A method for selecting a machining tool for a cavity feature based on process information mining

Abstract

The present application aims at the technical problems of the existing cavity feature machining tool selection method, such as coarse granularity, great experience dependence and weak process attribute correlation, and proposes a cavity feature machining tool selection method based on process information mining, belonging to the field of tool selection method. Firstly, the present method uses the medial axis transformation as the cavity feature tool path, and analyzes the mapping mechanism of the medial axis transformation and the numerical control machining process. Then, according to the numerical control machining rules, the intuitive and typical process information is mined, including the allowable tool set, the tool retraction times, the effective cutting area and the like. Finally, a multi-objective optimization model and a candidate tool set graph are constructed, and a hybrid optimization algorithm is used for solving, so as to obtain the optimal machining tool sequence. The present application has strong practicability, improves the intelligent level and reliability of tool selection, and has high calculation efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to a tool selection method, in particular to a cavity feature machining tool selection method based on process information mining. BACKGROUND

[0002] The document "Research on Computer Aided Tool Selection for Mold Cavity NC Machining", Computer Integrated Manufacturing Systems, Vol. 10, No. 2, pp. 226-229, 2004 discloses a cavity machining tool selection method. The method first summarizes the principles and methods of tool selection in each machining stage of the cavity, that is, rough machining removes most of the material as quickly as possible, semi-finishing mainly removes the steps left by rough machining, and finishing needs to ensure that the part size accuracy meets the requirements. Then, the tool type is determined according to the process type and the geometric shape of the cavity, such as flat bottom milling cutter for rough machining, ball head milling cutter for semi-finishing, etc. Finally, based on the principles of cavity entity layering and merging, the optimal tool combination is selected as the overall machining efficiency. However, the method described in the document mainly selects tools by analyzing the geometric shape parameters of the mold cavity, and fails to effectively associate the actual cutting motion process of the tool and its numerical control process connotation elements (tool diameter, tool path, number of tool retraction, etc.), resulting in a coarse tool granularity and poor process practicability, which makes it difficult to effectively support the numerical control machining process design in the production workshop.

[0003] The medial axis transformation of the cavity feature has particularity, which divides the cavity region into two symmetrical parts, which is similar to the process of tool movement along the predefined trajectory and then cutting the feature to remove material. Therefore, by introducing the medial axis transformation of the cavity feature as the tool path, the cavity feature geometric properties and numerical control machining process can be mapped, valuable process information can be mined, and subsequent tool selection can be supported. SUMMARY

[0004] The present application provides a cavity feature machining tool selection method based on process information mining to overcome the deficiencies of coarse granularity and poor process practicability in the prior art.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0006] A cavity feature machining tool selection method based on process information mining, comprising the following steps:

[0007] Step 1: Calculate the set of continuous curves and straight lines formed by connecting the centers of all the inscribed circles tangent to the boundary curve of the cavity feature in turn, and define the set of continuous curves and straight lines as the medial axis of the cavity feature;

[0008] Step 2: Based on the maximum and minimum diameters of all inscribed circles, with the goal of completing the overall roughing of the feature area without interference, determine the maximum and minimum diameters of the cutting tools, and obtain the allowable tool set based on the maximum and minimum diameters of the cutting tools;

[0009] Step 3: Determine the effective toolpath based on the continuous curve or straight line connecting the centers of the local inscribed circles, and obtain the number of tool retractions based on the number of effective toolpaths; calculate the effective cutting area obtained by a single tool along each effective toolpath using OpenGL interface functions, and sum the effective cutting areas obtained along each effective toolpath to obtain the overall effective cutting area of ​​a single tool.

[0010] Step 4: Find the intersection of the selectable tool set and the allowable tool set in the current machining center, and add the tool with the smallest diameter to obtain the candidate tool set. With the goal of maximizing the overall effective cutting area and minimizing the overall number of tool retractions when selecting tools from the optimal tool set in sequence, use the hybrid ant colony simulated annealing algorithm to optimize the candidate tool set and obtain the optimal tool set.

