Part machining process route intelligent generation method

Abstract

The application discloses a kind of part machining process route intelligent generation method, comprising: obtaining part processing information;Part processing feature set is constructed, and form candidate processing process chain set;Adopt fuzzy comprehensive evaluation method, establish part feature processing process chain matching model, including constructing process chain matching evaluation index, establish processing process chain matching evaluation function, obtain matching processing process chain set;Processing step sequence cost comprehensive evaluation function is constructed, consider processing sequence constraint, and construct part processing process route planning model;Part processing process route is planned using discrete artificial seagull optimization algorithm, and optimized part processing process route is obtained;The method can realize the intelligent generation of part processing process route, improve process route generation efficiency, ensure the quality of process route generation.

Description

Technical Field

[0001] This invention relates to the field of parts machining technology, and specifically to a method for intelligently generating machining process routes for parts. Background Technology

[0002] Currently, mechanical products are becoming increasingly diverse in materials, richer in functions, and more complex in structure, which places higher demands on the decision-making of machining methods for mechanical parts. Existing methods mainly involve first extracting and classifying different local machining features of existing parts, establishing a corresponding feature machining method library for each category of machining features, then using feature recognition methods to determine the local structural features on the target part, searching for the corresponding machining methods in the library for the identified local features, and finally, further reasoning and decision-making by directly reusing the machining methods of the identified object or combining them with feature machining requirement constraints, and some progress has been made.

[0003] The sequencing of machining steps in process route planning is one of the key factors affecting the overall design level of the process route. Because its decision-making process is influenced by many factors, such as the diversity of part characteristics and machining methods, the experiential nature of process decisions, and the complexity of the production environment, the process of sequencing steps becomes extremely complex. Moreover, for many manufacturing enterprises, high-efficiency, low-cost, and high-quality production methods have a significant impact on their survival, competitiveness, and development; therefore, this is also a focus of scholarly research.

[0004] Existing research on process route generation mainly focuses on the sequencing of machining steps. There is limited attention paid to matching machining process chains to machining features; the few methods mentioned that do exist simply select the chain based on machining feasibility. Furthermore, research on the representation and generation methods of sequence constraints between steps is also scarce. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention proposes an intelligent method for generating machining process routes for parts. This method acquires part machining information, obtains a set of matching process chains by constructing a feature-based machining process chain matching algorithm based on fuzzy comprehensive evaluation, constructs a comprehensive cost evaluation function for machining step sequences, considers machining sequence constraints, and establishes a part machining process route planning model, and plans the part machining process route based on the discrete artificial seagull optimization algorithm, thus achieving intelligent generation of the part machining process route and obtaining an optimized part machining process route. Compared with traditional manual process route compilation, this invention aims to achieve intelligent generation of part machining process routes, improve the efficiency of machining process route generation, ensure the quality of generated machining process routes, and simultaneously reduce machining and time costs.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A method for intelligently generating machining process routes for parts includes the following steps:

[0008] S1: Based on the 3D model of the part, obtain the part's machining information, including the part's geometric shape information and the part's process requirements information;

[0009] S2: Based on the part's geometric shape information and part's process requirements information, construct a set of part processing features; based on the part processing features in the set of part processing features, form a set of candidate processing technology chains;

[0010] S3: Using the fuzzy comprehensive evaluation method, a matching model of the machining process chain for part features is established, including constructing a matching evaluation index for the process chain, establishing a matching evaluation function for the machining process chain, and obtaining a set of matching machining process chains.

[0011] S4: Summarize, uniformly number, and arrange the processing steps of all matching processing process chains in the matching processing process chain set in a specific order to form a processing step sequence. Take the processing order of each processing step in the processing step sequence as the decision variable and the lowest processing step sequence cost as the optimization objective. Construct a comprehensive evaluation function for processing step sequence cost, consider processing order constraints, and establish a part processing process route planning model.

[0012] S5: Based on the machining process route planning model, the discrete artificial seagull optimization algorithm is used to plan the part machining process route. A discrete strategy is added to the seagull optimization algorithm, and two mutation strategies, Sine mapping strategy and nearest neighbor learning strategy, are added within the seagull population to obtain the optimized part machining process route.

[0013] Furthermore, each part to be processed includes several processing features, constituting a set of part processing features, as shown in equation (1):

[0014] F = {f1, f2, ..., f n} (1)

[0015] In equation (1), f i This represents the i-th machining feature of the part to be processed, where n is the total number of machining features of the part to be processed, and f is the i-th machining feature of the part to be processed. i As shown in equation (2):

[0016] f i ={x i ,y i ,z i} (2)

[0017] In equation (2), x i y represents the machining accuracy of the i-th machining feature of a part, describing the degree to which the actual dimensions of the machined feature match the dimensions specified in the drawing. i z represents the surface roughness of the i-th machining feature of a part, describing the smoothness of the surface after machining. i It represents other processing requirements for the i-th processing feature of a part, and describes the requirements unique to certain processing features, such as the diameter of a hole feature, or whether a planar feature is an end face.

[0018] Furthermore, the established part feature machining process chain matching model adopts the fuzzy comprehensive evaluation method to evaluate and analyze the machining process chain matching process of a single machining feature, and selects a matching machining process chain from the set of candidate machining process chains for each machining feature of the part.

[0019] The specific steps for obtaining the matching processing technology chain set include:

[0020] Based on each processing feature in the set of processing features of the part to be processed, a corresponding set of candidate processing technology chains is constructed, as shown in equation (3):

[0021] PC i ={pc1,pc2,…,pc k ,…,pc m}, k=1,2,...,m (3)

[0022] In equation (3), PC i This indicates that the i-th machining feature f of the part to be machined is... i The constructed set of candidate processing technology chains, in which pc k Let k represent the k-th processing chain in the set of candidate processing chains, as shown in equation (4):

[0023]

[0024] In equation (4), o q Indicates pc k In the qth processing step, p k Indicates pc k The total number of processing steps;

[0025] Based on the set of part processing features of the parts to be processed, and according to the processing technology requirements, a process chain matching evaluation index is constructed, thereby establishing a processing process chain matching evaluation function. The process chain matching evaluation index includes processing feature-level evaluation index and material type evaluation index. The processing feature-level evaluation index includes processing accuracy, processing surface roughness and other processing requirements, and the material type evaluation index includes material type.

[0026] Among them, the processing feature level evaluation index is related to the processing features of the part. Based on the processing accuracy, surface roughness and other processing requirements in the processing feature level evaluation index, a set of processing feature evaluation factors is constructed, as shown in Equation (5):

[0027] U = {u1, u2, u3} (5)

[0028] In equation (5), u g (g=1,2,3) represents the influencing factors in the set of processing feature evaluation factors, where u1 represents processing accuracy, u2 represents the surface roughness of the processed surface, and u3 represents other processing requirements;

[0029] Construct membership functions for each influencing factor in the evaluation of processing characteristics, including membership functions for processing accuracy, surface roughness, and other processing requirements, and calculate the membership index values ​​for each influencing factor.

