Visual-based end-mill flank wear detection method and image acquisition device thereof

Abstract

The application discloses a kind of visual-based end mill flank wear detection method and its image acquisition device, method includes fixing tool on the tool holder of image acquisition device, and the wear area of tool is observed by microscope of image acquisition device and S is photographed End mill flank wear picture under multiple shooting angles;Contrast the topography of tool flank face in the picture shot before and after wear, measure the wear of tool flank face in the picture shot, obtain S group of tool wear data under different shooting angles;Establish the mathematical model of tool wear and shooting angle, solve the only fitting function describing the change of tool wear with shooting angle;Draw the image of fitting function y (x) in the shooting range K, solve image peak value.The maximum wear of tool can be obtained simply, intuitively and accurately by the application.Under the premise of image acquisition automation, automatic measurement of tool wear can be realized, and has high engineering application value.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of milling cutter wear detection, in particular to a visual-based end face wear detection method of an end mill and an image acquisition device thereof. BACKGROUND

[0002] Milling cutters are widely used in the machining of parts in the fields of aerospace, mold production, medical devices, etc. Tool wear is the main factor affecting the surface quality and machining efficiency of parts in the milling process. Once the tool is worn, more cutting force and cutting heat will be generated in the subsequent machining, and in severe cases, the vibration of the workpiece will be intensified, reducing the machining accuracy of the workpiece and the production efficiency of the equipment. Judging tool wear degree according to machining experience may result in premature or late tool replacement. Premature tool replacement cannot maximize tool life, and late tool replacement may cause serious accidents such as part scrap and machine precision decline. Therefore, tool wear detection during workpiece machining is crucial.

[0003] Current tool wear detection methods are divided into direct measurement and indirect measurement. Indirect measurement mainly detects based on cutting force, cutting temperature, electrical signal or acoustic emission signal, etc. These methods are more dependent on the installation accuracy of sensors, and the vibration in the machining process will interfere with the detection performance of the sensors, and even cause measurement distortion. The most widely used and mature method is still the direct measurement method, which includes resistance measurement, tool-workpiece spacing measurement, optical measurement and computer image measurement, etc. Compared with image measurement, the remaining direct measurement methods have the disadvantages of poor adaptability and complex process. Computer image measurement is a fast, non-contact and high-precision detection method based on vision, which can accurately detect different forms of wear on each cutting edge and has good application prospects.

[0004] Tool flank wear value usually refers to the wear value along the direction perpendicular to the cutting edge. The camera shooting angle needs to be perpendicular to the wear area. Different shooting angles will cause differences in the measurement value of the wear area, that is, the shooting angle directly affects the detection result of the side edge wear parameter. This leads to the following problems in actual production: on the one hand, it is difficult to obtain accurate wear values, and on the other hand, the adjustment of the shooting angle reduces the wear measurement efficiency. SUMMARY

[0005] The present application aims to provide a visual-based end face wear detection method of an end mill and a matching image acquisition device, effectively reducing the risk of low surface quality, unqualified and scrap of workpieces caused by tool wear, ensuring that the tool is processed in a normal state, accurately measuring the wear of the irregular wear area of the side edge, and improving the detection efficiency and accuracy.

[0006] Technical solution: A visual-based end mill flank wear detection method, comprising the following steps:

[0007] Step 1: Fix the tool on the tool clamp of the image acquisition device, observe the tool wear area through the microscope of the image acquisition device, and take S pictures of the end mill flank wear at multiple shooting angles;

[0008] Step 2: Compare the topography of the tool flank before and after wear in the taken pictures, measure the wear of the tool flank in the taken pictures, and obtain S sets of tool wear data at different shooting angles;

[0009] Step 3: Use polynomial fitting method to fit the data, establish a mathematical model of tool wear and shooting angle, and solve the unique fitting function describing the change of tool wear with shooting angle;

[0010] Step 4: Draw the image of the fitting function y(x) in the shooting range K, solve the peak value, and the peak value is the numerical solution of the maximum value of the tool flank wear width.

[0011] Further, in step 1, the wear picture shooting includes the following steps:

[0012] S11: Observe the tool wear area through the microscope, adjust the microscope, and ensure that it can completely shoot the tool flank wear zone under the premise of taking a shooting angle as 0°, which is recorded as the reference angle. Rotate the tool, change the shooting angle, clockwise rotation is recorded as positive, counterclockwise rotation is recorded as negative, and determine the shooting range K, which is as follows:

[0013]

[0014] Where, γ i is the shooting interval angle between the ith picture and the i+1th picture;

[0015] S12: Control the tool clamp to rotate the tool, according to the number of photographs S and the rotation interval γ i Rotate and photograph the tool in turn to obtain S pictures of the tool flank wear at different shooting angles.

