Sheet metal production engineering data automatic compilation system based on industrial automation

Abstract

The invention discloses an industrial automation-based sheet metal production engineering data automatic compilation system, which comprises a data acquisition module, a data preprocessing module, a feature extraction module, an engineering data compilation module, a database module and a data output module, the data acquisition module is in communication connection with the sheet metal raw material detection equipment, the product design terminal, the production equipment and the production data recording device and is used for acquiring sheet metal raw material parameters, product design parameters, production equipment parameters and historical production data. According to the sheet metal production engineering data automatic compilation system based on industrial automation, various data related to sheet metal production are comprehensively acquired through the data acquisition module, and accurate and effective data support is provided for engineering data compilation in combination with data preprocessing and feature extraction technologies; and by utilizing a machine learning model and a preset process rule in the engineering data compiling module, automatic compiling of sheet metal production engineering data is realized, and a large amount of manual intervention is not needed.

Description

An Automatic Data Compilation System for Sheet Metal Production Engineering Based on Industrial Automation Technical Field

[0001] This invention relates to the field of industrial automation production technology, specifically to an automatic data compilation system for sheet metal production engineering based on industrial automation. Background Technology

[0002] Sheet metal production is an important part of the machinery manufacturing industry, and is widely used in many industries such as aerospace, automobile manufacturing, electronic equipment, and construction. The sheet metal production process involves multiple steps such as blanking, bending, stamping, welding, and surface treatment. Each step requires corresponding engineering data as guidance, such as process cards, process cards, bills of materials, and equipment processing parameters. The accuracy and completeness of the engineering data directly affect the quality, production efficiency, and production cost of sheet metal products.

[0003] Currently, the compilation of sheet metal production engineering data mostly relies on manual labor. Technicians need to manually compile various engineering data based on product design drawings, raw material characteristics, production equipment conditions, and production experience. This manual compilation method has many drawbacks: First, the compilation efficiency is low, requiring a lot of time and manpower for complex sheet metal products; second, the compilation accuracy is difficult to guarantee, as data errors and omissions are prone to occur during manual compilation, leading to product quality defects and production interruptions; third, it is difficult to adapt to the needs of industrial automated production, as manually compiled engineering data cannot be synchronized to automated production equipment in real time, affecting the continuity of the production process and the level of automation; fourth, data reusability is poor, as engineering data from different products is difficult to share and reference effectively, resulting in repetitive work.

[0004] With the development of industrial automation technology, sheet metal production is gradually transforming towards intelligence and automation, which puts forward higher requirements for the efficiency and accuracy of engineering data compilation. In the existing technology, although some software tools have emerged to assist in the compilation of engineering data, most of them require a large amount of manual parameter input and adjustment, resulting in a low degree of automation and the inability to achieve full-process automatic compilation of engineering data. In addition, existing tools are unable to make full use of historical and real-time production data and cannot dynamically optimize engineering data according to the actual production situation, resulting in poor adaptability of engineering data.

[0005] Therefore, we propose an automatic data compilation system for sheet metal production engineering based on industrial automation to solve the problems mentioned above. Summary of the Invention

[0006] The purpose of this invention is to provide an automatic data compilation system for sheet metal production engineering based on industrial automation, so as to solve the problems of low efficiency, poor accuracy and low degree of automation in the compilation of sheet metal production engineering data in the prior art as mentioned in the background.

[0007] To achieve the above objectives, the present invention provides the following technical solution: an automatic data compilation system for sheet metal production engineering based on industrial automation, comprising a data acquisition module, a data preprocessing module, a feature extraction module, an engineering data compilation module, a database module, and a data output module. The data acquisition module is communicatively connected to sheet metal raw material testing equipment, a product design terminal, production equipment, and a production data recording device, and is used to collect sheet metal raw material parameters, product design parameters, production equipment parameters, and historical production data. The data preprocessing module is connected to the data acquisition module and is used to clean, denoise, and standardize the collected raw data. The feature extraction module is connected to... A data preprocessing module is connected to extract key features affecting the compilation of engineering data from the preprocessed data. The engineering data compilation module is connected to both the feature extraction module and the database module. Based on the extracted key features, combined with preset sheet metal production process rules and machine learning models, it automatically completes the compilation of sheet metal production engineering data. The database module stores the original data, preprocessed data, key feature data, preset process rules, machine learning model parameters, and the compiled engineering data. The data output module is connected to the engineering data compilation module and outputs the compiled engineering data to sheet metal production automation equipment or production management system.