[0011] To optimize the above technical solution, the specific measures also include:

[0012] Furthermore, step 1 specifically includes:

[0013] Based on the central axis transformation, calculate the coordinates O of the center of all inscribed circles tangent to the feature boundary curve of the cavity. i (x i ,y i ) and diameter d i Connecting the centers of all inscribed circles sequentially forms a set of continuous curves and straight lines. This set of continuous curves and straight lines is defined as the central axis P of the cavity feature, expressed as:

[0014]

[0015] Where M is the total number of inscribed circles, and O i (x i ,y i Let d be the coordinates of the center of the i-th incircle among all incircles. i Let be the diameter of the i-th inscribed circle. The coordinates of the center of the circle and the diameter of the inscribed circle are calculated using the equation of the median transformation.

[0016] Furthermore, step 2 specifically involves:

[0017] Calculate the maximum tool diameter D L :

[0018] D L =int(d max -2δ)

[0019] wherein int() is an integer function, d max is the maximum value of the diameter of the overall inscribed circle, and δ is a rough machining allowance;

[0020] The minimum tool diameter D S is calculated:

[0021] D S = int(d min - 2δ)

[0022] wherein d min is the minimum value of the diameter of the overall inscribed circle;

[0023] The set of all feasible tools for the cavity feature is defined as the allowable tool set A(F) T , which is expressed as:

[0024] A(F) T = {D q | D q = D L - q, q = 0, 1, 2, 3..., (D L - D S )}

[0025] wherein D q is the diameter of the tool in the allowable tool set.

[0026] Further, step 3 is specifically:

[0027] a series of local inscribed circles {ω(O q (x k ,y k ), d k )| d k ≥ D k + 2δ, 0 ≤ k ≤ K} with diameters not less than (D q + 2δ), K is the number of local inscribed circles, d k is the diameter of the local inscribed circle, and the centers of the local inscribed circles {O k (x k ,y k )} are sequentially connected to form a continuous curve or straight line, wherein D q is the diameter of the tool in the allowable tool set, and δ is a rough machining allowance;

[0028] The continuous curve or straight line is taken as an effective tool path L q (D j )(0 ≤ j ≤ Q) of the tool with a diameter of D q in the allowable tool set, wherein Q is the number of effective tool paths generated in the tool cutting process;

[0029] The effective cutting area S(D q ) of a tool with diameter D j (D q ) is calculated by OpenGL interface functions. q j q j q q q

[0030]

[0031] N(D q ) is the number of retraction times.

[0032] Further, step 4 includes:

[0033] Step 4.1: the optional tool set in the current machining center is A(F) C , and the alternative tool set T C = D S U(A(F) T ∩A(F) C ) = {D1, D2, …, D i , …, D S}, where D S is the minimum tool diameter, A(F) T is the allowable tool set, and D i is the diameter of the tool in the alternative tool set; the tool diameters are sorted in descending order: D1 > D2 > … > D i … > D S

[0034] Step 4.2: establish an alternative tool directed graph CTP, and abstract the diameter D i of each tool in the alternative tool set as a vertex in the alternative tool directed graph CTP. Use vertex D0 to represent the initial blank state of the cavity feature, and the corresponding tool diameter is +∞. The remaining vertices represent the diameters of the machining tools in the alternative tool set T C . There is a directed edge between any two vertices, and the direction is from the larger tool to the smaller tool, which can be expressed as:

[0035] <ctp> ::＝(V, <e>})

[0036]

[0037]

[0038] wherein V represents the alternative tool has to the vertex of the graph CTP, E represents the alternative tool has to the directed edge of the graph CTP, m is the number of alternative tool has to the vertex of the graph CTP, the weight value indicates that the diameter D i large tool is selected first, and then the diameter D j small tool is selected, the weight value Dis(D i ,D j ) is the effective cutting area S(D i ) and the number of tool retractions N(D j ) after the diameter D i→j large tool is used to complete the machining, and then the diameter D i→j small tool is used;

[0039] Step 4.3: normalize the weight value Dis(D i ,D j );

[0040] Step 4.4: in the alternative tool has to the graph CTP, take the vertex D0 as the starting point and the vertex D S as the end point, the path from the starting point to the end point corresponds to a feasible tool sequence, a plurality of optimal path results are obtained by searching through the ant colony algorithm, the optimal path results are input into the simulated annealing algorithm for global optimization convergence, and the final global optimal solution is obtained as the optimal tool set.

