Intelligent monitoring and early warning method and system based on welding

Abstract

The invention discloses an intelligent monitoring and early warning method and system based on welding, and relates to the technical field of intelligent early warning, and the method specifically comprises the steps: obtaining the real-time physical data of a welding workshop; mapping the real-time physical data to a digital twinborn simulation model for real-time simulation, and identifying risk factors by using the digital twinborn simulation model and a neural network; and analyzing the risk factors, judging whether potential safety hazards causing injury to the personnel exist or not, and calling a preset early warning scheme based on a judgment result. The method for identifying the potential safety hazard comprises the following steps of: carrying out analogue simulation on the digital twin simulation model according to real-time physical data to obtain welding simulation data; and the welding simulation data are input into a hidden danger recognition model, dangerous factors are recognized and analyzed, and whether potential safety hazards causing harm to personnel exist or not is judged. According to the method, the physical data of the welding workshop are acquired in real time and mapped to the digital twinborn simulation model for simulation, so that the monitoring efficiency is improved, and the result accuracy is enhanced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent early warning technology, and more specifically to an intelligent monitoring and early warning method and system based on welding. Background Art

[0002] At present, welding technology plays an indispensable role in modern industrial manufacturing, and its application is everywhere, from building structures to automobile manufacturing, and aerospace. However, in the welding process, traditional monitoring methods often rely on manual inspection and offline detection, which is not only inefficient but also difficult to ensure real-time and accuracy. Especially in some complex or critical welding tasks, such as welding of large steel structures, pressure vessels, and piping systems, if defects are not discovered in time, it may lead to major safety hazards or even catastrophic accidents.

[0003] However, with the advancement of science and technology, although some online monitoring methods based on sensor technology and computer vision have been proposed and applied to welding process monitoring, these solutions still have limitations. On the one hand, the existing intelligent monitoring system focuses on the evaluation of welding quality, and the safety warning function of the welding environment is relatively weak; on the other hand, different welding processes and materials have huge differences in the requirements for monitoring parameters. The existing system lacks sufficient flexibility to adapt to the safety needs under various working conditions, and there is still room for improvement in data processing speed and warning response time.

[0004] Therefore, developing an intelligent system that can monitor the welding process in real time and has efficient early warning capabilities is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0005] In view of this, the present invention provides a welding-based intelligent monitoring and early warning method and system to overcome the above-mentioned defects.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] An intelligent monitoring and early warning method based on welding, the specific steps are:

[0008] Get real-time physical data of the welding workshop;

[0009] The real-time physical data is mapped into a digital twin simulation model for real-time simulation, and the digital twin simulation model and the neural network are used to identify risk factors; the risk factors are analyzed to determine whether there are potential safety hazards that may cause harm to personnel, and a preset early warning plan is called based on the determination result.

[0010] Optionally, the real-time physical data includes welding equipment operation data, material data, environmental data and operator data.

[0011] Optionally, the steps of constructing the digital twin simulation model are:

[0012] Acquire physical data of the welding workshop, and construct a welding process data set based on the physical data;

[0013] Based on the physical data of the workshop and the welding process data set, data modeling is performed for each physical entity to obtain an entity matrix including a multi-level structure and dynamic attribute changes;

[0014] Constructing a spatial layout matrix according to the spatial position of each of the physical entities;

[0015] constructing an environmental state matrix based on the welding process data set;

[0016] Obtain welding process constraints and manufacturing resource constraints information, automatically identify constraints through natural language processing technology, and generate a production constraint matrix;

[0017] Acquire twin modeling eigenvalues ​​according to the entity matrix and the spatial layout matrix;

[0018] Acquire optimized eigenvalues ​​according to the environmental state matrix and the production constraint matrix;

[0019] Based on the twin modeling eigenvalues ​​and the optimized eigenvalues, the digital twin simulation model is constructed.

[0020] Optionally, the steps of constructing the welding process data set are:

[0021] Preprocessing the physical data to generate preprocessed data;

[0022] Building a causal relationship model according to the preprocessed data, and screening safety influencing factors based on the causal relationship model;

[0023] Using a deep learning fusion model and a multimodal attention mechanism, the pre-processed data is fused based on the safety factors to generate time series data reflecting changes in the welding process state;

[0024] The time series data is classified and stored according to the welding process to construct the welding process data set.

