Saw tooth camera device and its usage method applied to automatic circular saw blade grinding machine

Abstract

This invention relates to a saw tooth camera device and its usage method for an automatic circular saw blade grinding machine. The device includes a high-pressure air chamber, a circumferential air curtain block, a supplementary light lamp bead hole, a camera fixing through hole, a vacuum extraction tube, a high-pressure air blowing tube, a connecting rod, a return spring, and a positioning coil. In use, two sets of devices are installed on both sides of the grinding wheel respectively, and the operation proceeds as follows: saw tooth imaging and parameter calculation, saw tooth wear calculation, saw tooth parameter selection and evaluation, saw tooth parameter optimization, saw tooth grinding accuracy verification, and saw blade flatness evaluation. The device of this invention features high reliability in drainage operation, no damage to the saw blade, simple and stable structure, and good scalability. The usage method of this invention has the ability to automatically and online realize saw tooth wear detection, saw tooth shape optimization, and saw blade flatness detection. Adding the device of this invention to an automatic grinding machine and using the method of this invention can greatly enhance the grinding accuracy and intelligence of the automatic grinding machine.

Description

Technical Field

[0001] This invention relates to a saw tooth camera device, and more particularly to a saw tooth camera device applied to an automatic circular saw blade grinding machine and its method of use. Background Technology

[0002] In actual manufacturing operations, materials such as sheet metal, cast iron, glass, stone, and plastics are frequently cut. Like turning, milling, planing, grinding, and fitting, cutting is a fundamental operation in manufacturing. A circular saw blade (or simply "saw blade") is a commonly used tool in the cutting process. A saw blade consists of a central body and outer teeth. During operation, the saw blade rotates at high speed, and the material is cut by the teeth located at the edge of the blade.

[0003] Automatic tooth grinding machines are commonly used equipment in industries related to the production and use of circular saw blades. They utilize grinding wheels as abrasives to process the tooth shape of the saw blades. The tooth grinding machine accompanies the entire lifecycle of the saw blade: during saw blade production, the final process is tooth preparation, which involves grinding teeth of a specific shape onto the edge of the saw blade; after a period of use, the teeth on the edge of the saw blade become dull, at which point an automatic tooth grinding machine is needed to re-grind the teeth to maintain their sharpness.

[0004] There are many types of saw teeth on circular saw blades, including arc teeth, wolf teeth, triangular teeth, high and low teeth, and chamfered teeth. The saw tooth parameters for each type of tooth are very important because they directly affect the saw blade's lifespan, cutting efficiency, cutting accuracy, cutting energy consumption, and other performance indicators.

[0005] Currently available automatic saw grinding machines cannot perform online real-time detection of tooth profile parameters during the saw tooth grinding process. Because the grinding process generates a large amount of heat, coolants such as soapberry liquid are typically used to dissipate heat and prevent damage to the circular saw blade. Consequently, a significant amount of coolant remains on the saw blade during grinding, making it impossible for laser or image-based detection methods to accurately measure the tooth profile parameters. The current standard practice is to disassemble the circular saw blade after grinding and then perform offline measurements using image inspection instruments, coordinate measuring machines, or other similar tools.

[0006] Because it is impossible to detect the saw tooth profile parameters online in real time, the saw tooth grinding process of the automatic tooth grinding machine cannot form a closed-loop feedback control. This means that the automatic tooth grinding machine cannot make adaptive tooth grinding adjustments according to the actual situation of the circular saw blade (such as adjusting the tooth grinding feed amount, tooth grinding feed speed, etc.), and therefore cannot guarantee the saw tooth grinding accuracy. Summary of the Invention

[0007] To overcome the shortcomings of existing technologies, this invention provides a saw tooth camera device and its usage method for an automatic circular saw blade grinding machine. The device can identify saw tooth parameters online through saw tooth shooting and image processing. Furthermore, the device uses a triple drainage structure design to prevent coolant from entering the saw tooth area being shot, thereby ensuring the accuracy of saw tooth profile parameter detection.

[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0009] A saw tooth camera device for use in an automatic circular saw blade grinding machine includes a high-pressure air chamber, a circumferential air curtain block, a supplementary light lamp hole, a camera fixing through hole, a vacuum extraction tube, a high-pressure blowing tube, a connecting rod, a return spring, and a positioning coil. The high-pressure air chamber is connected to the circumferential air curtain block. The camera fixing through hole is located at the center of the high-pressure air chamber and the circumferential air curtain block, and the supplementary light lamp hole is located at the bottom of the circumferential air curtain block. The vacuum extraction tube and the high-pressure blowing tube both penetrate the high-pressure air chamber and the circumferential air curtain block, and are respectively installed on both sides of the camera fixing through hole. The connecting rod is located above the high-pressure air chamber, and its two ends are fixed to the vacuum extraction tube and the high-pressure blowing tube respectively by interference fit. The return spring is installed in the upper cavity wall of the high-pressure air chamber and is coaxially installed with the vacuum extraction tube. The positioning coil is installed in the upper cavity wall of the high-pressure air chamber and is coaxially installed with the high-pressure blowing tube.

