A numerical control machine tool adaptive machining method suitable for multi-process parts

Abstract

The application belongs to the technical field of intelligent manufacturing, and discloses a numerical control machine tool adaptive machining method suitable for multi-process parts, which comprises the following steps: firstly, data acquisition and transmission are carried out, the spindle power data of the numerical control machine tool during part machining are collected, and the data are sent to an industrial computer; then, the spindle power data of the numerical control machine tool are segmented on the industrial computer by using a time sequence data segmentation method based on a sliding window and a top-down strategy, so that segmented process power data of the multi-process part are obtained; then, a power reference value is set for each segment of process power data, and the best feed speed is calculated based on a fuzzy logic controller; finally, adaptive machining of the numerical control machine tool is carried out. According to the method, the corresponding power reference values are set for different processes of the complex multi-process part, the machining efficiency is improved, the part machining quality and the tool life are ensured, the numerical control machine tool adaptive machining is not affected, and the practicality of the numerical control machine tool adaptive machining is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent manufacturing, and relates to a numerical control machine tool adaptive machining method suitable for multi-process parts. BACKGROUND

[0002] In the traditional numerical control machining process, the cutting parameters are determined in the programming stage before machining. These parameters are often determined according to computer-aided manufacturing software. In order to avoid defective products during part machining, the cutting parameters are relatively conservative, and the cutting parameters cannot be adjusted online according to the actual machining conditions. In fact, using fixed cutting parameters cannot adapt to the actual machining conditions: on the one hand, if the cutting parameters are set conservatively, the machining efficiency is low, and it is difficult to fully exert the performance of the numerical control equipment; on the other hand, if the cutting parameters are set too large, the cutting force, cutting temperature and cutting power are too high, which leads to large tool deformation, fast tool wear rate, and reduced workpiece surface quality, and in severe cases, it can also cause damage to the tool and even the spindle. In the actual numerical control machining process, experienced operators often need to adjust the cutting parameters, which increases the work intensity of the operators. Numerical control machine tool adaptive machining can adjust the cutting parameters in real time according to the machine tool machining load, which is an intelligent manufacturing technology to improve the intelligent degree of numerical control machine tools, improve the part machining efficiency, and reduce the dependence on experienced operators.

[0003] In the aspect of adaptive machining of CNC machine tools, domestic scholars have carried out relevant research. In the patent "Machine tool adaptive control method considering chatter suppression" (Application number: CN202010522482.2), the feed speed and spindle speed are optimized and adjusted in real time based on neural network algorithm and weighted wavelet packet entropy method, which can effectively improve the machining efficiency and machining quality; In the patent "Numerical control machine tool and control method and device for adaptive machining thereof" (Application number: CN202010930080.6), the cutting force is predicted by using a deep learning model, and the optimal feed rate is autonomously found according to the predicted cutting force, realizing adaptive machining of the numerical control machine tool; In the patent "Machine tool adaptive control method based on GA-BP neural network algorithm" (Application number: CN201910732917.3), the feed speed and spindle speed are optimized and adjusted in real time based on neural network algorithm, which can effectively improve the machining efficiency and machining quality; In the patent "Real-time adaptive optimization method for numerical control machining parameters" (Application number: 201410719430.9), a model corresponding to the spindle motor current and the actual cutting force is established, and the spindle speed and feed speed are optimized by using fuzzy algorithm, which effectively improves the machining quality and machining efficiency, reduces tool and machine tool damage, and reduces production cost; In the paper "On-line monitoring of composite material drilling surface roughness and adaptive optimization of machining parameters" (Journal of Mechanical Engineering, 2020, 56(02): 27-34+42.), a hole wall roughness on-line monitoring model based on support vector regression is established, and the drilling parameters are optimized under the current monitoring state by using simulated annealing algorithm to ensure the drilling quality; In the paper "Research on constant power constraint adaptive machining method combined with chatter control" (Machinery Manufacturing and Automation, 2018, 47(05): 41-44+52.), based on fuzzy theory and variable speed cutting chatter suppression theory, the adjustment principle of machine tool machining parameters is formulated with machining efficiency as the target, realizing constant power constraint adaptive machining combined with chatter control.

