Data acquisition method for standard structured data set architecture of intelligent SOP

Abstract

The invention discloses a data acquisition method for a standard structured dataset architecture of an intelligent SOP, the standard structured dataset architecture comprises a plurality of grades, data acquired by each grade comprises names, models and descriptions, formulas and processes below the names, the descriptions comprise feature descriptions and function descriptions, and the descriptions comprise the name models and the descriptions, the formulas and the processes. The formula comprises a bill of material table and a plurality of grades, and each grade comprises a corresponding name model in the grade and description, formula and process of the name model; the technology comprises production configuration and a technological process, and comprises a method used for a machine adaptive system and intelligent SOP to collect data in a guide mode. Through a systematic standard structured data set architecture and a data acquisition mode, the universality of the intelligent SOP among different products is realized, an enterprise can perform content supplementation according to a guide on the basis of an existing architecture, and the manpower, material and time consumption of technicians on the construction work of the intelligent SOP is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and specifically to a data collection method for a standard structured data set architecture for an intelligent SOP, which integrates multimodal sensory data through a hierarchical data architecture and combines artificial intelligence (AI) technology to achieve standardization, dynamic optimization and real-time adaptation of production processes. Background Art

[0002] In the manufacturing industry, especially in the field of modern intelligent manufacturing, SOP (Standard Operating Procedure) is of great significance. It can not only improve work efficiency, unify standards, reduce risks, and ensure quality, but also facilitate training and management. However, at present, the SOPs of most companies are still paper-based, and only a small number of companies use electronic screens to display SOPs. Very few large companies have customized intelligent SOP solutions. Most of the existing SOP systems cannot meet the requirements of intelligent manufacturing and lack higher flexibility, intelligence and automation support.

[0003] In conclusion, in the current intelligent manufacturing field, SOP needs to be made intelligent, simplified and standardized. The data collection method of the standard structured data set architecture still has the following problems: 1. Data isolation and the inability to integrate multimodal sensory data limit the application of AI technology in intelligent manufacturing.

[0004] 2. Poor flexibility. Different products need to rebuild the SOP standard operating procedures. The customization process is time-consuming and costly. Static SOP cannot respond to changes in the production environment (such as new equipment and process adjustments) in real time.

[0005] 3. The data collection process is complex, and the existing system is too dependent on third-party customized software and cannot be deployed on a large scale.

[0006] 4. Unable to share data. The current SOP format is not smart enough and cannot respond to changes in the production process in real time.

[0007] In the prior art, the patent with publication number CN112711239A proposes a SOP system based on electronic screen display, but it lacks structured data architecture and intelligent data collection mechanism.

[0008] In summary, the existing SOP method cannot meet the needs of the intelligent manufacturing field. A more intelligent, efficient, comprehensive, and simple standard structured data set architecture and data collection method is needed, combined with AI technology to improve the level of intelligent manufacturing. Summary of the invention

[0009] The purpose of the present invention is to provide an intelligent SOP data set architecture and collection method, which simplifies data input, improves data quality, and supports AI-driven dynamic SOP generation through standardized hierarchical data models and wizard-style interactive processes, solving the problems of data isolation, high customization cost, poor flexibility, and high training cost for replacement of old and new employees in traditional systems. At the same time, process update data can also be shared remotely in different places. In the future, the work experience data of the old master collected by the intelligent SOP can be used to generate a "process skill package" with one click, realizing the "second-level posting" of robots. From the rapid deployment of robots in a single scene to the construction of a cross-industry skill library, the unmanned digital craftsman army will eventually be gradually realized.

[0010] To achieve the above object, the present invention provides the following technical solutions: 1. Build a hierarchical standard structured data set architecture to integrate product description, formula, process and robot sensory data; 2. Generate standardized SOPs through guided data collection and associate multimodal sensory data; 3. Train AI models based on sensory data to achieve autonomous robot operation and human-machine collaboration.