[0008] Preferably, the data acquisition module includes a raw material parameter acquisition unit, a design parameter acquisition unit, an equipment parameter acquisition unit, and a historical data acquisition unit; the raw material parameter acquisition unit is used to acquire the material, thickness, tensile strength, yield strength, and chemical composition of sheet metal raw materials; the design parameter acquisition unit is used to acquire the dimensional parameters, shape parameters, precision requirements, and assembly requirements of sheet metal products; the equipment parameter acquisition unit is used to acquire the model, processing range, processing precision, operating speed, and energy consumption parameters of sheet metal production equipment; and the historical data acquisition unit is used to acquire engineering data, production efficiency data, product qualification rate data, and equipment failure data from historical production processes.

[0009] Preferably, the preprocessing process of the data preprocessing module includes: using an outlier detection algorithm to identify and remove outliers in the original data; using a smoothing filter algorithm to denoise the noisy data; and using a min-max normalization method to convert data of different dimensions to the same numerical range.

[0010] Preferably, the feature extraction module uses principal component analysis or random forest algorithm to extract key features, including key performance parameters of raw materials, key size parameters of products, key operating parameters of equipment, and key quality parameters of historical production.

[0011] Preferably, the engineering data compilation module includes a process rule storage unit, a model training unit, and a data compilation unit; the process rule storage unit is used to store preset sheet metal production process rules, including blanking process rules, bending process rules, stamping process rules, welding process rules, and surface treatment process rules; the model training unit is used to train and optimize the machine learning model based on historical data and extracted key features; the data compilation unit is used to automatically generate sheet metal production process cards, process cards, bills of materials, and equipment processing parameters based on the trained machine learning model and preset process rules.

[0012] Preferably, the machine learning model uses a BP neural network or a support vector machine, with the key features extracted as input and the various compilation parameters of the sheet metal production engineering data as output.

[0013] Preferably, the database module adopts a distributed database, which supports real-time data storage, query and update, and also has data backup and recovery functions.

[0014] Preferably, the data output module supports multiple data output formats, including XML format, Excel format and database interface format, and can transmit engineering data to production equipment or management system via wired or wireless communication.

[0015] Preferably, an automatic data compilation system for sheet metal production engineering based on industrial automation further includes a data verification module. The data verification module is connected to the engineering data compilation module and the database module respectively. It is used to compare and verify the compiled engineering data with historical qualified engineering data and preset standards. If the verification fails, it is fed back to the engineering data compilation module for re-compilation.

[0016] Preferably, an automatic data compilation system for sheet metal production engineering based on industrial automation also includes a visualization interaction module. The visualization interaction module is connected to the database module and is used to display various types of data, compilation process and compilation results. At the same time, it supports users to manually adjust preset process rules and model parameters.

[0017] Compared with the prior art, the beneficial effects of the present invention are: (1) The automatic compilation system for sheet metal production engineering data based on industrial automation comprehensively collects various types of data related to sheet metal production through the data acquisition module, and provides accurate and effective data support for engineering data compilation by combining data preprocessing and feature extraction technology; by utilizing the machine learning model and preset process rules in the engineering data compilation module, the automatic compilation of sheet metal production engineering data is realized without a large amount of manual intervention, which significantly improves the compilation efficiency and reduces labor costs; (2) The automatic compilation system for sheet metal production engineering data based on industrial automation removes outliers, removes noise and standardizes data through the data preprocessing module, which improves the data quality; by using a scientific feature extraction algorithm to select key features, combined with the trained and optimized machine learning model, the accuracy of engineering data compilation is improved, the error caused by manual compilation is reduced, and the quality of sheet metal products is improved. Quality and production stability; (3) The automatic compilation system for sheet metal production engineering data based on industrial automation adopts distributed storage technology in the database module, supports real-time storage and efficient query of a large amount of data, has data backup and recovery functions, and ensures data security and integrity; the data output module supports multiple formats and transmission methods, and can realize seamless connection between engineering data and automated production equipment and management system, meet the needs of industrial automated production, and improve the automation level and intelligence level of sheet metal production; (4) The automatic compilation system for sheet metal production engineering data based on industrial automation has an added data verification module that can perform double verification on the compiled engineering data (compare with historical qualified data and compare with standard requirements), further ensuring the accuracy of engineering data; the visualization interaction module makes it convenient for users to intuitively understand the data and compilation process, supports manual adjustment of parameters, and improves the flexibility and operability of the system. Detailed Implementation

