Intelligent cooperative control and precision self-optimization method and system for dash panel sawing jig

Abstract

The invention relates to an intelligent cooperative control and precision self-optimization method and system for a dash panel sawing jig, relates to the field of machining equipment, and solves the problems that the angle of the jig cannot be adjusted in real time according to the actual condition of a saw blade in the machining process, the sawing angle is easy to be inaccurate, the machining precision of a dash panel is further influenced, and the machining precision of the dash panel is influenced. The method comprises the steps that according to the position and angle information of a saw blade and threaded hole deviation data, an electric rotating table is driven to adjust the angle of a jig, the electric rotating table is made to be matched with the angle of the saw blade, and meanwhile an electric sliding table in the X direction and an electric sliding table in the Y direction are controlled to move the jig according to the threaded hole deviation value; the relative position deviation of the jig and the equipment is reduced to be within a preset distance; and the jig is started to execute saw cutting according to preset saw cutting parameters, wherein the saw cutting parameters comprise the rotating speed and the feeding speed. The method has the following beneficial effects that intelligent cooperation and precision self-optimization of the sawing jig are achieved, and the machining precision, efficiency and stability of the dash panel are improved.

Description

Technical Field

[0001] The present invention relates to the field of mechanical processing equipment, and in particular to an intelligent collaborative control and precision self-optimization method and system for a dash panel sawing jig. Background Art

[0002] With the vigorous development of modern manufacturing industry, the market has put forward increasingly stringent requirements on the processing accuracy of parts and components. As an indispensable and important part of the entire manufacturing process, the precision control and intelligent level of the front panel sawing process have become the focus of attention in the industry.

[0003] In the current front panel sawing field, conventional fixed fixtures are still used in most cases, and the sawing task is carried out following a relatively simple operation process. Taking the patent with the authorization announcement number CN206230089 and the patent name "A five-axis CNC double-head saw" as an example, in terms of fixture positioning, it mainly relies on the initial installation and debugging to determine the position and angle, and lacks a dynamic adjustment mechanism in the actual processing process.

[0004] With respect to the above-mentioned related technologies, the inventors found that there are the following defects: Since the jig cannot adjust the angle according to the actual situation of the saw blade in real time during the processing, it is easy to cause inaccurate sawing angles, which in turn affects the processing accuracy of the front panel, resulting in dimensional deviations, uneven cuts and other problems, reducing the quality and pass rate of the product. Secondly, the positioning deviations of the fixed holes cannot be discovered and corrected in time. As the processing proceeds, these deviations will gradually accumulate, making it difficult for the subsequent assembly process to proceed smoothly, increasing the scrap rate and rework costs. In addition, more manual intervention is not only inefficient, but also easily affected by human factors, such as the skill level and fatigue level of the operator, resulting in poor stability and consistency of the processing process, making it difficult to meet the needs of modern manufacturing for high-precision, high-efficiency and high-stability production. Summary of the invention

[0005] The first technical problem to be solved by the present invention is to provide a method for intelligent collaborative control and precision self-optimization of a dash panel sawing jig, which can realize intelligent collaboration and precision self-optimization of the sawing jig and improve the processing accuracy, efficiency and stability of the dash panel.

[0006] The second technical problem to be solved by the present invention is to provide a system capable of realizing the intelligent coordinated control and precision self-optimization method of the above-mentioned dash panel sawing jig in response to the above-mentioned existing technical status.

[0007] The technical solution adopted by the present invention to solve the above-mentioned first technical problem is:

[0008] The intelligent collaborative control and precision self-optimization method of the dash panel sawing jig includes:

[0009] Obtain image information of the saw blade and the surrounding area of ​​the fixed hole;

[0010] Based on the acquired image information, the target detection algorithm is used to identify the position and outline of the saw blade in the image. At the same time, the edge detection and feature extraction technology are used to accurately calculate the angle information of the saw blade. In addition, the image measurement algorithm is used to measure the actual position deviation data of the threaded hole in combination with the camera calibration parameters.

[0011] According to the saw blade position, angle information and threaded hole deviation data, the electric rotary table is driven to adjust the angle of the fixture so that the electric rotary table matches the saw blade angle. At the same time, the electric slide table in the X and Y directions is controlled to move the fixture according to the threaded hole deviation so that the relative position deviation between the fixture and the equipment is reduced to within the preset distance.

[0012] The jig is started to perform sawing according to preset sawing parameters, which include rotation speed and feed speed.

[0013] By adopting the above technical solutions, through precise image information processing and intelligent algorithms, the key data of the saw blade and the fixture can be accurately obtained, and the angle and position of the fixture can be adjusted quickly and accurately to make it perfectly match the saw blade, effectively reducing installation and debugging time and errors, ensuring sawing accuracy and stability, improving production efficiency and product quality, reducing scrap rate and production costs, and improving corporate benefits.

[0014] Optionally, a method for intelligent collaborative control and precision self-optimization of a dash panel sawing jig further includes:

[0015] Monitor and obtain the surface image of the saw blade, and use image analysis algorithms to analyze and obtain the wear-related features of the saw blade, including tooth wear, surface scratches and tiny cracks;

[0016] Input the wear-related features of the saw blade into a pre-built saw blade wear analysis model, and output the wear degree value of the saw blade;

[0017] If the wear degree value of the saw blade exceeds a first preset wear degree threshold value, the sawing operation is stopped and the saw blade is forced to be replaced;

[0018] If the wear degree value of the saw blade is less than the first preset wear degree threshold value and greater than the second preset wear degree threshold value, a warning message is issued to remind the operator to prepare to replace the saw blade;

[0019] If the wear degree value of the saw blade is less than the second preset wear degree threshold value, no notification is made.

[0020] By adopting the above technical solution, through monitoring and analyzing the surface image of the saw blade, the degree of wear can be accurately grasped, and corresponding measures can be taken based on threshold judgment. This can effectively avoid problems such as reduced sawing quality and equipment damage caused by excessive wear of the saw blade, ensure production continuity and stability, reduce the defective rate, and improve production efficiency. At the same time, the saw blade replacement can be reasonably arranged to reduce unnecessary downtime and waste of resources, thereby reducing production costs.

[0021] Optionally, the method further includes the following steps in parallel with issuing a warning message to remind the operator to prepare to replace the saw blade:

[0022] Obtain the current production task data and calculate the progress percentage of the current production task according to the time nodes and task quantities in the preset production plan;

[0023] According to the influence of the range of saw blade wear on sawing efficiency, the extra time required to saw each front panel and the remaining stable working time of the saw blade are predicted under the current saw blade wear state;

[0024] Compare the remaining stable working time of the saw blade with the actual remaining processing time;

[0025] If the remaining stable working time is greater than the actual remaining processing time, the completion of the current production task will be used as the opportunity to notify the operator of the replacement;

[0026] If the remaining stable working time is less than the actual remaining processing time, the remaining task volume will be calculated based on the remaining stable working time of the saw blade, and the equipment will be stopped when the remaining task volume is reached, and the operator will be notified to replace it.

[0027] By adopting the above technical solution, by accurately calculating the production progress and saw blade status, the timing of saw blade replacement can be planned more scientifically to avoid production interruptions and delays. Predicting the sawing time in advance helps to arrange production plans reasonably, improve production efficiency, reduce the defective rate caused by untimely saw blade replacement, and make full use of the remaining life of the saw blade, reduce production costs, and improve the stability and efficiency of the entire sawing production process.

[0028] Optionally, the prediction of the additional time required to saw each front panel and the remaining stable working time of the saw blade under the current saw blade wear state includes:

[0029] Analyze whether the current production task includes different types of front panels;

[0030] If not, maintain the original setting;

[0031] If yes, then collect the physical property data of different types of front panels of the same batch and the performance data of the saw blade when sawing different types of front panels, where the physical property data include the hardness, toughness and density of the material, and the performance data include the time required to saw each type of front panel at normal sawing speed, and the relationship between the degree of saw blade wear and the number of sawed pieces;

[0032] Input the physical property data of each type of front panel and the current wear degree of the saw blade into the sawing time model, and output the additional time required to saw the corresponding type of front panel;

[0033] Based on the current and limit wear of the saw blade and the number of each type of fender remaining in service, calculate how long the saw blade can still operate stably.

[0034] By adopting the above technical solution and accurately analyzing different types of front panels, the sawing time and the remaining life of the saw blade can be predicted more accurately, making the production plan more reasonable, avoiding production interruptions caused by saw blade wear, improving production efficiency, reducing defective rates, optimizing the timing of saw blade replacement, reducing production costs, improving the stability of the production process and resource utilization, and enhancing the competitiveness of enterprises in the market.

[0035] Optionally, the method further includes the following steps: inputting the wear-related features of the saw blade into a pre-built saw blade wear analysis model, and outputting the wear degree value of the saw blade, which are specifically as follows:

[0036] Monitor and obtain the surface image of the saw blade, and use image analysis algorithms to analyze and obtain the wear-related features of the saw blade, including tooth wear, surface scratches and tiny cracks;

[0037] Collect sawing force and vibration data, and use time synchronization equipment to align data of different modes in time;

[0038] Input the sawing force, vibration data and saw blade surface image into the trained time series-based wear analysis model, and output the wear trend of the saw blade when it continues to perform sawing tasks within a preset time range in the future, and output the prediction results;

[0039] If the wear degree analysis model based on the time series predicts that the wear degree value of the saw blade after the first preset time range exceeds the first preset wear degree threshold, an early warning message is issued to remind the operator to prepare to replace the saw blade.

[0040] By adopting the above technical solutions, through multi-dimensional data collection and advanced model analysis, we can not only accurately determine the current degree of saw blade wear, but also predict future wear trends and warn of replacement needs in advance. This helps to scientifically arrange production, avoid product quality problems and equipment failures caused by excessive saw blade wear, reduce downtime losses, improve production efficiency, reduce maintenance costs, ensure production stability and continuity, and improve corporate benefits.

[0041] Optionally, the method further includes the following steps in parallel with issuing a warning message to remind the operator to prepare to replace the saw blade:

[0042] Obtain the priority of each task in the current production plan and the corresponding quality requirement information. The priorities include urgent, important, and general;

[0043] The wear degree analysis model based on time series predicts the wear degree value of the saw blade after the first preset time range, the priority of each task in the current production plan and the corresponding quality requirement information are input into the adjustment plan analysis model to output a specific adjustment plan.

