenishment management unit is used to promptly remind and replenish products with insufficient inventory; the workshop management module includes a planning unit, an organization and command unit, a supervision unit and a production unit. The planning unit is used to formulate production plans for all products in the workshop, the organization and command unit is used to arrange the personnel required for production, the supervision unit is used to supervise the work status of the workshop personnel, and the production unit is used to process, manufacture and assemble the products; the quality management module includes The inspection unit and the defective product processing unit. The inspection unit is used to simulate the products produced according to the usage scenarios and divide the products into qualified and unqualified parts. The defective product processing unit is used to inspect, repair and scrap the unqualified products produced; the pre-sales / after-sales management module includes a user demand analysis unit, a product consulting service unit and an on-site repair and debugging unit. Although the warehouse management module, workshop management module, quality management module and pre-sales / after-sales management module realize unified management and control of the storage, production, quality inspection and after-sales stages in product manufacturing, there is a lack of effective and independent supervision of the entire cycle of product raw materials and their storage, product production, quality inspection, storage and after-sales, and it is impossible to monitor each link of product manufacturing online to ensure that the product manufacturing standards are met. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In order to solve the above-mentioned problem that the full-cycle quality management of raw material procurement, raw material storage, product production, product quality inspection, product storage, and product after-sales stages involved in product manufacturing mainly depends on product information collection, and each link of the product full-cycle quality management is too scattered and simple, lacks independent supervision, and cannot be monitored online to ensure that product manufacturing meets manufacturing standards, which increases the probability of product quality defects. The purpose of achieving the above-mentioned full-cycle quality monitoring of product manufacturing, ensuring that the production links meet manufacturing standards, and reducing product quality defects is achieved.

[0006] (2) Technical solution

[0007] The present invention is achieved through the following technical solutions: a smart factory product manufacturing full-cycle data interaction management system, the system includes a raw material procurement management module, a raw material storage management module, a product production management module, a product quality inspection management module, a product storage management module, a product after-sales management module, and a product management cloud platform;

[0008] The raw material procurement management module includes a raw material feature acquisition unit and a raw material feature tracking and judgment unit. The raw material feature acquisition unit acquires the purchased raw material feature data, and the raw material feature tracking and judgment unit uses the acquired purchased raw material feature data to match and judge with the purchased raw material standard feature data; the raw material storage management module includes a raw material warehouse feature acquisition unit and a raw material warehouse storage identification unit. The raw material warehouse feature acquisition unit collects the raw material warehouse storage feature data, and the raw material warehouse storage identification unit uses the raw material warehouse storage feature data to compare and identify with the raw material warehouse storage standard feature data;

[0009] The product production management module includes a production line static inspection unit, a production line dynamic inspection unit, and a production line human-machine cooperation inspection unit. The production line static inspection unit collects the static equipment installation and placement status image data of the product processing production line and the production line standard equipment installation and placement status image data for inspection. The production line dynamic inspection unit collects the dynamic operation of the equipment and the linkage status image data of the product processing production line and verifies it with the dynamic operation and linkage status image data of the standard equipment of the production line. The production line human-machine cooperation inspection unit collects the human-machine cooperation equipment and operator collaborative operation status image data of the product processing production line and verifies it with the production line standard equipment and operator collaborative operation status image data;

[0010] The product quality inspection management module includes an offline product quality feature collection unit and an offline quality feature comparison unit. The offline product quality feature collection unit obtains offline product quality feature data, and the offline quality feature comparison unit uses the offline product quality feature data to match and judge with the offline product standard quality feature data; the product storage management module includes a product warehouse feature collection unit and a product warehouse storage identification unit. The product warehouse feature collection unit obtains product warehouse storage feature data, and the product warehouse storage identification unit uses the product warehouse storage feature data to compare and analyze with the product warehouse storage standard feature data;

[0011] The product after-sales management module includes a product failure feedback unit and a product failure repair recording unit. The product failure feedback unit collects after-sales product failure feedback data online, and the product failure repair recording unit records the cause data of product failures found after product repair;

[0012] The product management cloud platform collaboratively manages the product manufacturing full-cycle management module through data interaction and monitors the orderly and normal operation of the product manufacturing execution module.

[0013] Preferably, the product management cloud platform is connected to the raw material procurement management module, the raw material storage management module, the product production management module, the product quality inspection management module, the product storage management module, and the product after-sales management module via Internet of Things communication, and the purchased raw material characteristic data includes one or more characteristic data of raw material origin, color, shape, density, melting point, boiling point, solubility, acidity and alkalinity, oxidizability, reducibility, stability, thickener, antioxidant, flame retardant, and packaging integrity, hydrophilicity, particle shape, porosity, ore particle size, and particle size;

[0014] The raw material feature tracking and judgment unit uses the obtained purchased raw material feature data and the purchased raw material standard feature data to perform matching and judgment operation steps as follows:

[0015] S11. Establish a data set of purchased raw material characteristics. in Represents different types of purchased raw material feature data, x represents the purchased raw material feature type number, x = 1, 2, 3, ... n, k represents the number of purchased raw material feature data collected, k = 1, 2, 3, ... m;

[0016] S12. Calculate the mean value of the characteristic data of purchased raw materials.

[0017] S13. Average value of purchased raw material characteristic data The range of standard characteristic data of purchased raw materials [S ax Q ax ] Belonging relationship judgment, where S axIndicates the minimum value of the standard characteristic data range of the purchased raw materials, Q ax Indicates the maximum value of the purchase material standard characteristic data range, x represents the purchase material characteristic type number, x = 1, 2, 3, ... n;

[0018] when If the characteristic data of the purchased raw materials meet the standard characteristic data requirements of the purchased raw materials, the product management cloud platform will prompt the purchased raw materials to enter the storage or processing stage;

[0019] when If it indicates that the characteristic data of the purchased raw materials do not meet the standard characteristic requirements of the purchased raw materials, the product management cloud platform will prompt that the purchased raw materials do not meet the standards and cannot enter the storage and processing stages.

[0020] Preferably, the raw material warehouse storage characteristic data includes one or more of temperature, humidity, air dust content, raw material partition spacing, lighting intensity, ventilation speed, number of fire points, number of monitoring points, and number of access channels;

[0021] The steps for the raw material warehouse storage identification unit to compare and identify the raw material warehouse storage feature data with the raw material warehouse storage standard feature data are as follows:

[0022] S21. Establish a raw material warehouse storage feature data set, in Represents storage characteristic data of different types of raw material warehouses, x represents the type number of the raw material warehouse storage characteristic, x = 1, 2, 3, ... n, k represents the number of raw material warehouse storage characteristic data collected, k = 1, 2, 3, ... m;

[0023] S22. Calculate the mean value of the raw material warehouse storage characteristic data set.