[0011] The present application discloses a data collection method for a standard structured data set architecture for intelligent SOP, including a standard structured data set architecture for collecting and storing data, the architecture is divided into several levels, and the collected data includes product description, formula and process; each hierarchical data is processed by AI intelligence, combined with the sensory data of the robot, to form a comprehensive and accurate product and production process data model, providing more accurate data support for intelligent SOP; The standard structured data set architecture includes several levels, and the data collected in each level include the name and its description, formula and process. The description includes feature description and function description. The formula includes a bill of materials table and several levels, and each level includes the corresponding name and its description, formula and process in the level. The process includes production configuration and process flow. The present application also includes a method for collecting data by a wizard for a machine adaptive system and an intelligent SOP, including guiding a user to input information such as a product name, function description, feature description, formula, production configuration, process flow, etc.; at the same time, the system automatically prompts the user to collect sensory data such as touch, smell, hearing, taste, etc. related to the product to provide data support for subsequent intelligent processing, including the following steps: S1: The user names the product and defines the model through the wizard; S2: Guide users to describe the product, including functional description and feature description; S21: Guide users to collect sensory data such as touch, smell, hearing, taste, etc. related to production, for the purpose of improving and supplementing the data; S3: Guide the user to input the content of the product formula, including the components and parts of the product and their specifications, quantity, and supplier; S4: Guide the user to input the production configuration content in the process of the product. The production configuration includes the necessary conditions for producing the product, including equipment, fixtures, and raw materials; S5: Guide the user to input the process flow content of the product. Each step in the process flow includes the feature description and function description of the process. S6: Integrate the user input content and store the data into the database through the standard function interface.

[0012] Preferably, the present application also includes a multimodal data acquisition system, which includes sensors such as touch, smell, hearing, and taste, and can collect real-time sensory data of robots and workers in the production process, such as force data, gas concentration, audio data, chemical composition detection, etc., for real-time recording of details of the production process.

[0013] Preferably, in S1, a wizard is used to guide the user to fill in the name and model of the product, where the model includes the model definition.

[0014] Preferably, in S2, the user uses a wizard to describe the functions and features of the product, and the input of the functional description includes describing its functions in words; the input of the feature description includes one or more combinations of 3D drawings, engineering drawings, visual data, auditory data, olfactory data, taste data, and tactile data.

[0015] Among them, visual data include images, point clouds, depth maps and other data from various visual sensors; auditory data include MP3, WAV, AAC and other data from microphones and language recognition modules; olfactory data include gas composition, concentration and other data from gas sensors; taste data includes compound content and other data from taste sensors that detect chemical components; tactile data include contact force, pressure, texture, hardness and other data from various tactile sensors.

[0016] Preferably, S3 includes the following contents: using a wizard to guide the user to input the formula contents: the formula includes a bill of materials, the bill of materials includes several levels, each level includes several component structures, the highest level is the product structure, the lowest level is the part structure, the composition of each level includes several next levels, until the next level is all part structures.

[0017] Preferably, S3 includes the following contents: the user's input of the formula includes the basic product and accessories list table and the formula table such as specification model, quantity, supplier, etc.

[0018] Preferably, the bill of materials includes three levels, including products, components and parts.

[0019] Preferably, S4 and S5 include the following contents: using a wizard to guide the user to configure the production configuration and process flow in the process: the production configuration includes the production conditions of the product and its components, including the production environment, production equipment, production tools and production raw materials; the S5 process flow includes a functional description of the process and a characteristic description of the process, and the input of the functional description of the process includes a description of its function in words; the characteristic description input of the process includes one or more combinations of kinematic state data, dynamic state data, and characteristic description data of product components and parts.

[0020] The kinematic state data includes motion coordinate position, posture, etc., which are derived from various position sensors, inertial sensors, visual sensors, and motion simulation software data for trajectory planning. The dynamic state data includes linear acceleration and deceleration, angular acceleration and deceleration, torque, etc., which are derived from force perception, gyroscopes, and position perception related sensors. The feature description data of products, components, and parts are derived from the feature description data collected in S2.

