Automatic parameter adjustment system and method based on visual identification and 5G communication

Abstract

The invention discloses an automatic parameter adjustment system and method based on visual identification and 5G communication, and belongs to the field of automatic parameter adjusting.The automatic parameter adjustment method comprises the steps that a virtual model is generated after shape information reading is conducted on manufacturing raw materials, simulation manufacturing is conducted on the virtual model, and a tool bit track generated in the simulation manufacturing process is recorded; synchronously simulating manufacturing in actual manufacturing, calculating the position difference of the tool bit at the same moment, and taking a result as an error value; setting a judgment error threshold value, adding a judgment set, adding the elements meeting the judgment error threshold value into the judgment set, and judging an error trend according to the adding time of the elements in the judgment set; by calculating the position difference of the simultaneous nicking tool bit, the error calculation is remarkably improved, so that the error compensation calculation is improved, the error is reduced, the manufacturing precision is improved, whether the manufacturing condition of the next product is met or not is predicted, the situation of unqualified manufacturing is effectively avoided, and the improvement of the manufacturing precision and the cost control of product manufacturing are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic parameter adjustment, and specifically to an automatic parameter adjustment system and method based on visual recognition and 5G communication. Background Art

[0002] Automation technology can significantly improve production efficiency, reduce costs, enhance product quality and innovation capabilities, while also solving labor shortages and reducing the pressure of physical labor. In industrial production, automation technology applications include production line automation, robotics, automated warehousing systems, and intelligent control systems. The application of these technologies can greatly improve production efficiency, reduce manpower input, and reduce quality problems caused by human operational errors. In smart homes, automation is achieved by using sensors and programming routines to control lighting, thermostats, and security systems. Energy is saved, safety is enhanced, and convenience is improved by replacing manually managed tasks with automation. With the continuous development and application of automation technology, automation technology has brought more opportunities and development space.

[0003] In automation technology, in the field of CNC machine tools, the accuracy of parameters determines the accuracy of manufactured products. Conventional CNC machine tool manufacturing generally detects the errors of manufactured products, and then calculates the compensation based on data analysis. The compensation results are used to adjust the CNC machine tools to improve the accuracy of manufactured products. The more accurate the error detection data, the more accurate the calculated compensation value, and the higher the accuracy of the manufactured product. How to detect errors in product manufacturing projects more carefully and accurately, and reduce the problem of unqualified manufacturing accuracy and increased manufacturing costs, has become a top priority. Summary of the invention

[0004] The purpose of the present invention is to provide an automatic parameter adjustment system and method based on visual recognition and 5G communication to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: an automatic parameter adjustment method based on visual recognition and 5G communication;

[0006] The automatic parameter adjustment method comprises the following steps:

[0007] Step S1, reading the morphological information of the manufacturing raw material through visual recognition, the morphological information also includes the spatial position of the manufacturing raw material, and constructing a model based on the read morphological information of the manufacturing raw material;

[0008] Step S2, performing calculation and analysis based on the model morphology constructed in step S1 and the target product morphology, and simulating manufacturing to obtain the tool head trajectory of simulated manufacturing and output it;

[0009] Step S3, receiving the simulated tool head trajectory outputted in step S2 to control the actual tool head movement, actually manufacturing the manufacturing material, and recording and storing the actual manufactured tool head trajectory;

[0010] Step S4, extracting the tool head trajectory obtained by the simulated manufacturing according to the time node and comparing and calculating the motion trajectory of the tool head in the actual manufacturing, and using the calculation result as the error value generated by the manufacturing raw material in the manufacturing process, the manufacturing node includes the manufacturing start node, the manufacturing middle node and the manufacturing end node;

[0011] Step S5, calculating and analyzing the error values ​​of unqualified products in the historical manufacturing data, obtaining the minimum error value of unqualified manufacturing, and setting it as the error threshold;

[0012] Step S6, set an error judgment threshold, compare the error value generated by the manufacturing raw materials during the manufacturing process with the error judgment threshold, add the error value that meets the error judgment threshold to the judgment set, judge whether the error value generated by the next manufacturing raw material is less than the error threshold, and predict whether the error value generated by the manufacturing raw materials after the manufacturing is completed meets the error condition for the next manufacturing.