[0024] S23, mean value of storage characteristic data of raw material warehouse The range of standard characteristic data of raw material warehouse storage [S bx Q bx ] Belonging relationship judgment, where S bx Indicates the minimum value of the raw material warehouse storage standard characteristic data range, Q bx Indicates the maximum value of the raw material warehouse storage standard characteristic data range, x represents the purchased raw material characteristic type number, x = 1, 2, 3, ... n;

[0025] when If the raw material warehouse storage characteristic data meets the requirements of the raw material warehouse storage standard characteristic data, the product management cloud platform will prompt the purchased raw materials to enter the warehouse storage stage;

[0026] when This indicates that the raw material warehouse storage characteristic data does not meet the raw material warehouse storage standard characteristic data requirements. The product management cloud platform will prompt that the warehouse storage characteristics do not meet the standards and cannot enter the warehouse storage stage.

[0027] Preferably, the production line static inspection unit collects the static equipment installation and placement state image data of the product processing production line and the production line standard equipment installation and placement state image data for inspection operation steps as follows:

[0028] S31. Collect static image data of equipment installation and placement status on the production line, and establish an image data set of equipment installation and placement status image data in represents image data of the installation and placement status of equipment at different positions, x represents the number of image data of the installation and placement status of equipment at different positions, x=1,2,3,…n, k represents the number of image data collected of the installation and placement status of equipment at different positions, k=1,2,3,…m;

[0029] S32, using convolutional neural network to identify the image data of the equipment installation and placement status,

[0030] Input image preprocessing: collect the image data of the equipment installation and placement status, The image data is scaled to a fixed size and the image is normalized and preprocessed;

[0031] Convolution layer processing: Convolution processing is applied to the pre-processed image data of the equipment installation and placement status. The convolution formula is:

[0032] Where F(ω) represents the convolution formula for performing convolution processing on the pre-processed equipment installation and placement status image data, f represents the integral function of the convolution processing formula F(ω), ω represents the adjustment parameter, i represents the object of the pre-processed equipment installation and placement status image data for convolution processing, e is a constant, and t is the input variable of the convolution formula, i.e., the pre-processed equipment installation and placement status image data at different positions, After extracting the convolution, a new feature map of the installation and placement status of equipment in different locations is formed;

[0033] Activation function layer processing: The ReLU activation function is used to perform nonlinear transformation on the feature map of the installation and placement status of devices in different positions output by the convolution layer. The ReLU activation function is:

[0034] Where λ∈(0,1), at this time x is the feature map data of the installation status of the equipment in different positions after convolution,

[0035] Pooling layer processing: downsampling the feature map data of the equipment installation and placement status at different locations after linear transformation;

[0036] Fully connected layer processing: Flatten the downsampled feature map data of the device installation and placement status at different locations into vectors and connect them to the fully connected layer for classification or regression.

[0037] Output results: Identify and output the image data of the equipment installation and placement status at different locations and the image data of the standard equipment installation and placement status of the production line output by the fully connected layer;

[0038] S33. When the output image data of the installation and placement status of the equipment in different positions is recognized, the product management cloud platform controls the production line to enter the dynamic inspection stage;

[0039] When the output image data of the installation and placement status of equipment in different positions cannot be recognized, the product management cloud platform controls the production line and cannot enter the dynamic inspection stage.

[0040] Preferably, the production line dynamic inspection unit collects dynamic equipment operation and inter-equipment linkage status image data of the product processing production line and inspects it with the dynamic equipment operation and inter-equipment linkage status image data of the production line standard equipment as follows:

[0041] S41. Collect dynamic operation of equipment and linkage status image data of production line, and establish dynamic operation and linkage status image data set of equipment in Indicates the image data of dynamic operation and linkage status of devices at different locations, x represents the number of the image data of dynamic operation and linkage status of devices at different locations, x = 1, 2, 3, ... n, k represents the number of image data collected of dynamic operation and linkage status of devices at different locations, k = 1, 2, 3, ... m;

[0042] S42, using the convolutional neural network image recognition method in S32, through the steps of input image preprocessing, convolution layer processing, activation function layer processing, pooling layer processing, full connection layer processing, and output result processing, to identify and output the image data of the dynamic operation of the equipment and the linkage status between the equipment and the dynamic operation of the standard equipment of the production line and the linkage status between the equipment.

[0043] S43. When the output image data of the dynamic operation of devices at different locations and the linkage status between devices are recognized, the product management cloud platform controls the production line to enter the human-machine cooperation inspection stage;

[0044] When the output image data of the dynamic operation of equipment in different positions and the linkage status between equipment cannot be recognized, the production line controlled by the product management cloud platform cannot enter the human-machine cooperation inspection stage.

[0045] Preferably, the production line human-machine cooperation inspection unit collects image data of the human-machine cooperation equipment and operator collaborative operation status of the product processing production line and inspects it with the image data of the collaborative operation status of the production line standard equipment and operator as follows:

[0046] S51. Collect image data of equipment and operator collaborative operation status of production line human-machine cooperation, and establish image data set of equipment and operator collaborative operation status in represents image data of collaborative operation status of equipment and operators at different positions, x represents the number of image data of collaborative operation status of equipment and operators at different positions, x=1, 2, 3, ... n, k represents the number of image data of collaborative operation status of equipment and operators at different positions, k=1, 2, 3, ... m;

[0047] S52, using the convolutional neural network image recognition method in S32 to perform input image preprocessing, convolution layer processing, activation function layer processing, pooling layer processing, fully connected layer processing, and output result processing steps to identify and output the image data of the collaborative operation status of the equipment and operator and the image data of the collaborative operation status of the standard equipment and operator of the production line;

[0048] S53: When the output image data of the dynamic operation of the equipment at different positions and the linkage status between the equipment are recognized, the product management cloud platform controls the production line to enter the product processing stage.

[0049] When the output image data of the dynamic operation of equipment in different positions and the linkage status between equipment cannot be recognized, the production line controlled by the product management cloud platform cannot enter the product processing stage.

[0050] Preferably, the quality characteristic data of the off-line products include one or more of appearance inspection, size, weight, hardness, stretching, compression, tensile strength, elongation at break, chemical composition, detection of harmful substances, service life of material components, product stability, product reliability, and durability;

[0051] The steps for the off-line quality feature comparison unit to perform a match judgment using the off-line product quality feature data and the off-line product standard quality feature data are as follows:

[0052] S61. Establish a data set of quality characteristics of off-line products. in Represents the quality characteristic data of different types of off-line products, x represents the data type number of the off-line product quality characteristic data, x = 1, 2, 3, ... n, k represents the number of off-line product quality characteristic data collected, k = 1, 2, 3, ... m;

[0053] S62, solve the mean value set of the quality characteristic data of the off-line products,

[0054] S63, average value of quality characteristic data of off-line products The range of the standard quality characteristic data of the offline products [S hx Q hx ] Belonging relationship judgment, where S hx Indicates the minimum value of the standard quality characteristic data range of the offline product, Q hx Indicates the maximum value of the standard quality characteristic data range of the offline product, x represents the data type number of the offline product quality characteristic data, x = 1, 2, 3, ... n;

[0055] when If the quality characteristic data of the off-line products meet the standard quality characteristic data requirements of the off-line products, the product management cloud platform will prompt that the processed products can be taken off-line for sale or put into the product warehouse storage stage;

[0056] when This indicates that the quality characteristic data of the off-line products do not meet the standard quality characteristic data requirements of the off-line products. The product management cloud platform will prompt that the processed products do not meet the quality characteristics and cannot be taken off-line for sale or enter the product warehouse storage stage.