[0030] The membership function for the machining accuracy is shown in equation (6):

[0031]

[0032] In equation (6), x i f represents the i-th machining feature of the part to be machined. i Required machining accuracy level value, This represents the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k The membership function of machining accuracy. Indicates the process chain to be selected (pc) k The highest precision level that can be processed. Indicates the process chain to be selected (pc) k The lowest possible precision level that can be processed; the higher the precision level, the smaller the value.

[0033] The membership function of the surface roughness of the machined surface is shown in equation (7):

[0034]

[0035] In equation (7), y i f represents the i-th machining feature of the part to be machined. i Required surface roughness value for machining. This represents the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k The membership function of the surface roughness of the machined surface. Indicates the process chain to be selected (pc)k The maximum surface roughness value that can be machined. Indicates the process chain to be selected (pc) k The minimum surface roughness value that can be machined; the greater the surface roughness, the greater the roughness value.

[0036] Other processing requirements and membership functions are shown in equation (8):

[0037]

[0038] In equation (8), z i f represents the i-th machining feature of the part to be machined. i Other processing requirements, This represents the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k Other processing requirements membership function;

[0039] The material type evaluation index is used to evaluate the degree of matching between the material types that can be processed in the selected processing chain and the material types of the parts to be processed.

[0040] Based on the material type in the material type evaluation index, a material type membership function is constructed, as shown in equation (9):

[0041]

[0042] In equation (9), ma represents the type of material required for the machining of the part. This represents the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k The material type membership function, when the i-th machining feature f of the part to be machined... i The k-th candidate processing chain pc k When the matable material type includes ma, machining is feasible, and the membership function value is 1. This applies to the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k When the ma type is not a ma type, processing is not feasible, and the membership function value is 0.

[0043] Based on the membership functions of machining accuracy, surface roughness, other machining requirements, and material type, a weighted sum is performed as shown in equation (10) to construct a machining process chain matching evaluation function:

[0044]

[0045] In equation (10), This represents the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k The membership index value of machining accuracy This represents the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k The membership index value of the surface roughness of the machined surface. This represents the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k Other processing requirements membership index values, This represents the i-th machining feature f of the part to be machined. i The k-th candidate processing chain pc k Material type membership index values, w1, w2, w3, w4 represent The corresponding weight coefficients, and w1+w2+w3+w4=1;

[0046] For the set of candidate processing technology chains PC i All machining process chains are constructed using equation (10) to create a machining process chain matching evaluation function, and the corresponding set of machining process chain matching evaluation function values ​​{Score1,...,Score1,...,Score2,...,Score3,...,Score4,...,Score2 ... k ,...,Score m The elements in the set of processing technology chain matching evaluation function values ​​are sorted in descending order. The processing technology chain corresponding to the maximum value element in the set of processing technology chain matching evaluation function values ​​is the i-th processing feature f of the part to be processed. i Matching processing technology chain;

[0047] For other processing features in the set of processing features of the part to be processed, repeat the above steps of constructing the processing process chain matching evaluation function to obtain the matching processing process chains of all n processing features of the part to be processed, and form a set of matching processing process chains.

[0048] Furthermore, the processing steps of all matching processing chains in the set of matching processing chains are summarized and arranged in a specific order to form a processing step sequence, the processing step sequence AJS as shown in equation (11).

[0049] AJS={Job1|〈a1,b1〉,Job2|〈a2,b2〉,…,Job N |〈a N ,b N >} (11)

[0050] In equation (11), N represents the total number of steps in the processing step sequence AJS, Job h This represents the h-th machining step of the part, tuple <a h ,b h > indicates Job h The processing feature priority a of the corresponding process step h and processing step priority b h ;

[0051] During the part machining process, the machining features are divided into three categories: precision datum machining features, auxiliary machining features, and ordinary machining features. Precision datum machining features refer to the features that need to be machined first as a reference for the machining accuracy of other machining features. Auxiliary machining features refer to features that are attached to other machining features and need to be machined after the machining features to which they are attached have been completed. Ordinary machining features refer to features other than precision datum machining features and auxiliary machining features.

[0052] The processing feature priority a h This includes: each machining feature should be processed in the order of precision datum machining features, ordinary machining features, and additional machining features;

[0053] If the job is a work step h If it belongs to the precision datum machining feature, then its corresponding machining feature priority is a. h =1, if the job step h If it belongs to a normal processing feature, then its corresponding processing feature priority is a. h =2, if the job step h If it belongs to the category of additional processing features, then its corresponding processing feature priority a h =3;

[0054] The processing step priority b h This includes: for each machining feature, the machining steps should be performed in the order of roughing, semi-finishing, finishing, and super-finishing;

[0055] If the job is a work step h For rough machining, the priority of the corresponding machining step is b. h =1, if the job step h For semi-finishing, the priority of the corresponding machining step is b. h =2, if the job step h For finishing, the priority of the corresponding machining step is b. h =3, if the job step h For ultra-precision machining, the priority of the corresponding machining step is b. h =4.

[0056] The decision variables for the part machining process route planning problem are: the sequence of machining steps for the part to be machined, and each machining step (Job) in AJS. h The processing order is arranged as h = 1, 2, ..., N.

[0057] During part machining, given a fixed machining process and equipment, frequent changes to machine tools, cutting tools, fixtures, and machining methods will increase machining time and costs, and compromise machining accuracy. Therefore, the goal of sequencing the machining steps is to generate a sequence of steps with the fewest changes in machining equipment and methods while satisfying machining sequence constraints. This sequence serves as the optimized machining process route for the part. In part machining, when two machining steps use the same machining methods, machine tools, cutting tools, and fixtures, they should be arranged together for machining as much as possible to reduce changes in machining methods and equipment.

[0058] Furthermore, the comprehensive cost evaluation function for the processing step sequence is constructed as shown in equation (12):

[0059]

[0060] In Equation (12), MCC represents the machine tool replacement cost, FCC represents the machining fixture replacement cost, CCC represents the machining tool replacement cost, PCC represents the machining method replacement cost, and ω1, ω2, ω3, ω4 represent the weight coefficients corresponding to MCC, FCC, CCC, and PCC, respectively, and ω1+ω2+ω3+ω4=1.

[0061] During the machining process, frequent changes to machining equipment and methods, such as machine tools, cutting tools, and fixtures, will increase machining time and costs and make it difficult to guarantee machining accuracy. Therefore, the number of changes in machining equipment and methods should be minimized.

[0062] The processing order constraint of the processing step sequence in Equation (12) means that each processing step in the processing step sequence must satisfy the processing order constraint of the processing feature priority and the processing step priority.

[0063] The replacement cost (MCC) of the machine tool is shown in equation (13):

[0064]

[0065] In equation (13), mcc h Job for the h-th machining step of the part h The machine tool replacement cost, where N is the total number of steps included in the machining sequence;

[0066] mcch The calculation is shown in equation (14):

[0067]

[0068] In equation (14), mc h To process the h-th processing step, Job h The machine tools used at that time;

[0069] In process route planning, the more times the machining fixtures are changed during the part processing, the more time is consumed and the longer it takes to complete the sequence of machining steps for that part. Therefore, the number of times the machining fixtures are changed should be minimized.