[0041] Further, step 4.3 is specifically:

[0042] Step 4.3.1: calculate the weight value of all directed edges {Dis(D i ,D j )={S(D i→j ),N(D i→j )}(0≦i≦n-1,i<j≦n), wherein {S(D i→j )} is the effective cutting area set, {N(D i→j )} is the tool retraction number set, and n is the number of directed edges;

[0043] Step 4.3.2: take the maximum cutting area S max in the cutting area set {S(D i→j )} and the maximum tool retraction number N max in the tool retraction number set {N(D i→j )}, respectively, and the cutting area set {S(D i→j )} and the tool retraction number set {N(D i→j The elements in the formula are normalized as follows:

[0044]

[0045] In the formula, S(D i→j )' is the normalized effective cutting area, and N(D i→j )' is the normalized number of retracts.

[0046] Step 4.3.3: Define the normalized weight Dis(D i , D j )' as follows

[0047] Dis(D i , D j )' = ω1*(1-S(D i→j )')) + ω2*N(D i→j )'

[0048] ω1+ω2=1

[0049] where ω1 and ω2 represent the weights of the importance of the tool path and the number of retracts, respectively.

[0050] Further, the ant colony algorithm is specifically as follows:

[0051] Randomly select a vertex D i as the access vertex of the ant Q, put the vertex D i into the taboo list, and automatically calculate the next available vertex set allowed Q (D i ), according to the search rules of the ant colony algorithm, the ant selects the access vertex D j ∈allowed Q (D i ), then puts the access vertex D j into the taboo list, calculates the next available vertex set allowed Q (D j ), and repeats the search until the ant reaches the end vertex D S , records the search path, randomly selects a new vertex as the access vertex of the ant, and repeats the ant colony search process until the iteration termination condition is reached, to obtain a plurality of optimal path results.

[0052] Further, the input of the optimal path results into the simulated annealing algorithm for global optimization convergence to obtain the final global optimal solution as the optimal tool set is specifically as follows:

[0053] Input the tool sequences corresponding to the X optimal paths into the simulated annealing algorithm, and input the obtained R groups of tool sequences {X1, X2, … X R} are sequentially iterated, wherein, for a certain tool sequence X i , a new solution X i is generated by probabilistic position tool variation C , a tool with a diameter of D0 is randomly selected from the alternative tool set T to replace a position tool in the tool sequence X i , and D0 and D S are excluded from further global search, and after the R-group tool sequence calculation is completed, the optimal solution obtained is taken as the optimal tool set.

[0054] The beneficial effects of the present application are:

[0055] Compared with the prior art, the present application effectively solves the technical problems of coarse granularity, strong experience dependence and weak process attribute correlation in the existing cavity feature machining tool selection method. The main contributions of the present application include: ①Introducing the medial axis transformation as the motion trajectory of the cavity feature tool machining process, reasonably correlating and mapping the geometric attributes and numerical control machining, mining the process information that deeply reflects the machining process, and having strong practicability; ②Based on the obtained process information such as the number of tool retractions and the effective cutting area, the alternative tool set in the tool magazine is represented in the form of a weighted directed graph, the tool sequence selection optimization problem is converted into a graph search problem, and a hybrid optimization algorithm is used for solving, thereby improving the intelligent level and reliability of tool selection, and having higher calculation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 is a flowchart of the method of the present application;

[0057] Figure 2 is a schematic diagram of the effective cutting area of the method of the present application;

[0058] Figure 3 is an alternative tool directed graph of the method of the present application;

[0059] Figure 4 is a tool sequence optimization solution flowchart. DETAILED DESCRIPTION

[0060] The present application will now be further described in detail in conjunction with the accompanying drawings.

[0061] In an embodiment, the present application proposes a cavity feature machining tool selection method based on process information mining, and a flowchart of the method is shown in Figure 1 , which specifically comprises:

[0062] Step 1: Calculate the continuous curve and straight line set formed by sequentially connecting the centers of all the inscribed circles tangent to the cavity feature boundary curve, and define the continuous curve and straight line set as the medial axis of the cavity feature;

[0063] According to the medial axis transform, the coordinates of the center of the inscribed circle tangent to the feature boundary curve of the cavity are calculated i (x i ,y i ) and the diameter d i , the center of the inscribed circle is sequentially connected to form a continuous curve and a straight line set, and the continuous curve and the straight line set are defined as the medial axis line P of the feature of the cavity, which is represented as:

[0064]

[0065] Wherein, M is the number of inscribed circles, which can be defined by oneself, O i (x i ,y i ) is the center coordinates of the i-th inscribed circle, and d i is the diameter of the i-th inscribed circle. The center coordinates and the diameter of the inscribed circle are calculated by the medial axis transform equation.