[0057] Preferably, the product warehouse storage characteristic data includes one or more of product stacking height, number of placement partitions, temperature, humidity, air dust content, lighting intensity, number of fire points, number of monitoring points, and number of access channels;

[0058] The steps for the product warehouse storage identification unit to compare and analyze the product warehouse storage feature data with the product warehouse storage standard feature data are as follows:

[0059] S71. Establish product warehouse storage feature data set, in Represents the storage characteristic data of different types of product warehouses, x represents the data type number of the product warehouse storage characteristic data, x = 1, 2, 3, ... n, k represents the number of product warehouse storage characteristic data collected, k = 1, 2, 3, ... m;

[0060] S72. Calculate the mean value of the product warehouse storage characteristic data set.

[0061] S73, mean value of product warehouse storage characteristic data The range of product warehouse storage standard characteristic data [S gx Q gx ] Belonging relationship judgment, where S gx Indicates the minimum value of the product warehouse storage standard characteristic data range, Q gxIndicates the maximum value of the product warehouse storage standard feature data range interval, x represents the product warehouse storage feature data type number, x = 1, 2, 3, ... n;

[0062] when If the product warehouse storage characteristic data meets the requirements of the product warehouse storage standard characteristic data, the product management cloud platform will prompt that the offline product can enter the product warehouse storage stage;

[0063] when This indicates that the product warehouse storage feature data does not meet the product warehouse storage standard feature data requirements. The product management cloud platform will prompt that the product warehouse features do not meet the standards and the offline products cannot enter the product warehouse storage stage.

[0064] Preferably, the steps for the product failure feedback unit to collect after-sales product failure feedback data online and the product failure repair recording unit to record product failure cause data found after product repair are as follows:

[0065] S81. Collect after-sales product failure feedback data H online through the product failure feedback system, and record online product failure cause data H' discovered after product repair.

[0066] A method for operating a data interactive management system for the entire manufacturing cycle of a smart factory product, the method comprising the following steps:

[0067] Step 1: During the raw material procurement management phase, the characteristic data of different types of purchased raw materials are obtained and numbered to create a data set. The characteristic data of different purchased raw materials are averaged and matched with the standard characteristic data of purchased raw materials. If the match is successful, the purchased raw materials enter the warehousing or processing process;

[0068] Step 2: Raw material storage management: Collect the storage characteristic data of different types of raw material warehouses, number them and create a data set. Then, average the storage characteristic data of different raw material warehouses and compare and identify them with the standard storage characteristic data of raw material warehouses. If the comparison and identification is successful, the purchased raw materials will enter the warehousing process;

[0069] Step 3: During the product production management phase, static image data of equipment installation and placement status, dynamic image data of equipment dynamic operation and linkage status between equipment, and image data of human-machine coordinated equipment and operator collaborative operation status are collected and numbered to establish a data set. A convolutional neural network is used to identify and verify the image data of equipment installation and placement status with the image data of standard equipment installation and placement status of the production line, the image data of equipment dynamic operation and linkage status between equipment with the image data of standard equipment dynamic operation and linkage status between equipment of the production line, and the image data of collaborative operation status of equipment and operators with the image data of collaborative operation status of standard equipment and operators of the production line. If the identification and verification is successful, the next execution process is entered;

[0070] Step 4: Product quality inspection and management phase: obtain the quality characteristic data of the off-line products. Obtain the quality characteristic data of different types of off-line products, number them and create a data set. The quality characteristic data of the off-line products are averaged and matched with the standard quality characteristic data of the off-line products. If the match is successful, the product is off-line and enters the sales or warehousing process.

[0071] Step 5: Product storage management phase: collect product warehouse storage feature data. Obtain the warehouse storage feature data of different types of products, number them and create a data set. Then, average the product warehouse storage feature data and compare it with the product warehouse storage standard feature data. If the comparison is successful, the product will be taken off the production line and enter the storage process.

[0072] Step 6: During the product after-sales management phase, collect after-sales product failure feedback data online and record the cause of product failures discovered after product repair.

[0073] (3) Beneficial effects

[0074] The present invention provides a data interaction management system for the entire manufacturing cycle of smart factories. It has the following beneficial effects:

[0075] 1. Through the raw material procurement management module, raw material storage management module, product production management module, product quality inspection management module, product storage management module, and product after-sales management module, the corresponding module feature data is collected for identification and matching. Through the matching results, the product management cloud platform monitors and manages the product manufacturing process in an orderly manner, and realizes the full-cycle quality monitoring of raw material procurement, raw material storage, product production, product quality inspection, product storage, and product after-sales in the product manufacturing process under data interaction, realizes autonomous monitoring of the entire product life cycle, and reduces product quality defects.

[0076] 2. The raw material procurement management module classifies and establishes sets of purchased raw material feature data, and performs average matching processing on different purchased raw material feature data; the raw material storage management module classifies and establishes sets of raw material warehouse storage feature data, and performs average identification processing on different raw material warehouse storage feature data; the product production management module classifies and establishes sets of static equipment installation and placement status image data, dynamic equipment dynamic operation and equipment linkage status image data, human-machine cooperation equipment and operator collaborative operation status image data, and uses convolutional neural networks for identification and processing, thereby realizing autonomous monitoring of raw material procurement, raw material warehousing, and production line production links, ensuring that the production links meet manufacturing standards and reduce product quality defects.

