Safety Detection Method for Die Area in Hot Stamping Production Line of High-Strength Steel

Abstract

The present invention discloses a method for safety detection in the die area of a high-strength steel hot stamping production line, which includes an industrial control computer. An image recognition software is installed on the industrial control computer. The industrial control computer controls a manipulator to feed a blank into a die and take out the formed part after stamping. The method comprises the following steps: 1) detection of the feeding position; 2) double-blank detection, which simultaneously verifies whether there is a double blank through a near-infrared camera and a far-infrared thermal imager; 3) detection of whether the formed part is taken away. The present invention uses a visible light industrial camera, a near-infrared industrial camera, and a far-infrared thermal imager to take pictures to detect the positioning of hot-formed workpieces, double blanks, and whether the formed parts are taken out in a high-strength steel hot stamping production line. It can automatically achieve high efficiency and good accuracy, save labor costs, reduce labor intensity, avoid die losses caused by damaged dies due to misaligned steel plates in the prior art and production downtime losses, reduce the occurrence of defective products, improve product quality, and improve production efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of hot forming of automotive parts, and particularly to a method for safety detection in a die area in a high-strength steel hot stamping production line. Background Art

[0002] The automotive industry is a pillar industry in the development of the national economy, with a strong industrial correlation degree, and is regarded as an important symbol of a country's economic development level. The stamping process is the top priority among the four major automotive processes and also the first of the four major processes. According to incomplete statistics, more than 40% of the car parts are metal sheet stamping parts. Among them, the ultra-high-strength steel hot stamping technology can manufacture parts with a tensile strength of up to 1500 MPa, having the dual advantages of achieving vehicle lightweight and improving safety. The demand for hot stamping parts maintains a rapid upward trend and is widely used in the manufacturing of automotive parts such as automotive A-pillars, B-pillars, bumpers, and roof longitudinal beams. The rapid growth of market demand has triggered a construction boom of domestic hot forming production lines.

[0003] During the hot stamping process, the temperature of the raw material sheet, i.e., the steel plate, is 900 °C. Only a manipulator can be used to feed multiple sheets (1 - 5 sheets) into the die. The press presses down to close the die, maintains the pressure for more than 10 seconds, and at the same time rapidly cools down to below 150 °C through the circulating water in the die. Then the press lifts the die to open, and the manipulator takes out the stamped formed part, and then the production cycle repeats. Although the degree of automation is very high during the whole production process and there are many positioning and limiting mechanical mechanisms in the die, there are still cases where 5‰ - 10‰ of the feeding is off-target, 0.01‰ of double sheets are stacked on top of each other, and about 1‰ of the formed parts cannot be accurately taken out. If any of the above three situations occur during the stamping process, it will cause damage to the die at best and scrapping of the die at worst. The cost of the die is very high, resulting in a relatively large loss. Moreover, die damage requires production line stoppage for treatment, affecting production efficiency.

[0004] Currently, manufacturers ensure the normal production by manual observation of operators beside the press. However, workers may be fatigued, especially in the early morning, and it is particularly easy to have insufficient observation. In case of an abnormality, the press still performs the downward pressing action, resulting in production stoppage. Summary of the Invention

[0005] The object of the present invention is to solve the above problems and provide a method for safety detection in the die area of a high-strength steel hot stamping production line. By means of machine vision, the present invention automatically judges whether the heated raw material is accurately placed at the correct position in the die, automatically judges whether there are two sheets stacked on top of each other, and automatically judges whether the formed part is completely taken away. If the above three situations do not occur, the normal production proceeds. If the placement is incorrect, the press is stopped to prevent the production of defective products and damage to the die. At the same time, an audible and visual alarm is given to remind the on-site workers to handle it. After the abnormal situation is handled, the production continues, reducing the labor intensity, improving the production efficiency, ensuring the product qualification rate, improving the product quality, avoiding the losses caused by die damage, and increasing the economic benefits.