[0021] Preferably, S6 includes the following content: S61: Obtain the user's input description of the product as the overall introduction of the product; S62: Obtaining the user's input content for the product formula, and using the input content in each classification as an introduction to the structure; S63: Obtain the user's input content on the product process, and intelligently divide and match the process flow and production configuration input content as an introduction to the corresponding process.

[0022] S64: Intelligent SOP Generation: Based on hierarchical data and multimodal input, the AI ​​model automatically generates executable SOPs, including the following: Visual operation instructions (action video, 3D animation); Real-time quality inspection rules (such as visual comparison of part positions, gas concentration threshold alarms); Dynamic parameter adjustment (optimize assembly forces based on haptic data). S65: Exception handling mechanism: Real-time monitoring of production data. If deviations from standards are found (such as pressure exceeding the limit, abnormal odor), an early warning is triggered and a log is recorded; The AI ​​model optimizes process parameters through learning from historical data and reduces human intervention.

[0023] Preferably, in S63, after constructing the introduction of the corresponding process based on the production configuration content input by the user, when the user runs the standard operating procedure, if new production equipment or personnel are added, the standard operating procedure will identify the production equipment or personnel and match them with the production configuration content, and assign the corresponding process flow to the production equipment.

[0024] Preferably, in S65, when the user runs the standard operating procedure, the standard operating procedure performs feature recognition of the user through motion capture, and provides feedback when the user's operation does not conform to the collected data of the current process flow.

[0025] The present application also discloses a data collection wizard device for intelligent SOP, including a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, they are used to implement the above-mentioned method for the wizard for intelligent SOP to collect its data.

[0026] The present application also discloses a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the above-mentioned method for a wizard for intelligent SOP to collect its data is implemented.

[0027] Beneficial effects: 1. Through the systematic storage framework and data collection method, the universality of intelligent SOPs between different products can be achieved. When enterprises develop new products, they can supplement the content according to the wizard based on the existing architecture, which greatly reduces the manpower investment. Technical personnel can complete the construction of intelligent SOPs more efficiently. 2. Through the data collection wizard, the construction cycle and difficulty of SOP are greatly shortened, and material and time costs are saved. For ordinary enterprises, this is undoubtedly a huge advantage, which can effectively reduce the operating costs of enterprises and improve their market competitiveness; 3. Quality assurance: Multimodal data binding and real-time verification improve SOP accuracy; 4. Dynamic adaptation: When new equipment is added or the process changes, the system automatically adjusts the process to reduce downtime; 5. Data sharing: process data can be shared remotely, such as product process changes in the Shanghai factory, which can be synchronized to the Thailand factory in real time; 6. Generation of process skill package: The work experience data of veteran workers collected through intelligent SOP can be used to generate a "process skill package" with one click, enabling the robot to "go to work in seconds". BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flowchart of a wizard-style data collection method for collecting data of a wizard for an intelligent SOP of the present invention; Figure 2 It is a standard structured data set architecture for intelligent SOP of the present invention; Figure 3 It is a schematic diagram of the architecture of the product part in the standard structured data set architecture of the present invention; Figure 4 It is a schematic diagram of the architecture of the components in the standard structured data set architecture of the present invention; Figure 5 It is a schematic diagram of the architecture of the parts in the standard structured data set architecture of the present invention; Figure 6 It is a schematic diagram of the logical framework of the AI ​​dynamically generating intelligent SOP of the present invention; Figure 7 It is a schematic diagram of a data collection wizard device for intelligent SOP of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below through the accompanying drawings and embodiments. However, it should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the scope of the present invention. In addition, in the following description, the description of known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0030] See also Figures 1 to 5 The embodiment of the present invention provides a data collection method for a standard structured data set architecture for intelligent SOP, including a standard structured data set architecture, wherein Figures 3-5 As shown, it is a schematic diagram of the product, component, and part architecture of the standard structured data set architecture of the present invention. The data collected by the standard structured data set architecture includes descriptions, formulas, and processes.

[0031] Description: The description includes feature description and function description; the feature description describes the characteristics of each level, including the structure, appearance, size, material, sound, gas composition, chemical composition and other aspects of each level; the function description describes the function of each level, including the various functions and performance of each level.