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention provides a technical solution: an automatic data compilation system for sheet metal production engineering based on industrial automation, comprising a data acquisition module, a data preprocessing module, a feature extraction module, an engineering data compilation module, a database module, and a data output module. The data acquisition module is communicatively connected to sheet metal raw material testing equipment, a product design terminal, production equipment, and a production data recording device, and is used to collect sheet metal raw material parameters, product design parameters, production equipment parameters, and historical production data. Further, the data acquisition module includes a raw material parameter acquisition unit, a design parameter acquisition unit, an equipment parameter acquisition unit, and a historical data acquisition unit; the raw material parameter acquisition unit is used to collect the material, thickness, tensile strength, etc., of the sheet metal raw materials. Yield strength and chemical composition; the design parameter acquisition unit is used to collect the dimensional parameters, shape parameters, precision requirements, and assembly requirements of sheet metal products; the equipment parameter acquisition unit is used to collect the model, processing range, processing precision, operating speed, and energy consumption parameters of sheet metal production equipment; the historical data acquisition unit is used to collect engineering data, production efficiency data, product qualification rate data, and equipment failure data from historical production processes; specifically, the raw material parameter acquisition unit, through connection to equipment such as a metal material analyzer and thickness measuring instrument, collects the material of sheet metal raw materials (such as SPCC cold-rolled steel sheet) as SPCC, thickness as 2mm, tensile strength as 320MPa, yield strength as 190MPa, and chemical composition (C: 0.10%, S...). i: 0.05%, Mn: 0.45%); The design parameter acquisition unit, connected to the CAD design terminal, acquires the dimensional parameters (length 200mm, width 150mm, key hole diameter 10mm), shape parameters (including 3 90° bending angles), accuracy requirements (dimensional tolerance ±0.1mm), and assembly requirements (connection to the dashboard clip) of sheet metal products (such as automotive dashboard brackets); The equipment parameter acquisition unit, connected to the PLC controller, acquires the processing range (1500×3000mm), processing accuracy (±0.05mm), and operating speed (30m / min) of the laser cutting machine (model G3015), and the bending thickness (0-) of the bending machine (model WC67Y-160T / 4000). 16mm, bending angle 0-180°; The historical data acquisition unit connects to the production data management system to collect historical engineering data, production efficiency data (average production cycle 20min / piece), product qualification rate data (98.5%), and equipment failure data (average 2 failures per month) for this type of sheet metal product over the past year; The data preprocessing module connects to the data acquisition module to clean, denoise, and standardize the collected raw data; Further, the preprocessing process of the data preprocessing module includes: using outlier detection algorithms (such as Grubbs' criterion, box plot method) to identify and remove outliers in the raw data; and using smoothing filtering algorithms (such as moving average filtering, median filtering) to denoise the noisy data;The min-max standardization method is used to transform data of different dimensions to the numerical range of [0,1] or [-1,1]. The standardization formula is: x'=(x-x_min) / (x_max-x_min), where x is the original data, x_min is the minimum value, x_max is the maximum value, and x' is the standardized data. Specifically, the data preprocessing module processes the collected raw data: a box plot method is used to identify and remove one outlier (3.5mm, far exceeding the normal range of 1.9-2.1mm) in the raw material thickness data; a moving average filtering algorithm is used to process the laser-cut... The cutting machine operating speed data is denoised to smooth fluctuations. The min-max normalization method is used to convert data of different dimensions, such as thickness (2mm), tensile strength (320MPa), and processing speed (30m / min), to the range [0,1]. For example, thickness normalization results in 0.5, and tensile strength normalization results in 0.6. The feature extraction module is connected to the data preprocessing module to extract key features affecting the compilation of engineering data from the preprocessed data. Furthermore, the feature extraction module uses Principal Component Analysis (PCA) or Random Forest algorithm to extract key features. When using PCA, the key features are... By calculating the covariance matrix, solving for eigenvalues ​​and eigenvectors, principal components with variance contributions greater than a preset threshold are selected as key features. When using the random forest algorithm, the importance scores of each feature are calculated, and features with scores greater than a preset threshold are selected as key features. The extracted key features include key performance parameters of raw materials (such as thickness and tensile strength), key dimensional parameters of products (such as key hole diameter and bending angle), key operating parameters of equipment (such as processing speed and processing accuracy), and key quality parameters of historical production (such as pass rate and scrap rate). Specifically, the feature extraction module uses the random forest algorithm to extract key features. By calculating the importance scores of each feature, the top 8 features with the highest scores are selected: raw material thickness, tensile strength, key hole diameter of the product, bending angle, laser cutting machine processing accuracy, processing speed, historical pass rate, and bending thickness range of the bending machine, as key features affecting the compilation of engineering data. The engineering data compilation module is connected to the feature extraction module and the database module respectively. It is used to automatically complete the compilation of sheet metal production engineering data based on the extracted key features, combined with preset sheet metal production process rules and machine learning models. Furthermore, the engineering data compilation module includes a process rule storage unit, a model training unit, and a data compilation unit.The process rule storage unit stores preset sheet metal production process rules, including blanking process rules (such as selecting shearing, laser cutting, or plasma cutting methods based on material thickness and size), bending process rules (such as determining bending sequence and bending radius based on bending angle and material thickness), stamping process rules (such as determining stamping dies and stamping pressure based on stamping shape and precision), welding process rules (such as selecting arc welding, spot welding, or laser welding methods based on material properties), and surface treatment process rules (such as selecting spraying, electroplating, or phosphating treatment based on product requirements). The model training unit is used to train machine learning based on historical data and extracted key features. The learning model is trained and optimized. Cross-validation is used during training to avoid overfitting, and model performance is optimized by adjusting model parameters (such as learning rate, number of iterations, and number of hidden layer nodes). The data compilation unit automatically generates process cards, material lists, and equipment processing parameters for sheet metal production based on the trained machine learning model and preset process rules. Furthermore, the machine learning model