[0013] Furthermore, visual recognition is used to read the morphological information of the manufacturing raw materials, the read image information is converted into data information, the data information is preprocessed and feature extracted, and the processed data is described through reverse features to generate a virtual model in the computer with the same proportions as the manufacturing raw materials. The model produced by the computer makes it easier to analyze the parameters and parameter control required for actual manufacturing.

[0014] Furthermore, the manufacturing material morphology information obtained in step S1 is processed and reversely described to generate a virtual model. The morphology and spatial position of the model are consistent with the actual manufacturing material. Simulation manufacturing is performed according to the morphology of the target product, and the trajectory of the tool head during the simulation manufacturing process is obtained and output.

[0015] A spatial rectangular coordinate system is established, and the spatial position of the simulated tool head corresponding to the time node is extracted and output according to the corresponding manufacturing time node.

[0016] Furthermore, the motion trajectory of the tool head in the actual manufacturing process is controlled by the simulated tool head trajectory output by step S2, the manufacturing raw materials are processed according to the specified trajectory, a spatial rectangular coordinate system is established, and the spatial position of the tool head corresponding to the manufacturing time node is recorded and stored. At the same time node, the position difference between the actual tool head and the simulated tool head is taken as the error. The position of the simulated tool head from the origin and the position of the actual tool head from the origin are calculated, and the absolute value of the difference between the two values ​​is taken as the error value; and the historical data is statistically analyzed and calculated, the error values ​​of unqualified products are statistically analyzed and calculated, and the minimum error value of unqualified manufacturing is taken as the manufacturing error threshold, which is set as the error threshold

[0017] Furthermore, the trajectory of the tool head is calculated as follows:

[0018] The position of the tool head trajectory in the spatial rectangular coordinate system is recorded, and the node at the start of manufacturing is set to T. The tool head manufactured virtually by the model is located at point M1 in the spatial rectangular coordinate system at time T, with coordinates (X1, Y1, Z1). The tool head actually manufactured is located at point M2 in the spatial rectangular coordinate system at time T, with coordinates (X2, Y2, Z2). The position difference between point M1 and point M2 at time T is α, and the calculation process of the difference α is as follows:

[0019] The calculation formula from a point to the origin in a spatial rectangular coordinate system is as follows:

[0020]

[0021] Where d is the distance from the point to the origin, and xyz represents the X-axis position, Y-axis position, and Z-axis position of the point in the spatial rectangular coordinate system respectively;

[0022] The position of point M1 from the origin is calculated as:

[0023]

[0024] The position of point M2 from the origin is calculated as:

[0025]

[0026] At time T, the position difference α between point M1 and point M2 is calculated as follows:

[0027] α=|d M1 -d M2 |

[0028] Further limit α, the position difference α between point M1 and point M2 at any time node is less than the error threshold

[0029] The error value of manufacturing raw materials after manufacturing is calculated using the following formula:

[0030]

[0031] in, is the error value of the manufacturing material after the manufacturing is completed, n is the number of position difference values ​​α calculated during the manufacturing process of the manufacturing material, and X represents the value of the i-th element in the set of position difference values ​​α. Among them, any time node appears If the current manufacturing raw materials do not meet the error range, the manufacturing will fail.

[0032] Assume the error judgment threshold is Will continue to grow to the error threshold The error value data will be counted and will continue to grow to the error threshold The minimum error value is set as the error judgment threshold The value reaches the error judgment threshold When , the subsequent error values ​​are counted and the error judgment threshold is reached continuously The error value is set as the dangerous error data;

[0033] Create an error set Q, the error values ​​for completing manufacturing using manufacturing materials is the element of error set Q, and the error value of each manufacturing raw material after manufacturing is completed Add the error set Q in sequence and set the corresponding subscript position for each element. When each element is added to the error set Q, it is consistent with the error judgment threshold. Make comparisons;

[0034] Create a judgment set, set the judgment set to The error set Q is greater than the error judgment threshold The elements and corresponding subscript positions are added to the judgment set, and the error of manufacturing raw materials after manufacturing is recorded as T a , product manufacturing error T a Error judgment threshold Compare, greater than the error judgment threshold Add the elements of to the judgment set Medium, less than the error judgment threshold Elements, first set the element to K and then add it to the set In, when the collection When the elements added to are not K continuously, it is judged that the error of the products currently manufactured continuously increases; when the set When the element added is K continuously, it is judged that the manufacturing error is within the error range.