[0032] Recipe part: The formula includes a BOM table and several levels, each of which includes the corresponding process, formula and description within the level; the BOM table includes the various component structures of the product, which are gradually refined from the highest level product structure to the lowest level part structure through a hierarchical approach.

[0033] The classification includes three levels: products, components and parts. These three levels can cover all products. Through the division of these three levels, the BOM table is clearer.

[0034] Product layer: defines the product’s functional description, characteristic data (visual, auditory, tactile, olfactory, taste), BOM and process flow; Component layer: defines the functional description of the component, characteristic data (visual, auditory, tactile, olfactory, taste), detailed component sub-formula (nested parts or sub-components), single-station production configuration and process flow; Part layer: defines the functional description of the part, characteristic data (visual, auditory, tactile, olfactory, taste), formula (material and physical properties), single-station process parameters and real-time sensor feedback; Multimodal data binding: Each level associates sensory data (such as part images, assembly audio, and pressure detection values) for AI-driven quality verification and process optimization.

[0035] Process part: The process includes production configuration and process flow; the production configuration includes the production conditions of the product and its components, including but not limited to production environment requirements such as temperature, humidity, cleanliness, and production equipment requirements such as assembly lines, robots, and machine tools.

[0036] Process definition and input specifications The process flow includes two parts: process function description and process feature description. The process flow characteristics need to define the assembly steps of each product in detail, and forcibly associate action videos and feature pictures to achieve visual guidance of standardized operations.

[0037] The specific requirements are as follows: Production Configuration Including production environment (temperature, humidity, cleanliness), equipment parameters (robot model, machine tool accuracy) and ergonomic data (operating platform height, safety area).

[0038] Example: In electronic product assembly, an anti-static environment (humidity ≤ 30%) and a robot arm reach range (≥ 1.5 m) are required.

[0039] Process Each process must include the following: Functional description: a textual description of the goal and operation requirements of this step; Feature description: Feature description is divided into 2 categories: Category 1 is derived from the previously collected feature data of parts, components or parts, which is used for part comparison and quality verification. Category 2 is derived from action videos: upload standardized operation videos to show the correct execution process (such as part assembly sequence, tool usage).

[0040] The process flow includes the process flow of products, components and parts, each of which has its own specific functions and characteristics. The method includes the following steps: S1: The user uses the wizard to describe the first step of the process, describing the function, goal, and task of the step; S2: The user describes the process features of the first step through the wizard. The process features include two options: feature data and action video. For example: import the feature data of products, components, and parts; or upload or shoot an action video.

[0041] S3: Repeat S1 and S2 to add more process steps.

[0042] S4: Integrate the content input by the user and store the data into the database through the standard function interface; the system integrates the description, process, formula and other information input by the user, and realizes data standardization processing through the modular data interface; in the subsequent processing, the intelligent SOP (standard operating procedure) is built through the data content in the database. This intelligent SOP can provide guidance for the company's production management through the collected information to ensure that the product production process is standardized and efficient.

[0043] In a feasible embodiment, the method comprises the following steps: S1: Product definition: guide users to enter product name and model, upload function description text and feature data (3D model, test report); S2: Recipe building: supports importing BOM tables from Excel or manually defining product-component-part hierarchical relationships, and automatically parsing them into structured data; S3: Process configuration: Define the production environment (temperature, humidity, cleanliness), equipment parameters (robot model, accuracy); Bind the standardized action videos and sensor data collection rules of the process flow; Supports step-by-step input of process functions (text description) and features (visual comparison rules, tactile thresholds); S4: Data verification: Automatically detect the integrity of input data (such as missing BOM table levels, sensory data not bound), and prompt the user to make corrections; S5: AI integration: The system inputs data into the AI ​​model through API, generates dynamic SOPs and embeds real-time operation guidance; S6: Deployment and Adaptation: Supports automatic identification of parameters (such as machine arm span, coordinate range) when adding new equipment, and dynamically adjusts the process flow. Multi-format input support: Text description, Excel table, 3D model file (STEP, STL), sensor data (CSV, JSON); Visual data supports images (JPG, PNG), point clouds (PLY), and depth maps (TIFF); Auditory data supports audio (WAV, MP3), voice commands (text-to-speech). The data verification includes BOM table level integrity check, sensory data format verification and process parameter compliance determination.