[0010] The high-pressure air chamber has an annular cavity surrounding the camera fixing hole at its center, and an air inlet is opened on each of the two side walls of the high-pressure air chamber. The upper half of the circumferential air curtain block has several cylindrical through holes distributed around the periphery of the camera fixing hole, and the cylindrical through holes are connected to the annular cavity of the high-pressure air chamber. The lower half of the circumferential air curtain block has an annular through hole with an inclined cross-section.

[0011] Preferably, the high-pressure air chamber and the circumferential air curtain block are two cylinders with the same bottom diameter, and they are coaxially installed.

[0012] Preferably, the supplementary light lamp holes are located on the bottom surface of the circumferential air curtain block and are evenly distributed around the outer circumference of the camera fixing through hole.

[0013] Preferably, the bottom of the high-pressure blowing tube is 0.5 mm to 1 mm higher than the bottom of the vacuum extraction tube.

[0014] Preferably, the high-pressure air pipe corresponding to the positioning coil is covered with a metal outer tube, and the height of the metal outer tube is equal to that of the positioning coil.

[0015] Preferably, the vacuum extraction tube has an extraction port at the bottom, the extraction port being stepped in shape, the extraction port including a first step line and a second step line, the width of the first step line and the second step line being equal; the width of the first step line and the second step line is 1 mm to 3 mm.

[0016] The present invention also provides a method for using the saw tooth camera device applied to the above-mentioned automatic saw tooth grinding machine for circular saw blades. Two sets of saw tooth camera devices are installed on both sides of the grinding wheel respectively, and the operation is carried out according to the following process: saw tooth shooting and parameter calculation, saw tooth wear calculation, saw tooth parameter selection and evaluation, saw tooth parameter optimization, saw tooth grinding accuracy verification, and saw blade flatness evaluation.

[0017] Sawtooth imaging and parameter calculation: This includes imaging the sawtooth before and after grinding and calculating the tooth shape parameters. The tooth shape parameter calculation process includes sawtooth edge recognition, sawtooth feature point extraction, and sawtooth parameter calculation.

[0018] Sawtooth wear calculation: Subtract the standard tooth profile from the tooth profile before grinding to calculate the wear values ​​of the rake angle, clearance angle, and tooth back of the sawtooth before grinding;

[0019] Saw tooth parameter selection evaluation: Using the support vector machine (SVM) model, the three parameters of average wear value of the front angle, average wear value of the back angle, and average wear value of the tooth back are used as input parameters of the SVM model. The appropriateness of saw tooth parameter selection and the inappropriateness of saw tooth parameter selection are used as output parameters of the SVM model to judge whether the saw tooth parameters are suitable for the saw blade working conditions.

[0020] Sawtooth parameter optimization: Adjust the average wear value of the rake angle, the average wear value of the clearance angle, and the average wear value of the tooth back of the tooth to be refurbished;

[0021] Sawtooth grinding accuracy verification: Subtract the standard tooth profile from the ground tooth profile to calculate the grinding error of the sawtooth tooth profile parameters;

[0022] Saw blade flatness assessment: Based on the tooth width and tooth depth parameters after grinding, calculate the flatness of different teeth on the same saw blade after grinding, and then determine whether the ground saw blade is qualified.

[0023] As a preferred method, the sawtooth edge recognition uses algorithms including the Canny operator detection to extract the lines of the outermost edge of the sawtooth; sawtooth feature point extraction extracts five feature points that are easy to identify from the image; sawtooth parameter calculation calculates five parameters: back angle α, front angle γ, tooth back Rb, tooth width T, and tooth depth H, using the five feature points.

[0024] Saw tooth wear calculation involves subtracting the rake angle, clearance angle, and tooth back parameters of each saw tooth before regrinding from the standard tooth profile's three parameters. Then, the average of the resulting rake angle difference, clearance angle difference, and tooth back difference values ​​for all saw teeth is calculated. The resulting average rake angle wear value Δγ, average clearance angle wear value Δα, and average tooth back wear value ΔRb represent the amount of wear on the rake angle, clearance angle, and tooth back parameters of the circular saw blade during the previous cutting operation.

[0025] As a preferred embodiment, the SVM model is represented as follows:

[0026] w T x+b=0

[0027] In the above formula, x represents the input parameters of the SVM model, w represents the normal vector of the classification hyperplane, T represents the transpose of the vector, and b represents the offset of the hyperplane relative to the origin; among them, the input parameters x are three parameters: the average wear value of the front angle Δγ, the average wear value of the rear angle Δα, and the average wear value of the tooth back ΔRb.

[0028] Assuming the output parameter of the SVM model is y, the output parameter y can only take two values: y = +1 indicates that the sawtooth parameter selection is appropriate, and y = -1 indicates that the sawtooth parameter selection is inappropriate.

