Pocket milling tool path optimization method based on digital filtering

Abstract

A pocket milling tool path optimization method based on digital filtering, comprising the following steps: reading tool path information; discretizing a tool path at a constant interval along a pocket machining tool path to obtain discrete tool location points; calculating the feeding speed at each discrete tool location point in each motion axis direction under a theoretical scenario to obtain a speed change curve of the tool path; using a Gaussian filter to smooth the tool path speed curve to obtain a feeding speed curve after smoothing; on the basis of the speed at each discrete point on the speed curve after smoothing, re-calculating new tool location points; and combining discrete points obtained by re-calculation, so as to form an optimized pocket milling tool path. A tool path optimization method based on digital filtering technology is used to smooth the speed change curve of each axis during movement of a machine tool, thereby reducing the occurrence of the speed reduction of the machine tool, improving the pocket milling efficiency, making the pocket machining tool path smoother, and making the method more suitable for the acceleration and deceleration characteristics of the machine tool.

Description

A tool path optimization method for slot milling based on digital filtering Technical Field

[0001] The present invention belongs to the field of intelligent manufacturing technology, and specifically relates to a slot milling tool path optimization method based on digital filtering. Background Art

[0002] Slot milling is a common method in structural component machining, particularly in aircraft structural component manufacturing, where over 80% of the workload is consumed by slot milling. High surface quality and minimal machining time are the goals of slot toolpath generation or optimization.

[0003] For slot milling, optimizing tool paths at corners is crucial for improving machining efficiency and quality. Currently, many commercial CAM software programs struggle to address corner issues when generating slot tool paths. To address this issue, numerous researchers have proposed a series of slot tool path optimization methods to optimize corner machining.

[0004] For example, the patent (patent number CN201310001740) discloses an integrated finishing method for the inner corner of a groove feature. The method performs cyclic processing at the corner according to the principle of constant contact angle and maximum contact angle, ensuring the stability of the cutting force during the corner processing; Hyun-Chul (2007) published a paper "Tool path modification for optimized pocket milling" in the academic journal "International Journal of Production Research" 2007, 45 (24), p5715-5729, disclosing a groove feature processing method, which ensures the stability of the cutting force through a stable material removal rate and avoids vibration during the processing; Banerjee et al. proposed a method for 2 in the academic journal "Computers & Industrial Engineering". 1 / The tool path generation algorithm with constant feed speed and constant cutting force for 2D groove cavity adopts the parallel offset idea, which reverses the tool path from the last layer to the previous layer to realize the control of actual cutting width, but the tool path is not smooth and continuous enough.

[0005] The above methods can optimize the machining tool path to a certain extent, but none of them fully quantitatively analyze the dynamic response capability of the machine tool, and then optimize the machining tool path based on the dynamic characteristics of the machine tool.

[0006] Summary of the Invention

[0007] To solve the above problems, the present invention provides a slot milling tool path optimization method based on digital filtering, which can make the slot machining tool path smoother, more adaptable to the acceleration and deceleration characteristics of the machine tool, and improve machining efficiency.

[0008] In order to achieve the above-mentioned invention objectives, the technical solutions provided by this application are as follows:

[0009] A slot milling tool path optimization method based on digital filtering includes the following steps:

[0010] Step 1: Read tool path information from the slot machining tool location file or NC program;

[0011] Step 2: Discretize the tool path along the groove cavity machining tool path at a constant spacing ΔS to obtain a series of discrete tool position points;

[0012] Step 3: According to the given feed rate of the tool trajectory, the feed rate in each motion axis direction at each discrete tool position is calculated under theoretical conditions, and the speed change curve of the tool path is obtained based on the feed rate at each discrete tool position;

[0013] Step 4: Use a Gaussian filter to smooth the tool path speed curve to obtain a smoothed feed speed curve;

[0014] Step 5: recalculate the new tool position point based on the speed of each discrete point on the smoothed speed curve;

[0015] Step 6: Combine the recalculated discrete points to form an optimized slot milling tool path.