[0044] The Dynamic SOP supports the following functions: 1. Real-time visual comparison of parts position and audio prompts of operation steps; 2. Dynamically adjust assembly parameters based on tactile data; 3. When the gas concentration exceeds the limit, a shutdown command is triggered and an abnormal log is recorded.

[0045] Among them, it also includes: sensory data collection. After collecting product descriptions and formulas, the wizard will guide users to collect sensory data such as touch, smell, hearing, and taste related to production, which will be used to improve and supplement the data; It includes a multimodal data acquisition system. Tactile sensors are installed on the operating parts of the robot to collect tactile data such as contact force, pressure, texture, and hardness in real time. Olfactory sensors are distributed in the production environment to continuously monitor gas composition and concentration. Microphones and language recognition modules are set near the equipment to collect audio data generated when the equipment is running. Taste sensors are used in the key links of detecting the chemical composition of products or materials.

[0046] Including the perception data module, including the following: 1. Tactile data: Data collected by the robot's tactile sensors, including contact force, pressure, texture, hardness, etc., can help the AI ​​model understand the operating environment of objects in production and ensure product accuracy.

[0047] 2. Olfactory data: The gas composition and concentration data collected by the gas sensor can be used to detect gas leaks and composition changes during the production process, providing precise process control and environmental monitoring.

[0048] 3. Auditory data: The audio data collected by the microphone and speech recognition module can identify the operating sounds of the equipment during the production process, thereby determining the operating status and potential failures of the machine.

[0049] 4. Taste data: Taste sensors that detect chemical composition can monitor changes in the chemical composition of products or materials to ensure product quality standards, including a chemical composition detector that can analyze taste by detecting chemical composition.

[0050] During the sensory data collection stage, after the user enters the product description and formula, data is collected through the multimodal data collection system; the tactile sensor detects and collects various force data when the robot, equipment and objects come into contact during the production process; the olfactory sensor monitors and collects the gas conditions in the production environment; the auditory sensor detects the sound of the collection equipment in operation; the taste sensor detects the chemical composition of the collection product or material; the data is synchronously collected and transmitted to the system to provide a data basis for subsequent processing.

[0051] In the description content part of S2 for products, components and parts, the feature description includes not only the appearance structure of the product, but also the internal structure, material composition and other aspects of the product; the functional description includes the main functions of the product and some additional functions, as well as the combination functions of the product with other corresponding products.

[0052] The user's input of description and process includes any one of the formats of text description, picture display, video demonstration, model file, motion trajectory, drawing file, data plus unit, and their combination.

[0053] In the content part of the formula in S3, the formula includes a BOM table, which includes three levels, including products, components, and parts. Products are composed of several components and several parts. The composition of components includes two modes: composed of other components and several parts, and composed of several parts. Among them, the collection of BOM tables includes Excel table import, manual input, other file import and other methods.

[0054] The user's input of the formula in S3 includes the process content, formula content and description content of each component structure; the process content includes the production process, processing method, etc. of the component structure; the formula content includes the parts composition, name, specification, quantity, etc. of the component structure; the description content includes the characteristics, functions, uses, etc. of the component structure.

[0055] In a feasible embodiment, the standard structured data set architecture of the recipe includes: a product function (ProductFunction: product layer), including: Product X = product function (name, function description, feature description, call component function, call part function, bill of materials, production configuration, process flow); For example: Product X = Product Function ("Product X", "Transportation Function", "Modular Design", [Part A], [Part B], {"Environment": "Dust-free Workshop"}, ["Packaging Step 1"]); Among them, the parameters of the product include: product name (name): product model; Function Description: a text description of the product function; Feature Description: text description, pictures, modeling files, etc. of product features; BOMList: includes components and parts information; Production Setup: production conditions, such as equipment and environment; Process Flow: including the function and feature description of each step; Call the component function (ComponentFunction).