[0044] By adopting the above technical solution, this parallel step has obvious advantages. After knowing the wear of the saw blade, combined with the production task priority and quality requirements, a targeted adjustment plan is generated through the model. This can not only ensure the smooth progress and quality compliance of urgent and important tasks, but also reasonably arrange general tasks to avoid production delays or quality decline due to saw blade problems. Optimize resource allocation, improve production efficiency, reduce scrap rate, reduce production costs, and enhance the flexibility and scientificity of enterprise production management.

[0045] Optionally, obtaining the preset sawing parameters includes:

[0046] Obtaining physical property data and wear degree of the front panel to be sawed;

[0047] The physical property data and wear degree of the front panel to be sawed are input into the sawing parameter analysis model, and the sawing parameters are output as the sawing parameters for this application.

[0048] By adopting the above technical solution, the physical properties of the front panel and the degree of saw blade wear are accurately considered, and the adaptation parameters are determined using the sawing parameter analysis model, which can effectively optimize the sawing process. Improve the sawing accuracy and reduce the scrap rate of the front panel; reduce the risk of excessive wear of the saw blade and extend its service life; improve the sawing efficiency, thereby enhancing production benefits, saving costs for the enterprise, and ensuring high-quality production and sustainable development.

[0049] The technical solution adopted by the present invention to solve the above second technical problem is:

[0050] The intelligent collaborative control and precision self-optimization system of the dash panel sawing jig includes a memory, a processor, and a program stored in the memory and executable on the processor. The program can be loaded and executed by the processor to implement the intelligent collaborative control and precision self-optimization method of the dash panel sawing jig as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flow chart of the intelligent collaborative control and precision self-optimization method of the dash panel sawing jig in the embodiment of the present application;

[0052] Figure 2 It is a flow chart of the intelligent collaborative control and precision self-optimization method of the dash panel sawing jig in the embodiment of the present application;

[0053] Figure 3 is a flowchart of steps in parallel with issuing a warning message to remind the operator to prepare to replace the saw blade in the embodiment of the present application;

[0054] Figure 4 It is a flow chart of predicting the additional time required to saw each front panel and the remaining stable working time of the saw blade under the current saw blade wear state in an embodiment of the present application. DETAILED DESCRIPTION

[0055] The present invention is further described in detail below with reference to the accompanying drawings.

[0056] The present application is further described in detail below in conjunction with the accompanying drawings.

[0057] Reference Figure 1 , is an intelligent collaborative control and precision self-optimization method for a dash panel sawing jig disclosed in this application, comprising:

[0058] Step S100, obtaining image information of the saw blade and the surrounding area of ​​the fixed hole.

[0059] Saw blade: In the dash panel sawing jig involved in this application, the saw blade is a key component used to directly cut the dash panel. It is usually made of a metal material with a certain hardness and wear resistance, with saw teeth on its edge. It contacts the dash panel through high-speed rotation, thereby realizing the function of sawing the dash panel into the corresponding shape and size as required. For example, the common carbide saw blade can maintain good sharpness and stability during the sawing process, and is suitable for sawing dash panels of various materials.

[0060] Area around fixed holes: This refers to the location of the holes on the dash panel sawing jig used to fix related components (such as fixing the jig itself to the work platform, etc.) and the area around them within a certain range. The position accuracy of these holes has an important impact on the installation and coordination between the entire jig and the equipment and the accuracy of subsequent sawing work, so it is necessary to obtain image information of the surrounding area in order to analyze its actual position deviation. For example, on some mechanical processing jigs, the jig is fixed to the workbench by bolts passing through the fixed holes. The accuracy of the fixed holes is directly related to whether the jig can be accurately in the preset sawing position.

[0061] In this step, the image information of the saw blade and the surrounding area of ​​the fixed hole is obtained mainly by means of an image acquisition device. A common image acquisition device may be an industrial camera, and its installation position needs to be carefully designed and debugged to ensure that the saw blade and the surrounding area of ​​the fixed hole can be clearly and completely captured.

[0062] The industrial camera should be installed in a suitable position to fully cover the saw blade and the surrounding area of ​​the fixed hole, and the shooting angle should be adjusted to ensure that the saw blade and the surrounding area of ​​the fixed hole can be fully presented in the shooting picture without obvious distortion. The shooting parameters of the industrial camera also need to be reasonably set, including exposure time, aperture size, frame rate, etc.

[0063] Step S200, based on the acquired image information, a target detection algorithm is used to identify the position and outline of the saw blade in the image, and edge detection and feature extraction techniques are used to accurately calculate the angle information of the saw blade. In addition, the camera calibration parameters are synchronously combined and an image measurement algorithm is used to measure the actual position deviation data of the threaded hole.

[0064] The target detection algorithm is a computer vision technology that aims to automatically identify the location of a specific target from an image or video and determine its category. In this step, first, the image information obtained in step S100 is input into the selected target detection algorithm. If the YOLOv5 algorithm is selected here, a large amount of image data with information such as the position and category of the saw blade already marked will be prepared as a training set, and the YOLOv5 model will be trained to learn the characteristic performance of the saw blade in different scenes, angles, etc. After the training is completed, the saw blade image to be analyzed is input into the trained model, and the model can output the position of the saw blade in the image (usually expressed in the form of coordinates of a rectangular box, such as the upper left corner coordinates and the lower right corner coordinates) and identify the category information of the saw blade, and then determine the approximate outline range of the saw blade.

[0065] Edge detection is a technology used in the field of image processing and computer vision to identify the edge of an object in an image. The position of the edge is determined by detecting the sudden change of the grayscale value of the pixel in the image. After obtaining the approximate location area of ​​the saw blade, the edge detection algorithm (taking the Canny edge detection algorithm as an example) is then used to process the image area where the saw blade is located. First, the image of the area is Gaussian filtered to remove possible noise interference in the image, such as some tiny noise points caused by light reflection, so that the image becomes relatively smooth. Then the gradient amplitude and direction of the image are calculated. Through the non-maximum suppression step, only the pixels with the local maximum gradient amplitude are retained as edge candidate points. After double threshold detection, the real edge pixels are further screened out, so that the edge contour of the saw blade is clearly outlined. At the same time, combined with feature extraction techniques such as local binary pattern (LBP), the saw blade image is processed to extract relevant features such as its texture and sawtooth distribution, in preparation for the subsequent accurate calculation of the saw blade angle.

[0066] Feature extraction technology refers to a method of extracting information that can represent the essential characteristics of the target object from the original image data. These features can help distinguish different objects and conduct more in-depth analysis of the objects. In this step, the feature extraction technology can be used to obtain key features from the saw blade image that help to accurately calculate its angle. Specific texture features on the saw blade and the shape distribution features of the saw teeth are all extracted feature information. For example, the feature extraction method based on the local binary pattern (LBP) generates a binary-coded feature descriptor by comparing the grayscale value relationship between the pixel in the image and its neighboring pixels, thereby characterizing the relevant features in the saw blade image and assisting in the accurate calculation of the angle information.

[0067] Combining camera calibration parameters with image measurement algorithms: For the camera calibration parameters, the camera calibration operation has been carried out in advance by using a standard calibration plate (such as a chessboard calibration plate) according to the Zhang Zhengyou calibration method, and the camera's internal and external parameters and other parameters have been obtained. When measuring the actual position deviation data of the threaded hole, first determine the pixel coordinate position of the threaded hole in the image (which can be achieved through some image analysis software or the coordinate acquisition function provided in the algorithm), and then use the image measurement algorithm based on the obtained camera calibration parameters and the corresponding coordinate conversion formula to convert the pixel coordinate deviation of the threaded hole in the image into the position deviation value in the actual space, for example, to obtain specific deviation data such as how many millimeters the threaded hole is offset from the standard position in the X, Y, and Z directions.

[0068] Step S300, according to the saw blade position, angle information and threaded hole deviation data, drives the electric rotary table to adjust the angle of the fixture so that the electric rotary table matches the saw blade angle, and at the same time controls the electric slide in the X and Y directions to move the fixture according to the threaded hole deviation, so that the relative position deviation between the fixture and the equipment is reduced to within a preset distance.

[0069] The electric rotary table is a device that can realize rotational motion under the drive of a motor. In the front panel sawing jig system of the present application, it is mainly used to carry the jig and adjust its angle as required to match the angle of the saw blade, thereby ensuring that the relative angle between the saw blade and the jig is in the best state during the sawing process and improving the sawing accuracy. The process of adjusting the angle of the electric rotary table based on the saw blade information is as follows: After obtaining the saw blade position and angle information output by step S200, the system will transmit these data to the controller that controls the electric rotary table (usually a programmable logic controller, PLC, etc., which can receive data signals and drive the motor to operate according to the set program). The controller calculates the angle difference that the electric rotary table needs to adjust based on the angle information of the saw blade, and then sends the corresponding control instructions to the drive motor of the electric rotary table. For example, if the current angle of the saw blade deviates 5 degrees clockwise from the standard sawing angle, the controller will drive the motor of the electric rotary table to rotate it 5 degrees counterclockwise, so that the angle of the electric rotary table is accurately matched with the angle of the saw blade, ensuring the accuracy of the relative angle between the two during sawing.

[0070] Electric slide: It is a mechanical structure that realizes linear motion through motor drive. The electric slides set in the X and Y directions can drive the fixture to move in the corresponding horizontal direction. The X direction here can usually be understood as the horizontal direction, and the Y direction is the vertical direction. Through the coordinated movement of the electric slides in these two directions, the position of the fixture in the plane can be accurately adjusted, so that the relative position deviation between the fixture and the equipment is reduced to within the preset distance, ensuring the accuracy of the sawing work.

[0071] The relative position deviation between the fixture and the equipment refers to the position difference of the fixture in space relative to the original standard position of the sawing equipment during installation and use. This deviation will affect the accuracy and quality of sawing. It is generally measured by measuring the distance difference between the key reference points on the fixture (such as the center of the fixed hole, etc.) and the corresponding standard reference points on the equipment in the X, Y, and Z directions.