[0025] Optionally, the preprocessing includes cleaning, denoising, standardization and time alignment.

[0026] Optionally, the step of identifying the potential safety hazard is:

[0027] The digital twin simulation model performs simulation according to the real-time physical data to obtain welding simulation data;

[0028] The welding simulation data is input into a hidden danger identification model to identify and analyze the risk factors and determine whether there are hidden safety hazards that may cause harm to personnel.

[0029] Optionally, the hidden danger identification model is constructed based on a recurrent neural network, and a multi-attention graph convolutional network module is introduced to extract the position and behavior characteristics of personnel in the welding process to assist in judging safety hazards.

[0030] An intelligent monitoring and early warning system based on welding, comprising:

[0031] Real-time data acquisition module, used to obtain real-time physical data of the welding workshop;

[0032] The safety hazard identification module is used to map the real-time physical data into the digital twin simulation model for real-time simulation, and use the digital twin simulation model and neural network to identify dangerous factors; and analyze the dangerous factors to determine whether there are safety hazards that may cause harm to personnel, and call a preset early warning plan based on the judgment result.

[0033] It can be seen from the above technical solutions that the present invention provides an intelligent monitoring and early warning method and system based on welding, which has the following beneficial effects compared with the prior art:

[0034] Real-time and accuracy: By acquiring the physical data of the welding workshop in real time and mapping it into the digital twin simulation model for simulation, it can provide almost instant data analysis and risk assessment, which not only improves the efficiency of monitoring, but also enhances the accuracy of the results.

[0035] Active prevention of safety hazards: The hazard identification model combines the recurrent neural network (RNN) and multi-attention graph convolutional network modules, which can effectively extract the position and behavior characteristics of personnel during the welding process, assist in judging safety hazards, and enable the system to issue warnings in time before accidents occur, thereby achieving active prevention of potential safety risks and ensuring the safety of workers.

[0036] Improve production efficiency: The introduction of the intelligent monitoring and early warning system reduces downtime caused by safety accidents, promotes the continuity and stability of production, and thus improves overall production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0038] Figure 1 A schematic diagram of the overall method flow structure provided by the present invention;

[0039] Figure 2 A schematic diagram of the construction process of the digital twin simulation model provided by the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0041] The embodiment of the present invention discloses an intelligent monitoring and early warning method based on welding, such as Figure 1 As shown, the specific steps are:

[0042] Step 1: Obtain real-time physical data of the welding workshop;

[0043] Step 2: Map the real-time physical data to the digital twin simulation model for real-time simulation, and use the digital twin simulation model and neural network to identify dangerous factors; analyze the dangerous factors to determine whether there are potential safety hazards that may cause harm to personnel, and call the preset early warning plan based on the judgment results.

[0044] In one embodiment, the real-time physical data includes welding equipment operation data, material data, environmental data, and operator data.

[0045] In one embodiment, the steps of constructing the digital twin simulation model are as follows: Figure 2 As shown, specifically:

[0046] Step 211, obtaining physical data of the welding workshop, and constructing a welding process data set based on the physical data;

[0047] Step 212: Perform data modeling for each physical entity based on the physical data of the workshop and the welding process data set to obtain an entity matrix including a multi-level structure and dynamic attribute changes;

[0048] Step 213: construct a spatial layout matrix according to the spatial position of each physical entity;

[0049] Step 214, constructing an environmental state matrix based on the welding process data set;

[0050] Step 215: Obtain welding process constraint and manufacturing resource constraint information, automatically identify constraint conditions through natural language processing technology, and generate a production constraint matrix;

[0051] Step 216: Obtain twin modeling eigenvalues ​​according to the entity matrix and the space layout matrix;

[0052] Step 217: Obtain optimized eigenvalues ​​according to the environment state matrix and the production constraint matrix;

[0053] Step 218: Build a digital twin simulation model based on the twin modeling eigenvalues ​​and optimized eigenvalues.