[0077] 3. Through the product quality inspection management module, the quality characteristic data of the offline products are classified and established into sets, and the quality characteristic data of different offline products are averaged and matched; the product storage management module classifies the product warehouse storage characteristic data and establishes sets, and the storage characteristic data of different product warehouses are averaged and matched; the product after-sales management module collects the after-sales product failure feedback data online and records the cause data of product failures discovered after product repairs, thereby realizing autonomous monitoring and timely feedback of product quality inspection, product warehousing, and product after-sales links, improving the autonomous monitoring and execution of production standards in the product manufacturing process, and reducing the probability of product defective rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 This is a functional module diagram of the smart factory product manufacturing full-cycle data interaction management system provided by the present invention;

[0079] Figure 2 for Figure 1 The figure shows the operation flow chart of raw material procurement, raw material storage, product production, product quality inspection, and product storage in the full-cycle data interaction management system of smart factory product manufacturing. DETAILED DESCRIPTION

[0080] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0081] The following is an example of the smart factory product manufacturing full cycle data interaction management system:

[0082] See also Figure 1-Figure 2, a smart factory product manufacturing full-cycle data interactive management system, the system includes raw material procurement management module, raw material storage management module, product production management module, product quality inspection management module, product storage management module, product after-sales management module, and product management cloud platform;

[0083] The raw material procurement management module includes a raw material feature acquisition unit and a raw material feature tracking and judgment unit. The raw material feature acquisition unit acquires the purchased raw material feature data, and the raw material feature tracking and judgment unit uses the acquired purchased raw material feature data to match and judge with the purchased raw material standard feature data; the raw material storage management module includes a raw material warehouse feature acquisition unit and a raw material warehouse storage identification unit. The raw material warehouse feature acquisition unit collects the raw material warehouse storage feature data, and the raw material warehouse storage identification unit uses the raw material warehouse storage feature data to compare and identify with the raw material warehouse standard feature data;

[0084] The product production management module includes a production line static inspection unit, a production line dynamic inspection unit, and a production line human-machine coordination inspection unit. The production line static inspection unit collects image data of the static equipment installation and placement status of the product processing production line and the image data of the standard equipment installation and placement status of the production line for inspection. The production line dynamic inspection unit collects image data of the dynamic operation of the equipment and the linkage status between the equipment of the product processing production line and inspects it with the image data of the dynamic operation and the linkage status between the equipment of the production line standard equipment. The production line human-machine coordination inspection unit collects image data of the collaborative operation status of the equipment and operators in the human-machine coordination of the product processing production line and inspects it with the image data of the collaborative operation status of the standard equipment and operators of the production line;

[0085] The product quality inspection management module includes an offline product quality feature collection unit and an offline quality feature comparison unit. The offline product quality feature collection unit obtains offline product quality feature data, and the offline quality feature comparison unit uses the offline product quality feature data to match and judge with the offline product standard quality feature data; the product storage management module includes a product warehouse feature collection unit and a product warehouse storage identification unit. The product warehouse feature collection unit obtains product warehouse storage feature data, and the product warehouse storage identification unit uses the product warehouse storage feature data to compare and analyze with the product warehouse storage standard feature data;

[0086] The product after-sales management module includes a product failure feedback unit and a product failure repair record unit. The product failure feedback unit collects after-sales product failure feedback data online, and the product failure repair record unit records the cause data of product failures found after product repair;

[0087] The product management cloud platform collaboratively manages the product manufacturing full-cycle management module through data interaction and monitors the orderly and normal operation of the product manufacturing execution module.

[0088] Through the raw material procurement management module, raw material storage management module, product production management module, product quality inspection management module, product storage management module, and product after-sales management module, corresponding module feature data is collected for identification and matching. Through the matching results, the product management cloud platform monitors and manages the product manufacturing process in an orderly manner, and realizes full-cycle quality monitoring of raw material procurement, raw material storage, product production, product quality inspection, product storage, and product after-sales in the product manufacturing process under data interaction, realizes autonomous monitoring of the entire product life cycle, and reduces product quality defects.

[0089] For further information, see Figure 1-Figure 2 ,The product management cloud platform is connected to the raw material procurement management module, the raw material storage management module, the product production management module, the product quality inspection management module, the product storage management module, and the product after-sales management module through the Internet of Things communication. The purchased raw material characteristic data includes one or more characteristic data of raw material origin, color, shape, density, melting point, boiling point, solubility, acidity and alkalinity, oxidizability, reducibility, stability, thickener, antioxidant, flame retardant, and packaging integrity, hydrophilicity, particle shape, porosity, ore particle size, and particle size;

[0090] The raw material feature tracking and judgment unit uses the acquired purchased raw material feature data and the purchased raw material standard feature data to perform matching and judgment operations as follows:

[0091] S11. Establish a data set of purchased raw material characteristics. in Represents different types of purchased raw material feature data, x represents the purchased raw material feature type number, x = 1, 2, 3, ... n, k represents the number of purchased raw material feature data collected, k = 1, 2, 3, ... m;

[0092] S12. Calculate the mean value of the characteristic data of purchased raw materials.

[0093] S13. Average value of purchased raw material characteristic data The range of standard characteristic data of purchased raw materials [S ax Q ax ] Belonging relationship judgment, where S ax Indicates the minimum value of the standard characteristic data range of the purchased raw materials, Q ax Indicates the maximum value of the purchase material standard characteristic data range, x represents the purchase material characteristic type number, x = 1, 2, 3, ... n;

[0094] when If the characteristic data of the purchased raw materials meet the standard characteristic data requirements of the purchased raw materials, the product management cloud platform will prompt the purchased raw materials to enter the storage or processing stage;

[0095] when If it indicates that the characteristic data of the purchased raw materials do not meet the standard characteristic requirements of the purchased raw materials, the product management cloud platform will prompt that the purchased raw materials do not meet the standards and cannot enter the storage and processing stages.

[0096] The raw material warehouse storage characteristic data includes one or more of temperature, humidity, air dust content, raw material partition spacing, lighting intensity, ventilation speed, number of fire points, number of monitoring points, and number of access channels;

[0097] The raw material warehouse storage identification unit uses the raw material warehouse storage feature data and the raw material warehouse storage standard feature data to compare and identify the operation steps as follows:

[0098] S21. Establish a raw material warehouse storage feature data set, in Represents storage characteristic data of different types of raw material warehouses, x represents the type number of the raw material warehouse storage characteristic, x = 1, 2, 3, ... n, k represents the number of raw material warehouse storage characteristic data collected, k = 1, 2, 3, ... m;

[0099] S22. Calculate the mean value of the raw material warehouse storage characteristic data set.

[0100] S23, mean value of storage characteristic data of raw material warehouse The range of standard characteristic data of raw material warehouse storage [S bx Q bx ] Belonging relationship judgment, where S bx Indicates the minimum value of the raw material warehouse storage standard characteristic data range, Q bx Indicates the maximum value of the raw material warehouse storage standard characteristic data range, x represents the purchased raw material characteristic type number, x = 1, 2, 3, ... n;

[0101] when If the raw material warehouse storage characteristic data meets the requirements of the raw material warehouse storage standard characteristic data, the product management cloud platform will prompt the purchased raw materials to enter the warehouse storage stage;

[0102] when This indicates that the raw material warehouse storage characteristic data does not meet the raw material warehouse storage standard characteristic data requirements. The product management cloud platform will prompt that the warehouse storage characteristics do not meet the standards and cannot enter the warehouse storage stage.