uses either a backpropagation neural network (BP neural network) or a support vector machine (SVM). When using a BP neural network, the model includes an input layer, hidden layers, and an output layer. The number of nodes in the input layer is the same as the number of key features, and the number of nodes in the output layer is the same as the number of engineering data compilation parameters. The number of hidden layer nodes is consistent and determined through empirical formulas or trial and error. When using support vector machines, the data is mapped to a high-dimensional feature space by selecting an appropriate kernel function (such as a linear kernel, polynomial kernel, or radial basis kernel), constructing an optimal classification hyperplane or regression model to predict the compilation parameters of the engineering data. The input of the model is the extracted key features, and the output is the compilation parameters of the sheet metal production engineering data. Specifically, the process rule storage unit in the engineering data compilation module stores the preset sheet metal production process rules, such as the blanking process rule: when the material thickness is ≤2mm and the dimensional accuracy requirement is ±0.1mm, laser cutting is selected; bending... Process rules: For 90° bends with a material thickness of 2mm, the bending sequence is from the outside in, with a bending radius of 2mm. The model training unit trains a BP neural network model based on historical data from the past year and extracted key features. The model has 8 input layer nodes (number of key features), 12 hidden layer nodes (determined through trial and error), and 6 output layer nodes (2 process card parameters, 2 process card parameters, and 2 equipment processing parameters). A 5-fold cross-validation method is used to avoid overfitting. The learning rate is adjusted to 0.01, and the number of iterations is 1000. The prediction error of the trained model is less than 2%.Based on the trained BP neural network model and preset process rules, the data processing unit automatically generates process cards (process 1: laser cutting blanking, process 2: bending, process 3: drilling, process 4: surface spraying), process cards (laser cutting power 3000W, cutting speed 30m / min, bending pressure 120MPa), and bill of materials (SPCC steel plate 2mm×200mm×150mm). The system includes: 1 piece of coating material (0.5kg) and equipment processing parameters (laser cutting machine processing path, bending machine bending sequence); a database module for storing raw data, preprocessed data, key feature data, preset process rules, machine learning model parameters, and completed engineering data; furthermore, the database module uses a distributed database (such as Hadoop Distributed File System HDFS combined with HBase database) to support real-time data storage, querying, and updating, while also having data backup and recovery functions. The database module ensures data security and integrity through data backup strategies (such as scheduled full backups and incremental backups) and data recovery mechanisms (such as recovery based on backup files and recovery based on logs). Simultaneously, the database module supports data indexing to improve data query efficiency; specifically, the database module uses Hadoop Distributed File System HDFS combined with HBase database to store raw data, preprocessed data, key feature data, preset process rules, BP neural network model parameters, and completed engineering data. It ensures data security through scheduled full backups (every day at 3 AM) and incremental backups (every hour) and supports fast data querying; the data output module is connected to the engineering data compilation module to output the completed data. The engineering data is output to sheet metal production automation equipment or production management system. Furthermore, the data output module supports multiple data output formats, including XML, Excel, and database interface formats (such as JDBC and ODBC). Data transmission methods include wired communication (such as Ethernet and RS485 bus) and wireless communication (such as WiFi, Bluetooth, and 5G). The appropriate transmission method can be selected based on the production environment to transmit the completed engineering data to sheet metal production automation equipment (such as laser cutting machines, bending machines, and stamping machines) or production management system (such as MES system) in real time, achieving seamless integration between engineering data and the production process. Specifically, the data output module uses Ethernet communication to output the completed engineering data in XML format to automated production equipment such as laser cutting machines and bending machines, as well as the MES production management system. After receiving the data, the equipment automatically adjusts its parameters to prepare for production. Furthermore, an automatic compilation system for sheet metal production engineering data based on industrial automation also includes a data verification module. The data verification module is connected to both the engineering data compilation module and the database module. The verification process of the data verification module includes comparing the completed engineering data with historical qualified engineering data stored in the database and calculating the data similarity.The completed engineering data is compared with preset industry and enterprise standards to check whether the data meets the standard requirements. If the data similarity is lower than the preset threshold or does not meet the standard requirements, the verification fails. The data verification module feeds back the verification result to the engineering data compilation module, which adjusts the process rules or model parameters according to the feedback result and recompiles the engineering data. If the verification passes, the completed engineering data is stored in the database module and output through the data output module. Specifically, the data verification module compares the completed engineering data with historical qualified engineering data in the database, and the similarity is 96% (higher than the preset threshold of 90%). Compared with the enterprise standard, all parameters meet the requirements (e.g., laser cutting accuracy ±0.05mm meets the standard ±0.1mm), and the verification passes. Furthermore, an automatic compilation system for sheet metal production engineering data based on industrial automation also includes a visualization interaction module, which is connected to the database module. The visualization interaction module uses industrial configuration software (such as WinCC, KingSCADA) or... Web visualization technologies (such as ECharts and Highcharts) display various types of data (raw data, preprocessed data, key feature data), the compilation process (model training process, process rule matching process), and the compilation results (process cards, process cards, bill of materials) through charts (such as line charts, bar charts, pie charts), tables, flowcharts, etc. Simultaneously, the visualization interaction module allows users to manually adjust preset process rules and model parameters using input devices such as a mouse and keyboard. The adjusted data is synchronized to the database module in real time for use by the engineering data compilation module. Specifically, the visualization interaction module uses ECharts to display bar charts of key feature importance scores, flowcharts of engineering data compilation, and tables of completed process cards and process cards. Users can manually adjust laser cutting power parameters (from 3000W to 3200W) through the interface. The adjusted data is synchronized to the database in real time for use in subsequent data compilation. All content not described in detail in this specification belongs to existing technology known to those skilled in the art.