[0070] The FCC's cost for replacing the machining fixture is shown in equation (15):

[0071]

[0072] In equation (15), fcc h Job for the h-th machining step of the part h The cost of changing the machining fixture is calculated as shown in equation (16):

[0073]

[0074] In equation (16), fc h To process the h-th processing step, Job h The machining fixtures used at that time;

[0075] In process route planning, the more times the machining tools are changed during the part machining process, the more time is consumed and the longer it takes to complete the machining process sequence of the part. Therefore, the number of times the machining tools are changed should be minimized.

[0076] The tool replacement cost CCC is shown in equation (17):

[0077]

[0078] In equation (17), ccc h Job for the h-th machining step of the part h The cost of replacing machining tools is calculated as shown in equation (18):

[0079]

[0080] In equation (18), cut h To process the h-th processing step, Job h The cutting tools used in the process;

[0081] In process route planning, the more times the processing method is changed during the part processing, the more time it consumes and the longer it takes to complete the sequence of processing steps for that part. Therefore, the number of processing method changes should be minimized.

[0082] The replacement cost PCC for the processing method is shown in equation (19):

[0083]

[0084] In equation (19), pcc h Job for the h-th machining step of the part h The cost of changing the processing method is calculated as shown in equation (20):

[0085]

[0086] In equation (20), pc h To process the h-th processing step, Job h The processing method used at that time.

[0087] Furthermore, the process route planning model is shown in equation (21):

[0088] minf(Seq)=mincost(Seq) (21)

[0089] In Equation (21), Seq represents the set of all candidate machining step sequences, and cost(Seq) represents the machining step sequence cost calculated according to the comprehensive evaluation function of machining step sequence cost. The machining process route planning model selects the machining step sequence with the minimum machining step sequence cost from the set of all candidate machining step sequences as the optimized part machining process route.

[0090] The Discrete Artificial Seagull Optimization Algorithm is a biologically inspired intelligent optimization algorithm, primarily inspired by the migration and aggression behaviors of seagulls in nature.

[0091] Furthermore, in the discrete artificial seagull optimization algorithm, seagull l is defined as a candidate processing step sequence X of the part to be processed. l (t), which is initialized as an N-dimensional vector, where N is the total number of processing steps in the processing step sequence. The vector representation of the seagull l is shown in equation (22):

[0092] X l (t)=(x l1 (t),x l2 (t),…,x lN (t)) (22)

[0093] In equation (22), x lj(t) represents the sequence of processing steps to be selected, X. l The processing steps arranged in a certain order in (t), t=0,1,2,…,iter max Iter represents the current evolution iteration number of the Discrete Artificial Seagull Optimization Algorithm. max This represents the maximum number of evolution iterations preset in the Discrete Artificial Seagull Optimization Algorithm.

[0094] Furthermore, the specific steps for obtaining the optimized part machining process route using the discrete artificial seagull optimization algorithm include:

[0095] S81: Parameter settings and population initialization for the discrete artificial seagull optimization algorithm, setting the population size p and the maximum number of evolution iterations iter. max Seagull attack curve parameters uc, vc, set X best Record the seagulls with the minimum processing step sequence cost that have appeared in the current evolutionary iterative search process;

[0096] S82: A Sine mapping strategy is used to generate a continuous-value encoded processing step sequence for an initial population size p of seagulls. This continuous-value encoded processing step sequence is then discretized using a discretization strategy to generate the processing step sequence for the initial generation t=0 evolutionary search seagull population. The processing step sequence X corresponding to each seagull in the initial population is then calculated sequentially. l The processing step sequence cost (X) of (t) l (t)), and the seagull with the minimum processing step sequence cost is taken as X. best ;

[0097] S83: Evolutionary iteration of the discrete artificial seagull optimization algorithm, specifically including:

[0098] S831: The migration behavior of the seagull population was investigated, and the migration behavior of the processing step sequence represented by seagull l was carried out through equations (23) to (27):

[0099] C l (t)=A l (t)×X l (t) (23)

[0100]

[0101] B l (t)=2×(A l (t)) 2 ×rd l (25)

[0102] M l (t)=B l (t)×(Xbest (t)-X l (t)) (26)

[0103] D l (t)=|C l (t)+M l (t)| (27)

[0104] In equations (23) to (27), C l (t) represents the sequence of processing steps represented by seagulls that do not collide with other seagulls, A l (t) represents the movement behavior of seagull l in the evolutionary search solution space, f c It is a linearly decaying function used to control A. l The shift frequency of (t) is typically set to 2, B. l (t) is a random movement behavior, where rd l It is a random number whose value is between [0,1]. l (t) represents the seagull X with the minimum processing step sequence cost in the current evolutionary iteration of the seagull population. best (t) to approach;

[0105] S832: Aggressive behavior against seagull populations, the aggressive behavior of the processing step sequence represented by seagull l is carried out through equations (28) to (32):

[0106] x' l (t)=r l (t)×cos(kc(t)) (28)

[0107] y' l (t)=r l (t)×sin(kc(t)) (29)

[0108] z' l (t)=rc l ×kc(t) (30)

[0109] r l (t)=uc×e kc(t)·vc ×I (31)

[0110] X l (t+1)=(D l (t)⊙x' l (t)⊙y' l (t)⊙z' l (t))+X best (t) (32)

[0111] In equations (28) to (32), x'l (t), y' l (t) and z' l (t) represents the seagull's attack behavior in the x, y, z dimensions. Since the seagull's attack proceeds in a spiral pattern, rc l Let be the radius of the spiral, kc(t)∈[0,2π] be a constant that defines the shape of the spiral, I be an N-dimensional vector with all elements having a value of 1, and uc and vc be the seagull attack curve parameters set in S81.

[0112] S833: Use a nearest neighbor learning strategy to update the seagull population after performing migration and aggression behaviors;

[0113] S84: Discretize the continuous-value encoded processing step sequence of the seagull population after updating it using the nearest neighbor learning strategy, and obtain the processing step sequence of the seagull population. Calculate the processing step sequence X corresponding to each seagull in the population one by one. l The cost of the processing step sequence (t+1) is cost(X). l (t+1));

[0114] S85: Cost(X) based on the processing step sequence cost for each seagull. l (t+1)), update the Seagull X with the minimum processing step sequence cost in the part machining process route planning process. best The specific update rule is as follows: if the seagull X in the current evolutionary iteration population has the lowest processing step sequence cost... best (t+1) compared to X best If the cost value of the processing step sequence is small, then let X... best =X best (t+1), otherwise do not update X. best ;

[0115] S86: Current evolution iteration number t < maximum evolution iteration number iter max Proceed to S83; otherwise, stop the part machining process route planning algorithm and output the machining step sequence X with the minimum machining step sequence cost. best As an optimization of the parts processing route.

