Steel laser cutting path intelligent planning system based on deep learning

Abstract

The invention discloses a steel laser cutting path intelligent planning system based on deep learning, and relates to the technical field of path planning. Comprising an equipment verification module for acquiring basic information of laser cutting equipment in a target production line to obtain basic information items, and verifying the basic information items in a verification mode to obtain equipment verification items; and the energy efficiency calculation module is used for acquiring the energy efficiency index of the laser cutting equipment to obtain an equipment index item. The steel analysis module carries out area grade division on steel on the basis of thickness distribution, the path matching module further accurately matches a steel grade area with an equipment grade, it is ensured that high-power and high-precision equipment preferentially processes thick plates and high-precision cutting tasks, and universal equipment is responsible for the adaptive area; equipment overload or idle capacity caused by'one-time cutting 'distribution is avoided, the cutting efficiency in unit time is remarkably improved, and the cutting quality defect caused by mismatching of equipment performance is reduced.

Description

Technical Field

[0001] This invention relates to the field of path planning technology, specifically to a deep learning-based intelligent path planning system for laser cutting of steel. Background Technology

[0002] Laser cutting of steel is a processing technology that uses a high-energy laser beam to precisely cut steel. It concentrates laser energy to locally melt or vaporize the steel, generating a high-intensity laser beam. After being focused by a focusing lens, it forms an extremely small spot, concentrating the laser energy on a single point on the steel surface. This causes the material to rapidly heat up to its melting or vaporization point, forming a cutting kerf. Before laser cutting, the trajectory and sequence of the laser beam along the steel surface are determined. The purpose of this process is to maximize cutting efficiency, ensure cutting quality, reduce material waste, and avoid errors and downtime during cutting; in other words, the cutting path is planned.

[0003] A laser cutting control system and method for lens assemblies, disclosed in patent publication number CN119596842A, ensures the integrity of information during the cutting process by acquiring cutting task data. Classifying cutting targets from the cutting task data effectively distinguishes different cutting needs, and the resulting target classification provides a basis for subsequent identification. Identifying cutting targets based on the target classification results accurately extracts the targets to be cut, and the generated cutting target data provides a foundation for feature matching. Feature matching of the lens assembly model based on the cutting target data generates candidate cutting target models, which provide a clear reference for subsequent arrangement processing. Applying arrangement constraints to the candidate cutting models ensures the feasibility of the cutting process.

[0004] When planning cutting paths for the above-mentioned and similar technical solutions, since there are more than one laser cutting device within a production line or a production plant, the allocation of cutting tasks can lead to situations where some devices cannot complete the tasks on time or in the required quantity due to factors such as laser power, equipment lifespan, and order priority. Furthermore, when cutting irregular structures from the same piece of steel at different locations, the cutting efficiency of different cutting devices can vary due to their own factors when the thickness of the irregular structure is different. Consequently, the planned cutting path may not be suitable for all laser cutting devices within a production line or a production plant. Summary of the Invention

[0005] The purpose of this invention is to provide a deep learning-based intelligent planning system for laser cutting paths of steel, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a deep learning-based intelligent path planning system for laser cutting of steel, comprising: Equipment verification module: Obtain basic information of laser cutting equipment in the target production line, obtain basic information items, verify the basic information items through verification methods, and obtain equipment verification items; Energy efficiency calculation module: Obtain the energy efficiency index of laser cutting equipment, obtain the equipment index item, build a dynamic profile library of equipment capabilities based on the equipment index item, create at least two energy efficiency levels, and assign the energy efficiency index of laser cutting equipment to the equipment level item based on the dynamic profile library of equipment capabilities. Steel analysis module: Obtains cutting information of the steel to be cut, including thickness information, to obtain steel information items. Based on the classification method, the steel information items are classified into grades to obtain steel grade items. Path matching module: Matches steel grade items with equipment grade items to obtain the path belonging range of the laser cutting equipment, obtains the first path information, obtains the cutting attribute information of the steel to be cut, performs path matching on the steel to be cut based on the cutting attribute information, and then allocates the first path. Based on the allocated cutting path and the laser cutting equipment with the corresponding equipment grade item, the module generates the path and obtains the second path information, thereby realizing the collaborative allocation path generation of multiple machines.