[0035] Further, an automated parameter adjustment system based on visual recognition and 5G communication includes a morphology recognition module, a 5G transmission module, a simulation manufacturing module, a numerical control module and an error module;

[0036] The shape recognition module includes a visual recognition unit and a data sorting unit; the visual recognition unit is used to recognize and read the initial shape of the manufacturing raw material, transmit the read information to the data sorting unit, analyze the image through the data sorting unit, and sort the analyzed data;

[0037] The 5G transmission module includes a data acceleration unit and a data loading unit; the data acceleration unit is used to slice and compress data to speed up the output speed, and the data loading unit increases the amount of data transmitted at the same time to achieve multiple data transmission at the same time. The 5G transmission module is used for data transmission in each module.

[0038] The simulation manufacturing module includes a 3D model generation unit and a manufacturing simulation unit; the 3D model generation unit uses the information read and sorted by the shape recognition module to perform modeling in the same proportion to generate a model consistent with the real object; the manufacturing simulation unit uses the CNC model to manufacture the manufacturing raw material model until a complete product model is generated. The simulation manufacturing and actual manufacturing are carried out simultaneously, and the tool head trajectory is consistent;

[0039] The CNC module includes a CNC control unit and a control motion unit; the CNC control unit contains calculated instructions, which drive the cutter head and the motor through computer instructions; the control motion unit executes the corresponding computer instructions to run the cutter head to manufacture the raw materials to form products, and completes the manufacturing control of the required morphology and structure of the product through the speed, position and force of the cutter head.

[0040] The error module includes a spatial coordinate unit, an error calculation unit, an error set Q unit and an error judgment unit; the spatial coordinate unit is used to record the position of the tool head at time T, and records the spatial coordinate position of the tool head every T time; the error calculation unit calculates the position of the tool head in simulated manufacturing and the position of the tool head in actual manufacturing at the same time T, and the absolute value of the difference between the positions is used as the error value; the error set Q unit performs a collective calculation of the error α in the finished product manufacturing process to obtain the error of the actual finished product manufacturing The error judgment unit is used to meet the judgment threshold Add the elements of , which is less than the judgment threshold First set it to K and then add it. When the number of elements added to is K consecutively, it is considered a reasonable error. When the element added is not K continuously, it is judged that the manufacturing conditions do not meet the next manufacturing and the parameters need to be readjusted.

[0041] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention reads the morphological information of the manufacturing raw materials, produces a virtual model by reverse feature description, analyzes and calculates the model and the designed product, simulates the manufacturing, records and stores the simulated tool head trajectory generated in the simulated manufacturing, performs synchronous simulation manufacturing again in the actual manufacturing process, calculates the position difference of the tool head position in the same time period in the spatial rectangular coordinate system, uses the calculation result as the error value, sets a judgment error threshold for the error value, adds the number reaching the error judgment threshold to the judgment set based on the time when the manufacturing raw materials are completed, judges the size of the manufacturing error value by judging the time when new elements are added to the set, and calculates the error of the position difference between the simulated tool head and the actual tool head at the same time, so that the error calculation can be more significantly improved, thereby improving the numerical control error compensation, reducing the error value, and improving the precision of product manufacturing, performing a threshold control and judgment threshold setting on the error value generated in the manufacturing process, setting the time when the elements of the new set are added to the judgment set, and predicting whether the equipment meets the manufacturing conditions of the next product, effectively avoiding the occurrence of unqualified manufacturing, thereby achieving the improvement of numerical control precision and cost control of product manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0043] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0044] 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.