[0116] Furthermore, the discrete strategy encodes the continuous value sequence X generated in the discrete artificial seagull optimization algorithm. l In (t), the element values ​​are arranged in ascending order. If the element values ​​are equal, the order is determined by the current element's position in the processing step sequence. Based on the element value sorting result, each processing step sequence X... lEach processing step element in (t) is assigned an integer index. This integer index is used to replace the original element value, thus generating a discrete processing step sequence. Each element x in this sequence... lj (t),j=1,2,...,N represents the part processing step number in the corresponding processing step sequence;

[0117] The Sine mapping is a chaotic mapping that can make the distribution of the initial seagull population generated by the discrete artificial seagull optimization algorithm more uniform, as shown in equation (33):

[0118] x l(v+1) (init) = μsin(πx) lv (init)).v=1,2,…,N-1; l=1,2,…,p; (33)

[0119] In equation (33), μ∈[0,4], is generally taken as 0.99, p is the population size of the seagulls, and x lv (init) represents the processing step sequence X represented by the randomly initialized seagull l. l The processing step number of the v-th processing step of (t) is encoded by a continuous value between [0,1]. l(v+1) (init) represents the processing step sequence X represented by the seagull l after Sine mapping. l The processing step number is the continuous value encoding of the vth processing step of (t);

[0120] The nearest neighbor learning strategy is shown in equations (34) to (37):

[0121] R l (t)=||X l (t+1)-X l (t)|| (34)

[0122] N l (t)={X ne (t)|||X l (t)-X ne (t)||≤R l (t)} (35)

[0123]

[0124]

[0125] In equations (34) to (37), R l (t) represents the nearest neighbor radius of the processing step sequence represented by the current l-th seagull, X. ne(t) represents the nearest neighbor sequence of the processing steps represented by the current l-th seagull, N. l (t) represents the set of nearest neighbor sequences of the processing step sequence represented by the current l-th seagull, rand(N) l (t) represents randomly selecting a nearest neighbor sequence from the set of nearest neighbor sequences, XNE l (t+1) represents the processing step sequence represented by the continuous value encoding of the l-th seagull candidate obtained from nearest neighbor sequence learning, X l (t+1) is the processing step sequence representing the continuous value encoding of the l-th seagull after being updated using the nearest neighbor learning strategy.

[0126] Compared with the prior art, the present invention has the following technical effects:

[0127] (1) Based on the processing requirements and considering the processing characteristics of the parts, a fuzzy evaluation method is adopted to establish a matching model of the processing process chain of the parts. By constructing the processing feature level evaluation index and the material type evaluation index, a processing process chain matching evaluation function is constructed to evaluate and analyze the processing process chain matching process of the processing features, and to match the optimal processing process chain for each processing feature of the parts. This method does not require complex manual intervention and improves the automation level of the parts processing process planning.

[0128] (2) According to the processing requirements, the matching processing process chains of each processing feature of the part to be processed are obtained to form a set of matching processing process chains and generate a sequence of processing steps to be selected. In order to optimize the sorting of processing steps in the sequence of processing steps to be selected, the replacement cost of processing equipment and processing methods required in the part processing process is considered. At the same time, under the constraint of processing sequence, a comprehensive evaluation function of processing step sequence cost is constructed. On this basis, a processing process route planning model is constructed to select the processing step sequence with the minimum processing step sequence cost from all the sets of processing step sequences to be selected as the optimized part processing process route.

[0129] (3) A discrete artificial seagull optimization algorithm is proposed and designed. A discrete strategy is added to the seagull optimization algorithm, and two mutation strategies, Sine mapping strategy and nearest neighbor learning strategy, are added to the seagull population evolution. The discrete artificial seagull optimization algorithm is applied to the part processing route planning problem. The discrete artificial seagull optimization algorithm can improve the population diversity and search efficiency of the population evolution iteration of the seagull algorithm, and improve the computational performance and efficiency of the algorithm to output the optimal or near-optimal optimized part processing route. Attached Figure Description

[0130] Figure 1 This is a flowchart of the present invention;

[0131] Figure 2A 3D model of the "double-disc bearing pipe" part;

[0132] Figure 3 A structural diagram of the process chain matching evaluation index for processing characteristics. Detailed Implementation

[0133] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the invention, any modifications, equivalent substitutions, improvements, etc., made by those skilled in the art to all other embodiments obtained without creative effort should be included within the protection scope of the present invention.

[0134] The core idea of ​​this invention is as follows: A set of candidate machining process chains is constructed based on the required machining features of the part. Then, a fuzzy comprehensive evaluation method is used to select a matching machining process chain for each machining feature of the part by constructing a matching evaluation index. This process selects matching machining process chains for all machining features of the part, forming a set of matching machining process chains for the required part. Further, the machining steps of all matched machining process chains are summarized into a machining step sequence. The processing order of each machining step in the sequence is used as the decision variable. Considering the cost of changing machining equipment and methods, as well as processing order constraints, a comprehensive evaluation function for the cost of the machining step sequence is constructed to establish a part machining process route planning model. Finally, a discrete artificial seagull optimization algorithm is used for iterative search to obtain the machining step sequence with the minimum cost as the optimized part machining process route. This achieves intelligent generation of the part machining process route, obtains a reasonable optimized order of machining steps in the process route, improves the efficiency of process route generation, reduces the processing time and cost, and ensures the quality and efficiency of the generated process route.

[0135] The following case study provides a detailed explanation of the algorithm, based on actual part processing requirements, models, and algorithms.

[0136] like Figure 1 As shown, a method for intelligently generating machining process routes for parts includes the following steps:

[0137] S1: Based on the 3D model of the part, obtain the part's machining information, including the part's geometric shape information and the part's process requirements information;

[0138] S2: Based on the part's geometric shape information and part's process requirements information, construct a set of part processing features; based on the part processing features in the set of part processing features, form a set of candidate processing technology chains;

[0139] S3: Using the fuzzy comprehensive evaluation method, a matching model of the machining process chain for part features is established, including constructing a matching evaluation index for the process chain, establishing a matching evaluation function for the machining process chain, and obtaining a set of matching machining process chains.

[0140] S4: Summarize, uniformly number, and arrange the processing steps of all matching processing process chains in the matching processing process chain set in a specific order to form a processing step sequence. Take the processing order of each processing step in the processing step sequence as the decision variable and the lowest processing step sequence cost as the optimization objective. Construct a comprehensive evaluation function for processing step sequence cost, consider processing order constraints, and establish a part processing process route planning model.

[0141] S5: Based on the machining process route planning model, the discrete artificial seagull optimization algorithm is used to plan the part machining process route. A discrete strategy is added to the seagull optimization algorithm, and two mutation strategies, Sine mapping strategy and nearest neighbor learning strategy, are added within the seagull population to obtain the optimized part machining process route.

[0142] like Figure 2 As shown, the part to be processed in this embodiment is a "double-disc bearing pipe". The steps and results for obtaining its processing feature information and the processing technology chain information of each processing feature are as follows:

[0143] like Figure 3 As shown, based on the set of part processing features of the parts to be processed, and according to the processing technology requirements, a process chain matching evaluation index is constructed, thereby establishing a processing process chain matching evaluation function. The process chain matching evaluation index includes processing feature-level evaluation index and material type evaluation index. The processing feature-level evaluation index includes processing accuracy, processing surface roughness and other processing requirements, and the material type evaluation index includes material type.