[0006] The technical solutions adopted by the present invention to solve its technical problems include the following three parts:

[0007] A method for safety detection in the die area of a high-strength steel hot stamping production line, including an industrial control computer, on which image recognition software is installed. The industrial control computer controls a manipulator to feed a sheet into a die and take out the formed part after stamping. The method includes the following steps:

[0008] 1) Detection of feeding position

[0009] Receive the picture of the die area taken by a near-infrared industrial camera through the image recognition software on the industrial control computer, and judge whether the sheet is placed correctly. If it is not placed correctly, the press will automatically stop and the operator will handle it;

[0010] 2) Detection of double sheets. Verify whether there are double sheets through a near-infrared camera and a far-infrared thermal imager at the same time, including:

[0011] A. Receive the picture of the die area taken by the far-infrared thermal imager through the image recognition software on the industrial control computer, obtain the temperature through the picture, and follow the temperature change to judge whether there are double sheets. If there are double sheets, the press will automatically stop and the operator will handle it;

[0012] B. Receive the picture of the die area taken by the near-infrared industrial camera through the image recognition software on the industrial control computer, and judge whether there are double sheets. If there are double sheets, the press will automatically stop and the operator will handle it;

[0013] 3) Detection of taking away the formed part, including:

[0014] A. Receive the picture of the blanking table taken by a visible light industrial camera through the image recognition software on the industrial control computer, and judge whether the formed part is completely taken out. If it is not completely taken out, the press will automatically stop and the operator will handle it;

[0015] B. Receive the die area pictures taken by the far-infrared thermal imager through the image recognition software on the industrial control computer, obtain the temperature from the pictures, and judge whether all the formed parts are taken out by following the temperature change. If not all are taken out, the press will automatically stop, and the operator will handle it.

[0016] The said feeding position detection includes the following steps:

[0017] Use a near-infrared industrial camera to take pictures of the blank at a furnace temperature of 850 degrees Celsius from the upper oblique side of the blank. The specific implementation method is as follows:

[0018] 1) Take pictures of the position image of the blank placed normally, transmit the image to the positioning program on the industrial control computer. The program finds the position information of the key position points in the image through visual algorithms, and saves the position information into the recipe template as a positioning template.

[0019] 2) When changing the recipe during the production process, the PLC transmits the recipe information to the positioning program on the industrial control computer. The program retrieves the positioning template information for backup according to the recipe information. Each time the manipulator places the 850°C blank into the mold, it sends a signal to the positioning program at the same time. The positioning program reads the image information on the current near-infrared camera, and the positioning program finds the key position information through visual algorithms and compares it with the key position information on the positioning template. If the similarity is less than a certain range, it means the blank is placed unevenly.

[0020] 3) The positioning software on the industrial control computer transmits the position information of the blank to the PLC of the press. The PLC of the press controls whether to press down according to the signal transmitted by the positioning software. When the signal obtained by the PLC is that the blank is placed correctly, the press presses down to continue production. When the signal obtained by the PLC is that the blank is placed unevenly, production stops, and the audible and visual alarm indicator reminds the operator to intervene and take out the unevenly placed blank. After the blank is taken out, production continues.

[0021] The key position information in step 2) includes two parts:

[0022] 2.1) The position area delimited when creating the positioning template. If any part of the blank exceeds the delimited position area after production feeding, it means the blank is placed unevenly, otherwise it is placed correctly.

[0023] 2.2) The positioning pin when creating the positioning template. If the top of the positioning pin can be found after production feeding, it means the blank is placed correctly, otherwise it is placed unevenly.

[0024] The said double-blank detection includes the following steps:

[0025] 1) The method for detecting double materials by a far-infrared thermal imager is that the double-material detection software on the industrial control computer obtains the material arrival information of the material sheet sent by the PLC → the thermal imager takes a picture → the double-material detection software separates the material sheet through the contour discovery algorithm → calculates the average temperature of the separated material sheet → compares it with the set average temperature, and if it exceeds or is lower than the set parameter range, it is double material;

[0026] 2) The method for detecting double materials by a near-infrared industrial camera is that the double-material detection software on the industrial control computer obtains the material arrival information of the material sheet sent by the PLC → the near-infrared industrial camera takes a picture → the double-material detection software separates the contour of the material sheet → calculates the average brightness of the separated material sheet → compares it with the set average brightness, and if it exceeds or is lower than the set parameter range, it is double material.