[0072] The process of controlling the electric slide to move the fixture based on the threaded hole deviation data is as follows: The actual position deviation data of the threaded hole measured in step S200 will also be transmitted to the controller that controls the X and Y electric slides. The controller generates corresponding control instructions and sends them to the drive motors of the X and Y electric slides based on the deviation of the threaded hole in the X and Y directions respectively (for example, the threaded hole deviates by 3 mm in the X direction and 2 mm in the Y direction relative to the standard position). After receiving the instruction, the motor of the X-direction electric slide will drive the slide to drive the fixture to move the corresponding distance in the X direction (here it is 3 mm). Similarly, the Y-direction electric slide will also move the fixture by 2 mm in the Y direction. Through such coordinated actions, the relative position deviation between the fixture and the equipment is gradually reduced to within the preset distance, ensuring that the fixture is in the accurate sawing working position on the equipment.

[0073] Step S400, starting the jig to perform sawing according to preset sawing parameters, where the sawing parameters include rotation speed and feed speed.

[0074] The preset sawing parameters are a series of parameters used to control the sawing process that are pre-set before the sawing of the front panel, based on the material characteristics of the front panel, the performance of the saw blade, and the expected sawing quality. They mainly include key indicators such as rotation speed and feed speed.

[0075] The jig is started to perform sawing operation after the preset sawing parameters (including rotation speed, feed speed, etc.) are determined in the above method. These parameter values ​​will be input into the control system that controls the sawing jig (generally an automated control system, such as a CNC system to achieve precise control). The control system sends corresponding control instructions to the drive motor of the saw blade according to the set rotation speed parameters, so that the motor drives the saw blade to rotate at the specified rotation speed; at the same time, according to the feed speed parameters, the moving device of the front panel relative to the saw blade is controlled (such as through a drive mechanism such as an electric slide), and the front panel is smoothly sent to the saw blade for sawing at the set feed speed, thereby realizing the sawing work according to the preset sawing parameters, ensuring that the sawing process can be completed efficiently under the premise of meeting quality requirements.

[0076] Reference Figure 2 , a method for intelligent collaborative control and precision self-optimization of a dash panel sawing jig also includes the following steps, which are as follows:

[0077] Step S500, monitoring and acquiring the surface image of the saw blade, and using an image analysis algorithm to analyze and acquire wear-related features of the saw blade, wherein the wear-related features include tooth wear, surface scratches, and tiny cracks.

[0078] Monitoring and acquiring the saw blade surface image means that during the sawing process, a specific image acquisition device is used to continuously photograph the surface area of ​​the saw blade to obtain image data that can reflect the real-time status of the saw blade surface.

[0079] Tooth wear: refers to the condition in which the shape of the teeth on the saw blade changes, such as the top of the tooth becomes blunt, a notch appears on the tooth edge, or the overall height of the tooth is reduced. These changes will directly affect the cutting ability and sawing quality of the saw blade. Surface scratches: refers to some linear marks on the surface of the saw blade. These marks may be caused by foreign particles in the front panel during the sawing process or accidental friction between the saw blade and other parts. Too many and deep surface scratches will affect the surface flatness of the saw blade, and then affect the smoothness of the sawing and the surface quality of the sawed front panel. Micro cracks: refers to extremely small cracks on the surface of the saw blade. They usually appear when the saw blade is subjected to a large sawing force for a long time or due to factors such as changes in the internal stress of the material. Once the micro cracks expand, they may cause serious consequences such as saw blade breakage, affecting the safety and continuity of the sawing work.

[0080] The application process of the image analysis algorithm is as follows: after obtaining the saw blade surface image, the image data is input into the selected image analysis algorithm for processing. Taking the algorithm based on the combination of texture feature analysis and morphological processing as an example, the texture feature analysis algorithm is first used to analyze the saw blade surface texture in the image. By calculating the texture feature parameters of different regions in the image (such as the contrast and correlation of the grayscale co-occurrence matrix), the difference between the normal saw blade surface texture and the saw blade texture in the current collected image is compared to preliminarily determine whether there are signs of wear and the areas where wear may be involved. Then, the image is processed using a morphological processing algorithm (such as corrosion, expansion and other operations) to further highlight possible surface scratches, tiny cracks and other features. For example, some small noise interference is removed through corrosion operations, and then scratches, cracks and other features are made more obvious through expansion operations, which facilitates accurate extraction and recognition. For the analysis of tooth wear, we can combine the edge detection technology of the image (similar to the Canny edge detection algorithm mentioned in the previous step) to clearly outline the edge contour of the sawtooth. By comparing it with the standard sawtooth shape template, we can analyze the degree of sawtooth wear. For example, we can measure the fillet radius of the top of the sawtooth, the integrity of the tooth blade and other indicators to determine the tooth wear condition.

[0081] Step S600: input the wear-related features of the saw blade into a pre-built saw blade wear analysis model, and output the wear degree value of the saw blade.

[0082] The saw blade wear analysis model is a mathematical model built based on a specific algorithm and a large amount of data training. Its purpose is to output the saw blade wear value based on the input saw blade related feature information. The model integrates many laws and relationships related to saw blade wear. By performing complex calculations and analysis on the input features, it simulates the actual wear condition of the saw blade, thereby providing a quantitative wear degree measurement indicator. For example, it may be a model built based on a machine learning algorithm (such as a neural network, support vector machine, etc.), which is trained using a large amount of saw blade feature data with a marked wear degree in the past, so that the model can learn the corresponding relationship between features and wear degree.

[0083] The wear degree value is a numerical indicator used to quantify the wear condition of the saw blade. It is calculated through a model, and its value directly reflects the severity of the saw blade wear. Usually a certain range of values ​​is set to correspond to different wear stages or treatment suggestions. For example, a wear degree value in the range of 0-0.3 may indicate that the saw blade is in a state of light wear, in the range of 0.3-0.7 it is moderate wear, and greater than 0.7 means severe wear, etc. Different wear degrees will determine whether the saw blade needs to be replaced, warned, or other corresponding measures.

[0084] Input the wear-related features into the model to obtain the wear degree value: After the wear-related features of the current saw blade are obtained through step S500, these features are sorted and quantified (for example, the bluntness of the saw teeth in the tooth wear condition is digitized, the total length of the surface scratches is counted, etc.) to make them meet the input format requirements of the saw blade wear analysis model. Then the sorted wear-related feature data is input into the trained saw blade wear analysis model. The model will output a corresponding wear degree value based on the mapping relationship between the features and the wear degree that it has learned in advance, and after a series of complex mathematical operations (for example, in a neural network, it is the weighted summation between neurons, activation function operations, etc.), this value represents the wear degree of the current saw blade.

[0085] Step S700: If the wear degree of the saw blade exceeds a first preset wear degree threshold, the sawing operation is stopped and the saw blade is forcibly replaced.

[0086] The first preset wear threshold: This is a key value set in advance based on multiple factors such as saw blade characteristics, sawing requirements, and past experience. It is used to determine whether the saw blade wear reaches the limit at which it must be forced to stop using and replace. For example, it is set to 0.8 (assuming the wear value range is 0-1). When the saw blade wear value exceeds this value, it means that the wear condition of the saw blade has seriously affected its normal sawing function, sawing quality, and may bring safety hazards. At this time, the saw blade is no longer suitable for continued sawing operations.

[0087] Step S800: If the wear degree value of the saw blade is less than the first preset wear degree threshold value and greater than the second preset wear degree threshold value, an early warning message is issued to remind the operator to prepare to replace the saw blade.

[0088] The second preset wear threshold: This is also a preset wear value, which is smaller than the first preset wear threshold, and is used to distinguish the transition stage from moderate wear to severe wear of the saw blade. When the saw blade wear value is between the first preset wear threshold and the second preset wear threshold, it means that the saw blade can still be used for a period of time, but it is close to the point where it needs to be replaced, so an early warning message is issued to remind the operator to prepare to replace the saw blade so that the operator can make corresponding arrangements in advance. For example, if the first preset wear threshold is set to 0.8, the second preset wear threshold may be set to 0.5, and the operation of this step is triggered when the saw blade wear value is between 0.5-0.8.

[0089] The wear degree judgment and warning operation are as follows: after obtaining the current wear degree value of the saw blade (from step S600), compare the value with the first preset wear degree threshold and the second preset wear degree threshold. Assuming that the first preset wear degree threshold is 0.8 and the second preset wear degree threshold is 0.5, if the current saw blade wear degree value is 0.6, the condition that "the wear degree value of the saw blade is less than the first preset wear degree threshold and greater than the second preset wear degree threshold" is satisfied. At this time, the control system of the sawing equipment will trigger the early warning mechanism, and send out early warning information to remind the operator to prepare to replace the saw blade through sound and light alarms, pop-up prompt boxes on the display screen, or send notifications to the operator's mobile device, so that the operator can arrange time in advance, prepare new saw blades, etc., to ensure that the saw blade is ready for replacement before the wear further deteriorates and reaches the mandatory replacement standard, so as to ensure the continuity and efficiency of the sawing work.

[0090] Step S900: If the wear degree value of the saw blade is less than the second preset wear degree threshold value, no notification is made.

[0091] Reference Figure 3 , a method for intelligent collaborative control and precision self-optimization of a dash panel sawing jig also includes steps in parallel with issuing an early warning message to remind an operator to prepare to replace a saw blade, specifically as follows:

[0092] Step SA00, obtain the current production task data, and calculate the progress percentage of the current production task according to the time nodes and task quantities in the preset production plan.

[0093] The current production task data refers to various detailed information related to the ongoing production work of sawing the front panels, including the total number of front panels that need to be sawed, the number of front panels that have been sawed, and the number of front panels that need to be sawed. It may also include the distribution of task quantities of different types of front panels, as well as time-related information such as the start time and expected completion time of each task. For example, the current production task is to saw 1,000 front panels, of which 300 have been completed and 700 are left. Among these 1,000 panels, 600 are model A front panels and 400 are model B front panels. At the same time, each front panel has a corresponding time node for the planned start and completion of sawing. These constitute the current production task data as a whole.

[0094] Preset production plan: This is a detailed plan made before the front panel sawing work is carried out, based on production order requirements, equipment capacity, manpower arrangements and other factors. It clearly stipulates the production tasks to be completed at each stage and the corresponding time arrangements, and is the guiding framework for the orderly progress of the entire production process. For example, it is planned to complete the sawing task of 2,000 front panels within a week, including sawing 500 panels per day from Monday to Wednesday, and sawing 250 panels per day from Thursday to Friday, and specifically divides the proportion of different models of front panels in daily production. This is the content of the preset production plan.