[0029] Each circular saw blade yields a data point (x, y). Using multiple circular saw blades of the same type, a sample set is obtained to train the SVM model. This sample set is represented as {(x1, y1), (x2, y2), ..., (x...}. i ,y i )}, where: x i ∈R d y i ∈{+1,-1} is the category label, i = 1, 2, ..., n;

[0030] After the SVM model is trained to maturity using the above sample set, it can be used to evaluate whether the selection of saw tooth parameters is appropriate. That is, the three parameters, namely the average wear value of the front angle Δγ, the average wear value of the rear angle Δα, and the average wear value of the tooth back ΔRb, are input into the SVM model, and then it is obtained whether the saw tooth parameters of the circular saw blade match the on-site working conditions, that is, whether the selection of saw tooth parameters is appropriate.

[0031] As a preferred option, the sawtooth parameters are optimized. If the above SVM model determines that the sawtooth parameter selection is inappropriate, then the sawtooth tooth profile parameters to be reground are adjusted. The specific adjustment principle is as follows: First, adjust the three parameters of the standard tooth profile: rake angle, clearance angle, and tooth back. Second, subtract the adjusted standard tooth profile parameters from the tooth profile parameters before reground. Finally, determine whether the adjusted average wear values ​​of the rake angle, clearance angle, and tooth back are equal. If they are equal, the adjustment ends; otherwise, repeat the above steps.

[0032] The saw tooth grinding accuracy is verified by subtracting the actual front angle, back angle, and tooth back angle values ​​of the ground saw teeth from the standard tooth profile values ​​to check whether the automatic tooth grinding machine has ground the saw teeth accurately. A threshold is set, and if the error of any parameter of any tooth on any circular saw blade exceeds 2%, the operator is required to intervene and further review the saw blade.

[0033] The flatness of a saw blade is evaluated using two parameters: tooth width (T) and tooth depth (H). The flatness of the saw blade is judged based on the deviation values ​​of the tooth width (T) and tooth depth (H) of all teeth on the blade. The judgment threshold is determined by three parameters: the diameter of the saw blade, the number of teeth, and the allowable range of flatness error. The deviation value of the tooth width (T) of all teeth should be within ±1%, and the deviation value of the tooth depth (H) of all teeth should be within ±2.4%. Otherwise, the operator should be required to intervene and further review the saw blade.

[0034] The beneficial effects of this invention are as follows:

[0035] (1) High reliability: The triple drainage structure design of external drainage, internal blowing and central pumping ensures that the coolant in the area to be tested can be completely removed, thereby ensuring the accuracy of the test.

[0036] (2) No damage to the saw blade: Since the liquid discharge and blowing process is completed by gas and the liquid extraction process is vacuumed, and no part comes into contact with the saw blade throughout the process, it will not cause scratches or other damage to the surface of the saw blade.

[0037] (3) Good scalability: The present invention has a simple structure and low cost. The required air source, vacuum pump and other equipment are also commonly used equipment in saw blade manufacturers. Therefore, the present invention has good scalability.

[0038] (4) Multiple detectable parameters: Through the device of the present invention and the method of using the device of the present invention, the saw tooth wear detection, saw tooth shape optimization and saw blade flatness detection can be completed automatically. All of the above processes are completed online during the grinding process of the circular saw blade and will not interrupt the grinding process of the circular saw blade.

[0039] (5) The device proposed in this invention can be used as an important component of an automatic tooth grinding machine, laying the foundation for adding closed-loop feedback control to the automatic tooth grinding machine, which can improve the grinding accuracy of the saw teeth by the automatic tooth grinding machine; the method of using the device proposed in this invention can greatly enhance the intelligence of the automatic tooth grinding machine for circular saw blades. Attached Figure Description

[0040] Figure 1 This is a partial cross-sectional structural diagram of the present invention. Figure 1 ;

[0041] Figure 2This is a partial cross-sectional structural diagram of the present invention. Figure 2 ;

[0042] Figure 3 This is a cross-sectional view of the present invention;

[0043] Figure 4 This is a schematic diagram of the working principle of the present invention;

[0044] Figure 5 This is a schematic diagram of the bottom structure of the vacuum extraction tube of the present invention;

[0045] Figure 6 This is a schematic diagram illustrating the calculation method of the sawtooth parameter scale of the present invention;

[0046] Figure 7 This is a schematic diagram of the arrangement structure of the present invention in actual use;

[0047] Figure 8 This is a flowchart of the method of using the device of the present invention;

[0048] Figure 9 This is a schematic diagram of the sawtooth parameter calculation method based on image recognition of the present invention;

[0049] Illustrations: 1. High-pressure air chamber, 1a. High-pressure air chamber inlet, 2. Circumferential air curtain block, 2a. Cylindrical through hole, 2b. Annular through hole, 3. Supplemental light lamp bead hole, 4. Camera fixing through hole, 5. Vacuum extraction tube, 5a. Extraction port, 5b. First step line, 5c. Second step line, 6. High-pressure blowing pipe, 6a. Metal outer tube, 7. Connecting rod, 8. Return spring, 9. Positioning coil, 10. Overall assembly of the device of this invention, 11. Circular saw blade, 12. Grinding wheel. Detailed Implementation

[0050] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the scope of protection of the present invention is not limited thereto.