[0056] Component Function (Component layer), including: Component A = component function (name, functional description, feature description, calling part function, bill of materials, production configuration, process flow); For example: Part A = Part Function ("Part A", "Transfer Function", "Plastic Material", [Part A], {"Temperature": "Normal Temperature"), ["Assembly Step 1"]); The parameters of the component include: Name: component name; Function Description: a text description of the component function; Feature Description: text description, pictures, modeling files, etc. of component features; BOMList: includes parts information; Production Setup: production conditions of components; Process Flow: including the function and feature description of each step; Call the part function (PartFunction).

[0057] Part function (PartFunction: part layer), including: Part A = part function (name, functional description, feature description, production configuration, process flow); For example: Part A = Part Function ("Part A", "Connection Function", "Metal Material", {"Temperature": "Normal Temperature"}, ["Connection Part B"]); The parameters of the parts include: Name: part name; Function Description: text description of the function of the part; Feature Description: text description, pictures, modeling files, etc. of part features; Production Setup: production conditions of parts; Process Flow: includes the function and feature description of each step.

[0058] In S4, the production environment in the production configuration also includes content such as the safety environment, environmental protection requirements, precautions, etc.; the production equipment includes information such as the equipment model, specifications, performance parameters, etc., so that production staff can better understand the use of the equipment and production capacity.

[0059] In S5, the user inputs the content of the sub-process including the function, description, process, formula, technical requirements and other aspects of the sub-process.

[0060] In said S6, the artificial intelligence first analyzes and organizes various information such as description, process and formula input by the user; For the description part, it recognizes input content in different formats, such as product features in pictures, functional demonstrations in videos, detailed descriptions in text, etc., and uses this part of the input content for the overall description of the product when building the smart SOP; For the process part, AI will collect the production configuration and process of each sub-process and perform intelligent matching. When building the intelligent SOP, the description of the obtained production configuration and process flow will be used in the introduction of the sub-process; For the formula part, artificial intelligence will intelligently assign the hierarchical relationship of the product component structure according to the hierarchical structure of the BOM table, and at the same time combine the process content, formula content and description content of each component structure to build an intelligent SOP.

[0061] Finally, the generated intelligent SOP will be presented to the user in a clear and complete form. During production, the intelligent SOP will identify the operations of the production staff according to the actual situation, and then give a standardized operation demonstration to facilitate the production staff to adjust and optimize, ensuring that it meets the company's production requirements.

[0062] The S6 includes the following sub-steps: S61: Obtain the user's input content for the description of the product as the overall introduction of the product, and collect the description content in various formats such as text description, picture display, video demonstration, modeling file, drawing, etc. input by the user for constructing the product description of the intelligent SOP; S62: Obtain the user's input content on the product process, and intelligently divide and match the process flow and production configuration input content as an introduction to the corresponding process; When users are running the smart SOP, if new production equipment is added, the smart SOP will automatically allocate the production equipment by identifying the production equipment and matching it with the production configuration content. There is no need for users to manually allocate, which simplifies the operation difficulty and speeds up the operation process. S63: Obtain the user's input content for the product formula, use the input content in each level as the introduction of the structure, organize the bill of materials table input by the user and the formula content of each level, clarify the process, formula and description information of each component structure, and serve as a management guide for materials in the production process; S64: Integrate the information obtained from S61, S62 and S63, and use artificial intelligence algorithms to build a complete intelligent SOP, including three main contents: production process, formula and description, to provide comprehensive standardized guidance for the company's production management; S65: When the user is actually running the intelligent SOP, the system will provide real-time technical guidance based on the user's operating conditions.

[0063] In a feasible embodiment, when running the intelligent SOP, for each process flow, the intelligent SOP performs feature recognition of the production staff through motion capture to determine which process flow the step belongs to, and at the same time, it can identify whether there are any operating errors in the process.