[0144] Refinement of S2:

[0145] For the "double-disc bearing pipe" part, the part's machining feature information is first obtained, as shown in Table 1.

[0146] Table 1. Set of machining features for the "double-disc bearing tube" part.

[0147]

[0148] For each machining feature of the "double-disc bearing pipe" part in Table 1, the corresponding set of candidate machining process chains summarized by process engineers can be obtained, including Tables 2 and 3. Table 2 shows the set of candidate machining process chains corresponding to the outer cylindrical surface feature, and Table 3 shows the set of candidate machining process chains corresponding to the hole feature.

[0149] Table 2. Information on the set of candidate machining processes for outer cylindrical surface features.

[0150]

[0151] Table 3. Information on the set of candidate processing technology chains for hole features.

[0152]

[0153]

[0154] Refinement of S3:

[0155] Based on the machining features of the outer cylindrical surface and holes of the "double-disc bearing pipe" part, the set of candidate machining process chains shown in Tables 2 and 3 are formed respectively. Then, the fuzzy comprehensive evaluation method is used to match the optimal machining process chain for each machining feature of the "double-disc bearing pipe" part.

[0156] Taking the outer circular surface feature “F06” as an example, denoted as machining feature f6, it is represented as:

[0157] f6 = {x6, y6, z6}

[0158] Where x6 represents the machining accuracy of machining feature f6, and x6 = 8; y6 represents the surface roughness of machining feature f6, and y6 = 1.2 is preferred; z6 represents other machining requirements of machining feature f6, and z6 = none is preferred, indicating that there are no other machining requirements; in addition, the material type m used for machining feature f6 is low carbon cast steel.

[0159] f6 corresponds to the set of 10 candidate processing technology chains PC6 shown in Table 2, represented as:

[0160] PC6 = {pc1, pc2, ..., pc} 10} = {101, 102, ..., 110}, where 101, 102, ..., 110 are the numbers of the candidate processing chain.

[0161] The calculation of the matching evaluation function value of the first candidate machining process chain pc1 = 101 with machining feature f6 is illustrated by example.

[0162] Substituting the machining accuracy x6 = 8, surface roughness y6 = 1.2, other machining requirements z6 = none, and m being low-carbon cast steel, into the membership function calculation formulas for each influencing factor in the machining characteristic evaluation factor set, including the machining accuracy membership function... Surface roughness membership function Other processing requirements membership function and material type membership function The membership function values ​​of processing feature f6 for the first candidate processing chain pc1 = 101 were calculated as follows:

[0163] Machining accuracy membership function: Surface roughness membership function: Other processing requirements membership function: Material type membership function:

[0164] Using the same method, the membership function values ​​corresponding to the 10 candidate machining process chains of machining feature f6 are shown in Table 4:

[0165] Table 4 shows the membership function values ​​corresponding to the 10 candidate process chains for processing feature f6.

[0166]

[0167] Based on the membership function values ​​of each influencing factor in the set of evaluation factors for processing features shown in Table 4, a processing process chain matching evaluation function is constructed for processing feature f6 to determine its candidate processing process chain:

[0168]

[0169] For the current calculation instance, w1 = w2 = w3 = w4 = 0.25;

[0170] The calculation results of the Score of the machining process chain matching evaluation function for each candidate machining process chain of machining feature f6 are shown in Table 5:

[0171] Table 5. Evaluation function values ​​for matching processing technology chains for each candidate processing technology chain.

[0172]

[0173] From the 10 candidate machining process chains, select the machining process chain with the largest matching evaluation function value as the matching machining process chain for machining feature f6, namely machining process chain 103 "rough turning-semi-finish turning-finish turning".

[0174] Using the above method, matching machining process chains can be selected for other machining features of the "double-disc bearing pipe" part, and the set of matching machining process chains for the part can be obtained. The results are shown in Table 6.

[0175] Table 6. Set of matching machining process chains for parts

[0176]

[0177]

[0178] Refinement of S4:

[0179] A part machining process route planning model is established to prepare for the machining process route planning of the "dual-disc support pipe" part. This includes summarizing, uniformly numbering, and arranging the machining steps of all matching machining process chains in the part's matching machining process chain set into a machining step sequence AJS, and matching the required machining equipment (machine tools, cutting tools, fixtures) and machining methods for the machining features of each machining step in the machining step sequence AJS. Based on the machine tool change cost, fixture change cost, cutting tool change cost, machining method change cost, and machining sequence constraints, the comprehensive cost evaluation function of the machining step sequence AJS is calculated. Using the machining sequence arrangement of each machining step in the machining step sequence as the decision variable, a part machining process route planning model is constructed, and the machining step sequence with the minimum cost is selected from all candidate machining step sequence sets as the optimized part machining process route.

[0180] First, a machining step sequence AJS is constructed. For all matching machining process chains of machining features F01-F20 in Table 6, the machining steps are summarized, uniformly numbered, and arranged in a specific order to form the machining step sequence AJS. The machining step sequence AJS is represented as follows:

[0181] AJS={Job1|〈a1,b1〉,Job2|〈a2,b2〉,...,Job h |〈a h ,b h >,...,Job 68 |〈a 68 ,b 68 >}

[0182] Among them, each processing step Job h The attribute values ​​for h = 1, 2, ..., 68 are shown in Table 7:

[0183] Table 7 Attribute values ​​for each machining step

[0184]

[0185]

[0186]

[0187] Each processing step Job h The machining equipment (machine tools, cutting tools, and fixtures) and machining methods used for h = 1, 2, ..., 68 are shown in Table 8.

[0188] Table 8. Machining equipment and methods used in each machining step.

[0189]

[0190]

[0191]

[0192]

[0193] The following calculates the comprehensive cost evaluation function of the processing step sequence, taking processing step sequence AJS1 as an example;

[0194] First, obtain the machine tool number sequence {mc} corresponding to the machining step sequence AJS1. h}、Machining fixture number sequence {fc h}、Machining tool number sequence {cut h}、Processing method number sequence {pc h}, calculate the machine tool replacement cost sequence {mcc h}、Care fixture replacement cost sequence {fcc h}、CNC machining tool replacement cost sequence {ccc h} and processing method change cost sequence {pcc h}, as shown in Tables 9-17:

[0195] Table 9 Processing Step Sequence AJS1

[0196]

[0197] Table 10 Machining Step Sequence AJS1 Machine Tool Number Sequence {mc h}

[0198]

[0199]

[0200] Table 11 Machine tool replacement cost sequence for machining step sequence AJS1 {mcc h}

[0201]

[0202] Table 12 Machining Fixture Number Sequence AJS1 for Machining Step Sequence {fc h}

[0203]

[0204] Table 13 Machining Fixture Change Cost Sequence for Machining Step Sequence AJS1 {fcc h}

[0205]

[0206]

[0207] Table 14 Machining Step Sequence AJS1 Machining Tool Number Sequence {cut h}

[0208]

[0209] Table 15 Machining Step Sequence AJS1 Tool Change Cost Sequence {ccc h}

[0210]

[0211] Table 16 Machining Step Sequence AJS1 Machining Method Number Sequence {pc h}

[0212]

[0213] Table 17 Machining Step Sequence AJS1 Machining Method Replacement Cost Sequence {pcc h}

[0214]

[0215] Based on the data in each table, calculate the relevant values ​​such as the machining equipment replacement cost and machining method replacement cost for machining step sequence AJS1, including: machining machine tool replacement cost MCC = 2, machining fixture replacement cost FCC = 2, machining tool replacement cost CCC = 34, and machining method replacement cost PCC = 49.