[0051] Reference Figure 1-3 A saw tooth camera device for use in an automatic circular saw blade grinding machine, the saw tooth camera device includes a high-pressure air chamber 1, a circumferential air curtain block 2, a supplementary light lamp bead hole 3, a camera fixing through hole 4, a vacuum extraction tube 5, a high-pressure air blowing tube 6, a connecting rod 7, a return spring 8, and a positioning coil 9.

[0052] Both the high-pressure air chamber 1 and the circumferential air curtain block 2 are cylindrical, and both have a cylindrical camera mounting hole 4 at their center. The bottom diameter of the high-pressure air chamber 1 is the same as that of the circumferential air curtain block 2, and they are coaxially mounted. An annular cavity surrounding the camera mounting hole 4 is formed in the center of the high-pressure air chamber 1, and an air inlet (high-pressure air chamber air inlet) 1a is formed on each of the two side walls of the high-pressure air chamber 1. The upper half of the circumferential air curtain block 2 has several cylindrical through holes 2a distributed around the periphery of the camera mounting hole 4, and these cylindrical through holes communicate with the annular cavity of the high-pressure air chamber. The lower half of the circumferential air curtain block 2 has an annular through hole 2b with an inclined cross-section to allow for the gas to be ejected obliquely outwards. Several supplementary lighting lamp holes 3 are evenly distributed around the outer periphery of the camera mounting hole 4 on the bottom surface of the circumferential air curtain block 2, allowing for the installation of camera supplementary lighting lamps for capturing clear images.

[0053] In this invention, the vacuum extraction tube 5 and the high-pressure blowing tube 6 both penetrate the high-pressure air chamber 1 and the circumferential air curtain block 2, and are respectively installed on both sides of the camera fixing through hole 4 in a straight line. At the same time, the bottom of the high-pressure blowing tube 6 is 0.5 mm to 1 mm higher than the bottom of the vacuum extraction tube 5. The connecting rod 7 is fixed above the high-pressure air chamber 1, and the fixing holes on both sides are respectively fixed to the vacuum extraction tube 5 and the high-pressure blowing tube 6 by interference fit. The return spring 8 is installed in the upper cavity wall of the high-pressure air chamber 1 and is coaxially installed with the vacuum extraction tube 5. The positioning coil 9 is installed in the upper cavity wall of the high-pressure air chamber 1 and is coaxially installed with the high-pressure blowing tube 6. The high-pressure blowing tube 6 is made of non-metallic material, and the part of the high-pressure blowing tube 6 corresponding to the positioning coil 9 is covered with a metal outer tube 6a, and the height of the metal outer tube 6a is equal to that of the positioning coil 9. In the initial state, the positioning coil 9 is level with the metal outer tube 6a.

[0054] The device of the present invention is driven by a motor, which is controlled by a controller.

[0055] When the device of this invention is working, the bottom of the vacuum extraction tube should be as close as possible to the circular saw blade, but not in contact with it. This is because the circular saw blade needs to rotate continuously during saw tooth grinding, and if the bottom of the vacuum extraction tube is in contact with the circular saw blade, it will cause scratches on the surface of the circular saw blade. This invention solves the problem of how to position the device and the circular saw blade through the design of the connecting rod, the return spring, and the positioning coil. The specific process is as follows:

[0056] First, the device of the present invention continuously approaches the circular saw blade. When the bottom of the vacuum extraction tube 5 contacts the circular saw blade, the vacuum extraction tube 5 will be pushed upward. At this time, the return spring 8 will be compressed. Since the high-pressure air blowing tube 6 is fixed to the vacuum extraction tube 5 through the connecting rod 7, the high-pressure air blowing tube 6 will also be pushed upward synchronously to prevent the bottom of the high-pressure air blowing tube from contacting and damaging the surface of the circular saw blade.

[0057] Secondly, the outer metal tube 6a of the high-pressure air pipe 6 is also pushed upwards, creating a misalignment between it and the positioning coil 9, thus changing the self-inductance value of the positioning coil 9. The detection circuit (e.g., an AC bridge) can detect the change in the self-inductance value of the positioning coil, and after the controller reads the signal value from the detection circuit, it immediately stops the drive motor, that is, it stops the device of the present invention from further approaching the circular saw blade.

[0058] Finally, the controller drives the motor to rotate in the opposite direction, that is, drives the device of the present invention to move away from the circular saw blade. At the same time, the controller synchronously detects the self-inductance value of the positioning coil. When the self-inductance value returns to its original value, the drive motor is immediately stopped. At this point, the positioning between the device of the present invention and the circular saw blade is completed, and the bottom of the vacuum extraction tube is close to the circular saw blade, but does not touch it.