[0016] Further, in step 2, the measurement of the taken pictures is the measurement and recording of the maximum value of the tool flank wear width or length. The obtained multiple sets of tool wear data are recorded as:

[0017] (x1, y1), (x2, y2),..., (x i , y i ),..., (x s , y s );

[0018] wherein x i is the shooting angle of the i-th picture, y i is the maximum value of the width or length of the tool flank wear band of the i-th picture.

[0019] Further, in step 3, the solution of the unique fitting function includes the following steps:

[0020] S21: Perform polynomial fitting processing on the S group of wear data, and assume that the fitting polynomial function formula is as follows:

[0021]

[0022] wherein t is the polynomial degree, a j is the polynomial coefficient;

[0023] S22: Use the least square principle to substitute the original data into the error sum of squares formula, which is as follows:

[0024]

[0025] wherein ε is the error sum of squares, then take the partial derivative of the unknown parameter, and simplify and arrange to obtain:

[0026] X T XQ=X T Y;

[0027] wherein,

[0028]

[0029] wherein x1, x2, …, x S are all different, the matrix X is a symmetric matrix, solve y(x) and obtain a unique solution.

[0030] An image acquisition device matched with the above-mentioned visual-based detection method of the flank wear of a ball end mill, comprising a tool clamp, a shooting and acquisition assembly, and an image processing assembly, the tool clamp is arranged on one side of the shooting and acquisition assembly, the tool is clamped on the tool clamp, the shooting and acquisition assembly comprises a microscope, which is arranged above the tool at intervals, and the image processing assembly is signal-connected with the microscope.

[0031] Further, the tool clamp comprises a scale disc, a rotating handle, and a three-jaw chuck, the three-jaw chuck is connected with the rotating handle, the circumferential direction of the tail end jaw part of the three-jaw chuck is provided with the scale disc, the three-jaw chuck is used for clamping the tool to be measured, and the scale disc comprises a scale pointer, which is located at the vertex of the scale disc.

[0032] Optimally, the photographing and collecting assembly further comprises a support and a reflector plate, the microscope is installed on the upper part of the support, the reflector plate is installed on the lower part of the support, the cutter is located between the microscope and the reflector plate, and the head of the microscope is provided with an LED lamp for providing a light source.

[0033] Optimally, the image processing assembly comprises a computer and a transmission line, and the computer is connected with the microscope through the transmission line.

[0034] Advantages: compared with the prior art, the method has the advantages that: a visual-based precise measurement method for the flank wear of a helical end mill and a matching image collecting device are provided, the method is suitable for irregular wear of the flank of the helical end mill in a machining process, compared with the existing direct measurement method, the method has the advantages of being simple and convenient, wide application range, and high detection precision and engineering application value, and the method does not need to obtain a large number of pictures for training. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 is a method flowchart of the application;

[0036] Figure 2 is a front view structural schematic diagram of the image collecting device of the application;

[0037] Figure 3 is a structural schematic diagram of a cutter clamp in the image collecting device of the application;

[0038] Figure 4 is a partial structural sectional view of a worm under the clamp of the cutter clamp of the application;

[0039] Figure 5 is a structural schematic diagram of the helical end mill of the embodiment of the application. DETAILED DESCRIPTION

[0040] The application will be further illustrated below in combination with the drawings and specific embodiments, and it should be understood that the embodiments are only used for illustrating the application and not for limiting the scope of the application.

[0041] A visual-based end mill flank wear detection method, as shown in Figure 1 , comprises the following steps:

[0042] Step 1: use the image collecting device to photograph the whole end mill flank wear images under 5 different photographing angles, and the specific process is as follows:

[0043] S11: fix the cutter in the image collecting device, and the cutter sample schematic diagram is shown in Figure 5, the tool wear area is observed by microscope 210, and on the premise of taking a complete picture of the tool flank wear zone, an arbitrary shooting angle is taken as 0°, which is recorded as the reference angle, the tool is rotated, the positive rotation is clockwise and the negative rotation is counterclockwise, and the shooting range K is determined, and the formula is as follows:

[0044]

[0045] wherein, γ i is the shooting interval angle of the ith photo and the i+1th photo.