[0103] The static inspection unit of the production line collects image data of the static equipment installation and placement status of the product processing production line and the image data of the standard equipment installation and placement status of the production line for inspection. The operation steps are as follows:

[0104] S31. Collect static image data of equipment installation and placement status on the production line, and establish an image data set of equipment installation and placement status image data in represents image data of the installation and placement status of equipment at different positions, x represents the number of image data of the installation and placement status of equipment at different positions, x=1,2,3,…n, k represents the number of image data collected of the installation and placement status of equipment at different positions, k=1,2,3,…m;

[0105] S32, using convolutional neural network to identify the image data of the equipment installation and placement status,

[0106] Input image preprocessing: collect the image data of the equipment installation and placement status, The image data is scaled to a fixed size and the image is normalized and preprocessed;

[0107] Convolution layer processing: Convolution processing is applied to the pre-processed image data of the equipment installation and placement status. The convolution formula is:

[0108] Where F(ω) represents the convolution formula for performing convolution processing on the pre-processed equipment installation and placement status image data, f represents the integral function of the convolution processing formula F(ω), ω represents the adjustment parameter, i represents the object of the pre-processed equipment installation and placement status image data for convolution processing, e is a constant, and t is the input variable of the convolution formula, i.e., the pre-processed equipment installation and placement status image data at different positions, After extracting the convolution, a new feature map of the installation and placement status of equipment in different locations is formed;

[0109] Activation function layer processing: The ReLU activation function is used to perform nonlinear transformation on the feature map of the installation and placement status of devices in different positions output by the convolution layer. The ReLU activation function is:

[0110] Where λ∈(0,1), at this time x is the feature map data of the installation status of the equipment in different positions after convolution,

[0111] Pooling layer processing: downsampling the feature map data of the equipment installation and placement status at different locations after linear transformation;

[0112] Fully connected layer processing: Flatten the downsampled feature map data of the device installation and placement status at different locations into vectors and connect them to the fully connected layer for classification or regression.

[0113] Output results: Identify and output the image data of the equipment installation and placement status at different locations and the image data of the standard equipment installation and placement status of the production line output by the fully connected layer;

[0114] S33. When the output image data of the installation and placement status of the equipment in different positions is recognized, the product management cloud platform controls the production line to enter the dynamic inspection stage;

[0115] When the output image data of the installation and placement status of equipment in different positions cannot be recognized, the product management cloud platform controls the production line and cannot enter the dynamic inspection stage.

[0116] The production line dynamic inspection unit collects dynamic equipment operation and inter-equipment linkage status image data of the product processing production line and verifies it with the dynamic operation and inter-equipment linkage status image data of the production line standard equipment. The operation steps are as follows:

[0117] S41. Collect dynamic operation of equipment and linkage status image data of production line, and establish dynamic operation and linkage status image data set of equipment in Indicates the image data of dynamic operation and linkage status of devices at different locations, x represents the number of the image data of dynamic operation and linkage status of devices at different locations, x = 1, 2, 3, ... n, k represents the number of image data collected of dynamic operation and linkage status of devices at different locations, k = 1, 2, 3, ... m;

[0118] S42, using the convolutional neural network image recognition method in S32, through the steps of input image preprocessing, convolution layer processing, activation function layer processing, pooling layer processing, full connection layer processing, and output result processing, to identify and output the image data of the dynamic operation of the equipment and the linkage status between the equipment and the dynamic operation of the standard equipment of the production line and the linkage status between the equipment.

[0119] S43. When the output image data of the dynamic operation of devices at different locations and the linkage status between devices are recognized, the product management cloud platform controls the production line to enter the human-machine cooperation inspection stage;

[0120] When the output image data of the dynamic operation of equipment in different positions and the linkage status between equipment cannot be recognized, the production line controlled by the product management cloud platform cannot enter the human-machine cooperation inspection stage.

[0121] The production line human-machine cooperation inspection unit collects image data of the human-machine cooperation equipment and operator collaborative operation status of the product processing production line and verifies it with the image data of the collaborative operation status of the standard production line equipment and operator. The operation steps are as follows:

[0122] S51. Collect image data of equipment and operator collaborative operation status of production line human-machine cooperation, and establish image data set of equipment and operator collaborative operation status in represents image data of collaborative operation status of equipment and operators at different positions, x represents the number of image data of collaborative operation status of equipment and operators at different positions, x=1, 2, 3, ... n, k represents the number of image data of collaborative operation status of equipment and operators at different positions, k=1, 2, 3, ... m;

[0123] S52, using the convolutional neural network image recognition method in S32 to perform input image preprocessing, convolution layer processing, activation function layer processing, pooling layer processing, fully connected layer processing, and output result processing steps to identify and output the image data of the collaborative operation status of the equipment and operator and the image data of the collaborative operation status of the standard equipment and operator of the production line;

[0124] S53: When the output image data of the dynamic operation of the equipment at different positions and the linkage status between the equipment are recognized, the product management cloud platform controls the production line to enter the product processing stage.

[0125] When the output image data of the dynamic operation of equipment in different positions and the linkage status between equipment cannot be recognized, the production line controlled by the product management cloud platform cannot enter the product processing stage.

[0126] The raw material procurement management module classifies the purchased raw material feature data into sets, and performs average matching processing on the feature data of different purchased raw materials; the raw material storage management module classifies the raw material warehouse storage feature data into sets, and performs average identification processing on the storage feature data of different raw material warehouses; the product production management module classifies the static equipment installation and placement status image data, dynamic equipment dynamic operation and equipment linkage status image data, human-machine cooperation equipment and operator collaborative operation status image data into sets, and uses convolutional neural networks for identification and processing, thereby realizing autonomous monitoring of raw material procurement, raw material warehousing, and production line production links, ensuring that the production links meet manufacturing standards and reduce product quality defects.

[0127] For further information, see Figure 1-Figure 2 ,Quality characteristic data of off-line products include one or more of appearance inspection, size, weight, hardness, stretching, compression, tensile strength, elongation at break, chemical composition, harmful substance detection, service life of material components, product stability, product reliability, and durability;

[0128] The off-line quality feature comparison unit uses the off-line product quality feature data and the off-line product standard quality feature data to perform a matching judgment operation as follows:

[0129] S61. Establish a data set of quality characteristics of off-line products. in Represents the quality characteristic data of different types of off-line products, x represents the data type number of the off-line product quality characteristic data, x = 1, 2, 3, ... n, k represents the number of off-line product quality characteristic data collected, k = 1, 2, 3, ... m;

[0130] S62, solve the mean value set of the quality characteristic data of the off-line products,

[0131] S63, average value of quality characteristic data of off-line products The range of the standard quality characteristic data of the offline products [S hx Q hx ] Belonging relationship judgment, where S hx Indicates the minimum value of the standard quality characteristic data range of the offline product, Q hx Indicates the maximum value of the standard quality characteristic data range of the offline product, x represents the data type number of the offline product quality characteristic data, x = 1, 2, 3, ... n;

[0132] when If the quality characteristic data of the off-line products meet the standard quality characteristic data requirements of the off-line products, the product management cloud platform will prompt that the processed products can be taken off-line for sale or put into the product warehouse storage stage;

[0133] when This indicates that the quality characteristic data of the off-line products do not meet the standard quality characteristic data requirements of the off-line products. The product management cloud platform will prompt that the processed products do not meet the quality characteristics and cannot be taken off-line for sale or enter the product warehouse storage stage.