[0059] like Figure 4 As shown, this invention, through a triple drainage structure design, completely removes the coolant from the automatic gear grinding machine's field of view. The detailed design concept and working method are shown in the following figure:

[0060] First, the outer annular air curtain removes the coolant. High-pressure gas 1 enters the annular cavity through the air inlet of the high-pressure gas chamber, and then enters the annular through-hole through the cylindrical through-hole inside the circumferential air curtain block, finally forming a circumferential annular air curtain that is sprayed obliquely downwards and outwards to prevent the coolant of the automatic gear grinding machine from entering the camera's field of view.

[0061] Secondly, use a high-pressure air hose for directional air blowing. An annular air curtain is insufficient to completely remove coolant, especially coolant droplets adhering to the circular saw blade's surface, which will pass through the annular air curtain as the saw blade rotates. In this case, use a high-pressure air hose to blow air outwards in the opposite direction of the saw blade's movement, expelling the coolant droplets brought in by the saw blade. The direction of the air blown out by the high-pressure air hose should be consistent with the direction of the air in the annular air curtain to create a combined force, and opposite to the direction of the saw blade's movement to achieve the best removal effect.

[0062] Finally, the vacuum extraction tube removes the droplets. After directional air blowing through the high-pressure air pipe, there are virtually no coolant droplets left. However, just in case, for example, a small droplet might leak out from the camera mounting hole, this invention includes a vacuum extraction tube. The air extraction direction of the vacuum extraction tube is consistent with the movement direction of the circular saw blade to achieve the best extraction effect.

[0063] This invention employs a triple drainage structure design—outer drainage, inner blowing, and central extraction—to ensure complete removal of coolant from the area to be inspected, thereby guaranteeing the camera's detection accuracy of the saw teeth. Simultaneously, during the drainage, blowing, and extraction processes, no components come into contact with the saw blade, thus preventing scratches or other damage to the saw blade surface. The device proposed in this invention can serve as a crucial component of an automatic gear grinding machine, laying the foundation for closed-loop feedback control and improving the grinding accuracy of the saw teeth.

[0064] Reference Figure 5 The vacuum extraction tube of the present invention has an extraction port 5a at its bottom. The extraction port 5a has a stepped shape and includes a first step line 5b and a second step line 5c. Figure 5 The first and second step lines shown have equal widths, ranging from 1 mm to 3 mm to accommodate most sawtooth sizes. In this invention, the first and second step lines are used to provide a scale for the image processing process, facilitating more accurate calculation of sawtooth parameters and saw blade distance. The specific principle is as follows: Figure 6 As shown.

[0065] Figure 6 The left half shows an image captured by the camera inside the device of this invention. Because the second step line at the bottom of the vacuum extraction tube is closer to the camera than the first step line, the width of the second step line is greater than the width of the first step line within the camera's field of view. Since the actual widths of the first and second step lines, the height difference between the first and second step lines, and the difference between the first step line and the circular saw blade are known, therefore... Figure 6 The right half shows the height value between the camera and the second step line. Figure 6 (As shown by the dotted line on the right half), the distance between the camera and the circular saw blade can be calculated based on geometric relationships. Furthermore, the actual value of the saw blade can be calculated based on the actual width of the first step line, the width of the first step line within the field of view, and the value of the sawtooth shape within the field of view.

[0066] Reference Figure 7 When using the device of the present invention, two devices can be used together. After the positioning between the device 10 of the present invention and the circular saw blade 11 is completed, one device 10 of the present invention is set in front of the grinding wheel 12 and another device 10 of the present invention is set in behind the grinding wheel. One device is in front of the grinding wheel to capture the saw tooth parameters before grinding, and the other device is behind the grinding wheel to capture the saw tooth parameters after grinding.

[0067] Reference Figure 8The above-mentioned method of using the saw tooth camera device applied to the automatic grinding machine for circular saw blades involves installing two sets of saw tooth camera devices on both sides of the grinding wheel and working according to the following process: saw tooth shooting and parameter calculation, saw tooth wear calculation, saw tooth parameter selection and evaluation, saw tooth parameter optimization, saw tooth grinding accuracy verification, and saw blade flatness evaluation; in order to complete a comprehensive evaluation of the circular saw blade and saw teeth.

[0068] The aforementioned saw tooth photography and parameter calculation steps include photographing the saw teeth before and after grinding and calculating the tooth shape parameters. Figure 9 This invention relates to a method for calculating jagged edges based on image recognition. The specific process includes three steps: jagged edge recognition, jagged feature point extraction, and jagged parameter calculation. The jagged edge recognition steps are as follows: Figure 9 As shown in Figure a, algorithms such as the Canny operator can be used to extract the outermost lines of the jagged edges. The jagged feature point extraction is as follows: Figure 9 As shown in b, five feature points (ABCDE) are extracted from the image. Among these feature points, point A, like point B, is not only the intersection of two straight lines but also the highest point; point E is the lowest point; points C and D are intersections of two lines, with a sudden change in the curvature of the arc. Therefore, all five points are easily identifiable from the image. The sawtooth parameter calculation steps involve using the five feature points ABCDE to calculate five parameters: back angle α, front angle γ, tooth back Rb, tooth width T, and tooth depth H.