[0045] See also Figure 1 The present invention provides a technical solution: an automatic parameter adjustment method based on visual recognition and 5G communication, which specifically includes the following steps:

[0046] Step S1, reading the morphological information of the manufacturing raw material through visual recognition, the morphological information also includes the spatial position of the manufacturing raw material, and constructing a model based on the read morphological information of the manufacturing raw material;

[0047] Step S2, performing calculation and analysis based on the model morphology constructed in step S1 and the target product morphology, and simulating manufacturing to obtain the tool head trajectory of simulated manufacturing and output it;

[0048] Step S3, receiving the simulated tool head trajectory outputted in step S2 to control the actual tool head movement, actually manufacturing the manufacturing material, and recording and storing the actual manufactured tool head trajectory;

[0049] Step S4, extracting the tool head trajectory obtained by the simulated manufacturing according to the time node and comparing and calculating the motion trajectory of the tool head in the actual manufacturing, and using the calculation result as the error value generated by the manufacturing raw material in the manufacturing process, the manufacturing node includes the manufacturing start node, the manufacturing middle node and the manufacturing end node;

[0050] Step S5, calculating and analyzing the error values ​​of unqualified products in the historical manufacturing data, obtaining the minimum error value of unqualified manufacturing, and setting it as the error threshold;

[0051] Step S6, set an error judgment threshold, compare the error value generated by the manufacturing raw materials during the manufacturing process with the error judgment threshold, add the error value that meets the error judgment threshold to the judgment set, judge whether the error value generated by the next manufacturing raw material is less than the error threshold, and predict whether the error value generated by the manufacturing raw materials after the manufacturing is completed meets the error condition for the next manufacturing.

[0052] Use visual recognition to read the morphological information of the manufacturing raw materials, convert the read image information and spatial position information into data information, pre-process and extract features of the data information, and generate a virtual model with the same proportion as the manufacturing raw materials in the computer through reverse feature description of the processed data;

[0053] Perform simulation manufacturing according to the morphology of the target product, obtain the trajectory of the tool head during the simulation manufacturing process, and output the tool head trajectory. According to the output tool head trajectory, control the movement trajectory of the tool head during the actual manufacturing process, complete the processing of the manufacturing raw materials according to the specified trajectory movement, establish a spatial rectangular coordinate system, and record and store the spatial position of the tool head corresponding to the manufacturing time node;

[0054] The position of the tool head trajectory in the spatial rectangular coordinate system is recorded, and the node at the start of manufacturing is set to T. The tool head manufactured virtually by the model is located at point M1 in the spatial rectangular coordinate system at time T, with coordinates (X1, Y1, Z1). The tool head actually manufactured is located at point M2 in the spatial rectangular coordinate system at time T, with coordinates (X2, Y2, Z2). The position difference between point M1 and point M2 at time T is α, and the calculation process of the difference α is as follows:

[0055] The calculation formula from a point to the origin in a spatial rectangular coordinate system is as follows:

[0056]

[0057] Where d is the distance from the point to the origin, x, y, z represent the X-axis position, Y-axis position and Z-axis position of the point in the spatial rectangular coordinate system respectively;

[0058] The position of point M1 from the origin is calculated as:

[0059]

[0060] The position of point M2 from the origin is calculated as:

[0061]

[0062] At time T, the position difference α between point M1 and point M2 is calculated as follows:

[0063] α=|d M1 -d M2 |

[0064] Among them, the position difference α between point M1 and point M2 at any time node is less than the minimum error value of manufacturing failure;

[0065] The error value of manufacturing raw materials after manufacturing is calculated using the following formula:

[0066]

[0067] in, is the error value of the manufacturing raw material after the manufacturing is completed, n is the number of position difference values ​​α calculated during the manufacturing process of the manufacturing raw material, and x represents the value of the i-th element in the set of position difference values ​​α, and the error value of the manufacturing raw material after the manufacturing is completed is less than the minimum error value of unqualified manufacturing.

[0068] Statistical analysis is performed on the historical manufacturing data to calculate the error values ​​of unqualified products, and the minimum error value of unqualified products is taken as the manufacturing error threshold, which is set as the error threshold. Assume the error judgment threshold is Will continue to grow to the error threshold The error value data will be counted and will continue to grow to the error threshold The minimum error value is set as the error judgment threshold The value reaches the error judgment threshold When , the subsequent error values ​​are counted and the error judgment threshold is reached continuously The error value is set as the dangerous error data;

[0069] Create an error set Q, the error values ​​for completing manufacturing using manufacturing materials is the element of error set Q, and the error value of each manufacturing raw material after manufacturing is completed Add the error set Q in sequence and set the corresponding subscript position for each element. When each element is added to the error set Q, it is consistent with the error judgment threshold. Make comparisons;