[0134] Product warehouse storage characteristic data includes one or more of product stacking height, number of storage partitions, temperature, humidity, air dust content, lighting intensity, number of fire points, number of monitoring points, and number of access channels;

[0135] The product warehouse storage identification unit uses the product warehouse storage feature data and the product warehouse storage standard feature data for comparison and analysis. The operation steps are as follows:

[0136] S71. Establish product warehouse storage feature data set, in Represents the storage characteristic data of different types of product warehouses, x represents the data type number of the product warehouse storage characteristic data, x = 1, 2, 3, ... n, k represents the number of product warehouse storage characteristic data collected, k = 1, 2, 3, ... m;

[0137] S72. Calculate the mean value of the product warehouse storage characteristic data set.

[0138] S73, mean value of product warehouse storage characteristic data The range of product warehouse storage standard characteristic data [S gx Q gx ] Belonging relationship judgment, where S gx Indicates the minimum value of the product warehouse storage standard characteristic data range, Q gx Indicates the maximum value of the product warehouse storage standard feature data range interval, x represents the product warehouse storage feature data type number, x = 1, 2, 3, ... n;

[0139] when If the product warehouse storage characteristic data meets the requirements of the product warehouse storage standard characteristic data, the product management cloud platform will prompt that the offline product can enter the product warehouse storage stage;

[0140] when This indicates that the product warehouse storage feature data does not meet the product warehouse storage standard feature data requirements. The product management cloud platform will prompt that the product warehouse features do not meet the standards and the offline products cannot enter the product warehouse storage stage.

[0141] The steps for the product failure feedback unit to collect after-sales product failure feedback data online and the product failure repair recording unit to record product failure cause data found after product repair are as follows:

[0142] S81. Collect after-sales product failure feedback data H online through the product failure feedback system, and record online product failure cause data H' discovered after product repair.

[0143] Through the product quality inspection management module, the quality characteristic data of the offline products are classified and established into sets, and the quality characteristic data of different offline products are averaged and matched; the product storage management module classifies the product warehouse storage characteristic data and establishes sets, and the storage characteristic data of different product warehouses are averaged and matched; the product after-sales management module collects the after-sales product failure feedback data online and records the cause data of product failures discovered after product repairs, thereby realizing autonomous monitoring and timely feedback of product quality inspection, product warehousing, and product after-sales links, improving the autonomous monitoring and execution of production standards in the product manufacturing process, and reducing the probability of product defective rate.

[0144] A method for operating a data interactive management system for the entire manufacturing cycle of a smart factory product, the method comprising the following steps:

[0145] Step 1: During the raw material procurement management phase, the characteristic data of different types of purchased raw materials are obtained and numbered to create a data set. The characteristic data of different purchased raw materials are averaged and matched with the standard characteristic data of purchased raw materials. If the match is successful, the purchased raw materials enter the warehousing or processing process;

[0146] Step 2: Raw material storage management: Collect the storage characteristic data of different types of raw material warehouses, number them and create a data set. Then, average the storage characteristic data of different raw material warehouses and compare and identify them with the standard storage characteristic data of raw material warehouses. If the comparison and identification is successful, the purchased raw materials will enter the warehousing process;

[0147] Step 3: During the product production management phase, static image data of equipment installation and placement status, dynamic image data of equipment dynamic operation and linkage status between equipment, and image data of human-machine coordinated equipment and operator collaborative operation status are collected and numbered to establish a data set. A convolutional neural network is used to identify and verify the image data of equipment installation and placement status with the image data of standard equipment installation and placement status of the production line, the image data of equipment dynamic operation and linkage status between equipment with the image data of standard equipment dynamic operation and linkage status between equipment of the production line, and the image data of collaborative operation status of equipment and operators with the image data of collaborative operation status of standard equipment and operators of the production line. If the identification and verification is successful, the next execution process is entered;

[0148] Step 4: Product quality inspection and management phase: obtain the quality characteristic data of the off-line products. Obtain the quality characteristic data of different types of off-line products, number them and create a data set. The quality characteristic data of the off-line products are averaged and matched with the standard quality characteristic data of the off-line products. If the match is successful, the product is off-line and enters the sales or warehousing process.

[0149] Step 5: Product storage management phase: collect product warehouse storage feature data. Obtain the warehouse storage feature data of different types of products, number them and create a data set. Then, average the product warehouse storage feature data and compare it with the product warehouse storage standard feature data. If the comparison is successful, the product will be taken off the production line and enter the storage process.

[0150] Step 6: During the product after-sales management phase, collect after-sales product failure feedback data online and record the cause of product failures discovered after product repair.