[0069] The saw tooth wear calculation involves subtracting the standard tooth profile from the tooth profile before grinding to calculate the wear values ​​of the front angle, back angle, and tooth back of the saw tooth before grinding. Specifically, this means subtracting the front angle, back angle, and tooth back of each saw tooth before grinding from the initial parameter values ​​of the standard tooth profile (inputted by the operator on the grinding machine's touchscreen). Then, the average of the differences in front angle, back angle, and tooth back of all the saw teeth is calculated. The resulting average front angle wear value Δγ, average back angle wear value Δα, and average tooth back wear value ΔRb represent the amount of wear on the front angle, back angle, and tooth back of the circular saw blade during the previous round of cutting operations.

[0070] The aforementioned saw tooth parameter selection evaluation refers to using a support vector machine (SVM) model, taking the three parameters of the front angle average wear value Δγ, the rear angle average wear value Δα, and the tooth back average wear value ΔRb as input parameters of the SVM model, and taking the appropriate or inappropriate saw tooth parameter selection as output parameters of the SVM model (i.e., using the SVM model for binary classification), thereby judging whether the saw tooth parameters are suitable for the saw blade working conditions.

[0071] The mathematical principles of the SVM model of this invention are as follows:

[0072] The essence of the SVM model is to find a hyperplane in a multidimensional space that can separate data (especially nonlinear data) with the maximum margin, thereby achieving data classification. The equation of this hyperplane can be written in the following form:

[0073] w T x+b=0 (1)

[0074] In the above formula, x represents the input parameters of the SVM model, w represents the normal vector of the classification hyperplane, T represents the transpose of the vector, and b represents the offset of the hyperplane relative to the origin. The input parameters x consist of three parameters: the average wear value at the front angle Δγ, the average wear value at the rear angle Δα, and the average wear value on the tooth back ΔRb. Assuming the output parameter of the SVM model is y, in this invention, the output parameter y has only two values: y = +1 indicates that the sawtooth parameter selection is appropriate, and y = -1 indicates that the sawtooth parameter selection is inappropriate.

[0075] Each circular saw blade yields a data point (x, y). Therefore, by using a large number of circular saw blades of the same type, a sample set can be obtained to train the SVM model. The sample set can be represented as {(x1, y1), (x2, y2), ..., (x...}. i ,y i )}, where: x i ∈R d y i ∈{+1,-1} is the category label, i=1,2,…,n.

[0076] After the SVM model is trained to maturity using the above sample set, it can be used to evaluate whether the saw tooth parameters are appropriate. That is, the three parameters, namely the average wear value of the front angle Δγ, the average wear value of the rear angle Δα, and the average wear value of the tooth back ΔRb, are input into the SVM model. Then the SVM model can tell the operator whether the saw tooth parameters of the circular saw blade match the on-site working conditions, that is, whether the saw tooth parameters are appropriate.

[0077] The aforementioned sawtooth parameter optimization refers to adjusting the sawtooth profile parameters of the tooth to be re-grinded if the SVM model determines that the selected sawtooth parameters are inappropriate. Specifically, this adjustment involves adjusting the rake angle, clearance angle, and tooth back parameters. The specific adjustment principle is as follows: First, adjust the rake angle, clearance angle, and tooth back parameters of the standard tooth profile. Second, subtract the adjusted standard tooth profile parameters from the original tooth profile parameters. Finally, determine if the adjusted average wear values ​​of the rake angle, clearance angle, and tooth back are equal. If they are equal, the adjustment ends; otherwise, repeat the above steps.

[0078] The aforementioned saw tooth grinding accuracy verification refers to calculating the grinding error of the saw tooth parameters by subtracting the standard tooth profile from the ground tooth profile. Specifically, it involves subtracting the actual rake angle, clearance angle, and tooth back angle values ​​of the ground saw tooth from the standard tooth profile values ​​to verify the accuracy of the automatic tooth grinding machine's grinding of the saw teeth. A threshold can be set; if the error of any parameter of any saw tooth on any circular saw blade exceeds 2%, the operator is required to intervene and further review the saw blade.

[0079] The aforementioned saw blade flatness evaluation refers to calculating the flatness of different teeth on the same saw blade after grinding, based on the tooth width and tooth depth parameters. Specifically, the evaluation criteria are as follows: when the circular saw blade has good flatness, each tooth should be the same size within the camera's field of view; when the circular saw blade is warped, the upward-curving teeth (closer to the camera) will appear larger within the camera's field of view, while the downward-curving teeth (away from the camera) will appear smaller. In practical operation, this invention adopts... Figure 8 The two parameters, tooth width T and tooth depth H, are used to determine whether the flatness of the circular saw blade is up to standard. The deviation values ​​of tooth width T and tooth depth H of all the saw teeth on the blade are used to judge whether the flatness of the circular saw blade is up to standard. The judgment threshold is determined by three parameters: the diameter of the circular saw blade, the number of saw teeth, and the allowable range of flatness error. For example, for a circular saw blade with a diameter of 300 mm, 100 saw teeth, and a flatness error allowable range of ±10 mil, its tooth width T value is 9.42 mm and its tooth depth H value is 4.24 mm. The deviation value of tooth width T of all its saw teeth should be within ±1% (that is, within the field of view of the camera, the deviation of the maximum and minimum saw tooth width captured by the camera from the average tooth width should be controlled within 1%). At the same time, the deviation value of tooth depth H of all saw teeth should be within ±2.4%. Otherwise, the operator should be required to intervene and further review the saw blade.