[0007] Furthermore, the method for obtaining the basic information items includes: Obtain the quantity information of laser cutting equipment in the target production line to obtain the target quantity item; Based on the target quantity item, the serial number is determined to obtain the number information corresponding to the laser cutting equipment. The model information, power information and equipment hardware information corresponding to each number information are obtained to obtain the basic information item.

[0008] Furthermore, the verification includes power fluctuation range verification, and the verification methods include: A first sensor is set based on the laser exit optical path of the laser cutting equipment. The first sensor includes a broadband photodiode array. Acquire the laser wavelength and output power data of the laser cutting equipment, set the sensor coverage data, and obtain the first verification item; The power data of the laser cutting equipment is acquired, a first sampling threshold is set, and the power fluctuation data of the laser cutting equipment is acquired based on the first sampling threshold to obtain the power fluctuation range data.

[0009] Furthermore, the verification also includes mechanical positioning error verification, and the verification methods include: A second sensor, including a high-resolution camera, is set up based on the laser cutting equipment. Laser trajectory data is acquired based on the second sensor to obtain the first information item. Obtain the working area information of the laser cutting equipment to obtain the working range item; place the ruler device based on the working range item; obtain the lateral and longitudinal errors of the laser cutting equipment to obtain the second information item. Based on a comprehensive judgment of the first and second information items, the dynamic error of the laser cutting equipment is obtained, and mechanical positioning error data is obtained.

[0010] Furthermore, the method for obtaining the device index term includes: Based on the equipment calibration items, an equipment energy efficiency index formula is created. The energy efficiency index formula includes power fluctuation range data and mechanical positioning error data. The energy efficiency index of laser cutting equipment is obtained based on the equipment energy efficiency index formula, thus yielding the equipment index item.

[0011] Furthermore, the method for obtaining the device level item includes: Set at least two energy efficiency index ranges to obtain at least three energy efficiency index range items, each of which corresponds to an energy efficiency level. Based on the equipment index item, the comparison data between the laser cutting equipment and the energy efficiency index range item is obtained to obtain the equipment energy efficiency comparison item. Based on the equipment energy efficiency comparison item, the energy efficiency level is assigned to obtain the equipment level item.

[0012] Furthermore, the partitioning method includes: Obtain the total area data of the steel to be cut, and get the steel area item; Based on the steel area item, a partitioning threshold is set to partition the steel area item, resulting in a steel area partition set; Based on the steel area division set, the vertical thickness data of the steel to be cut is obtained, and the steel thickness dataset corresponding to the steel area division item is obtained. Define at least two steel thickness ranges, split the steel thickness dataset based on the steel thickness ranges to obtain steel grade region division data, and determine the grade of different regions of steel based on the steel grade region division data to obtain steel grade items.

[0013] Furthermore, the method for obtaining the first path information includes: Based on the steel grade item, the steel to be cut is divided into grade regions to obtain the grade range item; Based on the matching results between the steel grade item and the equipment grade item, the grade range item is compared with the equipment grade item to obtain the working area of ​​the cutting equipment, thus obtaining the equipment working area item. The path to be cut is divided based on the working area of ​​the equipment to obtain the first path information.