[0070] Create a judgment set, set the judgment set to The error set Q is greater than the error judgment threshold The elements and corresponding subscript positions are added to the judgment set, and the error of manufacturing raw materials after manufacturing is recorded as T a , product manufacturing error T a Error judgment threshold Compare, greater than the error judgment threshold Add the elements of to the judgment set Medium, less than the error judgment threshold Elements, first set the element to K and then add it to the set In, when the collection When the elements added to are not K continuously, it is judged that the error of the products currently manufactured continuously increases; when the set When the added element is K continuously, it is judged that the manufacturing error is within the error range;

[0071] An automatic parameter adjustment system based on visual recognition and 5G communication includes a shape recognition module, a 5G transmission module, a simulation manufacturing module, a numerical control module and an error module;

[0072] The simulation manufacturing module includes a 3D model generation unit and a manufacturing simulation unit; the 3D model generation unit uses the information read and sorted by the shape recognition module to perform modeling in the same proportion to generate a model consistent with the real object; the manufacturing simulation unit uses the CNC model to manufacture the manufacturing raw material model until a complete product model is generated. The simulation manufacturing and actual manufacturing are carried out simultaneously, and the tool head trajectory is consistent;

[0073] The CNC module includes a CNC control unit and a control motion unit; the CNC control unit contains calculated instructions, which drive the cutter head and the motor through computer instructions; the control motion unit executes the corresponding computer instructions to run the cutter head to manufacture the raw materials to form products, and completes the manufacturing control of the required morphology and structure of the product through the speed, position and force of the cutter head.

[0074] The error module includes a spatial coordinate unit, an error calculation unit, an error set Q unit and an error judgment unit; by calculating the difference in the spatial position of the tool head at the same time node, based on the comprehensive results of the error set Q and the judgment set, the error generated by manufacturing and the conditions required for predicting the next manufacturing of the target product are judged.

[0075] Working principle of the present invention: the manufacturing raw material is a roughly processed titanium alloy blank, and the target product is a turbine blade;

[0076] Step S1, reading the morphology information of the titanium alloy blank and its spatial position through visual recognition, and constructing a model according to the read information;

[0077] Step S2, performing calculation and analysis based on the constructed model morphology and the morphology of the target product turbine blade, and simulating manufacturing, obtaining and outputting the tool head trajectory of the simulated manufacturing;

[0078] Step S3, receiving the outputted tool head trajectory to control the actual tool head movement, processing the titanium alloy blank, and recording and storing the tool head trajectory;

[0079] Step S4, calculating the position difference between the actual tool head and the simulated tool head at the manufacturing start node, the manufacturing middle node and the manufacturing end node;

[0080] At the node at the end of manufacturing, the actual tool head position is (2, 6, 10), and the simulated tool head position is (2.01, 6, 10). The distance between the tool head position and the origin is calculated as follows:

[0081]

[0082] At the end of manufacturing, node d 实际 and d 模拟 The position difference is calculated as follows:

[0083] α=|d 实际 -d 模拟 |=0.002

[0084] Step S5: Error threshold is 0.1, judging the error threshold is 0.08, create error set Q, create judgment set

[0085] Step S6: α is recorded as the error, and the value is 0.002; where 0.002<0.08<0.1, then Add α to the error set Q, set the value of α to K, and add K to the judgment set In the judgment set The element in is K, which is judged to be a reasonable error.

