A tap parameter optimization method

Abstract

A tap parameter optimization method relates to the technical field of tap processing. The method solves the problem of unreasonable design of key parameters in traditional tap design. The method comprises the following steps: S1: determining the value range of the geometric parameters of the model of the tap to be optimized, and performing data processing on the geometric parameters; S2: obtaining a fitting equation of the geometric parameters after data processing in the tap model and corresponding generated performance parameters through orthogonal test; S3: performing weight distribution on the performance parameters generated by the geometric parameters after data processing; and S4: optimizing the geometric parameters after data processing and the performance parameters after weight distribution through a multi-island genetic algorithm. The tap parameter optimization method, device, medium, product and apparatus based on the multi-island genetic algorithm optimize the geometric parameters of the tap, and improve the performance of the tap in the actual processing process.

Description

Technical Field

[0001] This invention relates to the field of tapping technology, specifically to parameter optimization technology in tapping. Background Technology

[0002] In tap design, the rake angle, clearance angle, and number of cutting teeth are key parameters. An inappropriate rake angle can weaken the cutting edge, increasing the risk of tool breakage and affecting cutting efficiency and machining quality. An excessively large rake angle can cause vibration and reduce machining accuracy; a too-small rake angle increases cutting friction, accelerates tool wear, and may even lead to cutting difficulties. A too-small clearance angle increases friction between the cutting edge and the machined surface of the workpiece, leading to increased cutting force and temperature, accelerating tool wear, and even damaging the cutting edge. While a too-large clearance angle can reduce friction, it may weaken the cutting edge and reduce durability. Too many cutting teeth increase cutting resistance and load, leading to increased cutting force and temperature, accelerating tap wear, and even breakage, while also affecting thread surface quality and accuracy. Too few cutting teeth may increase the amount of material removed per tooth, exacerbating cutting edge wear, reducing durability, and resulting in an unstable cutting process.

[0003] Tap machining involves nonlinear factors and multi-objective optimization problems, such as reducing cutting forces, reducing tool wear, and increasing tool life. Traditional tap design relies on engineering experience and experimental data, which is not only time-consuming and costly but also makes it difficult to obtain a globally optimal solution. If the parameter design is unreasonable, the system or product performance will fail to meet expectations, increasing energy consumption, reducing work efficiency, and requiring additional resources to correct design flaws. This further increases production costs. Furthermore, product reliability and stability will be affected, leading to shorter service life, increased subsequent maintenance costs, and even safety hazards. Summary of the Invention

[0004] This invention aims to address the irrationality of key parameter design in traditional tap design, and proposes a method for tap parameter optimization.

[0005] The technical solution adopted by the present invention to solve the above-mentioned technical problems includes the following steps:

[0006] S1: Determine the range of values ​​for the geometric parameters of the model of the tap to be optimized, and perform data processing on the geometric parameters, including the tap rake angle γ. p Tap back angle α p and the number of teeth N in the chip cone;

[0007] S2: Obtain the fitting equations of the geometric parameters and corresponding performance parameters in the above tap model after data processing through orthogonal experiments. The corresponding performance parameters include cutting force F, cutting temperature T, and tool wear rate μ.

[0008] S3: Perform a comprehensive analysis and weight allocation on the performance parameters generated from the geometric parameters after data processing;

[0009] S4: Optimize the geometric parameters after data processing and the performance parameters after weight allocation using a multi-island genetic algorithm;

[0010] The geometric parameters after data processing are encoded into genetic data, and chromosome individuals are randomly generated to form n2 populations.

[0011] Substitute the gene data carried by the individual chromosome into the fitting equation to obtain the corresponding performance parameters;

[0012] The fitness of the obtained performance parameters is calculated;

[0013] Traditional genetic manipulations are performed on individual chromosomes within each population based on fitness.

[0014] Randomly select e individuals with different chromosomes for population migration to generate a new population of n².

[0015] Elite individuals are selected from the new population, and these elite individuals are those with the highest fitness chromosome.

[0016] S5: Determine whether the selected elite individual meets the termination condition. If the selected elite individual does not meet the termination condition, repeat S4 for gene iteration. If the selected elite individual meets the termination condition, decode the selected elite individual and perform inverse normalization.

[0017] S6: Output the parameters processed by inverse normalization as the optimal geometric parameters of the tap.