[0014] Furthermore, the cutting attribute information includes cutting patterns, and the second path information is obtained in the following ways: Based on the cutting requirements, the cutting pattern information of the steel to be cut is obtained, and the steel pattern item is obtained; Based on the steel texture item, the texture pre-study is performed on the steel to be cut. Based on the first path information, the texture pre-study result is obtained in the texture display of the first path information, and then the working texture of different laser cutting equipment on the steel to be cut is obtained, and the second path information is obtained.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This deep learning-based intelligent path planning system for laser steel cutting quantifies and verifies core performance parameters of laser cutting equipment, such as power fluctuations and mechanical positioning errors, through an equipment calibration module. The acquired data is then used for deep learning calculations, with the results fed back as data. Combined with an energy efficiency calculation module, a dynamic profile library of equipment capabilities is constructed, classifying equipment into three energy efficiency levels: high, medium, and low. Simultaneously, a steel analysis module categorizes steel into regional levels based on thickness distribution, and a path matching module further precisely matches steel level regions with equipment levels. This ensures that high-power, high-precision equipment prioritizes thick plate and high-precision cutting tasks, while general-purpose equipment handles suitable regions. This avoids "one-size-fits-all" allocation that could lead to equipment overload or idle capacity, significantly improving cutting efficiency per unit time and reducing cutting quality defects caused by equipment performance mismatch.

[0016] Meanwhile, through the "equipment-steel-path" three-dimensional matching mechanism, dynamic path allocation is achieved for different areas of the same piece of steel. The steel analysis module divides the area and collects thickness data, and generates differentiated paths based on the cutting pattern requirements. The path matching module generates second path information based on the equipment hardware characteristics, so that the equipment only processes the area that is suitable for its own performance, reducing the idle movement distance and power waste of the equipment. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall process of the present invention; Figure 2 This is a schematic diagram of the basic information item acquisition process of the present invention; Figure 3 This is a schematic diagram of the basic information item distribution structure of the present invention; Figure 4 This is a schematic diagram of the power fluctuation range data acquisition process of the present invention; Figure 5 This is a schematic diagram illustrating the classification of steel information items according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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] Differences in power and lifespan among laser cutting equipment are significant factors influencing task allocation. Simply distributing tasks evenly across all machines often overlooks performance differences. For example, equipment with lower power or longer lifespan may be unable to handle high-precision or high-intensity cutting tasks, leading to decreased cutting quality or delayed completion. Furthermore, order priority must be considered. Some orders may be more urgent and require allocation to higher-performance equipment to ensure timely delivery. Therefore, task allocation must comprehensively consider both equipment performance characteristics and order priority to develop a reasonable allocation strategy. Secondly, cutting irregular structures at different locations on the same piece of steel places higher demands on cutting path planning. Thickness differences at different locations on the steel result in varying cutting efficiencies among different machines. For instance, some machines may be better suited for cutting thicker steel, while others are better suited for thinner steel. Failure to fully consider these factors and blindly adopting a uniform cutting path may lead to… Some cutting equipment is inefficient or even unable to complete the task. This application provides a deep learning-based intelligent path planning system for steel laser cutting. The system uses an equipment verification module to quantitatively verify core performance parameters of laser cutting equipment, such as power fluctuations and mechanical positioning errors. Combined with an energy efficiency calculation module, it constructs a dynamic image library of equipment capabilities, classifying equipment into three energy efficiency levels: high, medium, and low. Simultaneously, a steel analysis module classifies steel into regional levels based on thickness distribution, performs deep learning on various data, and outputs the results in the form of calculations. The path matching module further precisely matches the steel level regions with the equipment level, ensuring that high-power, high-precision equipment prioritizes thick plate and high-precision cutting tasks, while general-purpose equipment handles suitable regions. This avoids overload or idle capacity caused by a "one-size-fits-all" approach, significantly improving cutting efficiency per unit time and reducing cutting quality defects caused by equipment performance mismatch. Figure 1 As shown, it includes an equipment calibration module, an energy efficiency calculation module, a steel analysis module, and a path matching module.

[0020] Equipment verification module: Obtains basic information on laser cutting equipment in the target production line and performs verification to obtain equipment verification items.

[0021] It is important to note that, such as Figure 2 As shown, the first step is to obtain the basic information of the laser cutting equipment in the production line to obtain the basic information items. The methods for obtaining the basic information items include: obtaining the quantity information of the laser cutting equipment in the target production line to obtain the target quantity item; performing serial number calibration based on the target quantity item to obtain the number information corresponding to the laser cutting equipment; and obtaining the model information, power information, and equipment hardware information corresponding to each number information to obtain the basic information items.