[0095] The current production task data, preset production plan and other related data are sourced from the production management system equipped by the production enterprise, which records in detail the relevant information of all production tasks, including the current production task data and the preset production plan content. For example, the ERP (Enterprise Resource Planning) system used by the enterprise has a production module, which will enter the details of the front panel sawing task corresponding to each production order, such as the planned number of front panels, the number of categories of different types of front panels, the time requirements corresponding to the tasks at each stage, etc. By connecting data with the system or accessing permissions, the current production task data and the preset production plan related content required for step SA00 can be obtained.

[0096] Step SB00, based on the influence of the saw blade wear degree value on the sawing efficiency, predict the additional time required to saw each front panel and the remaining stable working time of the saw blade under the current saw blade wear state.

[0097] The range of the wear degree value of the saw blade falls into: This is determined based on the previous step (such as the wear degree value of the saw blade obtained in step S600), and is divided according to the preset different wear degree thresholds (first preset wear degree threshold, second preset wear degree threshold, etc.). For example, if the first preset wear degree threshold is set to 0.8 and the second preset wear degree threshold is set to 0.5, then the wear degree value may fall into the range of less than 0.5 (mild wear range), 0.5-0.8 (medium wear range), and greater than 0.8 (heavy wear range). Different ranges represent different wear states of the saw blade.

[0098] Sawing efficiency: refers to the number of front panels cut per unit time or the time spent on sawing a certain number of front panels, etc., which reflects the speed and effect of sawing work. For example, originally 10 front panels could be sawed in 1 hour, but as the saw blade wears, only 8 panels may be sawed in 1 hour, which indicates that the sawing efficiency has changed, which is affected by many factors such as the wear state of the saw blade and the material of the front panel.

[0099] Impact value: This is a quantitative value that represents the impact of the saw blade wear range on the sawing efficiency. It is obtained through analysis of past experience, test data, etc., and is used to measure the extent of the reduction or change in sawing efficiency under different wear conditions. For example, in the moderate wear range (assuming a wear range of 0.5-0.8), it is found through analysis that the sawing efficiency will be reduced by 30% compared to when the saw blade was new, so this 30% can be used as the impact value of this range on the sawing efficiency.

[0100] The extra time required to saw each front panel: refers to the extra time it takes to saw a front panel when the saw blade is in a certain state of wear compared to when the saw blade is in good condition. For example, it takes 5 minutes to saw a front panel when the saw blade is new, but it may take 6 minutes after it is worn. The extra 1 minute is the extra time required to saw each front panel.

[0101] The remaining stable working time of the saw blade: Based on the current wear degree of the saw blade and its material, performance and other factors, it is estimated that the saw blade can continue to work while maintaining relatively stable sawing quality and efficiency. Once this time is exceeded, the increased wear of the saw blade may lead to a significant decline in sawing quality and malfunctions. For example, it is estimated that a saw blade can still work stably for 2 hours at the current wear degree, and these 2 hours are the remaining stable working time.

[0102] Predicting the extra time required to saw each front panel involves the following process: 1. Obtaining normal sawing time data: It is necessary to know the standard time it takes to saw a front panel when the saw blade is in good condition (usually a new saw blade or when the wear is very low). This can be obtained through the normal sawing efficiency data accumulated in previous production or by performing sawing tests on new saw blades. For example, after testing, it takes an average of 5 minutes for a new saw blade to saw a front panel. 2. Calculation based on the impact value: Calculate the extra time based on the impact value corresponding to the interval in which the saw blade wear value is determined. Assuming that the saw blade is in the moderate wear range (0.5-0.8), the impact value of this range on the sawing efficiency is 30%, and it takes 5 minutes for a new saw blade to saw a front panel, then the extra time = 5 minutes × 30% = 1.5 minutes, that is, the extra time required to saw each front panel in the current state is 1.5 minutes, and the total time becomes 6.5 minutes.

[0103] Calculate the remaining stable working time of the saw blade: Make a comprehensive judgment based on the material, specifications and current wear level of the saw blade. Saw blades of different materials (such as high-speed steel saw blades, carbide saw blades, etc.) have different stable working time under different wear levels. For example, when a carbide saw blade is moderately worn (assuming the wear level is 0.6), based on its material properties and previous experience with the use of the same type of saw blades, it is expected that the wear level will reach an unacceptable state after sawing 100 front panels. At the current sawing speed, 10 front panels can be sawed per hour, so the remaining stable working time is 100÷10=10 hours.

[0104] Step SC00, comparing the remaining stable working time of the saw blade with the actual remaining processing time.

[0105] Actual remaining processing time: This refers to the estimated time required from the current moment to complete all remaining front panel sawing tasks under the current production task arrangement. It comprehensively considers the number of remaining front panels, the current sawing efficiency (including the actual efficiency after the influence of saw blade wear factors) and other factors that may affect the processing time (such as equipment maintenance time, operator rest time, etc.). For example, there are 50 front panels left to be sawed. Considering that it takes 8 minutes to saw each front panel after the saw blade is worn (it may only take 5 minutes with a new saw blade), and it is estimated that equipment maintenance and personnel rest will take a total of 30 minutes, then the actual remaining processing time = 50 × 8 + 30 = 430 minutes.

[0106] Step SD00: If the remaining stable working time is greater than the actual remaining processing time, the completion of the current production task will be used as the opportunity to notify the operator of the replacement.

[0107] Remaining stable working time: As mentioned above, it is a comprehensive estimate based on multiple factors such as the current wear condition of the saw blade, material characteristics, and past experience. It is the length of time that the saw blade can continue to saw while maintaining relatively stable sawing quality and efficiency. For example, through analysis and evaluation, the current remaining stable working time of a certain saw blade is 3 hours, which means that under normal circumstances, it can still saw stably for this long.

[0108] The specific operation process is as follows:

[0109] Obtain data: Extract the specific values ​​of the remaining stable working time and the actual remaining processing time of the saw blade from the results calculated and analyzed in the previous steps. For example, the remaining stable working time is calculated to be 240 minutes (4 hours), and the actual remaining processing time is calculated to be 180 minutes (3 hours).

[0110] Comparison operation: Use the comparison operator to compare the two time data, that is, to determine whether the remaining stable working time is greater than the actual remaining processing time. Taking the above data as an example, because 240 minutes (remaining stable working time) is greater than 180 minutes (actual remaining processing time), the condition of "remaining stable working time is greater than actual remaining processing time" is met.

[0111] Decision execution: When this condition is met, the system will follow the preset rules (usually the control system or production management system of the sawing equipment will execute the relevant instructions) and will notify the operator to replace the saw blade after completing the current production task. This means that the saw blade can continue to be used normally before all the remaining production tasks are completed. Once these tasks are completed, the operator will be reminded to replace the saw blade in time to ensure that subsequent production work is carried out when the saw blade is in good condition. At the same time, it can also make full use of the remaining available time of the saw blade to avoid waste of resources caused by premature replacement. For example, the operator is informed by a pop-up prompt box on the device display, a text message notification, or an audible and visual alarm that the saw blade needs to be replaced after completing the current batch of front panel sawing tasks.

[0112] Step SE00, if the remaining stable working time is less than the actual remaining processing time, the remaining task volume is calculated based on the remaining stable working time of the saw blade, and the equipment will be stopped when the remaining task volume is reached, and the operator will be notified to replace it.

[0113] Remaining stable working time: This is the estimated time that the saw blade can continue to work while maintaining relatively stable sawing performance, taking into account factors such as the current degree of wear of the saw blade, material characteristics, past usage data, and sawing process requirements. For example, according to relevant models and data calculations, a saw blade can still stably saw the front panel for about 120 minutes in its current state, and these 120 minutes are its remaining stable working time.

[0114] Remaining task volume: the specific number or workload of the front panels that have not yet been sawed. This is one of the key indicators to measure the progress of production task completion and subsequent production arrangements. For example, the production plan originally required sawing 500 front panels, and 300 have been completed, so the remaining task volume is 200 front panels.

[0115] The remaining task volume calculation and decision execution are as follows:

[0116] Calculate the remaining task volume: Get the original total production task volume and the currently completed task volume from the production management system, and subtract the two to get the remaining task volume. For example, the production management system records show that this production task requires a total of 300 front panels to be sawed, and 120 panels have been completed, so the remaining task volume = 300-120 = 180 panels.

[0117] Comparison judgment and operation: Compare the remaining stable working time obtained with the actual remaining processing time. If the remaining stable working time is less than the actual remaining processing time, it means that the saw blade cannot maintain a stable working state in the remaining production tasks. At this time, it is necessary to analyze and calculate the remaining task volume based on the remaining stable working time of the saw blade, that is, according to the current sawing efficiency (considering the efficiency after the saw blade is worn), calculate the number of front panels that can be completed within the remaining stable working time, and use this number as the remaining task volume. For example, the remaining stable working time is 90 minutes, and it currently takes 7 minutes to saw a front panel (considering wear), then the number of front panels that can be completed within the remaining stable working time (that is, the remaining task volume) = 90÷7≈12.86, rounded down to 12 pieces. When the task of sawing these 12 front panels is completed, the system will automatically trigger an instruction to stop the equipment (generally executed by the equipment's control system, such as the programmable logic controller PLC sending a stop signal according to a preset program), and notify the operator to replace the saw blade through sound and light alarms, display prompts or text message notifications, so as to ensure the stability of the production process and product quality, and avoid production accidents or quality problems caused by excessive wear of the saw blade.

[0118] Reference Figure 4 , the prediction of the additional time required to saw each front panel and the remaining stable working time of the saw blade under the current saw blade wear state includes:

[0119] Step SB10, analyzing whether the current production task includes different types of dash panels. If not, executing step SB20; if yes, executing step SB30.

[0120] Different types of dash panels: dash panels are classified into many types due to differences in material, thickness, structural design or purpose. For example, there are aluminum alloy dash panels, steel dash panels, plastic dash panels, etc. according to different materials; or they can be divided into main frame dash panels, auxiliary support dash panels, etc. according to their specific uses and installation locations in the product. These different types of dash panels may have different effects on the degree of wear of the saw blade and the parameters required for sawing due to their different characteristics during the sawing process.