[0016] Furthermore, the specific steps of the method for calculating the feed speed in each motion axis direction at each discrete tool position point in step 3 are as follows:

[0017] 3.1) For the slot milling tool path, it is a three-axis machining process, and the X-axis and Y-axis motion axes are mainly involved in the machining process. Therefore, when calculating the feed rate at each discrete tool position, only the X-axis and Y-axis motion directions are considered;

[0018] 3.2) When calculating the feed rate, assume that the feed rate is set to a constant speed and the speed is recorded as V F ;

[0019] 3.3) Based on the discrete spacing ΔS, calculate the time interval Δt between adjacent discrete points = ΔS / V F ;

[0020] 3.4) The theoretical feed rate v in the X-axis and Y-axis directions at the i-th discrete point x (i) v y (i) can be expressed as follows, where X i 、Y irepresents the X-axis and Y-axis coordinates of the i-th discrete point, and n is the number of discrete points;

[0021] 3.5) Based on the calculated feed rates in the X-axis and Y-axis directions at each discrete point, draw a feed rate curve for the slot machining tool path in the X-axis and Y-axis directions;

[0022] Furthermore, the method steps for smoothing the tool path velocity curve using a Gaussian filter in step 4 are as follows:

[0023] 4.1) Select a suitable Gaussian filter and determine the key parameters of the filter;

[0024] 4.2) Calculate the time required for the machine tool to accelerate from a standstill to a given feed speed;

[0025] 4.3) Determine the filter cutoff wavelength;

[0026] 4.4) Calculate the feed rate curve using the convolution formula;

[0027] Furthermore, in step 5, the new tool position point is recalculated based on the speed of each discrete point on the smoothed speed curve. The specific method is as follows:

[0028] Each discrete point is calculated based on the new speed to obtain the new X-axis coordinate Q of the tool position point x (i) and Y-axis coordinate Q y (i) can be calculated by the following formula:

[0029] where v′ x (i), v′ y (i) is the feed rate of the i-th discrete point in the X-axis and Y-axis directions after filtering.

[0030] Furthermore, the Gaussian filter is represented by the following formula:

[0031] Among them, u represents the relative position from the center of the weighted function curve, λ c is the cut-off wavelength of the filter, and α can be calculated as follows,

[0032] Furthermore, in step 4.2), the acceleration of the machine tool from rest to a given feed speed is calculated as follows:

[0033] For CNC machine tools, acceleration and deceleration are performed according to the S-shaped acceleration and deceleration control mode during the processing. First, the acceleration of the machine tool from static to a given feed speed V is solved. F Time required T min , where A max is the maximum acceleration of a certain motion axis of the machine tool, Jmax is the maximum jerk of a certain motion axis of the machine tool; when When the acceleration has not reached the maximum value, deceleration is required. In this case, the corresponding acceleration time can be calculated by the following formula:

[0034] Furthermore, the filter cutoff wavelength in 4.3) can be calculated by the following formula:

[0035] Since the time required for the speed to drop from a given value to 0 is also T min , so 2T min As the cutoff wavelength λ in the Gaussian filter c , that is, filter out the min The speed transition changes dramatically during the acceleration and deceleration reaction time, thereby smoothing the speed curve and the cutoff wavelength λ c It can be calculated by the following formula:

[0036] Furthermore, the calculation method of the feed speed curve described in 4.4) is as follows:

[0037] The filtered feed rate is calculated using the convolution formula. The filtered feed rate curve p(t) can be calculated by the following formula, where c(ε) is the theoretical feed rate curve:

[0038] Where ε is the time parameter of the theoretical feed rate curve;

[0039] The theoretical feed rate curve consists of a series of discrete points, i.e., t = kΔt, (k = 0, 1, ..., n-1), where n represents the number of discrete points. The feed rate at each discrete point after filtering can be calculated as follows:

[0040] Where i represents the i-th discrete point after filtering, p(i) represents the feed rate at the i-th discrete point after filtering, c i-k Represents the theoretical feed rate at the ikth point after filtering, which should satisfy λ c >50Δt. In addition, the filter parameter m is the interaction range parameter of each discrete point during convolution calculation, and m is calculated by the empirical formula:

[0041] The beneficial effects of the present invention are:

[0042] This application fully quantitatively analyzes the dynamic response capability of the machine tool, optimizes the machining tool path based on the dynamic characteristics of the machine tool, and proposes a method based on digital filtering technology to smooth the speed change curve of each axis of the machine tool during movement, thereby directly reducing the machine tool deceleration. By reducing the deceleration, the efficiency of the slot milling process is improved, and the slot machining tool path can be made smoother and more adapted to the acceleration and deceleration characteristics of the machine tool. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of the tool path optimization algorithm for slot milling based on digital filtering.