[0022] In the specific implementation process, such as Figure 3 As shown, a laser cutting factory currently has six laser cutting machines. The target quantity is obtained, and these six laser cutting machines are numbered 1-6. The model information, power information, and hardware information corresponding to each number are obtained. Machines 1, 2, and 3 are high-power cutting units, all with the model information TRUMPF TruLaser Cell 7040, a power of 12kW (dynamically adjustable), and hardware information including a dual-ring cooling design, TruDisk 6000 smart head, and magnetic levitation linear motor. Machines 4 and 5 are high-precision cutting units, both with the model information BYSTRONIC ByStar Fiber 10kW, a power of 10kW (constant power), and hardware information including a BNC Navigator control system and a multispectral confocal thickness gauge. Machine 6 is a general-purpose cutting unit, model information HAN'S LASER HG-F1530, a power of 6kW (dynamically adjustable), and hardware information including Han's SmartCut. 4.0, industrial edge gateway IoT module, etc., to obtain basic information items.

[0023] It is important to note that, such as Figure 4 As shown, the basic information items are verified through a verification method to obtain the equipment verification items. The verification includes power fluctuation range verification. The verification method includes: setting a first sensor based on the laser exit optical path of the laser cutting equipment. The first sensor includes a broadband photodiode array; acquiring the laser wavelength data and output power data of the laser cutting equipment, setting sensor coverage data, and obtaining the first verification item; acquiring the power data of the laser cutting equipment, setting a first sampling threshold, and acquiring the power fluctuation data of the laser cutting equipment based on the first sampling threshold to obtain the power fluctuation range data.

[0024] In the specific implementation process, it is now necessary to verify the power fluctuation range of a laser cutting device. A broadband photodiode array is set on the output optical path of the laser cutting device to capture the beam energy distribution in real time. The laser wavelength data and output power data of the laser cutting device are obtained as 2μm and 12kW, respectively. The sensor coverage data is wavelength coverage of 1-11μm. The first sampling threshold is set to 10, that is, 10 sampling data of the laser cutting device are obtained. The power data of the laser cutting device are 12.01kW, 12.02kW, 12.03kW, 12.01kW, 12.04kW, 12.03kW, 12.02kW, 12.02kW, 12.03kW, and 12.01kW. At this time, the average power of the laser cutting device is 12.03kW, and the fluctuation data is 0.25%, thus obtaining the power fluctuation range data.

[0025] It should be noted that the verification also includes mechanical positioning error verification. The verification method includes: setting a second sensor based on the laser cutting equipment, the second sensor including a high-resolution camera, acquiring laser trajectory data based on the second sensor to obtain the first information item; acquiring the working area information of the laser cutting equipment to obtain the working range item; placing a scale device based on the working range item to acquire the lateral and longitudinal errors of the laser cutting equipment to obtain the second information item; and making a comprehensive judgment based on the first and second information items to obtain the dynamic error of the laser cutting equipment and obtain the mechanical positioning error data.

[0026] It is important to note that after obtaining power fluctuation range data and mechanical positioning error data, deep learning calculations are performed on the data based on artificial intelligence to calculate the energy efficiency index of the laser cutting equipment.

[0027] In the specific implementation process, it is now necessary to verify the mechanical positioning error of a laser cutting device. A high-resolution camera is set up to acquire the laser trajectory data of the laser cutting device. At this time, according to the working range of the laser cutting device, a scale is placed. Based on the laser trajectory data, the lateral error and longitudinal error of the laser cutting device are both 1%. At this time, the average error is 1%, and thus the mechanical positioning error data is obtained.

[0028] Energy efficiency calculation module: Obtains the energy efficiency index of laser cutting equipment, builds a dynamic profile library of equipment capabilities, and assigns levels to the energy efficiency index of laser cutting equipment.