[0027] The method for separating the contour by the double-material detection software in step 2) is as follows:

[0028] 1.1) Set the gray value of each pixel point to the median of the gray values of all pixel points within a certain neighborhood window of this point through the median filtering algorithm in the double-material detection software to eliminate noise points in the image;

[0029] 1.2) Correct the pictures with too high or too low gray levels to enhance the contrast and edge enhancement;

[0030] 1.3) Calculate the gradient to find the position with the strongest change in gray intensity in a picture. The strongest change in gray intensity means that after obtaining the direction and magnitude of the gradient, the entire image should be scanned to remove those points that are not on the boundary and find the complete contour of the material sheet in the image;

[0031] 1.4) Perform a logical AND operation on the image within the contour and the original image to separate the material sheet from the background.

[0032] The detection of the removal of the formed part includes the following steps:

[0033] 1) Measure the temperature in the mold cavity by a thermal imager. By measuring the mold cavity temperature at a specific time, it can be judged whether the formed part has been removed. When the material sheet has not been removed, it is 50 - 100 degrees Celsius higher than when it is normally removed;

[0034] 2) Detect through a visible light industrial camera: The camera takes a complete picture of the formed part and transmits the picture to the industrial control computer, and the visual processing program receives the picture.

[0035] Through step 2) above, the visual processing program performs the following operations:

[0036] 2.1) Image enhancement: Through low-pass filtering, obtain the part with a large gradient change in the gray value and superimpose it with the original gray value, perform an addition operation, and highlight the part with a large gradient change in the image;

[0037] 2.2) Remove noise. For each pixel value in the image, a value is obtained by weighted averaging its own value and the other pixel values in its neighborhood, and then this value is assigned to the original pixel value.

[0038] 2.3) Perform Canny edge detection on the image.

[0039] 2.4) Find the contours. By searching based on key factors such as contour area, circumscribed rectangle, length, and width, the number of formed parts is found and compared with the pre-set number. If it is less than the pre-set number, it can be determined whether the formed parts have been completely taken away. If the formed parts have not been completely taken out, the press is stopped and an audible and visual alarm is given to prompt the on-site workers to take them out.

[0040] Advantages of the present invention

[0041] The present invention uses visible light industrial cameras, near-infrared industrial cameras, and far-infrared thermal imagers to take pictures to detect the positioning, double sheets, and whether the formed parts are taken out in the hot stamping production line of ultra-high strength steel. It can save labor costs and reduce labor intensity. In the prior art, there were originally 24-hour three-shift operators in front of each press, which is equivalent to saving the labor costs of three workers; since the temperature of the raw materials in the hot forming line is 900°C and the on-site working conditions are very poor, which affects the physical and mental health of the operators. After adopting the present invention, the operators do not need to be on duty at the scene all the time, reducing the labor intensity and being beneficial to the physical and mental health of the operators. It is especially suitable for the positioning detection of hot formed workpieces in the ultra-high strength steel hot stamping production line, avoiding the die losses caused by the damaged die due to the misaligned steel plate in the prior art and the production downtime losses, reducing the appearance of defective products, improving product quality, and improving production efficiency.

[0042] In addition, during the production process, the detection of the feeding position of the sheet, the detection of double sheets, and the detection of whether all the formed parts are taken out are all automatically realized by the computer using artificial intelligence, with high efficiency and good accuracy.

[0043] 2. The present invention improves the production efficiency of the whole line. By automatically positioning each hot forming line, the operating time is increased by an average of 1.5 hours per day, improving the economic benefits of the enterprise. The press is performed 3 times per minute and the comprehensive output per time is 400 yuan. At the same time, the comprehensive output per minute of each hot forming line is 1200 yuan, which is equivalent to Average Each production line increases the economic benefits by more than 90,000 yuan per day.

[0044] 3. The present invention can reduce die damage, reduce maintenance and production costs, and improve work efficiency and economic benefits. Each stamping with the raw material (steel plate) placed incorrectly will cause great damage to the die. The cost of each die is about 1 million yuan. The present invention can minimize die damage.