[0004] Through the analysis of adaptive machining technology of CNC machine tools, it is found that the current adaptive machining method of CNC machine tools can improve the machining efficiency by adjusting the cutting parameters, but the existing adaptive machining method regards the part machining process as the same process, and the target of cutting parameter regulation is a constant reference value, which is not suitable for complex multi-process part machining, and the actual application value is low, so a CNC machine tool adaptive machining method suitable for multi-process parts is needed. The invention aims at the above problems, and proposes a multi-process part process automatic division method, which can divide the part machining process according to different stages, and can set corresponding constant regulation reference values for different stages, adjust the cutting parameters in real time, and significantly improve the machining efficiency of complex multi-process parts. SUMMARY

[0005] In order to solve the problem that the common numerical control machine tool adaptive machining method does not consider setting corresponding control reference values for each process of a complex multi-process part, the application designs a numerical control machine tool adaptive machining method suitable for multi-process parts.

[0006] The technical scheme of the application:

[0007] A numerical control machine tool adaptive machining method suitable for multi-process parts, the steps are as follows:

[0008] S1: data acquisition and transmission: acquiring spindle power data when a numerical control machine tool processes a part, and sending the acquired spindle power data to an industrial computer;

[0009] S1.1: installing a power sensor at a proper position of the numerical control machine tool, and using the power sensor to acquire spindle power data in the processing process;

[0010] S1.2: sending the spindle power data to the industrial computer through a data transmission line;

[0011] S2: using the industrial computer to automatically divide the processes of the multi-process part: performing smoothing filtering processing on the spindle power data acquired in step S1 to obtain filtered spindle power data, and using a time series data segmentation method based on a sliding window and a top-down strategy (SWAB method) to segment the filtered spindle power data to obtain segmented process power data of the multi-process part;

[0012] S2.1: performing smoothing filtering processing on the spindle power data acquired in step S1 to obtain filtered spindle power data;

[0013] S2.2: using a time series data segmentation method based on a sliding window and a top-down strategy to segment the filtered spindle power data to obtain segmented process power data of the multi-process part, the time series data segmentation method based on a sliding window and a top-down strategy includes a sub-method data segmentation method based on a sliding window (SW method) and a sub-method data segmentation method from bottom to top (BU method);

[0014] S2.2.1: first, preset a maximum error threshold δ and a window length threshold w, and initialize the starting point of the filtered spindle power data as the left boundary point w_left of the sliding window;

[0015] S2.2.2: using the SW method to update the right boundary point w_right of the sliding window according to the maximum error threshold δ;

[0016] S2.2.3: using the BU method to segment the buffered spindle power data in the window according to the maximum error threshold δ to obtain a plurality of linear segments;

[0017] S2.2.4: Determine whether the number of linear segments within the sliding window has reached the preset window length threshold w. If it has not reached the preset window length threshold w, repeat steps S2.2.3 to S2.2.4; if it has reached the preset window length threshold w, continue running.

[0018] S2.2.5: Determine whether the segmentation task is completed. If the segmentation is not completed, first update the left boundary point w_left of the window to the right endpoint l_right of the first linear segment in the sliding window, then move the first linear segment out of the sliding window and record its endpoint as the segmentation point; if the segmentation is completed, continue to run.

[0019] S2.2.6: Divide the filtered spindle power data according to the segmentation points to obtain the segmented process power data of multi-process parts;

[0020] Furthermore, in step S2.2.1, the window length threshold w is set to 4 to 6.

[0021] S3: Calculate the optimal feed speed in real time using an industrial control computer: Set a constant power reference value for the segmented process power data in step S2, and select a dual-input single-output fuzzy logic controller as the solution model for the optimal feed speed. The input of the fuzzy logic controller is the power deviation and the rate of change of the power deviation, and the output is the optimal feed speed.

[0022] S3.1: Set a constant power reference value Pi for the power data of the segmented process in step S2. ref A dual-input, single-output fuzzy logic controller is selected as the optimal feed rate solution model. The inputs to the fuzzy logic controller are the power deviation e and the power deviation change rate ec. The formula for the power deviation e is: e = P o -Pi ref , where P o The filtered spindle power data; the formula for the power deviation change rate (ec) is: Where de is the differential of power deviation and dt is the differential of time;

[0023] S3.2: Use negative large NB, negative medium NM, negative small NS, zero ZO, positive small PS, positive medium PM, and positive large PB linguistic variables to describe power deviation e, power deviation change rate ec, and optimal feed rate f. c e, ec and f c Let the fuzzy universe of discourse be [-n, n], and its fuzzy set be denoted as {NB, NM, NS, ZO, PS, PM, PB}. Trigonometric functions are used as e, ec, and f. c Membership function;

[0024] S3.3: a fuzzy control rule table is formulated, a traditional fuzzy inference method (Mamdani fuzzy inference method) is selected to calculate fuzzy membership output, and an area barycenter method is used to calculate defuzzification to obtain the optimal feed speed f c ;

[0025] Further, in step S3.2, n in the fuzzy domain [-n, n] of e, ec and f c is a positive integer not greater than 7.