[0018] Furthermore, the range of values ​​for the geometric parameters of the model of the tap to be optimized described in S1 is as follows:

[0019]

[0020] Where, γ pmin For γ p The minimum value in the range, γ pmax For γ p The maximum value in the range, α pmin For α p The minimum value in the range, α pmax For α p The maximum value in the range, N min Let N be the minimum value in the range of N. max This represents the maximum value within the range of N.

[0021] Furthermore, in S1, the geometric parameters are processed using a dimensionless normalization method, and the normalization formula is as follows:

[0022]

[0023] Where x represents any geometric parameter to be normalized, x new x represents the result obtained after normalization. max This represents the maximum value in the range of x. min This represents the minimum value in the range of values, where x... new The value range is [0, 1], and the precision is set to 10. -3 .

[0024] Furthermore, the fitting equation between the geometric parameters after data processing and the corresponding generated performance parameters described in S2 is as follows:

[0025]

[0026] Among them, g1(γ) p ,α p (N) is the fitting equation corresponding to the cutting force F, g2(γ) p ,α p (N) represents the fitting equation corresponding to the tool wear rate μ, and g3(γ) p ,α p ,N) is the fitting equation corresponding to the cutting temperature T.

[0027] Furthermore, the weight allocation method described in S3 is implemented through a weight matrix, which is:

[0028] B = [A1 A2 A3],

[0029] A1 represents the weighting of cutting force, A2 represents the weighting of cutting temperature, and A3 represents the weighting of tool wear rate, where A1+A2+A3=1.

[0030] Furthermore, the performance index function for generating the weight matrix is:

[0031] W = A1F + A2T + A3μ,

[0032] Furthermore, the gene data encoding method described in S4 is to encode the anterior angle γ of the tap. p Tap back angle α p The number of teeth N in the cutting cone is encoded as X1, X2, X3, so each chromosome individual is {X1, X2, X3}, where the values ​​of X1, X2, and X3 are between [0, 1], and the precision is set to 10. -3 The population size is n2, and the number of individuals with chromosomes in each population is n1.

[0033] Furthermore, the fitness described in S4 is calculated using a fitness function, which is:

[0034]

[0035] Where R is fitness and W is a performance index function.

[0036] Furthermore, the traditional genetic operations described in S4 include gene selection, gene crossover, and gene mutation. The gene selection is achieved using a roulette wheel selection method, where the probability of a chromosome being selected in the population is proportional to the fitness function of the corresponding chromosome. The selection formula is:

[0037]

[0038] Among them, P i Let R be the probability that the i-th gene is selected. i Let be the fitness of the individual with chromosome i.

[0039] For each generation of individuals in each population, chromosomes are sorted in reverse order by adaptation: The corresponding individual chromosome sequence is Let the chromosomes of individuals be G1 and G2, G3 and G4, ... and Gene crossover is performed based on the crossover probability, and the formula for the gene crossover probability is:

[0040]

[0041] Among them, P c max P represents the upper bound of the probability of gene crossover. c min R represents the lower bound of the probability of gene crossover. max R represents the maximum fitness of the population. avg R represents the average fitness of the population. c max This represents the maximum fitness among the parent generations participating in the crossover; c0 is 9.903438.

[0042] Gene crossing is achieved through a two-point crossing method, where crossing points are set between X1 and X2, and between X2 and X3, and parts of the chromosomes of the two individuals are exchanged between the two set crossing points.

[0043] The gene mutation is achieved through basic positional mutation, and the gene mutation probability formula is:

[0044]

[0045] Among them, P m max P represents the upper bound of the probability of gene mutation. mmin R represents the lower bound of the probability of gene mutation. max R represents the maximum fitness of the population. avg R represents the average fitness of the population. m c0 represents the fitness of the individual with the variant chromosome, and is 9.903438.

[0046] Furthermore, the termination condition described in S5 is when the fitness of the elite individual and the fitness of the population no longer increase, or when the number of gene iterations t reaches the preset number of iterations T. In this case, the elite individual is decoded and denormalized, and the parameters of the denormalization process are output as the optimal geometric parameters of the tap.

[0047] Compared with the prior art, the present invention has the following advantages:

[0048] By optimizing the geometric parameters of the tap using the tap parameter optimization method, the performance of the tap in the actual machining process is improved. Compared with the unoptimized tap, the tool wear rate, cutting force and chip temperature are reduced by about 15% during the cutting process.

[0049] The taps optimized by the tap parameter optimization method have reduced the surface roughness of the actual machined threads, improved the dimensional accuracy, and shortened the machining cycle. At the same time, the taps are more durable, have a longer service life, and reduce the frequency of downtime for tap replacement.