[0066] Step 2: According to the maximum diameter and the minimum diameter of the inscribed circle, the maximum diameter and the minimum diameter of the tool are determined to complete the overall rough machining of the feature area without interference, and the allowable tool set is obtained according to the maximum diameter and the minimum diameter of the tool;

[0067] The maximum tool diameter D L is calculated:

[0068] D L =int(d max -2δ)

[0069] In the formula, int() is the integer function, d max is the maximum value of the diameter of the inscribed circle, and δ is the rough machining allowance;

[0070] The minimum tool diameter D S is calculated:

[0071] D S =int(d min -2δ)

[0072] In the formula, d min is the minimum value of the diameter of the inscribed circle; D S cannot be too small, otherwise the tool path is increased, causing unnecessary waste of time.

[0073] From the perspective of numerical control machining, only the tool with a diameter not less than d min -2δ (δ is the rough machining allowance) can complete the overall rough machining of the feature area along the medial axis line P without interference. On the other hand, when the tool diameter exceeds d max -2δ, the tool will interfere with the part when moving along any part of P, and the part machining size requirement cannot be met. This means that only the tool with diameter no greater than d max -2δ can be used to rough machine the feature part area along the center axis. Therefore, d max and d min The geometric openness of the cavity feature F is described, and the center axis transformation is effectively mapped to the process.

[0074] The set of all feasible tools for the cavity feature is defined as the allowable tool set A(F) T , which is expressed as:

[0075] A(F) T ={D q |D q =D L -q,q=0,1,2,3..., (D L -D S )}

[0076] In the formula, D q is the diameter of the tool in the allowable tool set.

[0077] The center axis transformation and the machining continuity mapping reflect the continuity of the tool movement process, which has an important influence on the machining result, and is often used as a reference index for the number of tool retraction.

[0078] Step 3: Determine the effective tool path according to the continuous curve or straight line formed by connecting the centers of the local inscribed circles, and obtain the number of tool retractions according to the number of effective tool paths; calculate the effective cutting area obtained by cutting along each effective tool path by a single tool through the OpenGL interface function, sum the effective cutting areas obtained by cutting along each effective tool path, and obtain the overall effective cutting area of the single tool; the effective cutting area is shown in Figure 2 .

[0079] A series of local inscribed circles with a diameter no less than (D q +2δ) {ω(O k (x k ,y k ), d k )|d k ≥D q +2δ, 0≤k≤K}, K is the number of local inscribed circles, d k is the diameter of the local inscribed circle, and the centers of the local inscribed circles {O k (x k ,y k )} are sequentially connected to form a continuous curve or straight line, wherein D q is the diameter of the tool in the allowable tool set, and δ is the rough machining allowance.

[0080] The continuous curve or straight line is taken as an effective tool path L q of a tool with a concentrated diameter D j (D q )(0≤j≤Q), wherein Q is the number of effective tool paths generated during tool cutting;

[0081] When a tool with a diameter D q completes a cutting movement along an effective tool path, a tool retracting operation needs to be performed to move to the next one, so the number of tool retractions N(D q ) during the machining of a tool with a diameter D q , that is, the number of effective tool paths generated.

[0082] The effective cutting area S j (D q ) of a tool with a radius D q along an effective tool path L j (D q ) is calculated through an OpenGL interface function;

[0083] The overall effective cutting area S(D q ) of a tool with a diameter D q is:

[0084]

[0085] In the formula, N(D q ) is the number of tool retractions.

[0086] Step 4: Intersect the selectable tool set in the current machining center with the allowable tool set, and add the tool with the smallest diameter to obtain a candidate tool set, so as to maximize the overall effective cutting area and minimize the overall number of tool retractions when selecting tools in the optimal tool set in order, and use a hybrid ant colony simulated annealing algorithm to optimize the candidate tool set to obtain the optimal tool set.