[0121] Whether different types of front panels are included can be obtained by querying the production management system.

[0122] Step SB20, maintain the original settings.

[0123] Step SB30, collects physical property data of different types of front panels of the same batch and performance data of saw blades when sawing different types of front panels, wherein the physical property data include hardness, toughness, and density of the material, and the performance data include the time required to saw each type of front panel at normal sawing speed, and the relationship between the degree of saw blade wear and the number of sawed pieces.

[0124] Physical property data of different types of front panels in the same batch: This refers to the physical property parameter information of various types of front panels involved in the current production task, which is closely related to the sawing process. For example, the hardness of the material reflects the ability of the front panel to resist saw blade cutting. The higher the hardness, the more difficult it is to saw and the faster the saw blade will wear. Toughness is related to whether the front panel is prone to cracks or breaks during the sawing process. Materials with low toughness may have problems such as edge collapse during sawing. Density will affect the mass and inertia of the front panel, and thus affect the stability during sawing and the required sawing force. For example, the hardness of an aluminum alloy front panel may be 80-120HB (Brinell hardness) and the density is about 2.7g / cm 3 , with a certain toughness; the hardness of the steel front panel may reach 150-250HB, and the density is about 7.8g / cm 3 , toughness is different from that of aluminum alloys. These different physical property data are crucial for the analysis of the sawing process and the evaluation of saw blade wear.

[0125] The performance data of saw blades when sawing different types of front panels: This refers to the performance index data related to the sawing effect and wear condition of the saw blades in the actual sawing process of different types of front panels. For example, the time required to saw each type of front panel at normal sawing speed directly reflects the sawing efficiency of the saw blade when dealing with different front panels. The longer the sawing time, the weaker the saw blade's cutting ability on this type of front panel. The relationship between the degree of saw blade wear and the number of sawed pieces shows the trend of saw blade wear as the number of sawed front panels of this type increases. For example, after sawing 50 aluminum alloy front panels, the degree of saw blade wear reaches a certain value (such as 0.5mm wear on the tooth top), while after sawing 30 steel front panels, the degree of saw blade wear may reach another value (such as 1mm wear on the tooth top). These data can help establish a saw blade wear prediction model, so as to more accurately evaluate the remaining life and replacement timing of the saw blade in the current production task.

[0126] Obtaining physical property data: Manufacturers of front panels usually obtain detailed physical property data of the materials used from raw material suppliers. These data will be clearly listed in the material specification or quality inspection report.

[0127] Saw blade performance data acquisition: In the actual production process of sawing front panels, the workshop operators will record the relevant information of each sawing task, including the type of front panels being sawed, the start time and end time of sawing, and the status of the saw blade before and after sawing (such as through a simple visual inspection or using a gauge to measure the wear of the saw blade). These records can be organized into a production log or spreadsheet for subsequent analysis. For example, after sawing a batch of a specific type of front panels each time, the operator records the total sawing time, whether the saw blade has obvious wear characteristics (such as blunt teeth, increased surface scratches, etc.), and whether there are any abnormal conditions during the sawing process (such as increased vibration of the saw blade, increased noise, etc.). As production continues, a large amount of these on-site data are accumulated, and the performance data of the saw blade when sawing different types of front panels can be analyzed, such as the time required to saw each type of front panel at normal sawing speed and the relationship between the degree of saw blade wear and the number of sawed pieces.

[0128] Step SB40, input the physical property data of each type of front panel and the current wear degree of the saw blade into the sawing time model, and output the additional time required for sawing the corresponding type of front panel.

[0129] Sawing time model: This is a model built based on mathematical algorithms and statistical principles. It aims to predict the additional time required to saw the corresponding type of front panel based on the input parameters such as the physical properties of the front panel and the current wear degree of the saw blade.

[0130] Model application and prediction involve the following processes:

[0131] Input current data for prediction: When it is necessary to predict the sawing time of a certain type of front panel in the current sawing task, the physical property data of the front panel (normalized hardness, toughness, density values) and the current wear data of the saw blade (also normalized) are used as input and input into the trained sawing time model. The model calculates and outputs the predicted value of the additional time required to saw the front panel based on internal mathematical operations and learned parameter relationships. For example, if the model is a neural network-based model, the input data will be processed by weighted summation, activation function operation, etc. through neurons in each layer of the network, and finally the predicted result of the sawing time will be obtained at the output layer; if it is a multivariate linear regression model, a linear combination calculation will be performed based on the input feature value and the trained coefficient to obtain the predicted value of the sawing time.

[0132] Step SB50, calculates the time that the saw blade can still work stably based on the current wear degree and limit wear degree of the saw blade and the number of each type of front panels in the remaining tasks.

[0133] The number of each type of fenders in the remaining tasks: This refers to the number of different types of fenders that still need to be sawed in the current unfinished production tasks. For example, the current production task requires sawing a total of 1,000 fenders, of which there are 400 type A fenders, 200 of which have been sawn, so the number of type A fenders in the remaining tasks is 200; there are 600 type B fenders, 300 of which have been sawn, so the number of type B fenders in the remaining tasks is 300. These data are crucial for accurately calculating the stable working time of the saw blade in the remaining production tasks, because different types of fenders have different degrees of wear on the saw blade, so the remaining life of the saw blade needs to be comprehensively evaluated based on their respective remaining quantities.

[0134] The process of calculating the time the saw blade can still work stably is as follows:

[0135] Calculation based on the wear model: Based on the current wear degree and the limit wear degree of the saw blade, combined with the number of each type of front panels in the remaining task and the wear rate data of the saw blade when sawing different types of front panels (obtained through previous sawing tests, empirical data or analysis based on the data collected in step SB30), a mathematical model is used to calculate the time the saw blade can still work stably. For example, a linear wear model can be used, assuming that the wear degree of the saw blade is linearly related to the number of front panels sawed (of course, the actual situation may be more complicated, but this simple model can be used to make a preliminary estimate within a certain range). First, the amount of wear that the saw blade can still withstand is calculated based on the current wear degree and the limit wear degree of the saw blade, and then the increase in wear of the saw blade after sawing these remaining front panels is calculated based on the remaining number of each type of front panels and the corresponding saw blade wear rate, and the two are compared to determine whether the saw blade can complete the remaining task. If it can be completed, the stable working time of the saw blade in the remaining task is further calculated. For example, if the current wear degree of the saw blade is 30% (assuming that it is expressed as a proportion of the wear amount to the limit wear amount) and the limit wear degree is 80%, then the wear amount that the saw blade can still withstand is 50%. It is also known that the number of type A front panels in the remaining tasks is 200, and the wear rate of the saw blade cutting each type A front panel is 0.1% (that is, cutting a type A front panel will increase the wear degree of the saw blade by 0.1%), and the number of type B front panels in the remaining tasks is 100, and the wear rate of the saw blade cutting each type B front panel is 0.2%. After cutting all the remaining front panels, the increase in the wear of the saw blade is 200×0.1%+100×0.2%=40%. Since 40%<50%, it means that the saw blade can complete the remaining tasks. Then, based on the current sawing efficiency (the actual sawing efficiency adjusted in combination with the wear condition of the saw blade) and the total number of remaining front panels, calculate the time that the saw blade can still work stably. Assuming that the current sawing efficiency is 50 panels per hour (actual efficiency after saw blade wear is taken into account), and the total number of remaining panels is 300, the saw blade can still work stably for 300÷50=6 hours.

[0136] A method for intelligent collaborative control and precision self-optimization of a dash panel sawing jig also includes the steps of inputting wear-related features of a saw blade into a pre-built saw blade wear analysis model and outputting a wear degree value of the saw blade, specifically as follows:

[0137] Step Sa00, monitor and obtain the surface image of the saw blade, and use image analysis algorithm to analyze and obtain the wear-related features of the saw blade, the wear-related features include tooth wear, surface scratches and tiny cracks.

[0138] Monitoring and obtaining saw blade surface images: During the sawing process, a special image acquisition device (such as an industrial camera) is used to continuously or periodically photograph the surface of the saw blade to obtain image information that can reflect the surface status of the saw blade. These images will serve as the basic data source for subsequent analysis of saw blade wear. For example, an industrial camera photographs the saw blade at a certain frame rate (such as 5 frames per second) and captures the details of the saw blade surface from different angles (such as the front, side, etc.), including the shape of the teeth, the flatness of the surface, and whether there are scratches or cracks.

[0139] The application process of the image analysis algorithm is as follows:

[0140] Image preprocessing: Before the collected saw blade surface image is input into the image analysis algorithm, the image is preprocessed to improve the image quality and enhance the recognizability of features. The preprocessing steps may include image denoising (using median filtering, Gaussian filtering and other methods to remove noise points in the image to make the image smoother), image enhancement (using histogram equalization, contrast stretching and other methods to improve the contrast and brightness distribution of the image and highlight the features of the saw blade surface) and image grayscale (converting color images to grayscale images to simplify subsequent calculations and analysis processes while retaining the main feature information of the image). For example, a 3x3 median filter kernel is used to denoise the image, and then the histogram equalization method is used to enhance the image to make the texture and wear features of the saw blade surface clearer and more recognizable.

[0141] Feature extraction and recognition: Use the selected image analysis algorithm to perform feature extraction and recognition operations on the preprocessed image. First, use the edge detection algorithm (such as the Canny edge detection algorithm) to process the image. By calculating the gradient amplitude and direction of the image, the edge profile of the saw blade is determined, thereby extracting the shape information of the sawtooth and preliminarily judging the wear of the tooth shape, such as by measuring the change in the fillet radius of the top of the sawtooth, the degree of reduction in the tooth height and other parameters to quantify the degree of wear of the tooth shape. Then, use the algorithm based on texture feature extraction (such as the gray level co-occurrence matrix method) to analyze the texture features of the saw blade surface image. By calculating the texture feature parameters (such as contrast, correlation, energy, etc.), it is determined whether there are abnormal changes in the texture caused by wear, such as the texture becoming rough and disordered, so as to further determine the degree of wear and surface state of the saw blade. At the same time, the image is processed in combination with morphological processing algorithms (such as corrosion and expansion operations) to highlight features such as surface scratches and tiny cracks. For example, by setting appropriate structural elements and corrosion and expansion parameters, surface scratches and tiny cracks can be made more obvious in the image, which is convenient for subsequent measurement and analysis. By measuring parameters such as the length and depth of scratches and the length and width of cracks, these wear-related characteristics can be quantitatively described, providing detailed data support for accurately evaluating the degree of wear of the saw blade.