[0216] Taking ω1 = 0.5, ω2 = 0.2, ω3 = 0.2 and ω4 = 0.1, and combining the processing sequence constraints, the comprehensive cost evaluation function of the processing step sequence AJS1, Cost = 13.1, can be calculated.

[0217] Based on the comprehensive evaluation function of processing step sequence cost, and taking the processing order of each processing step in the processing step sequence as the decision variable, a part processing process route planning model is constructed. From all the candidate processing step sequence sets, the processing step sequence with the minimum processing step sequence cost is selected as the optimized part processing process route.

[0218] S5 refinement:

[0219] A part machining process route planning method based on discrete artificial seagull optimization algorithm is adopted. Taking the machining process route planning of the part "double disk bearing pipe" as an example, the calculation process is as follows.

[0220] The initialization parameter settings in the discrete artificial seagull optimization algorithm include: setting the seagull population size p=50, and the maximum number of evolution iterations itermax =100, Seagull attack curve parameters uc=1, vc=1.

[0221] The processing steps for generating continuous value encodings for initializing the seagull population were generated using a Sine mapping strategy, as shown in Table 18:

[0222] Table 18. Processing steps sequence for initializing continuous value encoding of the seagull population generated by the Sine mapping strategy.

[0223]

[0224]

[0225]

[0226] The processing steps sequence of the continuous-value encoding for initializing the seagull population was discretized using a discretization strategy to generate the processing steps sequence of the first-generation t=0 evolutionary search seagull population, as shown in Table 19:

[0227] Table 19 shows the processing steps sequence corresponding to the discretized seagull population.

[0228]

[0229]

[0230]

[0231] Through continuous evolution and iteration of the artificially bred seagull population, each generation of the seagull population updates its migration and aggression behaviors using a nearest neighbor learning strategy; when the maximum number of evolutionary iterations is reached, iter max After the value reaches 100, the evolutionary iterative planning algorithm stops, and the processing step sequence with the minimum processing step sequence cost is output as the optimized part processing technology route.

[0232] Table 20 shows the processing steps sequence corresponding to the seagull population after the 100th generation of evolution:

[0233] Table 20 shows the processing steps sequence corresponding to the seagull population after the 100th generation of evolution.

[0234]

[0235]

[0236]

[0237]

[0238] Finally, output the machining step sequence X with the minimum machining step sequence cost. best Table 21 shows the optimized part machining process route, Optimized Part Machining Process Route X best The corresponding processing step sequence cost cost(X) best =10.6.

[0239] Table 21 shows the optimized part machining process route X output by the planning algorithm. best

[0240]

[0241] The above example of machining process route planning for the "double-disc support pipe" part specifically includes: First, obtaining all machining features of the "double-disc support pipe" part, such as annular planes, outer circular surfaces, and holes; Second, taking the outer circular surface feature and hole feature as examples, constructing their corresponding sets of candidate machining process chains; Third, taking the outer circular surface feature "F06" as an example, illustrating the use of fuzzy comprehensive evaluation method to construct a machining process chain matching evaluation function, calculating the process chain matching evaluation function value of the candidate machining process chain, and selecting matching machining process chains for this feature; selecting matching machining process chains for other machining features of the "double-disc support pipe" part respectively, obtaining a set of matching machining process chains for the part; Then, establishing a machining process route planning model for the part, including summarizing, uniformly numbering, and arranging the machining steps of all matching machining process chains in the set of matching machining process chains to form a machining step sequence, and providing... The machining process involves matching the required machining equipment (machine tools, cutting tools, and fixtures) and machining methods for each machining step in the machining sequence. Based on the machine tool replacement cost, fixture replacement cost, cutting tool replacement cost, and machining method replacement cost, as well as machining sequence constraints, a comprehensive cost evaluation function for the machining step sequence is calculated. Using the machining sequence of each machining step as the decision variable, a part machining process route planning model is constructed. The machining step sequence with the lowest cost is selected from all candidate machining step sequences as the optimized part machining process route. Finally, a part machining process route planning method based on the discrete artificial seagull optimization algorithm is used to plan the machining process route for the "double-disc bearing pipe" part, obtaining the machining step sequence with the lowest cost, which serves as the optimized part machining process route for the "double-disc bearing pipe" part. Further details are omitted.

[0242] Similarly, the method provided in this article can be used to plan the machining process routes for other mechanical parts; specific examples will not be given here.

[0243] In summary, the method provided by this invention is feasible and practical. It can effectively achieve the intelligent generation of the proposed machining process route for parts, obtain a reasonable optimized sorting of machining steps in the machining process route, improve the efficiency of machining process route generation, reduce the machining time and cost of the machining process route, and ensure the quality and efficiency of the generated machining process route.