[0087] Step 4.1: The selectable tool set in the current machining center is A(F) C , and the candidate tool set T C =D S U(A(F) T ∩A(F) C )={D1,D2,…,D i ,…,D S}, wherein D S is the smallest tool diameter, A(F) T is the allowable tool set, and D i is the diameter of the tool in the candidate tool set; sort the tool diameters in descending order: D1>D2>…>D i …>D S

[0088] Step 4.2: Establishing Candidate Tools Graph (CTP) Figure 3 As shown in FIG. 2, Candidate Tools Graph (CTP) is a directed graph that describes the set of candidate tools T C The weighted directed graph of feature morphology alternation in rough machining process. The diameters D i of each tool in the set of candidate tools T C are abstracted as vertices in the Candidate Tools Graph (CTP), with vertex D0 representing the initial blank state of the feature, corresponding to a tool diameter of +∞, and the rest of the vertices representing the diameters of the tools in the set of candidate tools T

[0089] <ctp> ::＝(V, <e>})

[0090]

[0091]

[0092] In the formula, V represents the vertices of the directed graph CTP of the candidate tools, E represents the directed edges of the directed graph CTP of the candidate tools, m is the number of vertices in the directed graph CTP of the candidate tools, and the weight is... This indicates that a diameter of D should be selected first. i After selecting a large cutting tool, choose one with a diameter of D. j Small knife, weight Dis(D) i D j ) is using a diameter of D i After the large cutting tool completes the machining, a diameter of D is selected. j The effective cutting area S(D) of a small cutting tool i→j ) and the number of retractions N(D) i→j );

[0093] Step 4.3: Normalize the weights Dis(D) i D j );

[0094] Step 4.3.1: Calculate the weights of all directed edges {Dis(D)} i D j )={S(D i→j ),N(D i→j )}(0≦i≦n-1,i<j≦n), where {S(D i→j {N(D)} represents the set of effective cutting areas. i→j Let} be the set of times the tool was withdrawn, and n be the number of directed edges;

[0095] Step 4.3.2: Take the cutting area set {S(D)} respectively. i→j The largest area S in )} max and the set of retraction times {N(D i→j The maximum number of retractions N in )} max For the set of cutting areas {S(D i→j The set of retraction counts {N(D)} and {N(D)} i→j The elements in )} are normalized as follows:

[0096]

[0097] In the formula, S(D) i→j )' represents the normalized effective cutting area, N(D) i→j )' represents the normalized number of retractions;

[0098] Step 4.3.3: Define the normalized weight Dis(D i , j ) as follows

[0099] Dis(D i , j ) = ω1*(1-S(D i→j )')) + ω2*N(D i→j )

[0100] ω1+ω2=1

[0101] wherein ω1 and ω2 represent the weights of the importance of the tool path and the number of tool retractions, respectively.

[0102] Step 4.4: In the alternative tool set graph CTP, take vertex D0 as the starting point and vertex D S as the end point, the path from the starting point to the end point corresponds to a feasible tool sequence, input the better path results into the simulated annealing algorithm for global optimization convergence, the optimization goal is to find the optimal subset from the alternative tool set T C , meet the total maximum effective cutting area when selecting the tools in the optimal subset in order and the total minimum number of tool retractions, and obtain the final global optimal solution as the optimal tool set.

[0103] The ant colony algorithm is as follows:

[0104] Randomly select vertex D i as the access point of the ant, put vertex D i into the taboo list, and automatically calculate the next available vertex set allowed Q (D i ), according to the search rules of the ant colony algorithm, the ant selects the access point D j ∈allowed Q (D i ), then puts the access point D j into the taboo list, calculates the next available vertex set allowed Q (D j ), and repeats the search until the ant reaches the end vertex D S , records the search path, randomly selects a new vertex as the access point of the ant, and repeats the ant colony search process until the iteration termination condition is reached, to obtain a number of better path results.