[0142] Step Sb00, collecting sawing force and vibration data, and using a time synchronization device to align data of different modes in time.

[0143] Sawing force: When the saw blade is sawing the front panel, the force generated by the interaction between the saw blade and the front panel, which reflects the extent of overcoming the material resistance during the sawing process.

[0144] Vibration data: refers to the relevant data corresponding to the vibration generated by the saw blade during the sawing process, including information such as the vibration frequency, amplitude and direction.

[0145] Time synchronization equipment: This is a device or technical means used to ensure that data from different sources can be accurately aligned in time. In this step, since the sawing force data and vibration data are synchronously collected by different sensors during the sawing process, but the collection frequency, response time, etc. of each sensor may be different, in order to accurately analyze the relationship between these different modal data and their changes over time, it is necessary to use time synchronization equipment to ensure that they are accurately corresponding on the time axis, so that the subsequent analysis results based on these data are more reliable. For example, the common high-precision clock synchronization module can provide a unified time reference for each sensor, so that all collected data can be marked and stored based on this unified time, which is convenient for subsequent data processing and analysis.

[0146] The applications of time synchronization equipment are as follows:

[0147] Time synchronization technology selection: Select the appropriate time synchronization technology according to the data acquisition system architecture of the sawing equipment and the requirements for time synchronization accuracy. Common synchronization methods include those based on the Network Time Protocol (NTP), which connects to the standard time server through the network, obtains accurate time information and distributes it to each device that needs to be synchronized (such as the data acquisition card of the force sensor, the data acquisition card of the vibration sensor, etc.); there are also synchronization methods based on high-precision clock modules (such as GPS timing modules, atomic clock modules, etc.), which can provide higher time synchronization accuracy and are generally suitable for occasions with extremely high requirements for data time accuracy. For example, in some large-scale, high-precision sawing production lines, GPS timing modules are used to ensure that the data acquisition of all sensors is accurately synchronized in time, and the error can be controlled at the microsecond level or even smaller.

[0148] Time synchronization setting and verification: When installing and configuring the time synchronization device, perform the corresponding settings according to its manual and operating guide, and connect each data acquisition device (data acquisition card connected to force sensor, vibration sensor, etc.) to the time base provided by the time synchronization device, so that they can receive a unified time signal for time marking. In order to verify the effect of time synchronization, some simple tests can be performed before collecting data, such as triggering each sensor to collect a short analog signal at the same time, and then checking whether the starting point of the collected data on the time axis is consistent, or by analyzing whether the time marks corresponding to some known synchronization events in the subsequent collected data are accurately aligned, to ensure that the time synchronization device is working properly, to ensure that the sawing force and vibration data can be accurately synchronized in time, and to provide a reliable data basis for subsequent joint analysis.

[0149] Step Sc00, input the sawing force, vibration data and saw blade surface image into the trained time series-based wear degree analysis model, and output the wear degree change trend of the saw blade when it continues to perform the sawing task within a preset time range in the future, and output the prediction result.

[0150] Wear degree analysis model based on time series: This is a model specifically used to analyze and predict the wear degree of saw blades that changes over time. It is built based on the principles and methods of time series analysis. The model collects, organizes and analyzes various relevant data of the saw blade at different time points (such as sawing force, vibration data, saw blade surface image features, etc.), and mines out the change patterns of these data over time and their intrinsic connection with the wear degree of the saw blade, so as to predict the wear degree trend of the saw blade within a specific time range in the future. For example, it may be a model built based on time series prediction algorithms such as ARIMA (autoregressive moving average model) and LSTM (long short-term memory network). Through learning and training of historical data, the model can automatically capture the time dependency and trend characteristics in the data, and then predict the future wear degree of the saw blade.

[0151] Future preset time range: This refers to a pre-set time interval after the current moment, during which the model will predict the change in the degree of wear of the saw blade. For example, the preset time range may be the next 1 hour, 2 hours, or 100 sawing operations, etc. The specific duration or number of operations will be determined based on the actual needs of the sawing production task and the usage of the saw blade. During this time period, through the prediction of the model, the wear trend of the saw blade can be understood in advance, so that corresponding measures (such as adjusting the sawing parameters, preparing to replace the saw blade, etc.) can be taken to ensure the smooth progress of the sawing work and the stability of product quality.

[0152] Wear degree change trend: This refers to the change in the degree of wear of the saw blade over time or as the amount of sawing work increases, including information such as the rate of increase in the degree of wear, whether there is a mutation point, and the general direction of development. For example, the model may predict that in the future, the degree of wear of the saw blade will gradually increase at a rate of 5% per hour, or after sawing a certain number of front panels, the degree of wear will suddenly jump due to the increased wear of a key part of the saw blade. These are specific manifestations of the trend of wear degree changes. By understanding the trend of wear degree changes, the remaining service life of the saw blade can be predicted in advance, and production plans and saw blade maintenance and replacement work can be reasonably arranged.

[0153] Prediction results: This is a wear degree analysis model based on time series. After internal calculation and analysis, it outputs specific numerical values ​​or descriptive information about the wear degree change trend of the saw blade within the preset time range in the future based on the input information such as sawing force, vibration data, and saw blade surface image. For example, the prediction result may be a specific wear degree numerical sequence, indicating the value that the saw blade wear degree is expected to reach at each time point in the future (such as every 10 minutes or after every 10 sawings of the front panel); it may also be a qualitative description, such as "the saw blade wear degree will rise rapidly in the next hour, approaching the limit wear state", etc. These prediction results will provide important basis for subsequent decision-making and help operators take effective measures to deal with the wear of the saw blade in a timely manner.

[0154] The model application and prediction are as follows:

[0155] Real-time data acquisition and input: In the actual sawing production process, the sawing force and vibration data are collected in real time through sensors installed on the sawing equipment, and the image acquisition equipment is used to regularly obtain the saw blade surface image data (taken at a certain time interval or sawing number interval). After the same preprocessing and feature extraction operations as the training data are performed on these real-time collected data, they are input as input data into the trained time series-based wear degree analysis model. For example, the sawing force and vibration data are collected every 5 minutes, and the saw blade surface image is taken every 20 sawings of the front panel. After preprocessing and feature extraction, these data are combined into a time series data vector and input into the model.

[0156] Wear prediction and result output: The model calculates and outputs the prediction results of the wear trend of the saw blade within the preset time range in the future based on the input real-time data and the time series law and wear prediction model learned internally. If the model is built based on LSTM, the input data will go through a series of cyclic calculations and information transmission in the hidden layer of the model, combined with the weights and bias parameters learned during the model training process, and finally obtain the predicted value sequence of the saw blade wear in the output layer. This prediction result can be presented to the operator or production management system in the form of a chart (such as a line chart showing the trend of wear over time), a numerical list (listing the predicted wear value at each time node in the future), or a qualitative description (such as "wear will gradually increase, it is recommended to pay attention", etc.), so that they can take corresponding measures in time according to the prediction results, such as adjusting the sawing parameters to slow down the wear rate, preparing to replace the saw blade, etc., to ensure efficient and stable sawing work.

[0157] Step Sd00: If the wear degree analysis model based on the time series predicts that the wear degree value of the saw blade after the first preset time range exceeds the first preset wear degree threshold, an early warning message is issued to remind the operator to prepare to replace the saw blade.

[0158] Prediction results of the trend of wear degree change: This is output by the wear degree analysis model based on time series in step Sc00, and is related to the conclusions about how the wear degree of the saw blade will develop and change within the preset time range in the future. It covers the prediction of different situations such as whether the wear degree is gradually increasing, remaining stable, or experiencing a sudden change, and may also include specific information such as the speed of changes in each stage. For example, the prediction results show that the wear degree of the saw blade will increase at a uniform rate of about 5% every 30 minutes in the next 2 hours, or at a certain moment (such as after sawing a certain number of front panels), the wear degree will have a sudden increase trend. These are all specific presentations of the trend of the wear degree of the saw blade, and are important basis for subsequent decision-making.

[0159] Compare with preset wear thresholds: Obtain specific values ​​of preset wear thresholds from the management regulations of sawing production or equipment maintenance documents. These values ​​are usually determined by professional technicians based on the performance of the saw blade itself, a large amount of past sawing practical experience, and the quality requirements of this production task. Then, compare each predicted value in the wear trend prediction result output by step Sc00 (such as the wear prediction value corresponding to different time nodes) with the corresponding preset wear thresholds (such as light, moderate, and severe wear thresholds, etc.). For example, the preset moderate wear threshold is 50%. If the prediction result shows that the saw blade wear will reach 45% in the next 30 minutes, it means that it is approaching the moderate wear threshold, and further judgment and corresponding measures are needed.

[0160] Determine and implement countermeasures: Determine specific countermeasures based on the comparison between the wear degree trend prediction results and the preset wear degree threshold. Generally speaking, there are several common judgment and response logics:

[0161] Threshold not reached: If the predicted wear level is always lower than all relevant preset wear level thresholds within the preset time range in the future (for example, the light wear threshold is not reached), it means that the saw blade is in a good working condition. At this time, it is usually necessary to continue sawing according to the current sawing parameters and operating procedures. However, it is still necessary to continue to pay attention to the prediction results output by the model and maintain the monitoring status.

[0162] Threshold situation: When the predicted wear level is close to a preset wear level threshold (for example, there is a small gap from the moderate wear threshold, such as the difference is within 5%), some preventive measures need to be taken. For example, the sawing feed speed can be appropriately reduced, the saw blade speed can be adjusted to a more appropriate range (depending on the specific situation of the saw blade and the front panel), and the relevant operators can be notified to pay close attention to the status of the saw blade and prepare spare saw blades so that they can be replaced in time when the wear level reaches the threshold, so as to avoid sudden impact on sawing quality and production efficiency due to excessive wear of the saw blade.