Claims

1. A method for intelligently generating machining process routes for parts, characterized in that, The intelligent generation method for the process route includes the following steps: S1: Based on the 3D model of the part, obtain the part's machining information, including the part's geometric shape information and the part's process requirements information; S2: Based on the part's geometric shape information and part's process requirements information, construct a set of part processing features; based on the part processing features in the set of part processing features, form a set of candidate processing technology chains; S3: Using the fuzzy comprehensive evaluation method, a matching model of the machining process chain for part features is established, including constructing a matching evaluation index for the process chain, establishing a matching evaluation function for the machining process chain, and obtaining a set of matching machining process chains. S31: Construct a set of candidate machining process chains for each machining feature in the set of machining features of the part to be processed; S32: Based on the set of part processing features of the parts to be processed, construct process chain matching evaluation index according to the processing technology requirements, thereby establishing a processing process chain matching evaluation function. The process chain matching evaluation index includes processing feature-level evaluation index and material type evaluation index. The processing feature-level evaluation index includes processing accuracy, processing surface roughness and other processing requirements. The material type evaluation index includes material type. S33: Based on the machining accuracy, surface roughness and other machining requirements in the machining feature-level evaluation indicators, construct a set of machining feature evaluation factors; S34: Construct membership functions for each influencing factor in the evaluation of processing characteristics, including membership functions for processing accuracy, surface roughness, and other processing requirements, and calculate the membership index values ​​for each influencing factor. S35: Construct a material type membership function based on the material type in the material type evaluation index; S36: Based on the aforementioned machining accuracy membership function, machining surface roughness membership function, other machining requirement membership function, and material type membership function, construct a machining process chain matching evaluation function: S37: For the set of candidate processing technology chains All processing technology chains are constructed according to the method in S36 to construct the processing technology chain matching evaluation function, and the corresponding processing technology chain matching evaluation function value set is obtained. S38: Repeat S31~S37 to obtain the required parts for processing. A set of matching processing technology chains is formed by matching processing features; S4: Summarize, uniformly number, and arrange the processing steps of all matching processing process chains in the matching processing process chain set in a specific order to form a processing step sequence. Take the processing order of each processing step in the processing step sequence as the decision variable and the lowest processing step sequence cost as the optimization objective. Construct a comprehensive evaluation function for processing step sequence cost, consider processing order constraints, and establish a part processing process route planning model. S5: Based on the processing route planning model, the discrete artificial seagull optimization algorithm is used to plan the part processing route. A discrete strategy is added to the seagull optimization algorithm, and two mutation strategies, Sine mapping strategy and nearest neighbor learning strategy, are added to the seagull population to obtain the optimized part processing route. The discrete artificial seagull optimization algorithm in the seagull Defined as a sequence of potential machining steps for the part to be machined. It is initialized to a A 3D vector, where N is the total number of machining steps in the machining step sequence, and the seagull... The vector representation of is shown in equation (1): ； （1） In equation (1), Indicates the sequence of processing steps to be selected. The processing steps are arranged in a certain order, where j=1,2…N. , This represents the current evolution iteration number of the Discrete Artificial Seagull Optimization Algorithm. This represents the maximum number of evolution iterations preset in the Discrete Artificial Seagull Optimization Algorithm; The specific steps for obtaining the optimized part machining process route using the Discrete Artificial Seagull Optimization Algorithm include: S51: Parameter settings and population initialization for the discrete artificial seagull optimization algorithm; setting the population size. Maximum number of evolution iterations Seagull attack curve parameters , ,set up Record the seagulls with the minimum processing step sequence cost that have appeared in the current evolutionary iterative search process; S52: Use the Sine mapping strategy to generate the initial population size. The continuous value encoding processing steps sequence of the seagull population is initialized by discretizing the continuous value encoding processing steps sequence of the seagull population using a discretization strategy to generate the first generation. Evolutionary search is performed to determine the processing steps sequence of a seagull population, and the processing steps sequence corresponding to each seagull in the initial population is calculated one by one. Processing step sequence cost And using the seagull with the lowest processing step sequence cost as... ; S53: Iterative evolution of the discrete artificial seagull optimization algorithm, specifically including: S531: Monitoring migratory behavior of seagull populations, seagulls The processing steps sequence represented The migration behavior is carried out through equations (2) to (6): ； （2） ； （3） ； （4） ； （5） ； （6） In equations (2) to (6), This represents the sequence of processing steps for seagulls that do not collide with other seagulls. Represents seagulls Movement behavior in the evolutionary search solution space It is a linearly decaying function used to control The mobile frequency, It is a random movement behavior, in which It is a random number, and its value is in between, Represents seagulls The seagull with the minimum processing step sequence cost in the current evolutionary iteration of the seagull population. near; S532: Aggressive behavior against seagull populations, seagulls The attack behavior of the processing step sequence represented is carried out through equations (7) to (11): ； （7） ； （8） ； （9） ； （10） ； （11） In equations (7) to (11), , and It's a seagull. The dimensional attack behavior, because the seagull's attack proceeds in a spiral pattern, therefore... Let be the radius of the spiral. It is a constant that defines the shape of a spiral. It is an N-dimensional vector whose elements are all 1. , These are the seagull attack curve parameters set in S51; S533: Use a nearest neighbor learning strategy to update the seagull population after performing migration and aggression behaviors; S54: Discretize the continuous-value encoded processing step sequence of the seagull population after updating it using the nearest neighbor learning strategy, and obtain the processing step sequence of the seagull population. Calculate the processing step sequence corresponding to each seagull in the population one by one. Processing step sequence cost ; S55: Cost based on the processing steps sequence for each seagull. The seagull with the minimum machining step sequence cost during the update of the part machining process route planning. The specific update rule is: if the seagull in the current evolutionary iteration has the seagull with the lowest processing step sequence cost... Compare If the cost value of the processing step sequence is small, then let Otherwise, no update. ; S56: Current evolution iteration number Maximum number of evolution iterations Proceed to S53; otherwise, stop the part machining process route planning algorithm and output the machining step sequence with the minimum machining step sequence cost. As an optimization of the parts processing route.

2. The intelligent generation method for machining process routes of parts according to claim 1, characterized in that, Each part to be processed includes several processing features, which constitute a set of part processing features, as shown in equation (12): ； （12） In equation (12), Indicates the number of parts to be processed. One processing feature, The total number of machining features required for the part to be machined, and the number of machining features required for the part. Processing features As shown in equation (13): ； （13） In equation (13), Indicates the part number The machining accuracy of each machining feature describes the degree to which the actual dimensions of the machined feature match the dimensions specified in the drawing. Indicates the part number The surface roughness of a machining feature describes the smoothness of the surface after machining. Indicates the part number Other processing requirements for a processing feature, describing the specific requirements of certain processing features, including the aperture size of a hole feature and whether a planar feature is an end face.

3. The intelligent generation method for machining process routes of parts according to claim 2, characterized in that, The set of candidate processing technology chains is shown in equation (14): ； （14） In equation (14), Indicates the number of parts to be processed. Processing features The constructed set of candidate processing technology chains, among which Indicates the first step in the set of candidate processing technology chains. The processing chain is shown in equation (15): ； （15） In equation (15), express The Middle Each processing step express The total number of processing steps; The set of evaluation factors for processing characteristics is shown in equation (16): ； （16） In equation (16), where Indicates machining accuracy. Indicates the surface roughness of the machined surface. Indicates other processing requirements; The membership function for the machining accuracy is shown in equation (17): ； （17） In equation (17), Indicates the number of parts to be processed. Processing features Required machining accuracy level value, Indicates the number of parts to be processed. Processing features The first Selectable processing technology chain The membership function of machining accuracy. Indicates the process chain to be selected The highest precision level that can be processed. Indicates the process chain to be selected The lowest possible precision level that can be processed; the higher the precision level, the smaller the value. ; The membership function of the surface roughness of the machined surface is shown in equation (18): ； （18） In equation (18), Indicates the number of parts to be processed. Processing features Required surface roughness value for machining. Indicates the number of parts to be processed. Processing features The first Selectable processing technology chain The membership function of the surface roughness of the machined surface. Indicates the process chain to be selected The maximum surface roughness value that can be machined. Indicates the process chain to be selected The minimum surface roughness value that can be machined; the greater the surface roughness, the greater the roughness value. ; Other processing requirements and membership functions are shown in equation (19): ； （19） In equation (19), Indicates the number of parts to be processed. Processing features Other processing requirements, Indicates the number of parts to be processed. Processing features The first Selectable processing technology chain Other processing requirements membership function; The membership function of the material type is shown in equation (20): ； （20） In equation (20), Indicates the type of material required for the machining of the part. Indicates the number of parts to be processed. Processing features The first Selectable processing technology chain Material type membership function, when for the part to be processed Processing features The first Selectable processing technology chain The types of materials that can be processed include When processing is feasible, the membership function value is 1. When the part to be processed is... Processing features The first Selectable processing technology chain The types of materials that can be processed do not include When processing is not feasible, the membership function value is 0; the processing technology chain matching evaluation function is as shown in equation (21): ； （21） In equation (21), Indicates the number of parts to be processed. Processing features The first Selectable processing technology chain The membership index value of machining accuracy Indicates the number of parts to be processed. Processing features The first Selectable processing technology chain The membership index value of the surface roughness of the machined surface. Indicates the number of parts to be processed. Processing features The first Selectable processing technology chain Other processing requirements membership index values, This indicates the number of parts to be processed. Processing features The first Selectable processing technology chain Material type membership index value, express The corresponding weighting coefficients, and ; Obtain the set of corresponding processing technology chain matching evaluation function values. The elements in the set of processing technology chain matching evaluation function values ​​are sorted in descending order. The processing technology chain corresponding to the maximum value element in the set of processing technology chain matching evaluation function values ​​is the processing technology chain of the part to be processed. Processing features Matching processing technology chain.