[0029] It is important to note that the energy efficiency index of the laser cutting equipment is obtained to obtain the equipment index item. Based on the equipment index item, a dynamic profile library of equipment capabilities is constructed, and at least two energy efficiency levels are created. Based on the dynamic profile library of equipment capabilities, the energy efficiency index of the laser cutting equipment is assigned to the level to obtain the equipment level item.

[0030] It should be noted that the method for obtaining the equipment index item includes: creating an equipment energy efficiency index formula based on the equipment calibration items. The energy efficiency index formula includes power fluctuation range data and mechanical positioning error data; obtaining the energy efficiency calculation index of the laser cutting equipment based on the equipment energy efficiency index formula, thus obtaining the equipment index item.

[0031] Specifically, the energy efficiency index formula is: ,in Energy efficiency index, and The weights for power fluctuation and mechanical positioning error are both 0.5. and The data consist of power fluctuation range data and mechanical positioning error data. The energy efficiency of the laser cutting equipment is calculated using deep learning based on the energy efficiency index formula, and the calculation results are fed back in the form of data.

[0032] In the specific implementation process, it is now necessary to calculate the energy efficiency index of a certain laser cutting equipment. The power fluctuation range data of the laser cutting equipment is 0.25%, and the mechanical positioning error data is 1%. According to the energy efficiency index formula, the calculation result is 0.498 + 0.493 = 0.993.

[0033] It is important to note that three energy efficiency levels are created: high, medium, and low. Based on a dynamic image database of equipment capabilities, the energy efficiency index of the laser cutting equipment is assigned to these levels, resulting in equipment level items. The method for obtaining these equipment level items includes: setting three energy efficiency index ranges, resulting in three energy efficiency index range items: 0.9-1, 0.8-0.9, and ≤0.8. These energy efficiency index range items correspond to the energy efficiency levels, i.e., high level corresponds to the 0.9-1 range, medium level to the 0.8-0.9 range, and low level to the ≤0.8 range. Based on the equipment index items, comparison data between the laser cutting equipment and the energy efficiency index range items is obtained, resulting in equipment energy efficiency comparison items. Energy efficiency level assignment is then performed based on these comparison items, resulting in the equipment level items.

[0034] Steel Analysis Module: Obtains cutting information of the steel to be cut, classifies the cutting information into grades based on the classification method, and obtains the steel grade item.

[0035] It is important to note that the process involves obtaining cutting information for the steel to be cut, including thickness information, to obtain steel information items. These steel information items are then categorized into grades based on a specific method, resulting in steel grade items. This categorization method includes: obtaining the overall area data of the steel to be cut to obtain steel area items; dividing the steel area items based on a 1cm threshold to obtain a steel area categorization set; obtaining the vertical thickness data of the steel to be cut based on the steel area categorization set to obtain a steel thickness dataset corresponding to the steel area categorization items; defining three steel thickness ranges and splitting the steel thickness dataset based on these ranges to obtain steel grade region categorization data; and then determining the grade of different regions of the steel based on this steel grade region categorization data to obtain steel grade items.

[0036] In the specific implementation process, such as Figure 5 As shown, we now need to classify a piece of steel into grades. The total area of ​​the steel is 1 square meter, yielding steel information items. Based on a set classification threshold, the steel is divided into one hundred blocks, resulting in a steel area classification set. Then, the vertical thickness data of each steel area classification set is obtained, resulting in a steel thickness dataset. The thickness of the steel is 3mm for 60% of its area, 6mm for 20%, and 12mm for 20%. Three steel thickness ranges are defined: Grade 1, Grade 2, and Grade 3, corresponding to thicknesses of 0-5mm, 5mm-10mm, and above 10mm, respectively. Based on these thickness ranges, the steel thickness dataset is further split, resulting in Grade 1 having 60% of its area as Grade 1, Grade 2 as Grade 2, and Grade 3 as Grade 3.