[0064] Example 1: Automobile parts assembly scenario 1. Product Definition: Enter the product name "New Energy Vehicle Motor Controller", upload the functional description (voltage regulation, overheat protection) and feature data (3D model, high temperature resistant material report). 2. Recipe construction: Import the BOM table and parse it into a three-level structure: Product layer: motor controller; Component layer: circuit board module, heat sink assembly; Parts layer: capacitor (model C0805), heat sink (aluminum, size 50mm×50mm). 3. Process configuration: Definition of production environment: anti-static workshop (humidity ≤ 40%); Binding action video: record the standard operation of circuit board soldering (temperature control, solder joint detection); Set the tactile threshold: Assemble the pressure range (5-10N), exceeding the limit will trigger an alarm. 4. SOP generation and execution: AI generates dynamic SOPs to guide workers to follow the steps, and cameras detect the quality of solder joints in real time; When adding a new robot arm, the system automatically calibrates the coordinates and assigns the welding path. Example 2: Chemical production quality inspection scenario Use olfactory sensors to detect the gas composition in the reactor, and AI compares it with the standard odor spectrum to automatically determine the reaction endpoint and record it in the SOP log.

[0065] In a feasible embodiment, when users add new equipment or change new products and lines as needed during subsequent use, the intelligent SOP will automatically assign tasks based on the products currently being produced. It will also identify the machine code of the machine, read the machine's volume, arm span, range of motion and other human engineering data and sensor data, and automatically match the calibration coordinates. It will automatically generate motion trajectories, coordinates and postures that are suitable for the current workstation, without the need for manual intervention for manual programming or teaching, or customized AI motion training for the workstation alone.

[0066] An embodiment of a data acquisition wizard device for intelligent SOP of the present invention can be applied to any device with data processing capability, and the device with data processing capability can be a device or apparatus such as a computer. The device embodiment can be implemented by software, or by hardware or a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, the processor of any device with data processing capability reads the corresponding computer program instructions in the non-volatile memory into the internal memory and runs them. From the hardware level, if Figure 7 As shown, it is a hardware structure diagram of any device with data processing capability where a data acquisition wizard device for intelligent SOP of the present invention is located, except Figure 7 In addition to the processor, memory, network interface, and non-volatile memory shown, any device with data processing capabilities in the embodiment of the device may also include other hardware according to the actual function of the device with data processing capabilities, which will not be described in detail. The implementation process of the functions and effects of each unit in the above device is specifically detailed in the implementation process of the corresponding steps in the above method, which will not be described in detail here.

[0067] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of the present invention. Ordinary technicians in this field can understand and implement it without paying creative work.

[0068] An embodiment of the present invention further provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, a data collection wizard device for an intelligent SOP in the above embodiment is implemented.

[0069] The computer-readable storage medium may be an internal storage unit of any device with data processing capability described in any of the aforementioned embodiments, such as a hard disk or a memory. The computer-readable storage medium may also be an external storage device of any device with data processing capability, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), an SD card, a flash card (Flash Card), etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capability. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capability, and may also be used to temporarily store data that has been output or is to be output.