[0026] S4: adaptive control of the feed ratio of the machine tool by using the industrial computer: the optimal feed speed obtained in step S3 is used to solve the feed ratio control value on the industrial computer, and the feed ratio control value is sent to the numerical control system of the numerical control machine tool, so that adaptive machining of the numerical control machine tool is realized;

[0027] S4.1: the feed ratio control value of the numerical control machine tool is set as M f , and the calculation formula is Wherein, f c is the optimal feed speed, and f o is the feed speed set in the machining program;

[0028] S4.2: the industrial computer and the numerical control system of the numerical control machine tool are connected through a network cable and communicate in real time;

[0029] S4.3: the feed ratio control value M f is sent to the numerical control system, and adaptive control of the feed ratio of the machine tool is realized through the built-in PLC module of the machine tool.

[0030] Advantages of the present application:

[0031] (1) The adaptive machining method of the numerical control machine tool suitable for multi-process parts of the present application sets corresponding power reference values for different processes, which can ensure that the machining quality and tool life of the parts are not affected by the machining parameter control of the machine tool, and improve the practicality of adaptive machining of the numerical control machine tool.

[0032] (2) The adaptive machining method of the numerical control machine tool suitable for multi-process parts of the present application improves the machining efficiency of complex structure blank parts, and significantly saves production time cost.

[0033] (3) The adaptive machining method of the numerical control machine tool suitable for multi-process parts of the present application can adaptively adjust the feed ratio of the numerical control machine tool in the machining process, and improves the intelligent level of the numerical control machine tool. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 It is a schematic diagram of the arrangement of the power sensor of the embodiment;

[0035] Figure 2The membership function diagram of the power deviation e, the power deviation change rate ec and the feed speed control value f in the embodiment fuzzy logic controller c ;

[0036] Figure 3 The feed speed ratio control value M f over time diagram

[0037] Figure 4 The flow chart of the method of the application.

[0038] In the figure: 1 is a three-axis vertical machining center; 2 is a spindle servo motor; 3 is a workbench; 4 is a power sensor; 5 is a spindle. DETAILED DESCRIPTION

[0039] The specific embodiments of the application are further described below in combination with the drawings and technical solutions.

[0040] The numerical control machine tool used in the embodiment is a three-axis vertical machining center, and the control method of other numerical control machine tools refers to the embodiment.

[0041] Embodiment 1

[0042] A numerical control machine tool adaptive machining method suitable for multi-process parts is realized by the following steps:

[0043] S1: data acquisition and transmission, specifically, acquiring spindle power data when a numerical control machine tool processes a part, and sending the acquired spindle power data to an industrial computer; S1 is specifically divided into two steps,

[0044] S1.1: as shown in the figure, a power sensor 4 is installed at an appropriate position of the three-axis vertical machining center 1, the spindle 5 of the three-axis vertical machining center 1 in the embodiment is located above the workbench 3, a spindle servo motor 2 is connected above the spindle 5, the power sensor 4 is arranged on the spindle servo motor 2, three current induction coils thereof are respectively sleeved on three power lines U, V and W of the spindle motor, and the voltage is connected to the three power lines U, V and W in a parallel manner; Figure 1

[0045] S1.2: sending the spindle power data to the industrial computer through a data transmission line.

[0046] S2: using the industrial computer to automatically divide the processes of the multi-process part, specifically, performing smoothing filtering processing on the spindle power data obtained in step S1 to obtain filtered spindle power data, using a time series data segmentation method (SWAB method) based on a sliding window and a top-down strategy to segment the filtered spindle power data, and obtaining segmented process power data of the multi-process part;

[0047] ​S2 is specifically divided into two steps:

[0048] S2.1: smoothing filtering processing is performed on the spindle power data obtained in step S1 to obtain filtered spindle power data;

[0049] S2.2: using a time series data segmentation method based on a sliding window and a top-down strategy (SWAB method) to segment the filtered spindle power data to obtain segmented process power data of a multi-process part, the SWAB method includes a sub-method data segmentation method based on a sliding window (SW method), and a method pseudo code diagram of the SW method is as follows:

[0050]

[0051] and a sub-method bottom-up data segmentation method (BU method), and a method pseudo code diagram of the BU method is as follows:

[0052]

[0053] Specifically:

[0054] S2.2.1: in this embodiment, a maximum error threshold δ is preset as 50, a window length threshold w is 5, and an initial starting point of the filtered spindle power data is a left boundary point w_left of the sliding window;

[0055] S2.2.2: the SW method is used to update a right boundary point w_right of the sliding window according to the maximum error threshold δ;

[0056] S2.2.3: the BU method is used to segment the buffered spindle power data in the window according to the maximum error threshold δ to obtain a plurality of linear segments;

[0057] S2.2.4: it is judged whether the number of linear segments in the window reaches the preset window length threshold w, if not, steps S2.2.3 to S2.2.4 are repeatedly run; if yes, the following step is run;

[0058] S2.2.5: it is judged whether the segmentation task is completed, if not, the window left boundary point w_left is first updated as a right end point l_right of the first linear segment in the window, then the first linear segment in the window is moved out of the window and its end point is recorded as a segmentation point; if yes, the following step is run;

[0059] S2.2.6: the filtered spindle power data is segmented according to the segmentation point to obtain segmented process power data of a multi-process part.

[0060] The pseudo code of the process automatic division method of the multi-process part of the embodiment is shown as follows:

[0061]

[0062] S3: Real-time calculation of the optimal feed speed by the industrial computer, specifically, setting a constant power reference value for the segmented process power data in step S2, selecting a double-input single-output fuzzy logic controller as the optimal feed speed solving model, the input of the fuzzy logic controller being the power deviation and the power deviation change rate, and the output being the optimal feed speed;

[0063] The S3 specifically includes three steps:

[0064] S3.1: Setting a constant power reference value Pi for the segmented process in step S2 ref , selecting a double-input single-output fuzzy logic controller as the optimal feed speed solving model, the input of the fuzzy logic controller being the power deviation e and the power deviation change rate ec, the power deviation e formula being: e = P o -Pi ref , wherein P o is the main shaft power data; the power deviation change rate ec formula being:

[0065] S3.2: Using negative large, negative medium, negative small, zero, positive small, positive medium and positive large language variables to describe the power deviation e, the power deviation change rate ec and the optimal feed speed f c , the fuzzy domain of e, ec and f c being set as [-6, 6], the fuzzy set being represented as {NB, NM, NS, ZO, PS, PM, PB}, and using a triangular function as the membership function of e, ec and f c , as shown in Figure 2 ;

[0066] S3.3: Formulating a fuzzy control rule table based on expert experience, as shown in Table 1, selecting a Mamdani fuzzy reasoning method to calculate the fuzzy membership output, and using the area barycenter method to calculate the defuzzification, to obtain the optimal feed speed f c .

[0067] Table 1 Rule chart of the fuzzy control module

[0068]

[0069] S4: adaptive control of the feed ratio of the machine tool by the industrial computer, specifically, solving the feed ratio control value by the optimal feed speed on the industrial computer, sending the feed ratio control value to the numerical control system of the numerical control machine tool, thereby realizing adaptive machining of the numerical control machine tool.

[0070] S4 is specifically divided into three steps:

[0071] S4.1: the feed ratio control value of the numerical control machine tool is M f , and the calculation formula is Wherein, f c is the optimal feed speed, and f o is the feed speed set in the machining program. The feed ratio control value M f obtained by the embodiment is as shown in the table. Figure 3

[0072] S4.2: the industrial computer and the numerical control system of the numerical control machine tool are connected through a network cable and communicate in real time;

[0073] S4.3: the feed speed ratio control value M f is sent to the numerical control system, and adaptive control of the machine tool feed ratio is realized through the built-in PLC module of the machine tool.

[0074] The flowchart of the method of the embodiment is as shown in the figure. Figure 4

[0075] The technical effect of the embodiment is that the complex parts are divided into six process power data according to the spindle power data, and the part machining efficiency is improved by 23.4%, and the production time cost is saved.

[0076] It should be noted that the above specific embodiments of the present application are only used to illustrate the principles and processes of the present application, and do not constitute a limitation on the present application. Therefore, any modification and equivalent replacement made without departing from the spirit and scope of the present application shall be included in the protection scope of the present application.​​