[0086] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0087] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An automatic parameter adjustment method based on visual recognition and 5G communication, characterized in that: The parameter adjustment method comprises the following steps: Step S1, reading the morphological information of the manufacturing raw material through visual recognition, the morphological information also includes the spatial position of the manufacturing raw material, and constructing a model based on the read morphological information of the manufacturing raw material; Step S2, performing calculation and analysis based on the model morphology constructed in step S1 and the target product morphology, and simulating manufacturing to obtain the tool head trajectory of simulated manufacturing and output it; Step S3, receiving the simulated tool head trajectory outputted in step S2 to control the actual tool head movement, actually manufacturing the manufacturing material, and recording and storing the actual manufactured tool head trajectory; Step S4, extracting the tool head trajectory obtained by the simulated manufacturing according to the time node and comparing and calculating the motion trajectory of the tool head in the actual manufacturing, and using the calculation result as the error value generated by the manufacturing raw material in the manufacturing process, the manufacturing node includes the manufacturing start node, the manufacturing middle node and the manufacturing end node; Step S5, calculating and analyzing the error values ​​of unqualified products in the historical manufacturing data, obtaining the minimum error value of unqualified manufacturing, and setting it as the error threshold; Step S6, set an error judgment threshold, compare the error value generated by the manufacturing raw materials during the manufacturing process with the error judgment threshold, add the error value that meets the error judgment threshold to the judgment set, judge whether the error value generated by the next manufacturing raw material is less than the error threshold, and predict whether the error value generated by the manufacturing raw materials after the manufacturing is completed meets the error condition for the next manufacturing.

2. The automatic parameter adjustment method based on visual recognition and 5G communication according to claim 1, characterized in that: The step S1 comprises: Visual recognition is used to read the morphological information of manufacturing materials, and the read image information and spatial position information are converted into data information. The data information is preprocessed and feature extracted. The processed data is described through reverse features to generate a virtual model in the computer with the same proportion as the manufacturing material.

3. The automatic parameter adjustment method based on visual recognition and 5G communication according to claim 1, characterized in that: In step S2, the process of calculating and analyzing the model morphology constructed in step S1 and the target product morphology is as follows: The morphology information of the manufacturing material obtained in step S1 is processed and reversely described to generate a virtual model. The morphology and spatial position of the model are consistent with the actual manufacturing material. Simulation manufacturing is performed according to the morphology of the target product, and the trajectory of the tool head during the simulation manufacturing process is obtained and output; A spatial rectangular coordinate system is established, and the spatial position of the simulated tool head corresponding to the time node is extracted and output according to the corresponding manufacturing time node.

4. The automatic parameter adjustment method based on visual recognition and 5G communication according to claim 1, characterized in that: In step S3, the process of recording and storing the actual manufactured tool head trajectory is as follows: The simulated tool head trajectory outputted in step S2 is used to control the tool head's motion trajectory during the actual manufacturing process, and the processing of the manufacturing raw materials is completed according to the specified trajectory. A spatial rectangular coordinate system is established, and the spatial position of the tool head corresponding to the manufacturing time node is recorded and stored.

5. The automatic parameter adjustment method based on visual recognition and 5G communication according to claim 1, characterized in that: In step S4, the calculation process of using the calculation result as the error value generated during the manufacturing process of the raw material is as follows: The position of the tool head trajectory in the spatial rectangular coordinate system is recorded, and the node at the start of manufacturing is set to T. The tool head manufactured virtually by the model is located at point M1 in the spatial rectangular coordinate system at time T, with coordinates (X1, Y1, Z1). The tool head actually manufactured is located at point M2 in the spatial rectangular coordinate system at time T, with coordinates (X2, Y2, Z2). The position difference between point M1 and point M2 at time T is α, and the calculation process of the difference α is as follows: The calculation formula from a point to the origin in a spatial rectangular coordinate system is as follows: Where d is the distance from the point to the origin, x, y, z represent the X-axis position, Y-axis position and Z-axis position of the point in the spatial rectangular coordinate system respectively; The position of point M1 from the origin is calculated as: The position of point M2 from the origin is calculated as: At time T, the position difference α between point M1 and point M2 is calculated as follows: a=|d M1 -d M2 | Among them, the position difference α between point M1 and point M2 at any time node is less than the minimum error value of manufacturing failure; The error value of manufacturing raw materials after manufacturing is calculated using the following formula: in, is the error value of the manufacturing raw material after the manufacturing is completed, n is the number of position difference values ​​α calculated during the manufacturing process of the manufacturing raw material, and X represents the value of the i-th element in the set of position difference values ​​α, and the error value of the manufacturing raw material after the manufacturing is completed is less than the minimum error value of the unqualified manufacturing.

6. The automatic parameter adjustment method based on visual recognition and 5G communication according to claim 1, characterized in that: In step S5, the calculation and analysis process based on the error value of unqualified products in the historical manufacturing data is as follows: Statistical analysis is performed on the historical manufacturing data to calculate the error values ​​of unqualified products, and the minimum error value of unqualified products is taken as the manufacturing error threshold, which is set as the error threshold.