[0070] In a feasible embodiment, an intelligent SOP generation system is also included, and its system hardware architecture includes: Memory, which stores hierarchical data architecture, multimodal data, and AI models; A processor, executing the above-mentioned data collection method of a standard structured data set architecture for intelligent SOP, generating dynamic SOP and responding to production changes in real time; Sensor interface, connecting visual, tactile, and olfactory sensors to collect real-time data; The human-computer interaction module provides a graphical wizard interface and operation instructions. Usage process: When using this method, the user first follows the wizard to fill in the name of the product, and then enters the second level, including description, formula, and process. First, describe the functions and features of the product, including the input of pictures, text and other formats in this process; then enter the formula, and you can choose to import the BOM table through Excel and other formats, and supplement and expand it step by step according to the three levels of products, components, and parts, and supplement each component and part with more specific descriptions, formulas, and process content; then enter the process, through the input of production configuration and process flow content, supplement the configuration required for each step of the process, and improve it with pictures, videos, text and other formats to complete the content input of the intelligent SOP; then use artificial intelligence to build an intelligent SOP, and provide standardized and intelligent guidance for production staff when they are performing production and processing operations.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent substitution or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data collection method for a standard structured data set architecture for intelligent SOP, characterized by: The invention comprises a standard structured data set architecture for collecting and storing data, wherein the standard structured data set architecture comprises several levels, and the data collected by each level comprises a name and its description, formula and process, wherein the description comprises a feature description and a function description, and the formula comprises a bill of materials table and several levels, and each level comprises a corresponding name within the level and its description, formula and process; the process comprises a production configuration and a process flow, and comprises a method for collecting data by a wizard for a machine adaptive system and an intelligent SOP, and comprises the following steps: S1: The user names the product through the wizard; S2: Guide users to describe the product, including functional description and feature description; S3: Guide the user to input the content of the product formula, including the components and parts of the product and their specifications, quantity, and supplier; S4: Guide the user to input the production configuration content in the process of the product. The production configuration includes the necessary conditions for producing the product, including production equipment, fixtures, raw materials and other human engineering data; S5: Guide the user to input the process flow content of the product. Each step in the process flow includes the feature description and function description of the process flow. S6: Perform data verification and integrate the user input content, and store the data into the database through the standard function interface.

2. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 1, characterized in that: The feature description in S2 also includes: S21: guiding the user to collect sensory data of touch, smell, hearing and taste related to production, for the purpose of improving and supplementing the data.

3. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 1, characterized in that: In S2, the user uses a wizard to describe the function and features of the product. The input of the function description includes describing its function in words; the input of the feature description includes one or more combinations of 3D drawings, engineering drawings, visual data, auditory data, olfactory data, taste data, and tactile data.

4. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 1, characterized in that: The S3 includes the following contents: using a wizard to guide the user to input the formula contents: the formula includes a bill of materials, the bill of materials includes several levels, each level includes several component structures, the highest level is the product structure, the lowest level is the part structure, and the composition of each level includes several next levels, until the next level is all part structures.

5. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 4, characterized in that: The S3 includes the following contents: the user's input of the recipe includes the description of each component structure, the recipe content and the process content.

6. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 4, characterized in that: The bill of materials includes three levels, including products, components and parts.

7. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 1, characterized in that: The S4 and S5 include the following contents: using a wizard to guide the user to configure the production configuration and process flow in the process: the production configuration includes the production conditions of the product and its components, including the production environment and production equipment; the S5 process flow includes a functional description and a characteristic description, and the input of the functional description of the process flow includes a text description of its functions, goals, and tasks; the input of the characteristic description of the process flow includes two categories: one category is derived from the characteristic data of parts, components or parts collected previously, which is used for parts comparison and quality verification; the other category comes from action videos: upload standardized operation videos to demonstrate the correct execution of the process.

8. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 1, characterized in that: The S6 includes the following sub-steps: S61: Obtain the user's input description of the product as the overall introduction of the product; S62: Obtaining the user's input content for the product formula, and using the input content in each classification as an introduction to the structure; S63: Obtain the user's input content on the product process, and intelligently divide and match the process flow and production configuration input content as an introduction to the corresponding process; S64: storing the data in a database and constructing a standard operating procedure based on the above content; S65: When the user runs the standard operating procedure, a technical guide is provided according to the user's operation.

9. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 8, characterized in that: In S63, when the user is running the standard operating procedure, if a new production equipment is added, the standard operating procedure automatically generates an action trajectory and task allocation adapted to the current workstation based on the sensor data of the machine, matches it with the production configuration content, and allocates the corresponding process flow to the production equipment.

10. The data collection method for a standard structured data set architecture for intelligent SOP according to claim 8, characterized in that: In S65, when the user runs the standard operating procedure, the standard operating procedure recognizes the user's features through motion capture, and provides feedback when the user's operation does not conform to the collected data of the current process flow.

Images