4. The intelligent generation method for machining process routes of parts according to claim 3, characterized in that, The machining steps of all matching machining process chains in the set of matching machining process chains are summarized, uniformly numbered, and arranged in a specific order to form a machining step sequence. As shown in equation (22), ； （22） In equation (22), Represents the sequence of processing steps The total number of work steps This indicates the first machining step of the part. The part machining step number corresponding to each machining step, a binary tuple. express The priority of the machining features of the corresponding process step and processing step priority Where h = 1, 2, ..., N; During the part machining process, the machining features are divided into three categories: precision datum machining features, auxiliary machining features, and ordinary machining features. Precision datum machining features refer to the features that need to be machined first as a reference for the machining accuracy of other machining features. Auxiliary machining features refer to features attached to other machining features. Auxiliary machining features need to be machined after the machining features to which they are attached have been completed. Ordinary machining features refer to machining features other than precision datum machining features and auxiliary machining features. The processing feature priority This includes: each machining feature should be processed in the order of precision datum machining features, ordinary machining features, and additional machining features; If the process steps If a feature belongs to the precision datum machining feature, then its corresponding machining feature priority is... If the work steps If it belongs to a normal processing feature, then its corresponding processing feature priority is... If the work steps If it belongs to the category of additional processing features, then the priority of its corresponding processing features is... ; The priority of the processing steps This includes: for each machining feature, the machining steps should be performed in the order of roughing, semi-finishing, finishing, and super-finishing; If the process steps For rough machining, the priority of its corresponding machining step is... If the work steps If it is a semi-finishing process, then the priority of its corresponding machining step is... If the work steps For finishing, the priority of the corresponding machining step is... If the work steps For ultra-precision machining, the priority of the corresponding machining steps is... .

5. The intelligent generation method for machining process routes of parts according to claim 4, characterized in that, The comprehensive cost evaluation function for the processing step sequence is constructed as shown in equation (23): ； （23） In equation (23), This indicates the cost of replacing machine tools. This indicates the cost of replacing machining fixtures. This indicates the cost of replacing machining tools. Indicates the cost of changing processing methods. They represent The corresponding weighting coefficients, and have During the machining process, frequent changes of machine tools, cutting tools, fixtures, and machining methods will increase machining time and costs, and will not be conducive to ensuring machining accuracy. Therefore, the number of changes of machining equipment and machining methods should be minimized. The processing order constraint of the processing step sequence mentioned in Equation (23) means that each processing step in the processing step sequence must satisfy the processing order constraint of the processing feature priority and the processing step priority. The cost of replacing the machine tool As shown in equation (24): ； （24） In equation (24), For the part Each processing step The cost of replacing machine tools This represents the total number of steps included in the processing sequence. The calculation is shown in equation (25): ； （25） In equation (25), For processing the first Each processing step The machine tools used at that time; The cost of replacing the machining fixture As shown in equation (26): ； （26） In equation (26), For the part Each processing step The cost of changing the machining fixture is calculated as shown in equation (27): ； （27） In equation (27), For processing the first Each processing step The machining fixtures used at that time; The cost of replacing machining tools As shown in equation (28): ； （28） In equation (28), For the part Each processing step The cost of replacing machining tools is calculated as shown in equation (29): ； （29） In equation (29), For processing the first Each processing step The cutting tools used in the process; Cost of changing the processing method As shown in equation (30): ； （30） In equation (30), For the part Each processing step The cost of changing the processing method is calculated as shown in equation (31): ； （31） In equation (31), For processing the first Each processing step The processing method used at that time.

6. The intelligent generation method for machining process routes of parts according to claim 5, characterized in that, The process route planning model is shown in equation (32): ； （32） In equation (32), This represents the set of all possible processing step sequences. This represents the cost of the machining step sequence calculated according to the comprehensive evaluation function of the machining step sequence cost. The machining process route planning model selects the machining step sequence with the minimum cost from all candidate machining step sequences as the optimized part machining process route.

7. The intelligent generation method for machining process routes of parts according to claim 6, characterized in that, The discrete strategy is used to encode the processing steps sequence of continuous values ​​generated in the discrete artificial seagull optimization algorithm. The element values ​​are arranged in ascending order. If the element values ​​are equal, the order is determined by the current element's position in the processing step sequence. Based on the element value sorting result, each processing step sequence... Each processing step element is assigned an integer index, which is used to replace the original element value, thus generating a discrete processing step sequence. Each element in this sequence... This indicates the part machining step number in the corresponding machining step sequence; The Sine mapping is a chaotic mapping that can make the distribution of the initial seagull population generated by the discrete artificial seagull optimization algorithm more uniform, as shown in equation (33): ； （33） In equation (33), , For the population size of seagulls, Seagulls generated for random initialization The processing steps sequence represented The The values ​​of each processing step are in Processing step numbers with consecutive value codes between them. Seagulls mapped by Sine The processing steps sequence represented The The processing step number is a continuous value code for each processing step; The nearest neighbor learning strategy is shown in equations (34) to (37): ； （34） ； （35） ； （36） ； （37） In equations (34) to (37), Indicates the current number The nearest neighbor radius of the processing step sequence represented by each seagull, which is a continuous value encoding. Indicates the current number l The nearest neighbor sequence of the continuous value encoding of the processing steps represented by each seagull. Indicates the current number The set of nearest neighbor sequences for the continuous value encoded processing steps represented by each seagull. This means randomly selecting a nearest neighbor sequence from the set of nearest neighbor sequences. This represents the candidate sequence obtained from nearest neighbor sequence learning. The processing steps sequence represented by each seagull, with continuous value encoding. The updated version using the nearest neighbor learning strategy The processing steps sequence represented by each seagull is a continuous value encoding.

Images