[0046] S12: control the rotating handle 120 to make the tool rotate on the fixed shaft, record the angle of each rotation of the tool through the scale disc 110, ensure that the shooting range meets K≥70°, and the camera 1 takes a picture after the tool rotates by a specified angle γ, so as to realize the collection of tool wear images at different shooting angles, as shown in Figure 3 The microscope 210 is directly connected with the computer 310, and the collected images are directly presented on the computer 310, so as to facilitate subsequent image processing.

[0047] Step 2: compare the front and rear morphologies of the tool flank surface in the taken photos, measure the wear of the tool flank surface in the taken photos, and obtain 5 groups of tool flank surface wear data at different shooting angles. The measurement and recording of the taken photos are specifically the measurement and recording of the maximum width of the tool flank surface wear zone. The obtained 5 groups of tool wear data are recorded as:

[0048] (x1,y1),(x2,y2),(x3,y3),(x4,y4),(x5,y5)

[0049] wherein, x i , y i (i=1,2,3,4,5) are the shooting angle and the maximum width value of the tool flank surface wear zone of the ith photo, respectively.

[0050] Step 3: adopt a polynomial fitting method to process the data, establish a mathematical model of the tool wear and the shooting angle, and solve the unique fitting function describing the change of the tool wear with the shooting angle, and the specific process is as follows:

[0051] S21: perform quartic polynomial fitting processing on the 5 groups of wear data, and the fitting polynomial function formula is as follows:

[0052] y(x)=a0+a1x+a2x 2 +a3x 3 +a4x 4 (2)

[0053] wherein, a0, a1, a2, a3, a4 represent the polynomial coefficients.

[0054] S22: using the least square principle, the original data is substituted into the error sum of squares formula, the following formula can be obtained:

[0055]

[0056] Wherein, ε is the error sum of squares, using the derivative knowledge, the partial derivative of unknown parameters is solved, and the following formula can be obtained:

[0057] X T XQ=X T Y (4)

[0058] Record:

[0059]

[0060] Wherein, X T is the transpose matrix of matrix X, x1, x2, …, x5 are different from each other, matrix X is a symmetric matrix, the above formula can be solved and the solution is unique, after solving the polynomial coefficients a0, a1, a2, a3, a4, substitute formula (2) to obtain the fitting function.

[0061] Step 4: draw the image of the fitting function y(x) in the shooting range K, solve the peak value of the image, and the peak value is the numerical solution of the maximum value of the tool flank wear, wherein the shooting range K is converted into the range of independent variable x as follows:

[0062] min(x1,x i )≤x≤max(x1,x i )(i=1,2,...,5)。

[0063] The method can simply, intuitively and accurately obtain the maximum wear of the tool. Under the premise of automatic image acquisition, automatic measurement of tool wear can be realized, which has high engineering application value.

[0064] The image acquisition device matched with the above method, as shown in Figures 2 to 5 , including a tool clamp 100, a shooting acquisition assembly 200, and an image processing assembly 300. The tool clamp 100 is arranged on one side of the shooting acquisition assembly 200, the tool 400 is clamped on the tool clamp 100, the shooting acquisition assembly 200 includes a microscope 210, which is arranged at intervals above the tool, the image processing assembly 300 is connected with the microscope signal, the shooting acquisition assembly 200 focuses and shoots, and the tool flank wear images collected at different shooting angles are transmitted to the image processing assembly 300 for measurement.

[0065] The tool clamp 100 comprises a scale disc 110, a rotating handle 120 and a three-jaw chuck 130, the three-jaw chuck 130 is connected with the rotating handle 120, the three-jaw chuck 130 is used for clamping the tool 400 to be measured, and stability is ensured, the rotating handle 120 drives the three-jaw chuck 130 to rotate through a worm and gear structure, the rotating handle 120 drives the worm to rotate, the center of the worm is located on the same axis as the center of the tool 400 to be measured, so that the tool 400 is rotated in a fixed axis, and the tail end of the three-jaw chuck 130 is provided with the scale disc 110 in the circumferential direction, the scale disc 110 comprises a scale pointer 111, the scale pointer 111 is located at the top point of the scale disc 110, and the scale pointer is located at the highest point of the scale disc, so that the scale value at this time can be read.

[0066] The shooting and collecting assembly 200 further comprises a support 220 and a reflecting plate 230, the microscope 210 is installed on the upper part of the support 220, the reflecting plate 230 is installed on the lower part of the support 220, the tool 400 is located between the microscope 210 and the reflecting plate 230, and the head of the microscope 210 is provided with an LED lamp 211 for providing a light source. The microscope 210 is adjusted up and down through the knob 221 thereon, so that the tool wear area to be measured is accurately focused, the LED lamp 211 is used for providing a light source, and the reflecting plate 230 realizes a reflecting function.