[0045] 4. The present invention can automatically judge unexpected situations in the production process of high-strength steel through visual recognition software and can achieve automatic shutdown. In this way, on-site operators do not need to approach the press to observe production, which can effectively avoid personal injuries caused by equipment accidents and can also effectively avoid occupational injuries caused by dust, noise, etc., and has significant social benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic process flow diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0048] Combined with Figure 1 , a method for safety detection in the die area of a high-strength steel hot stamping production line, including an industrial control computer. The industrial control computer sends the matching rate information to the PLC through a communication module. An image recognition software is installed on the industrial control computer. The industrial control computer controls the manipulator to feed the blank into the die and take out the formed part after stamping, and includes the following steps:

[0049] 1) Detection of feeding position

[0050] Receive the die area picture taken by the near-infrared industrial camera through the image recognition software on the industrial control computer, judge whether the blank is placed correctly. If it is not placed correctly, the press will automatically shut down and the operator will handle it;

[0051] 2) Detection of double blanks, that is, detection of two blanks stacked on top of each other, including:

[0052] A. Receive the die area picture taken by the far-infrared thermal imager through the image recognition software on the industrial control computer, obtain the temperature through the picture, and judge whether it is a double blank by following the temperature change. If it is a double blank, the press will automatically shut down and the operator will handle it;

[0053] B. Receive the die area picture taken by the near-infrared industrial camera through the image recognition software on the industrial control computer, judge whether it is a double blank. If it is a double blank, the press will automatically shut down and the operator will handle it;

[0054] Verify whether it is a double blank through the near-infrared camera and the far-infrared thermal imager at the same time.

[0055] 3) Detection of taking away the formed part, including:

[0056] A. Receive the pictures of the blanking table taken by the visible light industrial camera through the image recognition software on the industrial control computer, and judge whether all the formed parts are taken out. If not all are taken out, the press will automatically stop, and the operator will handle it.

[0057] B. Receive the pictures of the die area taken by the far-infrared thermal imager through the image recognition software on the industrial control computer, obtain the temperature from the pictures, and follow the temperature change to judge whether all the formed parts are taken out. If not all are taken out, the press will automatically stop, and the operator will handle it.

[0058] In the above production process, the detection of the feeding position of the blank, the detection of double blanks, and the detection of whether all the formed parts are taken out are all automatically realized by the computer in an artificial intelligence manner, with high efficiency and good accuracy.

[0059] I. The detection of the feeding position includes the following steps:

[0060] Use a near-infrared industrial camera to take pictures of the blank with a furnace temperature of 850 degrees Celsius from the oblique upper part of the blank (the press slider cannot install a camera directly above). According to Wien's displacement law λmaxT = 2897.6, where T is the absolute temperature, 850 degrees Celsius is approximately equal to the absolute temperature of 1125K, and it is deduced that λmax = 2.57 microns, which belongs to the near-infrared band (0.75 - 3 microns). Therefore, using a near-infrared camera can present clear images and is little affected by ambient light. The specific implementation method is as follows:

[0061] 1) Take pictures of the position image of the blank placed normally, and transmit the image to the positioning program on the industrial control computer. The program finds the position information of the key position points in the image through visual algorithms, and saves the position information to the recipe template. (The recipe is the name in the hot forming production process and the process data of a workpiece production, such as the process data of the production of an A-pillar of a car and a recipe), as a positioning template.

[0062] 2) When changing the recipe during the production process, the PLC transmits the recipe information to the positioning program on the industrial control computer. The program retrieves the positioning template information for standby according to the recipe information. Each time the manipulator places the blank at 850°C into the mold, it sends a signal to the positioning program at the same time. The positioning program reads the image information on the current near-infrared camera. The positioning program finds the key position information through visual algorithms and compares it with the key position information on the positioning template. If the similarity is less than a certain range, it means the blank is placed offset.

[0063] The key position information includes two parts:

[0064] 2.1) The position area delimited when creating the positioning template. If any part of the blank exceeds the delimited position area after production feeding, it means the blank is placed offset, otherwise it is placed correctly.