[0050] Compared with traditional genetic algorithms, tap parameter optimization methods process more data and can facilitate data exchange between populations, which can reduce the probability of tap geometric and performance parameters getting trapped in local optima.

[0051] The tap parameter optimization method using an adaptive multi-island genetic algorithm achieves a 10%–20% faster convergence speed compared to the ordinary multi-island genetic algorithm, significantly improving the calculation speed for the optimal solutions to the geometric and performance parameters of the tap. Attached Figure Description

[0052] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0053] Figure 1 A flowchart of a tap parameter optimization method;

[0054] Figure 2 A schematic diagram of the gene crossover strategy for tap parameter optimization methods;

[0055] Figure 3 A schematic diagram of the gene mutation strategy for tap parameter optimization methods;

[0056] Figure 4A schematic diagram of the population migration strategy for tap parameter optimization methods;

[0057] Figure 5 A schematic diagram showing the number of teeth N in the chip cone section of the tap's geometry.

[0058] Figure 6 The tap rake angle γ is a geometric parameter of the tap. p and tap back angle α p A schematic diagram. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other, and the described embodiments are only some embodiments of the present invention, not all embodiments.

[0060] In the description of this invention, unless otherwise explicitly specified and limited, the terms "connected," "linked," and "fixed" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0061] A method for optimizing tap parameters, see [link to relevant documentation]. Figure 1 The specific steps are as follows:

[0062] S1: Determine the range of values ​​for the geometric parameters of the model of the tap to be optimized, and perform data processing on the geometric parameters, including the tap rake angle γ. p Tap back angle α p And the number of teeth N in the chip cone, see Figure 5 and Figure 6 ;

[0063] S2: Obtain the fitting equations of the geometric parameters and corresponding performance parameters in the above tap model after data processing through orthogonal experiments. The corresponding performance parameters include cutting force F, cutting temperature T, and tool wear rate μ.

[0064] S3: Perform a comprehensive analysis and weight allocation on the performance parameters generated from the geometric parameters after data processing;

[0065] S4: Optimize the geometric parameters after data processing and the performance parameters after weight allocation using a multi-island genetic algorithm;

[0066] The geometric parameters after data processing are encoded into genetic data, and chromosome individuals are randomly generated to form n2 populations.

[0067] Substitute the gene data carried by the individual chromosome into the fitting equation to obtain the corresponding performance parameters;

[0068] The fitness of the obtained performance parameters is calculated;

[0069] Traditional genetic manipulations are performed on individual chromosomes within each population based on fitness.

[0070] Randomly select e individuals with e chromosomes for population migration, see [link / reference]. Figure 4 , generate n² new populations;

[0071] Elite individuals are selected from the new population, and these elite individuals are those with the highest fitness chromosome.

[0072] S5: Determine whether the selected elite individual meets the termination condition. If the selected elite individual does not meet the termination condition, repeat S4 for gene iteration. If the selected elite individual meets the termination condition, decode the selected elite individual and perform inverse normalization.

[0073] S6: Output the parameters processed by inverse normalization as the optimal geometric parameters of the tap.

[0074] Specifically, Figure 4 In this context, the population migration refers to the random selection of e chromosome individuals from population 1 to population 4, the random selection of e chromosome individuals from population 2 to population 1, the random selection of e chromosome individuals from population 3 to population 2, and the random selection of e chromosome individuals from population 4 to population 3.

[0075] The range of values ​​for the geometric parameters of the model of the tap to be optimized described in S1 is as follows:

[0076]

[0077] Where, γ pmin For γ p The minimum value in the range, γ pmax For γ p The maximum value in the range, α pmin For α p The minimum value in the range, α pmax For α p The maximum value in the range, N min Let N be the minimum value in the range of N. max This represents the maximum value within the range of N.

[0078] In S1, the geometric parameters are processed using dimensionless normalization. The normalization formula is as follows:

[0079]

[0080] Where x represents any geometric parameter to be normalized, x new x represents the result obtained after normalization. max This represents the maximum value in the range of x. min This represents the minimum value in the range of values, where x... new The value range is [0, 1], and the precision is set to 10. -3 .

[0081] The fitting equations between the geometric parameters after data processing and the corresponding performance parameters described in S2 are as follows:

[0082]

[0083] Among them, g1(γ) p ,α p (N) is the fitting equation corresponding to the cutting force F, g2(γ) p ,α p (N) represents the fitting equation corresponding to the tool wear rate μ, and g3(γ) p ,α p ,N) is the fitting equation corresponding to the cutting temperature T.