[0020] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An automatic data compilation system for sheet metal production engineering based on industrial automation, comprising a data acquisition module, a data preprocessing module, a feature extraction module, an engineering data compilation module, a database module, and a data output module, characterized in that: The data acquisition module is communicatively connected to the sheet metal raw material testing equipment, product design terminal, production equipment, and production data recording device, and is used to collect sheet metal raw material parameters, product design parameters, production equipment parameters, and historical production data; the data preprocessing module is connected to the data acquisition module and is used to clean, denoise, and standardize the collected raw data. The feature extraction module is connected to the data preprocessing module and is used to extract key features that affect the compilation of engineering data from the preprocessed data. The engineering data compilation module is connected to both the feature extraction module and the database module and is used to automatically complete the compilation of sheet metal production engineering data based on the extracted key features, combined with preset sheet metal production process rules and machine learning models. The database module is used to store the original data, preprocessed data, key feature data, preset process rules, machine learning model parameters, and the compiled engineering data. The data output module is connected to the engineering data compilation module and is used to output the compiled engineering data to sheet metal production automation equipment or production management system.

2. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 1, characterized in that: The data acquisition module includes a raw material parameter acquisition unit, a design parameter acquisition unit, an equipment parameter acquisition unit, and a historical data acquisition unit. The raw material parameter acquisition unit is used to collect the material, thickness, tensile strength, yield strength, and chemical composition of sheet metal raw materials. The design parameter acquisition unit is used to collect the dimensional parameters, shape parameters, precision requirements, and assembly requirements of sheet metal products. The equipment parameter acquisition unit is used to collect the model, processing range, processing precision, operating speed, and energy consumption parameters of sheet metal production equipment. The historical data acquisition unit is used to collect engineering data, production efficiency data, product qualification rate data, and equipment failure data from historical production processes.

3. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 1, characterized in that: The preprocessing process of the data preprocessing module includes: using an outlier detection algorithm to identify and remove outliers in the original data; using a smoothing filter algorithm to denoise the noisy data; and using the min-max normalization method to convert data of different dimensions to the same numerical range.

4. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 1, characterized in that: The feature extraction module uses principal component analysis or random forest algorithm to extract key features, including key performance parameters of raw materials, key size parameters of products, key operating parameters of equipment, and key quality parameters of historical production.

5. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 1, characterized in that: The engineering data compilation module includes a process rule storage unit, a model training unit, and a data compilation unit. The process rule storage unit stores preset sheet metal production process rules, including blanking process rules, bending process rules, stamping process rules, welding process rules, and surface treatment process rules. The model training unit trains and optimizes the machine learning model based on historical data and extracted key features. The data compilation unit automatically generates sheet metal production process cards, process cards, bills of materials, and equipment processing parameters based on the trained machine learning model and preset process rules.

6. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 5, characterized in that: The machine learning model uses a BP neural network or a support vector machine. The input of the model is the extracted key features, and the output is the various compilation parameters of the sheet metal production engineering data.

7. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 1, characterized in that: The database module adopts a distributed database, which supports real-time data storage, querying and updating, and also has data backup and recovery functions.

8. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 1, characterized in that: The data output module supports multiple data output formats, including XML, Excel, and database interface formats, and can transmit engineering data to production equipment or management systems via wired or wireless communication.

9. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 1, characterized in that: It also includes a data verification module, which is connected to the engineering data compilation module and the database module respectively. The data verification module is used to compare and verify the compiled engineering data with historical qualified engineering data and preset standards. If the verification fails, it is fed back to the engineering data compilation module for re-compilation.

10. The automatic data compilation system for sheet metal production engineering based on industrial automation according to claim 1, characterized in that: It also includes a visualization interaction module, which is connected to the database module to display various types of data, compilation process and compilation results, while also supporting users to manually adjust preset process rules and model parameters.