[0065] 2.2) When creating the positioning template, mark the positioning pins (the positioning pin is a process in the hot forming production process. It can be understood that there are two or more holes on the blank, and two or more vertical thin metal rods are inserted into the corresponding holes respectively). After the production materials are fed, if the top of the positioning pin can be found, it means the blank is placed correctly; otherwise, it is placed incorrectly.

[0066] 3) The positioning software on the industrial control computer transmits the position information of the blank to the PLC (Programmable Logic Controller) of the press. The PLC of the press controls whether to press down according to the signal transmitted by the positioning software. When the signal obtained by the PLC indicates that the blank is placed correctly, the press presses down to continue production. When the signal obtained by the PLC indicates that the blank is placed incorrectly, production stops, and the audible and visual alarm indicator lights remind the on-site operator to intervene and take out the incorrectly placed blank. After the blank is taken out, production continues.

[0067] II. The double blank detection includes the following steps:

[0068] Double blanks are accidents that must be strictly avoided during the hot stamping production of high-strength steel. Double blanks will break the mold, causing the entire production line to stop for at least one day, and will also cause irreparable damage to the mold. The loss for one minute of downtime is approximately 1,200 yuan. Calculated in this way, the loss for one day of downtime is approximately more than 100,000 yuan. The loss of the mold is more than 1 million yuan. According to the on-site process, there are two reasons for the formation of double blanks: 1. When feeding the blank into the heating furnace, two blanks are fed, and the two blanks are stacked together. In this case, the temperature of the blank fed into the mold is much higher than that of a non-double blank. 2. When the blank is fed into the heating furnace, it is a single blank, but after it comes out of the furnace, the manipulator stacks the blank that was not properly taken away last time when feeding the material. In this case, the temperature of the blank fed into the mold is much lower than that of a non-double blank.

[0069] Verify whether there is a double blank through a near-infrared camera and a far-infrared thermal imager at the same time:

[0070] 1) The method of detecting double blanks through a far-infrared thermal imager is that the double blank detection software on the industrial control computer obtains the blank arrival information sent by the PLC → the thermal imager takes a picture → the double blank detection software separates the blank through the contour search algorithm → calculates the average temperature of the separated blank → compares it with the set average temperature. If it exceeds or is lower than the set parameter range, it is a double blank; the method for the double blank detection software to perform contour separation is as follows:

[0071] 1.1) Through the median filtering algorithm of the double blank detection software, set the gray value of each pixel point to the median of the gray values of all pixel points within a certain neighborhood window of this point to eliminate the noise points in the image;

[0072] 1.2) Correct the pictures with too high or too low gray levels, enhance the contrast, and enhance the edges;

[0073] 1.3) Calculate the gradient to find the position with the strongest change in grayscale intensity in an image. By the strongest change in grayscale intensity, it means that after obtaining the direction and magnitude of the gradient, a scan should be performed on the entire image to remove the non-boundary points and find the complete contour of the workpiece in the image;

[0074] 1.4) Perform a logical AND operation on the image within the contour and the original image to separate the workpiece from the background.

[0075] 2) The method for detecting double workpieces through a near-infrared industrial camera is that the double-workpiece detection software on the industrial control computer obtains the workpiece arrival information sent by the PLC → the near-infrared industrial camera takes a picture → the double-workpiece detection software separates through the contour search algorithm Discharge Sheet → Calculate the average brightness of the separated workpieces → Compare it with the set average brightness. If it exceeds or is lower than the set parameter range, it is a double workpiece; among them, the contour separation method of the double-workpiece detection software is the same as that of the far-infrared thermal imager.

[0076] III: The detection of the removal of the formed part includes the following steps:

[0077] If the formed part is not removed in time, when a new workpiece is put into the mold for stamping, it will damage the mold, causing the entire production line to stop for at least one day, and it will also cause irreparable damage to the mold. The loss for one minute of downtime is approximately 1,200 yuan. Calculated in this way, the loss for one day of downtime is approximately hundreds of thousands of yuan, and the mold loss is more than 1 million. Even more seriously, it may cause personal injuries. Therefore, it is necessary to strictly avoid the problem that the formed part cannot be removed in time.