[0163] Reaching or exceeding the threshold: Once the predicted wear reaches or exceeds a preset wear threshold, the operation must be carried out strictly in accordance with the established response plan. For example, if the wear reaches the severe wear threshold (assuming it is 80%), the sawing equipment should be stopped immediately, an alarm should be issued through the equipment control system (such as sound and light alarms to notify the operator), the saw blade should be replaced, and the sawing equipment should be fully inspected and maintained to check whether the excessive wear of the saw blade has affected other parts of the equipment, so as to ensure that subsequent sawing work can be resumed in a safe and stable state.

[0164] A method for intelligent collaborative control and precision self-optimization of a dash panel sawing jig also includes steps in parallel with issuing an early warning message to remind an operator to prepare to replace a saw blade, specifically as follows:

[0165] Step Se00, obtaining the priority of each task in the current production plan and the corresponding quality requirement information, the priorities include urgent, important, and general.

[0166] Task priority: refers to the classification of specific tasks in the current production plan according to their importance and urgency, usually divided into urgent, important, and general levels.

[0167] Urgent task: This means that the task needs to be handled immediately or must be completed in a very short time. If it cannot be completed on time, it may have a significant impact on the subsequent production links, delivery deadlines, and even the production operations of the entire enterprise. For example, a batch of front panels are used for the production of urgent orders, and the customer requires delivery in a very short time, so the corresponding sawing task is an urgent task.

[0168] Important tasks: These tasks play a key role in the overall production goals, product quality or the core business of the enterprise. Although they are not as time-sensitive as urgent tasks, they have a greater impact on the smooth progress of production and the final quality of products. For example, the sawing task of the front panels of certain key models has very high quality requirements and is directly related to the performance and quality of the subsequent assembled products. It is an important task.

[0169] General tasks: Relatively speaking, tasks with lower importance and urgency can be flexibly arranged according to actual conditions, without affecting the overall production schedule and the achievement of key goals. For example, some conventional specifications of front panel sawing tasks for inventory backup can be classified as general tasks.

[0170] Quality requirement information: It is a description of the quality standards for the sawing of the dash panel in each production task, including the specific index requirements for the dimensional accuracy, edge flatness, surface roughness, etc. of the dash panel after sawing. For example, for the sawing of the dash panel in an important task, the quality requirement information stipulates that its dimensional accuracy must be controlled within ±0.5mm, the error of edge flatness must not exceed 0.1mm, and the surface roughness Ra value must be less than 1.6μm. These detailed quality indicators are important bases for measuring whether the sawing work meets the standards.

[0171] Step Sf00, input the wear degree value of the saw blade predicted by the time series wear degree analysis model after the first preset time range, the priority of each task in the current production plan and the corresponding quality requirement information into the adjustment plan analysis model, and output the specific adjustment plan.

[0172] The first preset time range: This is a pre-set future time period, which is used to define the time span for wear degree prediction based on the time series wear degree analysis model. Its setting will comprehensively consider the rhythm of sawing production tasks, the normal wear change cycle of saw blades, and the convenience of taking countermeasures in advance. For example, it can be set to the next 1 hour, 3 hours, or the time corresponding to sawing a certain number of front panels, etc. The purpose is to grasp the wear degree of the saw blade in advance within this time interval, so as to make reasonable decisions and ensure the continuous and stable development of sawing work.

[0173] Adjustment plan analysis model: This is a model that aims to output a specific response adjustment plan for saw blade wear based on the premise of ensuring the smooth progress of production tasks, based on the input of various key information (including the predicted value of the future wear of the saw blade, the priority of each task in the current production plan, and the corresponding quality requirement information, etc.), through internal established logic, rules, and algorithm operations. This model may be built based on decision trees, expert systems, classification or regression algorithms in machine learning, etc. It integrates professional knowledge in the field of sawing production, past practical experience, and response strategies for different situations, so that it can generate appropriate adjustment plans based on different input combinations. For example, if the input information shows that the degree of wear of the saw blade will reach a high value in the next hour, and there is currently an urgent priority high-quality production task, the model may output adjustment plans such as replacing the saw blade immediately and adjusting the sawing parameters to temporarily ensure the sawing quality.

[0174] Specific adjustment plan: This is a collection of specific operational suggestions given by the adjustment plan analysis model after comprehensive analysis of the input information. It covers saw blade-related treatment methods (such as replacing the saw blade, continuing to use and strengthening monitoring, etc.), adjustment of sawing parameters (such as changing the saw blade speed, feed speed, etc.) and coordination of production plans (such as whether to give priority to certain tasks, whether to postpone some tasks, etc.), aiming to ensure that the sawing work can not only cope with the wear of the saw blade, but also meet the different priorities and quality requirements of various tasks in the current production plan. For example, the specific adjustment plan may be "replace the saw blade immediately, reduce the saw blade speed by 10%, give priority to emergency tasks, and postpone the sawing time of general tasks by 2 hours" and other clear operational instructions.

[0175] The acquisition of preset sawing parameters includes:

[0176] Step S410, obtaining physical property data and wear degree of the front panel to be sawed.

[0177] The physical property data of the dash panel to be sawed refers to the various physical property parameter information of the dash panel itself that is closely related to the sawing process. Common physical property data include the following aspects: 1. Material properties: For example, the dash panel is made of metal (such as aluminum alloy, steel, etc.), plastic or composite material, and the hardness, density, toughness, etc. of different materials are very different. For example, the aluminum alloy dash panel has relatively low hardness, low density, good toughness, and relatively small cutting force required when sawing; while the steel dash panel has high hardness and high density, which makes it more difficult to saw. 2. Size specifications: Covers the length, width, thickness and other size parameters of the dash panel. When sawing a larger or thicker dash panel, it is necessary to consider factors such as the cutting stroke and cutting stability of the saw blade, and it may take longer sawing time and appropriate sawing equipment parameters to ensure the sawing quality. 3. Surface condition: Including whether the dash panel surface is coated, roughness, etc. If there is a coating on the surface, attention should be paid to the shedding of the coating and its impact on the saw blade during sawing. Different surface roughness will also affect the friction between the saw blade and the front panel during sawing, thereby affecting the sawing effect.

[0178] Step S420, inputting the physical property data and the wear degree of the front panel to be sawed into the sawing parameter analysis model, and outputting the sawing parameters as the sawing parameters for this application.

[0179] Sawing parameter analysis model: This is a model built based on mathematical algorithms, machine learning technology or a combination of the two. It is designed to output the best sawing parameters for the current sawing task based on the input physical property data of the front panel to be sawed and the degree of wear of the saw blade and other related information through internal established logic, rules and data processing methods. The model integrates professional knowledge in the field of sawing, a large amount of past sawing practical experience, and the analysis and summary of sawing conditions under different materials and different wear conditions, so that it can accurately give appropriate sawing parameter recommendations based on different input conditions, such as the specific values ​​of key parameters such as saw blade speed, feed speed, and sawing depth, to ensure that the sawing work is completed efficiently and with high quality.

[0180] Construction and training of sawing parameter analysis model:

[0181] Model selection basis: Choose the appropriate model construction method based on factors such as the complexity of the sawing process, data characteristics, and requirements for model accuracy and interpretability. If there is a relatively obvious linear relationship between the sawing parameters and the input data (physical properties of the front panel and the degree of saw blade wear), and the model is expected to have good interpretability, then a multivariate linear regression model can be selected. This model predicts the appropriate values ​​of the sawing parameters by establishing a linear equation between the independent variable (input data) and the dependent variable (sawing parameters). If the sawing process involves many factors, complex relationships, nonlinear characteristics, and a large amount of historical sawing data is available for learning, models such as neural networks (such as multi-layer perceptrons, long short-term memory networks, etc.) and decision trees (such as random forests, gradient boosting trees, etc.) in machine learning are more advantageous. They can automatically mine complex nonlinear relationships from the data and learn the optimal sawing parameter combinations corresponding to different input conditions.

[0182] Data collection and organization: In order to train the sawing parameter analysis model, it is necessary to collect rich historical sawing data. These data should cover the actual sawing conditions of different types of fenders (with various physical property data) under various degrees of saw blade wear, and record the corresponding sawing parameters that are actually used and achieve good sawing results. For example, collect the sawing records of fenders of different materials (aluminum alloy, steel, plastic, etc.) and different sizes (different thicknesses, lengths and widths) in the past few years, as well as the specific settings of parameters such as saw blade speed, feed speed, and sawing depth corresponding to the saw blades at different wear stages (wear degree values ​​recorded through regular testing), and organize these data into a structured data set. Each data record contains fields related to the physical properties of the fender (such as material, thickness, hardness, etc.), saw blade wear degree values, and corresponding sawing parameter values, forming the basic data source for the training model.

[0183] Model training process: Divide the sorted data set into training set, validation set and test set according to a certain ratio (commonly 70% for training, 15% for validation and 15% for testing). Use the training set data to train the selected model. During the training process, for the multivariate linear regression model, the coefficients of the model (the weights corresponding to each variable) are adjusted through optimization algorithms such as the least squares method, so that the sawing parameter values ​​predicted by the model are as close as possible to the actual sawing parameters in the training set to reduce the prediction error; for machine learning models such as neural networks, the weights and bias parameters in the model are updated through the gradient descent algorithm and its variants (such as stochastic gradient descent, Adagrad, Adam, etc.), so that the difference between the predicted results of the sawing parameters output by the model and the actual results (measured by the loss function, such as the mean square error loss function, the cross entropy loss function, etc.) is continuously reduced, and the accuracy and generalization ability of the model are gradually improved. At the same time, the validation set data is used to verify and evaluate the model during the training process, and the performance of the model on data that has not participated in the training is observed. According to the verification results (such as by calculating evaluation indicators such as accuracy, root mean square error, mean absolute error, etc.), the model is optimized and adjusted (such as adjusting the model structure, increasing the amount of training data, adjusting hyperparameters, etc.) to ensure that the model has good performance. Finally, the test set data is used to perform a final test and evaluation on the trained model to ensure that the model can stably output reasonable and accurate sawing parameter prediction results on new data.