[0080] The saw tooth camera device proposed in this invention for use in automatic gear grinding machines features high reliability in fluid drainage, no damage to the saw blade, simple and stable structure, and good applicability. The method of using the saw tooth camera device proposed in this invention enables fully automatic online detection of saw tooth wear, optimization of saw tooth shape, and detection of saw blade flatness. Therefore, adding the device and method of this invention to an automatic gear grinding machine can greatly enhance the grinding accuracy and intelligence of the automatic gear grinding machine.

[0081] The above specific embodiments are used to explain and illustrate the present invention, but not to limit the present invention. Any modifications and changes made to the present invention within the spirit and scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A saw tooth camera device applied to an automatic circular saw blade grinding machine, characterized in that: The sawtooth camera device includes a high-pressure air chamber, a circumferential air curtain block, a supplementary light lamp hole, a camera fixing through hole, a vacuum extraction tube, a high-pressure blowing tube, a connecting rod, a return spring, and a positioning coil. The high-pressure air chamber is connected to the circumferential air curtain block. The camera fixing through hole is located at the center of the high-pressure air chamber and the circumferential air curtain block, and the supplementary light lamp hole is located at the bottom of the circumferential air curtain block. The vacuum extraction tube and the high-pressure blowing tube both penetrate the high-pressure air chamber and the circumferential air curtain block, and are respectively installed on both sides of the camera fixing through hole. The connecting rod is located above the high-pressure air chamber, and its two ends are fixed to the fixed vacuum extraction tube and the high-pressure blowing tube respectively by interference fit. The return spring is installed in the upper cavity wall of the high-pressure air chamber and is coaxially installed with the vacuum extraction tube. The positioning coil is installed in the upper cavity wall of the high-pressure air chamber and is coaxially installed with the high-pressure blowing tube. The high-pressure air chamber has an annular cavity surrounding the camera fixing hole at its center, and an air inlet is opened on each of the two side walls of the high-pressure air chamber. The upper half of the circumferential air curtain block has several cylindrical through holes distributed around the periphery of the camera fixing hole, and the cylindrical through holes are connected to the annular cavity of the high-pressure air chamber. The lower half of the circumferential air curtain block has an annular through hole with an inclined cross-section.

2. The saw tooth camera device applied to an automatic circular saw blade grinding machine according to claim 1, characterized in that: The high-pressure air chamber and the circumferential air curtain block are two cylinders with the same bottom diameter, and they are coaxially installed.

3. The saw tooth camera device applied to an automatic circular saw blade grinding machine according to claim 1, characterized in that: The supplementary light lamp holes are located on the bottom surface of the circumferential air curtain block and are evenly distributed around the outer circumference of the camera fixing through hole.

4. The saw tooth camera device applied to an automatic circular saw blade grinding machine according to claim 1, characterized in that: The bottom of the high-pressure blowing tube is 0.5 mm to 1 mm higher than the bottom of the vacuum extraction tube.

5. The saw tooth camera device applied to an automatic circular saw blade grinding machine according to claim 1, characterized in that: The high-pressure air blowing pipe corresponding to the positioning coil is covered with a metal outer tube, and the height of the metal outer tube is equal to that of the positioning coil.

6. The saw tooth camera device applied to an automatic circular saw blade grinding machine according to claim 1, characterized in that: The vacuum extraction tube has an extraction port at the bottom, which is stepped in shape and includes a first step line and a second step line. The width of the first step line and the second step line are equal, and the width of the first step line and the second step line is 1 mm to 3 mm.