[0044] Figure 2 shows the typical groove cavity features of aircraft structural parts.

[0045] Figure 3 shows the machining tool path before optimization.

[0046] Figure 4 shows the discrete tool position points of the tool path before optimization.

[0047] FIG5 is a feed speed curve of the machining tool path in the X-axis shown in FIG3 and FIG4 before optimization.

[0048] FIG6 is a feed speed curve of the tool path in the Y axis before optimization shown in FIG3 and FIG4.

[0049] FIG7 is a feed speed curve of the tool path in the X axis after optimization shown in FIG3 and FIG4.

[0050] FIG8 is a feed speed curve of the tool path in the Y axis after optimization shown in FIG3 and FIG4.

[0051] Figure 9 shows the discrete tool position points of the recalculated machining tool path after optimization.

[0052] Figure 10 shows the recalculated machining tool path after optimization.

[0053] Figure 11 shows an aluminum alloy aircraft structural component, where P1-P4 are four groove cavities.

[0054] Figures 12 to 15 are the machining tool paths of the four slot cavities P1 to P4 before the slot cavity optimization shown in Figure 11.

[0055] Figures 16 to 19 are the machining tool paths of the four slot cavities P1 to P4 after the slot cavity is optimized as shown in Figure 11.

[0056] Table 1 is a comparison table of the processing time before and after the optimization of the groove cavity of the structural part shown in Figure 11. DETAILED DESCRIPTION

[0057] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are intended to explain the present invention rather than to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0058] The specific implementation method of the present invention is described below with reference to the accompanying drawings and examples, but the present invention is not limited to this embodiment.

[0059] Example 1

[0060] As shown in Figure 1, a slot milling tool path optimization method based on digital filtering includes the following steps:

[0061] Step 1: Read tool path information from the slot machining tool location file or NC program;

[0062] Step 2: Discretize the tool path along the groove cavity machining tool path at a constant spacing ΔS to obtain a series of discrete tool position points;

[0063] Step 3: According to the given feed rate of the tool trajectory, the feed rate in each motion axis direction at each discrete tool position is calculated under theoretical conditions, and the speed change curve of the tool path is obtained based on the feed rate at each discrete tool position;

[0064] Step 4: Use a Gaussian filter to smooth the tool path speed curve to obtain a smoothed feed speed curve;

[0065] Step 5: recalculate the new tool position point based on the speed of each discrete point on the smoothed speed curve;

[0066] Step 6: Combine the recalculated discrete points to form an optimized slot milling tool path.

[0067] The specific steps of the method for calculating the feed speed in each motion axis direction at each discrete tool position point in step 3 are as follows:

[0068] 3.1) For the slot milling tool path, it is a three-axis machining process, and the X-axis and Y-axis motion axes are mainly involved in the machining process. Therefore, when calculating the feed rate at each discrete tool position, only the X-axis and Y-axis motion directions are considered;

[0069] 3.2) When calculating the feed rate, assume that the feed rate is set to a constant speed and the speed is recorded as V F ;

[0070] 3.3) Based on the discrete spacing ΔS, calculate the time interval Δt between adjacent discrete points = ΔS / V F ;

[0071] 3.4) The theoretical feed rate v in the X-axis and Y-axis directions at the i-th discrete point x (i) v y (i) can be expressed as follows, where X i 、Y i represents the X-axis and Y-axis coordinates of the i-th discrete point, and n is the number of discrete points;

[0072] 3.5) Based on the calculated feed rates in the X-axis and Y-axis directions at each discrete point, draw a feed rate curve for the slot machining tool path in the X-axis and Y-axis directions;

[0073] The method of using Gaussian filter to smooth the tool path velocity curve in step 4 is as follows:

[0074] 4.1) Select a suitable Gaussian filter and determine the key parameters of the filter;

[0075] 4.2) Calculate the time required for the machine tool to accelerate from a standstill to a given feed speed;

[0076] 4.3) Determine the filter cutoff wavelength;

[0077] 4.4) Calculate the feed rate curve using the convolution formula;

[0078] Furthermore, in step 5, the new tool position point is recalculated based on the speed of each discrete point on the smoothed speed curve. The specific method is as follows:

[0079] Each discrete point is calculated based on the new speed to obtain the new X-axis coordinate Q of the tool position point x (i) and Y-axis coordinate Q y (i) can be calculated by the following formula:

[0080] where v′ x (i), v′ y (i) is the feed rate of the i-th discrete point in the X-axis and Y-axis directions after filtering.