[0184] The input data and model application are as follows:

[0185] Data preprocessing and input: The physical property data of the front panel to be sawed and the wear degree data of the saw blade obtained in step S410 are preprocessed to meet the input requirements of the sawing parameter analysis model. Preprocessing may include data normalization (such as uniformly mapping physical property data of different magnitudes to a specific interval, commonly the interval of 0 to 1, to facilitate model calculation and comparison), missing value processing (if there is some missing data, it can be supplemented by mean filling, interpolation, etc.) and data format conversion (arranging the data into the specific data structure required by the model, such as vectors, matrices, etc.). After completing the preprocessing, the data are accurately input into the trained sawing parameter analysis model in accordance with the input order and format specified by the model.

[0186] Output and application of sawing parameters: After receiving the input data, the sawing parameter analysis model performs calculations based on the relationships and rules learned internally, and outputs the sawing parameters for this application, that is, the specific recommended values ​​of parameters such as saw blade speed, feed speed, and sawing depth. These output results can be displayed in a visual interface (for example, the recommended values ​​of each parameter are intuitively displayed in digital form on the operation panel of the sawing equipment, which is convenient for operators to set directly), or presented in the form of a text report (listing the recommended values ​​of each sawing parameter and related instructions in detail for operators to refer to and execute). The operator adjusts the corresponding parameter settings of the sawing equipment according to the output sawing parameters. During the actual sawing process, pay close attention to the sawing effect (such as sawing efficiency, sawing quality, etc.) and the working status of the saw blade. If the actual situation is found to be inconsistent with expectations (such as abnormal vibration during sawing, sawing quality not meeting the standards, etc.), the sawing parameters can be appropriately fine-tuned based on experience, and relevant situations can be fed back at the same time to further optimize the sawing parameter analysis model so that it can better adapt to the actual sawing scene and continuously improve the quality and efficiency of sawing work.

[0187] In addition, the present application may also take into account the increased wear of the saw blade, at which time the adjustment of the lubrication system should be initiated.

[0188] The quantification index of saw blade wear trend can be calculated by numerical wear degree and wear rate, as follows:

[0189] Numericalization of wear degree: Using a wear degree analysis model based on time series, the wear condition of the saw blade is quantified into a wear degree value (for example, 0 represents a brand new saw blade, 1 represents a completely scrapped saw blade, and the intermediate values ​​correspond to different degrees of wear in proportion). By analyzing the surface image of the saw blade, the proportion of the tooth wear height to the original tooth height, the total length and depth of the surface scratches, the number and length of small cracks and other characteristics are measured, and combined with the changing trends of sensor data such as sawing force, vibration and temperature, these multimodal information are converted into a continuous wear degree value using machine learning algorithms or pre-established mathematical models. For example, when the tooth wear height reaches 20% of the original tooth height, the total length of the surface scratches reaches 100mm, and the sawing force increases by 30% compared to the new saw blade, the vibration amplitude increases by 50%, and the temperature rises by 20°C, the wear degree value calculated by the model is 0.4.

[0190] Wear rate calculation: In the time series, the change in the wear degree value between two adjacent time points (for example, one time interval for every 100 front panels cut) is calculated to obtain the wear rate. Wear rate = (wear degree value at the current time point - wear degree value at the previous time point) / time interval. If the wear rate shows an upward trend in several consecutive time intervals and exceeds the preset threshold (for example, the wear rate rises from 0.02 to 0.05 per 100 front panels), it is judged that the saw blade wear will increase.

[0191] The quantitative conditions for starting the lubrication system optimization program are as follows:

[0192] Comprehensive wear index threshold setting: Set a comprehensive wear index threshold. When the wear degree value reaches 0.3 and the wear rate exceeds 0.03 for two consecutive time intervals, the cooling and lubrication parameter optimization program is started. This threshold is set based on a large amount of experimental data and actual production experience, and is obtained through statistical analysis of saw blade wear during sawing of front panels of different materials, ensuring that parameters are optimized in time before saw blade wear has a significant impact on production efficiency and product quality.

[0193] Abnormal temperature and sawing force thresholds: In addition to the wear degree and rate indicators, abnormal changes in saw blade temperature and sawing force are also considered. When the saw blade temperature continues to be more than 15°C higher than the normal working temperature (the normal working temperature is determined by the saw blade material and sawing process, for example, 80°C), and the sawing force exceeds 20% of the normal sawing force, even if the wear degree and rate have not reached the above thresholds, the cooling and lubrication parameter optimization program is started. This is because high temperature and high sawing force may accelerate saw blade wear, and taking optimization measures in advance can effectively protect the saw blade and maintain its stable working state.

[0194] Based on the same inventive concept, an embodiment of the present invention further provides an intelligent collaborative control and precision self-optimization system for a dash panel sawing fixture, including a memory and a processor, wherein the memory stores a program that can be executed on the processor to implement the following Figures 1 to 4 Procedure for either method.

[0195] The embodiments of this specific implementation method are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Therefore, all equivalent changes made based on the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. An intelligent collaborative control and precision self-optimization method for a dash panel sawing jig, characterized in that: include: Obtain image information of the saw blade and the surrounding area of ​​the fixed hole; Based on the acquired image information, the target detection algorithm is used to identify the position and outline of the saw blade in the image. At the same time, the edge detection and feature extraction technology are used to accurately calculate the angle information of the saw blade. In addition, the image measurement algorithm is used to measure the actual position deviation data of the threaded hole in combination with the camera calibration parameters. According to the saw blade position, angle information and threaded hole deviation data, the electric rotary table is driven to adjust the angle of the fixture so that the electric rotary table matches the saw blade angle. At the same time, the electric slide table in the X and Y directions is controlled to move the fixture according to the threaded hole deviation so that the relative position deviation between the fixture and the equipment is reduced to within the preset distance. The jig is started to perform sawing according to preset sawing parameters, which include rotation speed and feed speed.

2. The intelligent collaborative control and precision self-optimization method of the dash panel sawing jig according to claim 1 is characterized in that: Also includes: Monitor and obtain the surface image of the saw blade, and use image analysis algorithms to analyze and obtain the wear-related features of the saw blade, including tooth wear, surface scratches and tiny cracks; Input the wear-related features of the saw blade into a pre-built saw blade wear analysis model, and output the wear degree value of the saw blade; If the wear degree value of the saw blade exceeds a first preset wear degree threshold value, the sawing operation is stopped and the saw blade is forced to be replaced; If the wear degree value of the saw blade is less than the first preset wear degree threshold value and greater than the second preset wear degree threshold value, a warning message is issued to remind the operator to prepare to replace the saw blade; If the wear degree value of the saw blade is less than the second preset wear degree threshold value, no notification is made.

3. The intelligent collaborative control and precision self-optimization method of the dash panel sawing jig according to claim 2 is characterized in that: Also included are steps in parallel with issuing a warning message to remind the operator to prepare to replace the saw blade, as follows: Obtain the current production task data and calculate the progress percentage of the current production task according to the time nodes and task quantities in the preset production plan; According to the influence of the range of saw blade wear on sawing efficiency, the extra time required to saw each front panel and the remaining stable working time of the saw blade are predicted under the current saw blade wear state; Compare the remaining stable working time of the saw blade with the actual remaining processing time; If the remaining stable working time is greater than the actual remaining processing time, the completion of the current production task will be used as the opportunity to notify the operator of the replacement; If the remaining stable working time is less than the actual remaining processing time, the remaining task volume will be calculated based on the remaining stable working time of the saw blade, and the equipment will be stopped when the remaining task volume is reached, and the operator will be notified to replace it.

4. The intelligent collaborative control and precision self-optimization method of the dash panel sawing jig according to claim 3 is characterized in that: The prediction of the additional time required to saw each front panel and the remaining stable working time of the saw blade under the current saw blade wear state includes: Analyze whether the current production task includes different types of front panels; If not, maintain the original setting; If yes, then collect the physical property data of different types of front panels of the same batch and the performance data of the saw blade when sawing different types of front panels, where the physical property data include the hardness, toughness and density of the material, and the performance data include the time required to saw each type of front panel at normal sawing speed, and the relationship between the degree of saw blade wear and the number of sawed pieces; Input the physical property data of each type of front panel and the current wear degree of the saw blade into the sawing time model, and output the additional time required to saw the corresponding type of front panel; Based on the current and limit wear of the saw blade and the number of each type of fender remaining in service, calculate how long the saw blade can still operate stably.

5. The intelligent collaborative control and precision self-optimization method of the dash panel sawing jig according to claim 2 is characterized in that: The method also includes the following steps: inputting the wear-related features of the saw blade into a pre-built saw blade wear analysis model, and outputting the wear degree value of the saw blade, which are as follows: Monitor and obtain the surface image of the saw blade, and use image analysis algorithms to analyze and obtain the wear-related features of the saw blade, including tooth wear, surface scratches and tiny cracks; Collect sawing force and vibration data, and use time synchronization equipment to align data of different modes in time; Input the sawing force, vibration data and saw blade surface image into the trained time series-based wear analysis model, and output the wear trend of the saw blade when it continues to perform sawing tasks within a preset time range in the future, and output the prediction results; If the wear degree analysis model based on the time series predicts that the wear degree value of the saw blade after the first preset time range exceeds the first preset wear degree threshold, an early warning message is issued to remind the operator to prepare to replace the saw blade.

6. The intelligent collaborative control and precision self-optimization method of the dash panel sawing jig according to claim 5 is characterized in that: Also included are steps in parallel with issuing a warning message to remind the operator to prepare to replace the saw blade, as follows: Obtain the priority of each task in the current production plan and the corresponding quality requirement information. The priorities include urgent, important, and general; The wear degree analysis model based on time series predicts the wear degree value of the saw blade after the first preset time range, the priority of each task in the current production plan and the corresponding quality requirement information are input into the adjustment plan analysis model to output a specific adjustment plan.

7. The intelligent collaborative control and precision self-optimization method of the dash panel sawing jig according to claim 4 is characterized in that: The acquisition of preset sawing parameters includes: Obtaining physical property data and wear degree of the front panel to be sawed; The physical property data and wear degree of the front panel to be sawed are input into the sawing parameter analysis model, and the sawing parameters are output as the sawing parameters for this application.

8. An intelligent collaborative control and precision self-optimization system for a dash panel sawing jig, characterized in that: It includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein the program, when loaded and executed by the processor, can implement an intelligent collaborative control and precision self-optimization method for a front panel sawing jig as described in any one of claims 1 to 7.

Images