[0105] The better path results are input into the simulated annealing algorithm for global optimization convergence, and the final global optimal solution is obtained as the optimal tool set, which is as follows:

[0106] Input the tool sequences corresponding to X optimal paths into the simulated annealing algorithm, and for the obtained R sets of tool sequences {X1, X2, ... X... R The optimization iterations are performed sequentially, where for a given tool sequence X... i A new solution X is generated through probabilistic tool position variation. i ', from alternative tool set T C Randomly select a [blank] with a diameter of [blank]. The cutting tool, the replacement cutting tool sequence X i A tool at a certain position, D0 and D S In addition, a further global search is performed. After the calculation of R sets of tool sequences is completed, the optimal solution obtained is taken as the optimal tool set. The tool sequence optimization solution process is as follows: Figure 4 As shown.

[0107] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.< / e> < / ctp> < / e> < / ctp> ​

Claims

1. A method for selecting machining tools for cavity features based on process information mining, characterized in that, The method comprises the following steps: Step 1: connecting the centers of all the inscribed circles tangent to the feature boundary curve of the cavity to form a continuous curve and a straight line set, and defining the continuous curve and the straight line set as the medial axis of the cavity feature; Step 2: determining the maximum diameter and the minimum diameter of the tool according to the maximum diameter and the minimum diameter of the inscribed circle, and obtaining the allowable tool set according to the maximum diameter and the minimum diameter of the tool, and aiming at completing the overall rough machining of the feature area without interference; Step 3: determining the effective tool path according to the continuous curve or straight line formed by connecting the centers of the local inscribed circles, obtaining the number of tool retractions according to the number of effective tool paths, calculating the effective cutting area obtained by cutting along each effective tool path by using an OpenGL interface function, summing the effective cutting areas obtained by cutting along each effective tool path, and obtaining the overall effective cutting area of a single tool; Step 3 is specifically as follows: diameter of the local inscribed circle is not less than D q +2 δ number of local inscribed circles ω ( O k ( x k , y k ), d k )| d k ≥ D q +2 δ , 0≤ k ≤ K}, K number of local inscribed circles, d k diameter of the local inscribed circle, the center of which O k ( x k , y k )} are sequentially connected to form a continuous curve or a straight line, wherein D q diameter of the tool in the allowable tool set, δ roughing allowance; said continuous curve or straight line as a valid tool path for a tool having a concentrated diameter of D q L j ( D q ), 0≤ j ≤ Q , where Q is the number of valid tool paths generated during the tool cutting process.​ The effective cutting area of a tool with a diameter of 100 mm along an effective tool path of 1000 mm is calculated by the OpenGL interface function D q . The effective cutting area of a tool with a diameter of 100 mm along an effective tool path of 1000 mm is calculated by the OpenGL interface function L j ( D q ) 1000 mm. The effective cutting area of a tool with a diameter of 100 mm along an effective tool path of 1000 mm is calculated by the OpenGL interface function S j ( D q ) 1000 mm. The diameter of the tool is D q The overall effective cutting area of the tool is S ( D q ) is: In the formula, N is the number of retraction times; Step 4: obtaining the candidate tool set by intersecting the selectable tool set in the current machining center with the allowable tool set and adding the tool with the minimum diameter, and using a hybrid ant colony simulated annealing algorithm to optimize the candidate tool set to obtain the optimal tool set, aiming at maximizing the overall effective cutting area and minimizing the overall tool retraction times when the tools in the optimal tool set are selected in sequence for machining; Step 4 comprises: Step 4.1: The set of alternative tools available in the current machining center is A ( F ) C , resulting in the set of alternative tools T C = D S ( A ( F ) T ∩ A ( F ) C )={ D 1, D 2, …, D u ,…, D S}, where D S is the minimum tool diameter, A ( F ) T is the set of allowable tools, D u is the diameter of the u-th tool in the set of alternative tools; the tool diameters are ordered in decreasing order: D 1> D 2> …> D u …> D S Step 4.2: Establish the alternative tool set graph CTP, which has vertices representing each tool in the tool set and edges representing the diameter of each tool in the tool set S Abstract the vertices in the alternative tool set graph CTP as the tool set, and the edges as the diameter of each tool in the tool set D 0 represents the initial blank state of the cavity feature, and the corresponding tool diameter is +∞, and the remaining vertices represent the tool set T C The diameter of the machining tool in the tool set is represented by the formula wherein, V represents the alternative tool has to the graph CTP vertex, E represents the alternative tool has to the graph CTP directed edge, m is the alternative tool has to the graph CTP vertex number, weight shows the first selected diameter D u large tool and then select the diameter D v small tool, weight Dis ( D u , D v ) is the use of diameter D u large tool to complete the processing after the selection of the diameter D v small tool effective cutting area S ( D u→v ) and the number of retraction N ( D u→v ); Step 4.3: Normalization weight Dis ( D u , D v ) Step 4.4: In the alternative knife has to the graph CTP, with vertex D 0 as the starting point, and vertex D S as the end point, the path from the starting point to the end point corresponds to a feasible tool sequence, a number of better path results are obtained by searching through the ant colony algorithm, the better path results are input into the simulated annealing algorithm for global optimization convergence, and the final global optimal solution is obtained as the optimal tool set.