[0037] Path matching module: Matches the steel grade item with the equipment grade item to obtain the path belonging range of the laser cutting equipment and obtain the first path information; It should be noted that the method for obtaining the first path information includes: dividing the steel to be cut into grade regions based on the steel grade item to obtain the grade range item; comparing the grade range item with the equipment grade item based on the matching result of the steel grade item and the equipment grade item to obtain the working area of ​​the cutting equipment to obtain the equipment working area item; and dividing the steel to be cut into paths based on the equipment working area item to obtain the first path information.

[0038] In the specific implementation process, there are two laser cutting machines on a certain production line, namely machine A and machine B. The energy efficiency index of machine A is calculated to be 0.95 and that of machine B is 0.85 using the equipment energy efficiency index formula. At this time, it is necessary to classify a piece of steel. The total area of ​​the steel is 1㎡. The thickness of the steel is 3mm for 80% of the area and 7mm for 20% of the area. According to the set thickness range, the steel is classified as level 1 for 80% of the area and level 2 for 20% of the area. The working area of ​​the cutting equipment is obtained by comparing the equipment level with the grade range. The working area of ​​machine A is 80% and that of machine B is 20%, thus obtaining the first path information.

[0039] Obtain the cutting attribute information of the steel to be cut, assign a path to the first path, and obtain the second path information; It is important to note that the process involves obtaining the cutting attribute information of the steel to be cut, performing path matching based on the cutting attribute information, allocating the first path, generating the second path information based on the allocated cutting path and the laser cutting equipment corresponding to the equipment level, thus realizing collaborative path generation by multiple machines.

[0040] It should be noted that the cutting attribute information includes the cutting pattern. The second path information is obtained in the following ways: based on the cutting requirements, the cutting pattern information of the steel to be cut is obtained to obtain the steel pattern item; based on the steel pattern item, the pattern is pre-simulated on the steel to be cut; based on the first path information, the pattern display of the pattern pre-simulation result in the first path information is obtained, and then the working pattern of different laser cutting equipment on the steel to be cut is obtained, thus obtaining the second path information.

[0041] Specifically, different cutting requirements necessitate different patterns. Based on these requirements, the cutting pattern information of the steel to be cut is obtained, and a pattern simulation is performed on the steel. Since the first path information has already divided the steel into at least two parts, each corresponding to a different laser cutting device, the cutting pattern is simulated based on the first path information, and then cut using the corresponding laser cutting device. This yields the cutting path of the laser cutting device, i.e., the second path information. This achieves a collaborative path generation effect across multiple machines, allowing different laser cutting devices to cut patterns at different locations on the same piece of steel.

[0042] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended embodiments and their equivalents.

Claims

1. A deep learning-based intelligent path planning system for laser cutting of steel, comprising: Equipment verification module: Obtain basic information of laser cutting equipment in the target production line, obtain basic information items, verify the basic information items through verification methods, and obtain equipment verification items; Its features are: Energy efficiency calculation module: Obtain the energy efficiency index of laser cutting equipment, obtain the equipment index item, build a dynamic profile library of equipment capabilities based on the equipment index item, create at least two energy efficiency levels, and assign the energy efficiency index of laser cutting equipment to the equipment level item based on the dynamic profile library of equipment capabilities. Steel analysis module: Obtains cutting information of the steel to be cut, including thickness information, to obtain steel information items. Based on the classification method, the steel information items are classified into grades to obtain steel grade items. Path matching module: Matches steel grade items with equipment grade items to obtain the path belonging range of the laser cutting equipment, obtains the first path information, obtains the cutting attribute information of the steel to be cut, performs path matching on the steel to be cut based on the cutting attribute information, and then allocates the first path. Based on the allocated cutting path and the laser cutting equipment with the corresponding equipment grade item, the module generates the path and obtains the second path information, thereby realizing the collaborative allocation path generation of multiple machines.

2. The intelligent planning system for laser cutting path of steel based on deep learning according to claim 1, characterized in that: The methods for obtaining the basic information items include: Obtain the quantity information of laser cutting equipment in the target production line to obtain the target quantity item; Based on the target quantity item, the serial number is determined to obtain the number information corresponding to the laser cutting equipment. The model information, power information and equipment hardware information corresponding to each number information are obtained to obtain the basic information item.