Claims

1. A smart factory product manufacturing full cycle data interaction management system, characterized by: It includes a raw material procurement management module, including a raw material feature acquisition unit and a raw material feature tracking and judgment unit. The raw material feature acquisition unit acquires the purchased raw material feature data, and the raw material feature tracking and judgment unit uses the acquired purchased raw material feature data to match and judge with the purchased raw material standard feature data; The raw material storage management module includes a raw material warehouse feature acquisition unit and a raw material warehouse storage identification unit. The raw material warehouse feature acquisition unit collects raw material warehouse storage feature data, and the raw material warehouse storage identification unit compares and identifies the raw material warehouse storage feature data with the raw material warehouse storage standard feature data; The product production management module includes a static inspection unit for the production line, a dynamic inspection unit for the production line, and a human-machine coordination inspection unit for the production line. The static inspection unit for the production line collects image data of the static equipment installation and placement status of the product processing production line and compares it with the image data of the standard equipment installation and placement status of the production line for inspection. The dynamic inspection unit for the production line collects image data of the dynamic operation of the equipment and the linkage status between the equipment in the dynamic processing production line and compares it with the image data of the dynamic operation and the linkage status between the equipment in the standard production line. The human-machine coordination inspection unit for the production line collects image data of the equipment and the collaborative operation status of the operator in the human-machine coordination of the product processing production line and compares it with the image data of the collaborative operation status of the standard equipment and the operator in the production line; The product quality inspection management module includes an offline product quality feature collection unit and an offline product quality feature comparison unit. The offline product quality feature collection unit obtains offline product quality feature data, and the offline product quality feature comparison unit uses the offline product quality feature data to match the offline product standard quality feature data; The product storage management module includes a product warehouse feature collection unit and a product warehouse storage identification unit. The product warehouse feature collection unit obtains product warehouse storage feature data. The product warehouse storage identification unit compares and analyzes the product warehouse storage feature data with the product warehouse storage standard feature data, including establishing a product warehouse storage feature data set, solving the product warehouse storage feature data mean set, and comparing the product warehouse storage feature data mean with the product warehouse storage standard feature data range [S gx Q gx ] Belonging relationship judgment, S gx Indicates the minimum value of the product warehouse storage standard characteristic data range, Q gx Indicates the maximum value of the product warehouse storage standard feature data range, x indicates the product warehouse storage feature data type number, x = 1, 2, 3, ... n; when If the product warehouse storage characteristic data meets the requirements of the product warehouse storage standard characteristic data, the product management cloud platform will prompt that the offline product can enter the product warehouse storage stage; when This indicates that the product warehouse storage feature data does not meet the product warehouse storage standard feature data requirements. The product management cloud platform will prompt that the product cannot enter the product warehouse storage stage. The product after-sales management module includes a product failure feedback unit and a product failure repair record unit. The product failure feedback unit collects after-sales product failure feedback data online, and the product failure repair record unit records the cause data of product failures discovered after product repair; The product management cloud platform collaboratively manages the product manufacturing full-cycle management module through data interaction and monitors the orderly and normal operation of the product manufacturing execution module.

2. The smart factory product manufacturing full-cycle data interactive management system according to claim 1, characterized in that: The product management cloud platform is connected to the raw material procurement management module, raw material storage management module, product production management module, product quality inspection management module, product storage management module, and product after-sales management module through IoT communication; The raw material feature tracking and judgment unit uses the acquired purchased raw material feature data and the purchased raw material standard feature data to perform matching and judgment operations as follows: S11. Establish a data set of purchased raw material characteristics. in Represents different types of purchased raw material feature data, x represents the purchased raw material feature type number, x = 1, 2, 3, ... n, k represents the number of purchased raw material feature data collected, k = 1, 2, 3, ... m; S12. Calculate the mean value of the characteristic data of purchased raw materials. S13. Average value of purchased raw material characteristic data The range of standard characteristic data of purchased raw materials [S ax Q ax ] Belonging relationship judgment, where S ax Indicates the minimum value of the standard characteristic data range of the purchased raw materials, Q ax Indicates the maximum value of the purchase material standard characteristic data range, x represents the purchase material characteristic type number, x = 1, 2, 3, ... n; when If the characteristic data of the purchased raw materials meet the standard characteristic data requirements of the purchased raw materials, the product management cloud platform will prompt the purchased raw materials to enter the storage or processing stage; when If it indicates that the characteristic data of the purchased raw materials do not meet the standard characteristic requirements of the purchased raw materials, the product management cloud platform will prompt that the purchased raw materials do not meet the standards and cannot enter the storage and processing stages.

3. The smart factory product manufacturing full-cycle data interactive management system according to claim 2 is characterized by: The raw material warehouse storage identification unit uses the raw material warehouse storage feature data and the raw material warehouse storage standard feature data to compare and identify the operation steps as follows: S21. Establish a raw material warehouse storage feature data set, in Represents storage characteristic data of different types of raw material warehouses, x represents the type number of the raw material warehouse storage characteristic, x = 1, 2, 3, ... n, k represents the number of raw material warehouse storage characteristic data collected, k = 1, 2, 3, ... m; S22. Calculate the mean value of the raw material warehouse storage characteristic data set. S23, mean value of storage characteristic data of raw material warehouse The range of standard characteristic data of raw material warehouse storage [S bx Q bx ] Belonging relationship judgment, where S bx Indicates the minimum value of the raw material warehouse storage standard characteristic data range, Q bx Indicates the maximum value of the raw material warehouse storage standard characteristic data range, x represents the purchased raw material characteristic type number, x = 1, 2, 3, ... n; when If the raw material warehouse storage characteristic data meets the requirements of the raw material warehouse storage standard characteristic data, the product management cloud platform will prompt the purchased raw materials to enter the warehouse storage stage; when This indicates that the raw material warehouse storage characteristic data does not meet the raw material warehouse storage standard characteristic data requirements. The product management cloud platform will prompt that the warehouse storage characteristics do not meet the standards and cannot enter the warehouse storage stage.

4. The smart factory product manufacturing full-cycle data interactive management system according to claim 3 is characterized by: The static inspection unit of the production line collects image data of the static equipment installation and placement status of the product processing production line and the image data of the standard equipment installation and placement status of the production line for inspection. The operation steps are as follows: S31. Collect static image data of equipment installation and placement status on the production line and establish an image data set of equipment installation and placement status in represents image data of the installation and placement status of equipment at different positions, x represents the number of image data of the installation and placement status of equipment at different positions, x=1,2,3,…n, k represents the number of image data collected of the installation and placement status of equipment at different positions, k=1,2,3,…m; S32, using convolutional neural network to identify the image data of the equipment installation and placement status, Input image preprocessing: collect the image data of the equipment installation and placement status, The image data is scaled to a fixed size and the image is normalized and preprocessed; Convolution layer processing: Convolution processing is applied to the pre-processed image data of the equipment installation and placement status. The convolution formula is: Where F(ω) represents the convolution formula for performing convolution processing on the pre-processed equipment installation and placement status image data, f represents the integral function of the convolution processing formula F(ω), ω represents the adjustment parameter, i represents the object of the pre-processed equipment installation and placement status image data for convolution processing, e is a constant, and t is the input variable of the convolution formula, i.e., the pre-processed equipment installation and placement status image data at different positions, After extracting the convolution, a new feature map of the installation and placement status of equipment in different locations is formed; Activation function layer processing: The ReLU activation function is used to perform nonlinear transformation on the feature map of the installation and placement status of devices in different positions output by the convolution layer. The ReLU activation function is: Where λ∈(0,1), at this time x is the feature map data of the installation status of the equipment in different positions after convolution, Pooling layer processing: downsampling the feature map data of the equipment installation and placement status at different locations after linear transformation; Fully connected layer processing: Flatten the downsampled feature map data of the device installation and placement status at different locations into vectors and connect them to the fully connected layer for classification or regression. Output results: Identify and output the image data of the equipment installation and placement status at different locations and the image data of the standard equipment installation and placement status of the production line output by the fully connected layer; S33. When the output image data of the installation and placement status of the equipment in different positions is recognized, the product management cloud platform controls the production line to enter the dynamic inspection stage; When the output image data of the installation and placement status of equipment in different positions cannot be recognized, the product management cloud platform controls the production line and cannot enter the dynamic inspection stage.