Claims

1. A self-adaptive machining method for a numerical control machine tool suitable for multi-process parts, characterized in that, The steps are as follows: S1, data acquisition and transmission: collecting spindle power data when machining parts on a numerical control machine tool, and sending the obtained spindle power data to an industrial computer; S2, using the industrial computer to automatically divide the process of the multi-process part: smoothing filtering the spindle power data obtained in step S1 to obtain filtered spindle power data, using a time series data segmentation method based on a sliding window and a top-down strategy to segment the filtered spindle power data to obtain segmented process power data of the multi-process part; The specific steps are as follows: S2.1: smoothing filtering the spindle power data obtained in step S1 to obtain filtered spindle power data; S2.2: using a time series data segmentation method based on a sliding window and a top-down strategy to segment the filtered spindle power data to obtain segmented process power data of the multi-process part, the top-down strategy time series data segmentation method includes a sub-method data segmentation method based on a sliding window, i.e. SW method, and a sub-method data segmentation method from bottom to top, i.e. BU method; S2.2.1: first, preset the maximum error threshold δ and the window length threshold w, and initialize the starting point of the filtered spindle power data as the left boundary point w_left of the sliding window; S2.2.2: using the SW method to update the right boundary point w_right of the sliding window according to the maximum error threshold δ; S2.2.3: using the BU method to segment the buffered spindle power data in the window according to the maximum error threshold δ to obtain several linear segments; S2.2.4: determine whether the number of linear segments in the sliding window reaches the preset window length threshold w, if not, repeat steps S2.2.3 to S2.2.4; if yes, continue to run; S2.2.5: determine whether the segmentation task is completed, if not, first update the window left boundary point w_left to the right end point l_right of the first linear segment in the sliding window, then move the first linear segment in the sliding window out of the sliding window and record its end point as a segmented point; if yes, continue to run; S2.2.6: segment the filtered spindle power data according to the segmented point to obtain segmented process power data of the multi-process part; S3, using the industrial computer to calculate the optimal feed speed in real time: setting a constant power reference value for the segmented process power data in step S2, selecting a double-input single-output fuzzy logic controller as the optimal feed speed solving model, the input of the fuzzy logic controller is the power deviation and the power deviation change rate, and the output is the optimal feed speed; S4, using the industrial computer to adaptively control the feed ratio of the machine tool: using the optimal feed speed in step S3 to solve the feed ratio control value on the industrial computer, and sending the feed ratio control value to the numerical control system of the numerical control machine tool, thereby realizing adaptive machining of the numerical control machine tool.

2. The method of adaptive machining of CNC machine tools for multi-process parts as claimed in claim 1, wherein, The specific steps of step S1 are as follows: S1.1: install a power sensor at an appropriate position of the numerical control machine tool, and use the power sensor to collect spindle power data during the machining process; S1.2: send spindle power data to industrial computer through data transmission line.

3. The method of claim 1, wherein, In step S2.2.1, the window length threshold w is set to 4-6.

4. The method for adaptive machining of CNC machine tools for multi-process parts as claimed in claim 1, wherein, The specific steps of step S3 are as follows: S3.1: setting a constant power reference value Pi for the step S2 segment process power data ref The fuzzy logic controller with double input and single output is selected as the best feed speed solution model, the input of the fuzzy logic controller is the power deviation e and the power deviation change rate ec, the formula of the power deviation e is: e=P o -Pi ref Wherein, P o is the filtered main shaft power data; the formula of the power deviation change rate ec is: Wherein, de is the differential of the power deviation, dt is the differential of the time; S3.2: Power deviation e, power deviation change rate ecand optimal feed rate f are described using negative large NB, negative medium NM, negative small NS, zero ZO, positive small PS, positive medium PM and positive large PB linguistic variables c The fuzzy domains of e, ecand f c are set to [-n, n] and their fuzzy sets are represented as {NB, NM, NS, ZO, PS, PM, PB} using a triangular function as the membership function of e, ecand f c ; S3.3: Make fuzzy control rule table, select traditional fuzzy inference method to calculate fuzzy membership output, use area barycenter method to calculate de-fuzzification, get the best feed speed f c .

5. The method for adaptive machining of CNC machine tools for multi-process parts as claimed in claim 4, wherein, In step S3.2, e, ec and f c n in the fuzzy domain [-n, n] of f is a positive integer not greater than 7.

6. The method for adaptive machining of CNC machine tools for multi-process parts as claimed in claim 1 wherein, The specific steps of step S4 are as follows: S4.1: Set the feed ratio control value of the numerical control machine tool as M f , the calculation formula is Wherein, f c is the optimal feed speed, f o is the feed speed set by the machining program; S4.2: the industrial computer and the numerical control system of the numerical control machine tool are connected through a network cable for real-time communication; S4.3: Adjust the feed ratio control value M f The feed ratio control value M is sent to the numerical control system, and the adaptive control of the feed ratio of the machine tool is realized through the built-in PLC module of the machine tool.

Images