3. The intelligent planning system for laser cutting path of steel based on deep learning according to claim 1, characterized in that: The verification includes power fluctuation range verification, and the verification methods include: A first sensor is set based on the laser exit optical path of the laser cutting equipment. The first sensor includes a broadband photodiode array. Acquire the laser wavelength and output power data of the laser cutting equipment, set the sensor coverage data, and obtain the first verification item; The power data of the laser cutting equipment is acquired, a first sampling threshold is set, and the power fluctuation data of the laser cutting equipment is acquired based on the first sampling threshold to obtain the power fluctuation range data.

4. The intelligent planning system for laser cutting path of steel based on deep learning according to claim 3, characterized in that: The verification also includes mechanical positioning error verification, and the verification methods include: A second sensor, including a high-resolution camera, is set up based on the laser cutting equipment. Laser trajectory data is acquired based on the second sensor to obtain the first information item. Obtain the working area information of the laser cutting equipment to obtain the working range item; place the ruler device based on the working range item; obtain the lateral and longitudinal errors of the laser cutting equipment to obtain the second information item. Based on a comprehensive judgment of the first and second information items, the dynamic error of the laser cutting equipment is obtained, and mechanical positioning error data is obtained.

5. The intelligent planning system for laser cutting path of steel based on deep learning according to claim 4, characterized in that: The method for obtaining the equipment index item includes: Based on the equipment calibration items, an equipment energy efficiency index formula is created. The energy efficiency index formula includes power fluctuation range data and mechanical positioning error data. The energy efficiency index of laser cutting equipment is obtained based on the equipment energy efficiency index formula, thus yielding the equipment index item.

6. The intelligent planning system for laser cutting path of steel based on deep learning according to claim 1, characterized in that: The method for obtaining the equipment level item includes: Set at least two energy efficiency index ranges to obtain at least three energy efficiency index range items, each of which corresponds to an energy efficiency level. Based on the equipment index item, the comparison data between the laser cutting equipment and the energy efficiency index range item is obtained to obtain the equipment energy efficiency comparison item. Based on the equipment energy efficiency comparison item, the energy efficiency level is assigned to obtain the equipment level item.

7. The intelligent planning system for laser cutting path of steel based on deep learning according to claim 1, characterized in that: The partitioning method includes: Obtain the total area data of the steel to be cut, and get the steel area item; Based on the steel area item, a partitioning threshold is set to partition the steel area item, resulting in a steel area partition set; Based on the steel area division set, the vertical thickness data of the steel to be cut is obtained, and the steel thickness dataset corresponding to the steel area division item is obtained. Define at least two steel thickness ranges, split the steel thickness dataset based on the steel thickness ranges to obtain steel grade region division data, and determine the grade of different regions of steel based on the steel grade region division data to obtain steel grade items.

8. The intelligent planning system for laser cutting path of steel based on deep learning according to claim 1, characterized in that: The method for obtaining the first path information includes: Based on the steel grade item, the steel to be cut is divided into grade regions to obtain the grade range item; Based on the matching results between the steel grade item and the equipment grade item, the grade range item is compared with the equipment grade item to obtain the working area of ​​the cutting equipment, thus obtaining the equipment working area item. The path to be cut is divided based on the working area of ​​the equipment to obtain the first path information.

9. The intelligent planning system for laser cutting path of steel based on deep learning according to claim 1, characterized in that: The cutting attribute information includes cutting patterns, and the second path information is obtained in the following ways: Based on the cutting requirements, the cutting pattern information of the steel to be cut is obtained, and the steel pattern item is obtained; Based on the steel texture item, the texture pre-study is performed on the steel to be cut. Based on the first path information, the texture pre-study result is obtained in the texture display of the first path information, and then the working texture of different laser cutting equipment on the steel to be cut is obtained, and the second path information is obtained.

Images