[0054] In one embodiment, the steps of constructing the welding process data set are:

[0055] Step 2111, preprocessing the physical data to generate preprocessed data;

[0056] Step 2112: construct a causal relationship model according to the preprocessed data, and screen safety influencing factors based on the causal relationship model;

[0057] Step 2113: using a deep learning fusion model and a multimodal attention mechanism, the pre-processed data is fused based on safety influencing factors to generate time series data reflecting changes in the welding process state;

[0058] Step 2114: Classify and store the time series data according to the welding process to construct a welding process data set.

[0059] Furthermore, the steps for constructing the welding process dataset are as follows:

[0060] Clean, denoise, standardize and time-align physical data to generate physical data in a unified format;

[0061] Based on the physical principles of the welding process and preprocessed physical data, a causal relationship model between variables in the welding process is constructed using a directed acyclic graph (DAG), and the impact of different parameters on welding safety is analyzed through Do-Calculus or potential result framework; thus, it is determined which variables have a significant impact on welding safety, thereby screening out safety influencing factors;

[0062] A deep learning fusion model is selected according to the safety influencing factors. The deep learning fusion model is used to extract features from the preprocessed data based on the screened safety influencing factors. A multimodal attention mechanism is introduced to dynamically generate weights of different modal features to achieve weighted fusion of cross-modal information, thereby generating high-precision time series data that reflects the overall state of the welding process.

[0063] The fused time series data are classified and stored in the database according to the welding process, and an effective data management mechanism is established to facilitate subsequent analysis and decision-making.

[0064] Furthermore, physical data includes production factor data, production behavior data and scheduling rule data, among which production factors include the geometric shape of the welding workshop, the size of the workshop, the quantity of equipment and materials, operators, etc.; production behavior data includes workshop status transfer data; scheduling rules are the basis for ensuring the orderly operation of the workshop, and they stipulate the interactions and constraints between various elements in the workshop.

[0065] Physical data is acquired using multiple types of sampling devices, including position sensors, welding parameter sensors (current, voltage, moment, torque) and environmental sensors (temperature, smoke concentration, harmful gas concentration, wind speed, light intensity, sound) and cameras deployed in the welding workshop.

[0066] The physical data collected include:

[0067] Welding equipment data: including the welding equipment’s main body data and the equipment’s operating parameters such as current, voltage, power, and other operating parameters, as well as welding speed, welding time, and other process parameters, which are used to reflect the basic state of the welding process and the equipment’s working conditions; welding environment data: welding parameters such as temperature, air quality (such as oxygen content, concentration of harmful gases), wind speed, light intensity, and sound during welding;

[0068] Material and workpiece data: including information such as the type, specification, quality, etc. of welding materials, as well as the size, shape, material, etc. of the workpiece;

[0069] Personnel operation data: identity information and behavior data of operators during the welding process, such as operation time, operation method, operation frequency, etc.

[0070] In one embodiment, during storage, based on the characteristics of the welding process, the fused data is classified according to different stages of the welding process (such as preheating, welding, cooling, etc.), and a welding process data set containing multiple characteristics (such as temperature, humidity, smoke concentration, sound characteristics, current / voltage fluctuations, etc.) is constructed and saved in a database.

[0071] In one embodiment, in step 212, data modeling is performed for each physical entity based on the physical data of the workshop to obtain an entity matrix including a multi-level structure and dynamic attribute changes, wherein the multi-level structure includes a basic level of physical entities and a more detailed component level, and the dynamic attribute changes include attributes that change over time, such as equipment wear and deformation of workpieces, specifically:

[0072] First, according to the physical data of the welding workshop, the basic attributes of each physical entity, such as name, type, location, size, etc., are determined, and a basic data model is created for each physical entity. These models will serve as the basic elements of the entity matrix; if the physical entity is more complex, its internal structure is further refined, and its key components are identified and modeled. The attributes of the components include component type, function, and position offset relative to the entity. In the entity matrix, sub-elements are created for these components, and a hierarchical relationship is established with the corresponding physical entities.