5. The smart factory product manufacturing full-cycle data interactive management system according to claim 4 is characterized by: The production line dynamic inspection unit collects dynamic equipment operation and inter-equipment linkage status image data of the product processing production line and verifies it with the dynamic operation and inter-equipment linkage status image data of the production line standard equipment. The operation steps are as follows: S41. Collect dynamic operation of equipment and linkage status image data of production line, and establish dynamic operation and linkage status image data set of equipment in Indicates the image data of dynamic operation and linkage status of devices at different locations, x represents the number of the image data of dynamic operation and linkage status of devices at different locations, x = 1, 2, 3, ... n, k represents the number of image data collected of dynamic operation and linkage status of devices at different locations, k = 1, 2, 3, ... m; S42, using the convolutional neural network image recognition method in S32, through the steps of input image preprocessing, convolution layer processing, activation function layer processing, pooling layer processing, full connection layer processing, and output result processing, to identify and output the image data of the dynamic operation of the equipment and the linkage status between the equipment and the dynamic operation of the standard equipment of the production line and the linkage status between the equipment. S43. When the output image data of the dynamic operation of devices at different locations and the linkage status between devices are recognized, the product management cloud platform controls the production line to enter the human-machine cooperation inspection stage; When the output image data of the dynamic operation of equipment in different positions and the linkage status between equipment cannot be recognized, the production line controlled by the product management cloud platform cannot enter the human-machine cooperation inspection stage.

6. The smart factory product manufacturing full-cycle data interactive management system according to claim 5, characterized in that: The production line human-machine cooperation inspection unit collects image data of the human-machine cooperation equipment and operator collaborative operation status of the product processing production line and verifies it with the image data of the collaborative operation status of the standard production line equipment and operator. The operation steps are as follows: S51. Collect image data of equipment and operator collaborative operation status of production line human-machine cooperation, and establish image data set of equipment and operator collaborative operation status in represents image data of collaborative operation status of equipment and operators at different positions, x represents the number of image data of collaborative operation status of equipment and operators at different positions, x=1, 2, 3, ... n, k represents the number of image data of collaborative operation status of equipment and operators at different positions, k=1, 2, 3, ... m; S52, using the convolutional neural network image recognition method in S32 to perform input image preprocessing, convolution layer processing, activation function layer processing, pooling layer processing, fully connected layer processing, and output result processing steps to identify and output the image data of the collaborative operation status of the equipment and operator and the image data of the collaborative operation status of the standard equipment and operator of the production line; S53: When the output image data of the dynamic operation of the equipment at different positions and the linkage status between the equipment are recognized, the product management cloud platform controls the production line to enter the product processing stage. When the output image data of the dynamic operation of equipment in different positions and the linkage status between equipment cannot be recognized, the production line controlled by the product management cloud platform cannot enter the product processing stage.

7. The smart factory product manufacturing full-cycle data interactive management system according to claim 6, characterized in that: The off-line quality feature comparison unit uses the off-line product quality feature data and the off-line product standard quality feature data to perform a matching judgment operation as follows: S61. Establish a data set of quality characteristics of off-line products. in Represents the quality characteristic data of different types of off-line products, x represents the data type number of the off-line product quality characteristic data, x = 1, 2, 3, ... n, k represents the number of off-line product quality characteristic data collected, k = 1, 2, 3, ... m; S62, solve the mean value set of the quality characteristic data of the off-line products, S63, average value of quality characteristic data of off-line products The range of the standard quality characteristic data of the offline products [S hx Q hx ] Belonging relationship judgment, where S hx Indicates the minimum value of the standard quality characteristic data range of the offline product, Q hx Indicates the maximum value of the standard quality characteristic data range of the offline product, x represents the data type number of the offline product quality characteristic data, x = 1, 2, 3, ... n; when If the quality characteristic data of the off-line products meet the standard quality characteristic data requirements of the off-line products, the product management cloud platform will prompt that the processed products can be taken off-line for sale or put into the product warehouse storage stage; when This indicates that the quality characteristic data of the off-line products do not meet the standard quality characteristic data requirements of the off-line products. The product management cloud platform will prompt that the processed products do not meet the quality characteristics and cannot be taken off-line for sale or enter the product warehouse storage stage.

8. The smart factory product manufacturing full-cycle data interactive management system according to claim 7, characterized in that: The steps for the product failure feedback unit to collect online after-sales product failure feedback data and the product failure repair record unit to record the cause of product failures found after product repair are as follows: S81. Collect after-sales product failure feedback data H online through the product failure feedback system, and record online product failure cause data H′ discovered after product repair.

9. The method for operating a smart factory product manufacturing full-cycle data interaction management system according to any one of claims 1 to 8, characterized in that: The method comprises the following steps: Step 1: During the raw material procurement management phase, the characteristic data of different types of purchased raw materials are obtained and numbered to create a data set. The characteristic data of different purchased raw materials are averaged and matched with the standard characteristic data of purchased raw materials. If the match is successful, the purchased raw materials enter the warehousing or processing process; Step 2: Raw material storage management: Collect the storage characteristic data of different types of raw material warehouses, number them and create a data set. Then, average the storage characteristic data of different raw material warehouses and compare and identify them with the standard storage characteristic data of raw material warehouses. If the comparison and identification is successful, the purchased raw materials will enter the warehousing process; Step 3: During the product production management phase, static image data of equipment installation and placement status, dynamic image data of equipment dynamic operation and linkage status between equipment, and image data of human-machine coordinated equipment and operator collaborative operation status are collected and numbered to establish a data set. A convolutional neural network is used to identify and verify the image data of equipment installation and placement status with the image data of standard equipment installation and placement status of the production line, the image data of equipment dynamic operation and linkage status between equipment with the image data of standard equipment dynamic operation and linkage status between equipment of the production line, and the image data of collaborative operation status of equipment and operators with the image data of collaborative operation status of standard equipment and operators of the production line. If the identification and verification is successful, the next execution process is entered; Step 4: Product quality inspection and management phase: obtain the quality characteristic data of the off-line products. Obtain the quality characteristic data of different types of off-line products, number them and create a data set. The quality characteristic data of the off-line products are averaged and matched with the standard quality characteristic data of the off-line products. If the match is successful, the product is off-line and enters the sales or warehousing process. Step 5: Product storage management phase: collect product warehouse storage feature data. Obtain the warehouse storage feature data of different types of products, number them and create a data set. Then, average the product warehouse storage feature data and compare it with the product warehouse storage standard feature data. If the comparison is successful, the product will be taken off the production line and enter the storage process. Step 6: During the product after-sales management phase, collect after-sales product failure feedback data online and record the cause of product failures discovered after product repair.

Images