[0081] The Gaussian filter is represented by the following formula:

[0082] Among them, u represents the relative position from the center of the weighted function curve, λ c is the cut-off wavelength of the filter, and α can be calculated as follows,

[0083] In step 4.2), the acceleration of the machine tool from rest to a given feed speed is calculated as follows:

[0084] For CNC machine tools, acceleration and deceleration are performed according to the S-shaped acceleration and deceleration control mode during the processing. First, the acceleration of the machine tool from static to a given feed speed V is solved. F Time required T min , where A max is the maximum acceleration of a certain motion axis of the machine tool, J max is the maximum jerk of a certain motion axis of the machine tool;

[0085] when When the acceleration has not reached the maximum value, deceleration is required. In this case, the corresponding acceleration time can be calculated by the following formula:

[0086] The filter cutoff wavelength in 4.3) can be calculated by the following formula:

[0087] Since the time required for the speed to drop from a given value to 0 is also T min , so 2T min As the cutoff wavelength λ in the Gaussian filter c , that is, filter out the min The speed transition changes dramatically during the acceleration and deceleration reaction time, thereby smoothing the speed curve and the cutoff wavelength λ c It can be calculated by the following formula:

[0088] The calculation method of the feed speed curve described in 4.4) is as follows:

[0089] The filtered feed rate is calculated using the convolution formula. The filtered feed rate curve p(t) can be calculated by the following formula, where c(ε) is the theoretical feed rate curve:

[0090] Where ε is the time parameter of the theoretical feed rate curve;

[0091] The theoretical feed rate curve consists of a series of discrete points, i.e., t = kΔt, (k = 0, 1, ..., n-1), where n represents the number of discrete points. The feed rate at each discrete point after filtering can be calculated as follows:

[0092] Where i represents the i-th discrete point after filtering, p(i) represents the feed rate at the i-th discrete point after filtering, c i-k Represents the theoretical feed rate at the ikth point after filtering. When Δt is small enough, a smooth speed curve can be obtained. To ensure the filtering accuracy of the speed curve, λ should be satisfied. c>50Δt. In addition, the filter parameter m is the interaction range parameter of each discrete point during convolution calculation. The larger the m, the higher the accuracy, but the corresponding calculation efficiency will also decrease. In general, it can be calculated by the empirical formula:

[0093] Example 2

[0094] The typical aircraft structural part groove feature shown in Figure 2 is milled using a Φ20mm tool with a cutting width of 10mm.

[0095] Step 1: Read the tool path information from the slot machining tool location file or NC program, as shown in Figure 3;

[0096] Step 2: Discretize the tool path along the groove machining tool path at a constant spacing ΔS = 0.2 mm to obtain a series of discrete tool position points, as shown in Figure 4; Figure 4 shows the tool path information read from the groove machining tool position file or NC program, and Figure 4 shows the discrete tool position points of the tool path.

[0097] Step 3: Tool path feed rate F = 3000mm / min, that is, V F =50mm / s, calculate the feed speed in each motion axis direction at each discrete tool position under theoretical conditions;

[0098] Since slot milling is a three-axis process, only the X-axis and Y-axis motion directions need to be considered when calculating the feed rate of each discrete point; the time interval between adjacent discrete points Δt=ΔS / V F , then the theoretical feed speed v in the X-axis and Y-axis directions at the i-th discrete point x (i) v y (i) can be calculated by the following formula:

[0099] According to the above formula, the feed rate at each discrete point can be calculated and the tool path feed rate curve can be drawn, as shown in Figures 5 and 6. Figures 5 and 6 are the theoretical feed rate at each discrete point calculated based on the proposed formula, and the tool path theoretical feed rate curve is drawn in combination with the running time. The running time in the figure refers to the time it takes for the program to run to a discrete point.