[0078] Verify whether the formed part is removed through a visible light camera and a far-infrared thermal imager at the same time:

[0079] 1) Measure the temperature in the mold cavity through the thermal imager. By measuring the temperature of the mold cavity at a specific time, it can be judged whether the formed part is removed. When the workpiece is not removed, the temperature is 50 - 100 degrees Celsius higher than when it is normally removed;

[0080] 2) Detect through a visible light industrial camera. The camera takes a complete picture of the formed part and transmits the picture to the visual positioning program installed on the industrial control computer. It is necessary to install a lighting source at the same time. The camera is installed 3.5 meters directly above the blanking table. The visual processing program performs the following operations:

[0081] 2.1) Image enhancement: Through low-pass filtering, the part with a large gradient change in grayscale values is added to the original grayscale values, so as to highlight the part with a large gradient change in the image;

[0082] 2.2) Remove noise For each pixel value in the image, a value is obtained by weighted averaging its own value and the other pixel values in its neighborhood, and then this value is assigned to the original pixel value;

[0083] 2.3) Perform Canny edge detection on the image;

[0084] 2.4) Find the contours. By searching based on key factors such as the contour area, bounding rectangle, length, and width, the number of formed parts is found and compared with the pre-set number. If it is less than the pre-set number, it can be determined whether the formed parts have been completely removed. If the formed parts have not been completely taken out, the press is stopped, and an audible and visual alarm is given to prompt the on-site workers to take them out.

[0085] The above embodiments merely illustrate the principles and effects of the present invention for patents by way of example, rather than limiting the present invention for patents. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention for patents. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the relevant technical field without departing from the spirit and technical ideas disclosed by the present invention for patents are still covered by the claims of the present invention for patents.

Claims

1. A method for safety detection in the die area of a high-strength steel hot stamping production line, including an industrial control computer on which image recognition software is installed. The industrial control computer controls a manipulator to feed a blank into a die and take out the formed part after stamping. It is characterized in that, Including the following steps: 1) Detection of feeding position Receive the picture of the die area taken by the near-infrared industrial camera through the image recognition software on the industrial control computer, and judge whether the workpiece is placed correctly. If it is not placed correctly, the press will automatically stop, and the operator will handle it; 2) Double-workpiece detection. Simultaneously verify whether there are double workpieces through the near-infrared camera and the far-infrared thermal imager, including: A. Receive the picture of the die area taken by the far-infrared thermal imager through the image recognition software on the industrial control computer, obtain the temperature from the picture, and follow the temperature change to judge whether there are double workpieces. If there are double workpieces, the press will automatically stop, and the operator will handle it; B. Receive the picture of the die area taken by the near-infrared industrial camera through the image recognition software on the industrial control computer, and judge whether there are double workpieces. If there are double workpieces, the press will automatically stop, and the operator will handle it; 3) Detection of taking away the formed part, including: A. Receive the picture of the material taking table taken by the visible light industrial camera through the image recognition software on the industrial control computer, and judge whether all the formed parts are taken out. If not all are taken out, the press will automatically stop, and the operator will handle it; B. Receive the picture of the die area taken by the far-infrared thermal imager through the image recognition software on the industrial control computer, obtain the temperature from the picture, and follow the temperature change to judge whether all the formed parts are taken out. If not all are taken out, the press will automatically stop, and the operator will handle it; The detection of the feeding position includes the following steps: Use the near-infrared industrial camera to take pictures of the workpiece with a furnace temperature of 850 °C from the obliquely upper side of the workpiece. The specific implementation method is as follows: 1) Take pictures of the position image of the workpiece placed normally, and transmit the image to the positioning program on the industrial control computer. The program finds the position information of the key position points in the image through the vision algorithm, and saves the position information into the recipe template as a positioning template; 2) When changing the recipe during the production process, the PLC transmits the recipe information to the positioning program on the industrial control computer. The program retrieves the positioning template information for backup according to the recipe information. Each time the manipulator places the workpiece at 850 °C into the mold, it sends a signal to the positioning program at the same time. The positioning program reads the image information on the current near-infrared camera, and the positioning program finds the key position information through the vision algorithm and compares it with the key position information on the positioning template. If the similarity is less than a certain range, it means that the workpiece is placed obliquely; 3) The positioning software on the industrial control computer transmits the position information of the workpiece to the PLC of the press. The PLC of the press controls whether to press down according to the signal transmitted by the positioning software. When the signal obtained by the PLC is that the workpiece is placed correctly, the press presses down to continue production. When the signal obtained by the PLC is that the workpiece is placed obliquely, production stops, and the audible and visual alarm indicator lights remind the operator to intervene to take out the obliquely placed workpiece. After the workpiece is taken out, production continues.