7. A method of using the saw tooth camera device as described in claim 1, applied to an automatic circular saw blade grinding machine, characterized in that: Two sets of saw tooth camera devices are installed on both sides of the grinding wheel, and the following procedures are followed: saw tooth shooting and parameter calculation, saw tooth wear calculation, saw tooth parameter selection and evaluation, saw tooth parameter optimization, saw tooth grinding accuracy verification, and saw blade flatness evaluation. Sawtooth imaging and parameter calculation: This includes imaging the sawtooth before and after grinding and calculating the tooth shape parameters. The tooth shape parameter calculation process includes sawtooth edge recognition, sawtooth feature point extraction, and sawtooth parameter calculation. Sawtooth wear calculation: Subtract the standard tooth profile from the tooth profile before grinding to calculate the wear values ​​of the rake angle, clearance angle, and tooth back of the sawtooth before grinding; Saw tooth parameter selection evaluation: Using the support vector machine (SVM) model, the three parameters of average wear value of the front angle, average wear value of the back angle, and average wear value of the tooth back are used as input parameters of the SVM model. The appropriateness of saw tooth parameter selection and the inappropriateness of saw tooth parameter selection are used as output parameters of the SVM model to judge whether the saw tooth parameters are suitable for the saw blade working conditions. Sawtooth parameter optimization: Adjust the average wear value of the rake angle, the average wear value of the clearance angle, and the average wear value of the tooth back of the tooth to be refurbished; Sawtooth grinding accuracy verification: Subtract the standard tooth profile from the ground tooth profile to calculate the grinding error of the sawtooth tooth profile parameters; Saw blade flatness assessment: Based on the tooth width and tooth depth parameters after grinding, calculate the flatness of different teeth on the same saw blade after grinding, and then determine whether the ground saw blade is qualified.

8. The method of using the saw tooth camera device applied to an automatic circular saw blade grinding machine according to claim 7, characterized in that: The jagged edge recognition uses algorithms including the Canny operator to extract the lines of the jagged edge; jagged feature point extraction extracts five feature points that are easy to identify from the image; jagged parameter calculation uses the five feature points to calculate five parameters: back angle α, front angle γ, tooth back Rb, tooth width T, and tooth depth H. Saw tooth wear calculation involves subtracting the rake angle, clearance angle, and tooth back parameters of each saw tooth before regrinding from the standard tooth profile's three parameters. Then, the average of the resulting rake angle difference, clearance angle difference, and tooth back difference values ​​for all saw teeth is calculated. The resulting average rake angle wear value Δγ, average clearance angle wear value Δα, and average tooth back wear value ΔRb represent the amount of wear on the rake angle, clearance angle, and tooth back parameters of the circular saw blade during the previous cutting operation.

9. The method of using the saw tooth camera device applied to an automatic circular saw blade grinding machine according to claim 8, characterized in that... The SVM model is represented as follows: w T x+b=0 In the above formula, x represents the input parameters of the SVM model, w represents the normal vector of the classification hyperplane, T represents the transpose of the vector, and b represents the offset of the hyperplane relative to the origin; among them, the input parameters x are three parameters: the average wear value of the front angle Δγ, the average wear value of the rear angle Δα, and the average wear value of the tooth back ΔRb. Suppose the output parameter of the SVM model is y. The output parameter y can only take two values: y = +1 indicates that the sawtooth parameter selection is appropriate, and y = -1 indicates that the sawtooth parameter selection is inappropriate. Each circular saw blade yields a data point (x, y). Using multiple circular saw blades of the same type, a sample set is generated to train the SVM model. This sample set is represented as {(x1, y1), (x2, y2), ..., (x...}. i ,y i )},in: x i ∈R d y i ∈{+1,-1} is the category label, i = 1, 2, ..., n; After the SVM model is trained to maturity using the above sample set, it can be used to evaluate whether the selection of saw tooth parameters is appropriate. That is, the three parameters, namely the average wear value of the front angle Δγ, the average wear value of the rear angle Δα, and the average wear value of the tooth back ΔRb, are input into the SVM model, and then it is obtained whether the saw tooth parameters of the circular saw blade match the on-site working conditions, that is, whether the selection of saw tooth parameters is appropriate.

10. The method of using the saw tooth camera device applied to an automatic circular saw blade grinding machine according to claim 9, characterized in that: For sawtooth parameter optimization, if the SVM model determines that the sawtooth parameter selection is inappropriate, then the sawtooth tooth profile parameters to be reground are adjusted. The specific adjustment principle is as follows: First, adjust the three parameters of the standard tooth profile: rake angle, clearance angle, and tooth back. Second, subtract the adjusted standard tooth profile parameters from the tooth profile parameters before reground. Finally, determine whether the adjusted average wear values ​​of the rake angle, clearance angle, and tooth back are equal. If they are equal, the adjustment ends; otherwise, repeat the above steps. The saw tooth grinding accuracy is verified by subtracting the actual front angle, back angle, and tooth back angle values ​​of the ground saw teeth from the standard tooth profile values ​​to check whether the automatic tooth grinding machine has ground the saw teeth accurately. A threshold is set, and if the error of any parameter of any tooth on any circular saw blade exceeds 2%, the operator is required to intervene and further review the saw blade. The flatness of a saw blade is evaluated using two parameters: tooth width (T) and tooth depth (H). The flatness of the saw blade is judged based on the deviation values ​​of the tooth width (T) and tooth depth (H) of all teeth on the blade. The judgment threshold is determined by three parameters: the diameter of the saw blade, the number of teeth, and the allowable range of flatness error. The deviation value of the tooth width (T) of all teeth should be within ±1%, and the deviation value of the tooth depth (H) of all teeth should be within ±2.4%. Otherwise, the operator should be required to intervene and further review the saw blade.

Images