[0100] Step 4: Use Gaussian filter to smooth the tool path speed curve to obtain the smoothed feed speed curve; the Gaussian filter formula used can be expressed as follows:

[0101] Among them, u represents the relative position from the center of the weighted function curve, λ c And is one of the important parameters of the filter, namely the cut-off wavelength, α can be calculated by the following formula:

[0102] For CNC machine tools, acceleration and deceleration are performed according to the S-shaped acceleration and deceleration control mode during the processing. First, the acceleration of the machine tool from static to a given feed speed V is solved. F Time required T min , where A max is the maximum acceleration of a certain motion axis of the machine tool, J max is the maximum jerk of a certain motion axis of the machine tool;

[0103] when When the acceleration has not reached the maximum value, deceleration is required. In this case, the corresponding acceleration time can be calculated by the following formula:

[0104] Since the time required for the speed to drop from a given value to 0 is also T min , so 2T min As the cutoff wavelength λ in the Gaussian filter c , that is, filter out the min The speed transition changes dramatically during the acceleration and deceleration reaction time, thereby smoothing the speed curve and the cutoff wavelength λ c It can be calculated by the following formula:

[0105] The filtered feed rate is calculated using the convolution formula. The filtered feed rate curve p(t) can be calculated by the following formula, where c(ε) is the theoretical feed rate curve:

[0106] Where ε is the time parameter of the theoretical feed rate curve;

[0107] The theoretical feed rate curve consists of a series of discrete points, i.e., t = kΔt, (k = 0, 1, ..., n-1), where n represents the number of discrete points. The feed rate at each discrete point after filtering can be calculated as follows:

[0108] Where i represents the i-th discrete point after filtering, p(i) represents the feed rate at the i-th discrete point after filtering, c i-k Represents the theoretical feed speed at the ikth point after filtering. When it is small enough, a smooth speed curve can be obtained. In order to ensure the filtering accuracy of the speed curve, λ should be satisfied. c >50Δt. In addition, the filter parameter m is the interaction range parameter of each discrete point during convolution calculation. The larger the m, the higher the accuracy, but the corresponding calculation efficiency will also decrease. In general, it can be calculated by the empirical formula:

[0109] For the processing of the groove feature shown in Figure 2, the J max =5m / s3 ,A max =0.5m / s 3 , the cut-off wavelength λ can be calculated c =0.2s, Δt=0.004s, and the filtered feed rate curves are shown in Figures 7 and 8. Figures 7 and 8 calculate the feed rate at each discrete point after filtering based on the proposed formula, and draw the filtered tool path feed rate curves in combination with the running time. The running time in the figure refers to the time it takes for the program to run to a discrete point.

[0110] Step 5: recalculate the new tool position point based on the speed of each discrete point on the smoothed speed curve;

[0111] Each discrete point is calculated based on the new speed to obtain the new X-axis coordinate Q of the tool position point x (i) and Y-axis coordinate Q y (i) can be calculated by the following formula:

[0112] where v′ x (i),v′ y (i) is the feed rate of the i-th discrete point in the X-axis and Y-axis directions after filtering;

[0113] The discrete points after filtering obtained by the above calculation are shown in Figure 9;

[0114] Step 6: Combine the recalculated discrete points to form the optimized slot milling tool path, as shown in Figure 10.

[0115] Example 3

[0116] On the basis of Example 1 and Example 2, for the groove cavity features P1-P4 of the aluminum alloy structural part shown in Figure 11, the traditional unoptimized groove cavity milling tool path is shown in Figures 12, 13, 14 and 15; the tool path optimized by the method of the present invention is shown in Figures 16, 17, 18 and 19. The optimized tool path is smooth, which avoids frequent acceleration and deceleration of the machine tool during the processing. The comparison of the groove cavity milling tool path processing time before and after optimization is shown in Table 1 below, which reduces the processing time by an average of 18.75%, and can effectively improve the processing efficiency.

[0117] Table 1:

[0118] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A slot milling tool path optimization method based on digital filtering, characterized in that: The following steps are involved: Step 1: Read tool path information from the slot machining tool location file or NC program; Step 2: discretize the tool path along the groove machining tool path at a constant spacing ΔS to obtain a series of discrete tool position points; Step 3: According to the given feed speed of the tool trajectory, the feed speed of each motion axis direction at each discrete tool position point under theoretical conditions is calculated, and the speed change curve of the tool trajectory is obtained according to the feed speed at each discrete tool position point; Step 4: Use a Gaussian filter to smooth the tool path speed curve to obtain a smoothed feed speed curve; Step 5: recalculate the new tool position point according to the speed of each discrete point on the smoothed speed curve; Step 6: Combine the recalculated discrete points to form an optimized slot milling tool path.