2. The safety detection method for the die area in the high-strength steel hot stamping production line according to claim 1, characterized in that, The key position information in step 2) includes two parts: 2.1) The position area delimited when creating the positioning template. If any part of the workpiece exceeds the delimited position area after production feeding, it means that the workpiece is placed obliquely, otherwise it is placed correctly; 2.2) The positioning pin when creating the positioning template. If the top of the positioning pin can be found after production feeding, it means that the workpiece is placed correctly, otherwise it is placed obliquely.

3. The safety detection method for the die area in the high-strength steel hot stamping production line according to claim 1, characterized in that, The double-workpiece detection includes the following steps: 1) The method for detecting double materials by a far-infrared thermal imager is that the double-material detection software on the industrial control computer obtains the material arrival information of the material sheet sent by the PLC → the thermal imager takes a picture → the double-material detection software separates the material sheet through the contour search algorithm → calculates the average temperature of the separated material sheet → compares it with the set average temperature. If it exceeds or is lower than the set parameter range, it is double material; 2) The method for detecting double materials by a near-infrared industrial camera is that the double-material detection software on the industrial control computer obtains the material arrival information of the material sheet sent by the PLC → the near-infrared industrial camera takes a picture → the double-material detection software separates the material sheet by contour → calculates the average brightness of the separated material sheet → compares it with the set average brightness. If it exceeds or is lower than the set parameter range, it is double material.

4. The method for safety detection in the die area of the high-strength steel hot stamping production line according to claim 3, characterized in that, The method for separating the contour by the double-material detection software in step 2) is as follows: 1.1) Set the gray value of each pixel point to the median of the gray values of all pixel points within a certain neighborhood window of the pixel point through the median filtering algorithm in the double-material detection software to eliminate the noise points in the image; 1.2) Correct the pictures with too high or too low gray levels to enhance the contrast and edge enhancement; 1.3) Calculate the gradient to find the position with the strongest change in gray intensity in a picture. The strongest change in gray intensity means that after obtaining the direction and magnitude of the gradient, scan the entire image to remove those points that are not on the boundary and find the complete contour of the material sheet in the image; 1.4) Perform a logical AND operation on the image within the contour and the original image to separate the material sheet from the background.

5. The safety detection method for the die area in the high-strength steel hot stamping production line according to claim 1, characterized in that, The detection of the removal of the formed part includes the following steps: 1) Measure the temperature in the mold cavity by a thermal imager. By measuring the mold cavity temperature at a specific time, it can be judged whether the formed part has been removed. When the material sheet has not been removed, it is 50 - 100 degrees Celsius higher than when it is normally removed; 2) Detect through a visible light industrial camera. The camera takes a complete picture of the formed part and transmits the picture to the vision processing program on the industrial control computer, and the vision processing program receives the picture.

6. The safety detection method for the die area in the high-strength steel hot stamping production line according to claim 5, characterized in that, Through step 2), the vision processing program performs the following operations: 2.1) Image enhancement. Through low-pass filtering, the part with a large gradient change in the gray value and the original gray value are superimposed and an addition operation is performed to highlight the part with a large gradient change in the image; 2.2) Remove noise. For each pixel value in the image, a value is obtained by weighted averaging of itself and other pixel values in the neighborhood, and then this value is assigned to the original pixel value; 2.3) Perform canny edge detection on the image; 2.4) Search for contours. Through key factors such as contour area, circumscribed rectangle, length and width, find the number of formed parts and compare it with the pre-set number. If it is less than the pre-set number, it can be judged whether the formed part has been completely removed; if the formed part has not been completely removed, stop the press and give an audible and visual alarm to prompt the on-site workers to take it out.

Images