2. The method of claim 1, wherein the method is characterized by: Step 1 is specifically as follows: According to the central axis transformation, the coordinates of the centers of all the inscribed circles tangent to the feature boundary curve of the cavity are calculated O i ( x i , y i ) and the diameter d i , the centers of all the inscribed circles are sequentially connected to form a continuous curve and a straight line set, and the continuous curve and the straight line set are defined as the central axis of the cavity feature P , which is represented as: wherein, M is the number of all inscribed circles, O i ( x i , y i ) is the center coordinate of the i th inscribed circle among all inscribed circles, d i is the diameter of the i th inscribed circle, the center coordinate and the diameter of the inscribed circle are calculated by the central axis transformation equation.

3. The method of claim 1, wherein the method is characterized by: Step 2 is specifically as follows: calculating the maximum tool diameter D L : In the formula, int() is an integer function, d max is the maximum value of the diameter of the incircle, δ is a rough machining allowance; Calculating minimum tool diameter D S : wherein d min is the minimum value of the diameter of the incircle for the entire body The set of all feasible tools for a cavity feature is defined as the admissible tool set A ( F ) T is represented as: In the formula, D q D is the diameter of the tool to be allowed in the tool set.

4. The method of claim 1, wherein the method is characterized by: Step 4.3 is specifically as follows: Step 4.3.1: Calculate the weight of all directed edges Dis D u , D v} = { S D u→v ), N D u→v}, 0 ≦ u ≦ n -1, u v ≦ n , where { S D u→v} is the set of valid cutting areas, { N D u→v} is the set of retraction times, and n is the number of directed edges.​​​​​​ Step 4.3.2: Take the cutting area set { S ( D u→v The largest area in )} S max and the set of retraction times { N ( D u→v Maximum number of retractions in )} N max For the set of cutting areas { S ( D u→v The set of retraction counts { N ( D u→v The elements in )} are normalized as follows: wherein is the normalized effective cutting area, is the normalized number of retreats; Step 4.3.3: Define Normalization Weights As follows wherein, ω 1 and ω 2 represent the weight of the importance of the tool path and the number of tool retractions, respectively.

5. The method of claim 1, wherein, The ant colony algorithm is specifically as follows: Randomly select a vertex D u As an ant W Visit the vertex, and put the vertex D u Into the tabu list, while automatically calculating the next step vertex set allowed W ( D u ), according to the ant colony algorithm search rule, the ant selects to visit the vertex D v allowed W ( D u ), the vertex is put into the tabu list, and the next step vertex set allowed D v is calculated W ( D v ), the search is repeated until the ant reaches the terminal vertex D S , the search path is recorded, a new vertex is randomly selected as the ant visiting vertex, and the ant colony search process is repeated until the iteration termination condition is reached, and a plurality of optimal path results are obtained.

6. The method of claim 1, wherein the method further comprises: The input of the better path result into the simulated annealing algorithm for global optimization and convergence to obtain the final global optimal solution as the optimal tool set is specifically as follows: Will X The tool sequence corresponding to the optimal path is input into the simulated annealing algorithm, and the obtained tool sequence is then processed. R Tool sequence { X 1, X 2 ,…X R The optimization is iterated sequentially, with each iteration targeting a specific tool sequence. X r New solutions are generated through probabilistic tool position variation. X r ’ From the alternative tool set T C Randomly select a [blank] with a diameter of [blank]. D u ( D u X r ( ) cutting tools, replacement tool sequence X r A tool at a certain position in the middle, D 0 and D S In addition, perform a further global search. R After the tool sequence calculation is completed, the optimal solution obtained is taken as the optimal tool set.

Images