2. The method for optimizing tool paths for slot milling based on digital filtering according to claim 1, characterized in that: The specific steps of the method for calculating the feed speed in each motion axis direction at each discrete tool position point in step 3 are as follows: 3.1) Determine the X-axis and Y-axis movement directions of the slot milling tool path; 3.2) When calculating the feed speed, assume that the feed speed is set to move at a constant speed, and the speed is recorded as V F ; 3.3) Based on the discrete spacing ΔS, calculate the time interval Δt between adjacent discrete points = ΔS / V F ; 3.4) For the theoretical feed speed v in the X-axis and Y-axis directions at the i-th discrete point x (i) v y (i) can be expressed as follows, where X i , Y i represents the X-axis and Y-axis coordinates of the i-th discrete point, and n is the number of discrete points; 3.5) Based on the calculated feed speeds in the X-axis and Y-axis directions at each discrete point, draw the feed speed curve of the groove machining tool path in the X-axis and Y-axis directions.

3. The method for optimizing slot milling tool paths based on digital filtering according to claim 1, characterized in that: The method steps for smoothing the tool path velocity curve using a Gaussian filter in step 4 are as follows: 4.1) Select a Gaussian filter and determine the filter parameters; 4.2) Calculate the time required for the machine tool to accelerate from rest to a given feed speed; 4.3) Determine the filter cut-off wavelength; 4.4) Use the convolution formula to calculate the feed speed curve.

4. The method for optimizing slot milling tool paths based on digital filtering according to claim 1, characterized in that: In step 5, the new tool position point is recalculated based on the speed of each discrete point on the smoothed speed curve. The specific method is as follows: Each discrete point is calculated based on the new speed to obtain the new X-axis coordinate Q of the tool position point x (i) and the Y-axis coordinate Q y (i) can be calculated by the following formula: where v′ x (i),v′ y (i) is the feed speed of the i-th discrete point in the X-axis and Y-axis directions after filtering.

5. The method for optimizing slot milling tool paths based on digital filtering according to claim 3, characterized in that: The Gaussian filter is represented by the following formula: Among them, u represents the relative position from the center of the weighted function curve, λ c is the cut-off wavelength of the filter, and α can be calculated as follows, 6. The method for optimizing slot milling tool paths based on digital filtering according to claim 5, characterized in that: In step 4.2), the acceleration of the machine tool from static to a given feed speed is calculated as follows: During the machining process, the acceleration and deceleration are carried out according to the S-shaped acceleration and deceleration control mode. First, the acceleration of the machine tool from static to a given feed speed V is solved. F Time requiredT min , where A max is the maximum acceleration of a certain motion axis of the machine tool, J max For machine tool Maximum jerk of each motion axis; when When the acceleration has not reached the maximum value, deceleration is required. In this case, the corresponding acceleration time can be calculated by the following formula:

7. The method for optimizing slot milling tool paths based on digital filtering according to claim 6, characterized in that: The filter cut-off wavelength in 4.3) can be calculated by the following formula: The time required for the speed to drop from a given value to 0 is also T min , 2T min As the cutoff wavelength λ in the Gaussian filter c , that is, filter out the min The speed transition changes dramatically during the acceleration and deceleration reaction time, thereby smoothing the speed curve and the cutoff wavelength λ c It can be calculated by the following formula:

8. The method for optimizing slot milling tool paths based on digital filtering according to claim 7, characterized in that: The calculation method of the feed speed curve described in 4.4) is as follows: The filtered feed speed is calculated using the convolution formula. The filtered feed speed curve p(t) can be calculated by the following formula, where c(ε) is the theoretical feed speed curve: Where ε is the time parameter of the theoretical feed speed curve; The theoretical feed speed curve consists of a series of discrete points, that is, t = kΔt, (k = 0, 1, ..., n-1), where n represents the number of discrete points. The feed speed at each discrete point after filtering can be calculated by the following formula: Where i represents the i-th discrete point after filtering, p(i) represents the feed speed at the i-th discrete point after filtering, and c i-k Represents the theoretical feed speed at the ikth point after filtering, which should satisfy λ c >50Δt, in addition, the filter parameter m is the interaction range parameter of each discrete point during convolution calculation, and m is calculated by the empirical formula:

Images