[0073] Secondly, identify the dynamic properties of physical entities, which may change over time or due to other factors, such as the degree of wear of equipment and the deformation of workpieces; capture the changing patterns of each dynamic property and assign corresponding data fields to each dynamic property for storing real-time or historical data.

[0074] Finally, the data models of base level, component level and dynamic attributes are integrated into a unified entity matrix.

[0075] In one embodiment, a real-time data update mechanism is provided to obtain the latest physical data from the IoT devices or other data sources in the workshop. When new data arrives, the corresponding fields in the entity matrix are updated to reflect the current status of the physical entity.

[0076] In one embodiment, in step 215, welding process constraints and manufacturing resource constraint information are obtained, and the constraints are automatically identified through natural language processing technology, and a production constraint matrix including intelligent constraint identification is constructed. The production constraint matrix can adjust the constraints according to the actual situation and emergencies in the production process.

[0077] Furthermore, the welding process constraints and manufacturing resource constraints are comprehensively collected. The process constraints include but are not limited to welding parameters, process sequence, and quality standards. The manufacturing resource constraints include but are not limited to equipment availability, personnel scheduling, and material supply conditions.

[0078] Apply natural language processing technology to pre-process and parse the collected constraint information, including text cleaning, word segmentation, part-of-speech tagging, named entity recognition, relationship extraction, and semantic understanding, so as to automatically identify and understand the meaning and logic of the constraint conditions;

[0079] Based on the processing results, a production constraint matrix is ​​constructed. The matrix includes rows representing different constraint conditions or resource items, and columns representing specific content or attributes of the constraint. The identified constraint condition information is filled into the matrix to form a production constraint matrix.

[0080] The constraints in the production constraint matrix are parsed and classified to clearly identify and store different types of constraints such as process parameter constraints, equipment resource constraints, and personnel scheduling constraints.

[0081] In one embodiment, in step 216, the entity matrix and the spatial layout matrix are input into a feature extraction module, and the twin modeling feature values ​​are automatically extracted using a deep learning algorithm;

[0082] In one embodiment, in step 217, the environment state matrix and the production constraint matrix are input into the optimization analysis module for intelligent analysis and optimization to obtain the optimized eigenvalues.

[0083] In one embodiment, the steps of identifying potential safety hazards are:

[0084] The digital twin simulation model performs simulation based on real-time physical data to obtain welding simulation data;

[0085] The welding simulation data is input into the hidden danger identification model to identify and analyze the dangerous factors and determine whether there are any safety hazards that may cause harm to personnel.

[0086] In one embodiment, the hidden danger identification model is constructed based on a recurrent neural network, and a multi-attention graph convolutional network module is introduced to extract the position and behavior characteristics of personnel in the welding process to assist in judging safety hazards.

[0087] Furthermore, the welding simulation data is fed into a pre-trained hazard identification model based on a recurrent neural network architecture and constructed by combining historical welding accident cases, expert domain knowledge, and multimodal dataset analysis; the multimodal dataset is constructed in the same way as the welding process dataset, based on data related to welding accidents;

[0088] Use the hidden danger identification model to deeply analyze the input welding simulation data, identify and evaluate the potential dangerous factors in the welding process, and determine whether there are safety hazards that may cause harm to personnel;

[0089] When the hidden danger identification model detects a safety hazard, it automatically generates safety warning information and immediately feeds this warning information back to the staff so that they can take timely and appropriate preventive measures;

[0090] All problems identified by the detection and their corresponding solutions are recorded in detail and saved in an industrial database. The data in the database serves as a new training set for subsequent optimization and improvement of the hidden danger identification model.

[0091] Furthermore, in a large welding workshop, it was detected that an important welding operation was about to be carried out. According to the requirements of the welding process, the system first determined the safe welding area and set the corresponding monitoring threshold. At the beginning of the welding operation, an operator was detected by video data to accidentally approach the safe welding area. The identity of the operator was identified, and the operator's behavior data was input into the digital twin simulation model for simulation. The simulation data was input into the hidden danger identification model. The hidden danger identification model judged that the operator's behavior may constitute a safety hazard based on the real-time physical data and simulation results. An alarm message was immediately issued, and the alarm location and the specific location of the operator were displayed through a visual interface. At the same time, the system automatically stopped the welding equipment to prevent potential accidents.