7. The automatic parameter adjustment method based on visual recognition and 5G communication according to claim 6, characterized in that: In step S6, the process of setting the error judgment threshold is as follows: Assume the error judgment threshold is Will continue to grow to the error threshold The error value data will be counted and will continue to grow to the error threshold The minimum error value is set as the error judgment threshold The value reaches the error judgment threshold When , the subsequent error values ​​are counted and the error judgment threshold is reached continuously The error value is set as the dangerous error data; Create an error set Q, the error values ​​for completing manufacturing using manufacturing materials is the element of error set Q, and the error value of each manufacturing raw material after manufacturing is completed Add the error set Q in sequence and set the corresponding subscript position for each element. When each element is added to the error set Q, it is consistent with the error judgment threshold. Make comparisons; Create a judgment set, set the judgment set to The error set Q is greater than the error judgment threshold The elements and corresponding subscript positions are added to the judgment set, and the error of manufacturing raw materials after manufacturing is recorded as T a , product manufacturing error T a Error judgment threshold Compare, greater than the error judgment threshold Add the elements of to the judgment set Medium, less than the error judgment threshold Elements, first set the element to K and then add it to the set In, when the collection When the elements added to are not K continuously, it is judged that the error of the products currently manufactured continuously increases; when the set When the element added is K continuously, it is judged that the manufacturing error is within the error range.

8. An automatic parameter adjustment system based on visual recognition and 5G communication, applied to an automatic parameter adjustment method based on visual recognition and 5G communication according to any one of claims 1 to 7, characterized in that: The automatic parameter adjustment system includes: a shape recognition module and a 5G transmission module; The shape recognition module includes a visual recognition unit and a data sorting unit; the visual recognition unit is used to recognize and read the initial shape of the manufacturing raw material, transmit the read information to the data sorting unit, analyze the image through the data sorting unit, and sort the analyzed data; The 5G transmission module includes a data acceleration unit and a data loading unit; the data acceleration unit is used to slice and compress the data to speed up the output speed, the data loading unit increases the amount of data transmitted at the same time, and transmits multiple data at the same time, and the 5G transmission module is used for data transmission in each module.

9. The automatic parameter adjustment system based on visual recognition and 5G communication according to claim 8, characterized in that: The automatic parameter adjustment system also includes a simulation manufacturing module and a numerical control module; The simulation manufacturing module includes a three-dimensional model generation unit and a manufacturing simulation unit; the three-dimensional model generation unit performs proportional modeling on the information read and sorted by the shape recognition module to generate a model consistent with the real object; the manufacturing simulation unit manufactures the manufacturing raw material model through the numerical control model until a complete product model is generated, and the simulation manufacturing and actual manufacturing are carried out simultaneously, and the tool head trajectory is consistent; The numerical control module includes a CNC control unit and a control motion unit; the CNC control unit contains calculated instructions, and drives the cutter head and the motor through computer instructions; the control motion unit executes the corresponding computer instructions to run the cutter head to manufacture the raw materials to form the product, and completes the manufacturing control of the required morphological structure of the product by controlling the speed, position and force of the cutter head.

10. The automatic parameter adjustment system based on visual recognition and 5G communication according to claim 8, characterized in that: The automated parameter adjustment system further includes an error module; The error module includes a spatial coordinate unit, an error calculation unit, an error set Q unit and an error judgment unit; the spatial coordinate unit is used to record the position of the tool head at time T, and records the spatial coordinate position of the tool head every T time; the error calculation unit calculates the position of the simulated tool head and the actual tool head at the same time T, and the absolute value of the difference between the positions is used as the error value; The error set Q unit performs a collective calculation of the error α in the finished product manufacturing process to obtain the actual error of the finished product manufacturing. The error judgment unit is used to judge whether the error is greater than the judgment threshold. The elements are added directly to the collection Less than the judgment threshold First set it to K and then add it. When the number of elements added to is K consecutively, it is considered a reasonable error. When the element added is not K continuously, it is judged that the manufacturing conditions do not meet the next manufacturing and the parameters need to be readjusted.

Images