[0067] The image processing assembly 300 comprises a computer 310 and a transmission line 320, the computer 310 is signal-connected with the microscope 210 through the transmission line 320. The computer 310 is built-in with existing measurement software, the online measurement of the tool wear amount can be realized, and the tool image is directly presented in the computer. The tool is rotated in a fixed axis through the rotating handle, the wear photos of the tool blade surfaces at different angles are shot, are transmitted to the computer for wear measurement, and the wear conditions of the tool blades are obtained.

Claims

1. A visual-based flank wear detection method for an end mill, the method comprising: The method comprises the following steps: Step 1: fix the tool on the tool clamp of the image acquisition device, observe the tool wear area through the microscope of the image acquisition device, and take S pictures of the tool flank wear surface at multiple shooting angles; Step 2: compare the tool flank wear surface in the taken pictures before and after wear, measure the tool flank wear surface in the taken pictures, and obtain S groups of tool wear data at different shooting angles; Step 3: adopt a polynomial fitting method to perform fitting processing on the data, establish a mathematical model of tool wear and shooting angle, and solve a unique fitting function describing the change of tool wear with shooting angle; Step 4: draw a graph of the fitting function y(x) in the shooting range K, solve the peak value of the graph, and the peak value is the numerical solution of the maximum value of the tool flank wear width; In step 3, the solving of the unique fitting function comprises the following steps: S21: perform polynomial fitting processing on the S groups of wear data, and set the fitting polynomial function formula as follows: where t is the polynomial degree, a j is the polynomial coefficient; S22: use the least square principle, substitute the original data into the error square sum formula, and the formula is as follows: Wherein, ε is the error square sum, then the partial derivative of the unknown parameter is solved, and the simplification and arrangement can obtain: X T XQ= X T Y; Wherein, where x1, x2, …, x S are all different, and the matrix X is symmetric, solve y(x) and get a unique solution.

2. The visual-based end mill flank wear detection method of claim 1, wherein, In step 1, the wear picture shooting comprises the following steps: S11: observe the tool wear area through the microscope, adjust the microscope, under the premise of ensuring that the tool flank wear surface can be completely taken, take an arbitrary shooting angle as 0°, mark as the reference angle, rotate the tool, change the shooting angle, clockwise rotation is positive, and counterclockwise rotation is negative, determine the shooting range K, and the formula is as follows: wherein γ i is the shooting interval angle of the ith photo and the ith+1 photo; S12: control the tool holder to drive the tool to rotate, according to the number of times of photographing S and the rotation interval γ i The tool is rotated and photographed in sequence to obtain S tool relief surface wear photos at different shooting angles.

3. The vision-based end mill flank wear detection method of claim 1, wherein: In step 2, the measurement of the taken pictures is the measurement and recording of the maximum value of the tool flank wear width or length, and the obtained multiple groups of tool wear data are marked as: (x1, y1), (x2, y2),..., (x i , i y s , s ) wherein x i is the shooting angle of the i-th picture, y i is the maximum value of the width or length of the tool flank wear band of the i-th picture.

4. An image acquisition device adapted to the visual-based flank wear detection method of any one of claims 1 to 3, characterized in that: The tool clamp (100), the shooting and collecting assembly (200), and the image processing assembly (300) are arranged on one side of the shooting and collecting assembly (200), the tool (400) is clamped on the tool clamp (100), the shooting and collecting assembly (200) comprises a microscope (210) which is arranged above the tool in a spaced manner, and the image processing assembly (300) is connected with the microscope.

5. The image acquisition device of claim 4, wherein: The tool clamp (100) comprises a scale disc (110), a rotating handle (120), and a three-jaw chuck (130), the three-jaw chuck (130) is connected with the rotating handle (120), the circumferential direction of the tail end clamping jaw part of the three-jaw chuck (130) is provided with the scale disc (110), the three-jaw chuck (130) is used for clamping the tool (400) to be measured, and the scale disc (110) comprises a scale pointer (111) which is located at the top point of the scale disc (110).

6. The image acquisition device of claim 4, wherein: The shooting and collecting assembly (200) further comprises a support (220) and a reflector (230), the microscope (210) is installed on the upper part of the support (220), the reflector (230) is installed on the lower part of the support (220), the tool (400) is located between the microscope (210) and the reflector (230), and the head part of the microscope (210) is provided with an LED lamp (211) for providing a light source.

7. The image acquisition device of claim 4, wherein: The image processing assembly (300) comprises a computer (310) and a transmission line (320), the computer (310) being signal connected with the microscope (210) through the transmission line (320).

Images