[0092] After receiving the alarm, the operator immediately realized his dangerous behavior and quickly evacuated to a safe area. The system then recorded the incident and saved the relevant data in the industrial database as a new training set for subsequent optimization and improvement of the hidden danger identification model.

[0093] On the other hand, this embodiment discloses an intelligent monitoring and early warning system based on welding, including:

[0094] Real-time data acquisition module, used to obtain real-time physical data of the welding workshop;

[0095] The safety hazard identification module is used to map real-time physical data to the digital twin simulation model for real-time simulation, and use the digital twin simulation model and neural network to identify dangerous factors; and analyze the dangerous factors to determine whether there are safety hazards that may cause harm to personnel, and call the preset early warning plan based on the judgment results.

[0096] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0097] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent monitoring and early warning method based on welding, characterized in that: The specific steps are: Get real-time physical data of the welding workshop; The real-time physical data is mapped into a digital twin simulation model for real-time simulation, and the digital twin simulation model and the neural network are used to identify risk factors; the risk factors are analyzed to determine whether there are potential safety hazards that may cause harm to personnel, and a preset early warning plan is called based on the determination result.

2. According to claim 1, a welding-based intelligent monitoring and early warning method is characterized in that: The real-time physical data includes welding equipment operation data, material data, environmental data and operator data.

3. According to claim 1, the intelligent monitoring and early warning method based on welding is characterized in that: The steps for constructing the digital twin simulation model are: Acquire physical data of the welding workshop, and construct a welding process data set based on the physical data; Based on the physical data of the workshop and the welding process data set, data modeling is performed for each physical entity to obtain an entity matrix including a multi-level structure and dynamic attribute changes; Constructing a spatial layout matrix according to the spatial position of each of the physical entities; constructing an environmental state matrix based on the welding process data set; Obtain welding process constraints and manufacturing resource constraints information, automatically identify constraints through natural language processing technology, and generate a production constraint matrix; Acquire twin modeling eigenvalues ​​according to the entity matrix and the spatial layout matrix; Acquire optimized eigenvalues ​​according to the environmental state matrix and the production constraint matrix; Based on the twin modeling eigenvalues ​​and the optimized eigenvalues, the digital twin simulation model is constructed.

4. According to claim 3, the intelligent monitoring and early warning method based on welding is characterized in that: The steps for constructing the welding process data set are: Preprocessing the physical data to generate preprocessed data; Building a causal relationship model according to the preprocessed data, and screening safety influencing factors based on the causal relationship model; Using a deep learning fusion model and a multimodal attention mechanism, the pre-processed data is fused based on the safety factors to generate time series data reflecting changes in the welding process state; The time series data is classified and stored according to the welding process to construct the welding process data set.

5. The intelligent monitoring and early warning method based on welding according to claim 4 is characterized in that: The preprocessing includes cleaning, denoising, normalization and time alignment.

6. The intelligent monitoring and early warning method based on welding according to claim 1 is characterized in that: The steps for identifying the potential safety hazard are: The digital twin simulation model performs simulation according to the real-time physical data to obtain welding simulation data; The welding simulation data is input into a hidden danger identification model to identify and analyze the risk factors and determine whether there are hidden safety hazards that may cause harm to personnel.

7. The intelligent monitoring and early warning method based on welding according to claim 6 is characterized in that: The hidden danger identification model is built based on a recurrent neural network, and a multi-attention graph convolutional network module is introduced to extract the position and behavior characteristics of personnel in the welding process to assist in judging safety hazards.

8. An intelligent monitoring and early warning system based on welding, characterized in that: include: Real-time data acquisition module, used to obtain real-time physical data of the welding workshop; The safety hazard identification module is used to map the real-time physical data into the digital twin simulation model for real-time simulation, and use the digital twin simulation model and neural network to identify dangerous factors; and analyze the dangerous factors to determine whether there are safety hazards that may cause harm to personnel, and call a preset early warning plan based on the judgment result.

Images