[0084] The weight allocation method described in S3 is implemented through a weight matrix, which is as follows:

[0085] B = [A1 A2 A3],

[0086] Where A1 represents the weighting of cutting force, A2 represents the weighting of cutting temperature, and A3 represents the weighting of tool wear rate, with A1 + A2 + A3 = 1. While considering the comprehensive weighting of each performance characteristic, external factors in machining, including workpiece material and machining process, are also taken into account to make the comprehensive weighting more closely reflect the actual situation of the machining process.

[0087] The performance index function generated by the weight matrix is:

[0088] W = A1F + A2T + A3μ.

[0089] The fitness mentioned in S4 is calculated using a fitness function, which is:

[0090]

[0091] Where R is fitness and W is a performance index function.

[0092] The gene data encoding method described in S4 is to encode the anterior angle γ of the tap. p Tap back angle α p The number of teeth N in the cutting cone is encoded as X1, X2, X3, so each chromosome individual is {X1, X2, X3}, where the values ​​of X1, X2, and X3 are between [0, 1], and the precision is set to 10. -3 The population size is n2, and the number of individuals with chromosomes in each population is n1.

[0093] The traditional genetic operations described in S4 include gene selection, gene crossover, and gene mutation. Gene selection is achieved using a roulette wheel selection method. The probability of a chromosome being selected in the population is proportional to the fitness function of that chromosome. The selection formula is:

[0094]

[0095] Among them, P i Let R be the probability that the i-th gene is selected. i Let be the fitness of the individual with chromosome i.

[0096] For each generation of individuals in each population, chromosomes are sorted in reverse order by adaptation: The corresponding individual chromosome sequence is Let the chromosomes of individuals be G1 and G2, G3 and G4, ... and Crossover is performed according to the gene crossover probability. By arranging chromosome individuals according to their fitness, the gene crossover probability of chromosome individuals with low fitness can be increased, while that of chromosome individuals with high fitness can be crossovered with a lower gene crossover probability. This allows high-fitness chromosome individuals to be retained to the next generation, which can reduce the computation time of the algorithm.

[0097] The formula for gene crossover probability is:

[0098]

[0099] Among them, P c max P represents the upper bound of the probability of gene crossover. c min R represents the lower bound of the probability of gene crossover. max R represents the maximum fitness of the population. avg R represents the average fitness of the population. c max This represents the maximum fitness among the parent generations participating in the crossover; c0 is 9.903438.

[0100] Gene crossover is achieved through a two-point crossover method, see [link / reference] Figure 2On the chromosomes of two parent individuals, two different crossover points, 1 and 2, are selected as gene crossover points. These crossover points divide the chromosome into three segments. The portions of the chromosomes between the two parent individuals are exchanged, resulting in two offspring chromosomes.

[0101] The gene mutations are achieved through basic positional mutations. See [link to relevant documentation] Figure 3 A portion of a chromosome from a randomly selected individual is mutated, resulting in gene X2 becoming X2″. After this mutation, gene X2″ will replace the original gene X2 in the offspring. Randomly mutating a portion of genes from an individual chromosome yields a chromosome with gene crossover. The probability formula for gene mutation is:

[0102]

[0103] Among them, P m max P represents the upper bound of the probability of gene mutation. m min R represents the lower bound of the probability of gene mutation. max R represents the maximum fitness of the population. avg R represents the average fitness of the population. m c0 represents the fitness of the individual with the variant chromosome, and is 9.903438.

[0104] The termination condition described in S5 is when the fitness of the elite individual and the fitness of the population no longer increase, or when the number of gene iterations t reaches the preset number of iterations T. Then, the elite individual is decoded and denormalized, and the parameters of the denormalized process are output as the optimal geometric parameters of the tap.

[0105] Obviously, the above-disclosed embodiments of the present invention are merely for illustrating the present invention. The embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementations. Many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. It is neither necessary nor possible to exhaustively describe all implementation methods herein.

Claims

1. A tap parameter optimization method, characterized by, The specific steps are as follows: S1: determine the value range of the geometric parameters of the model of the tap to be optimized, and perform data processing on the geometric parameters, the geometric parameters including tap rake angle γ p , tap clearance angle α p , and chip cone tooth number N; S2: obtaining a fitting equation of the geometric parameters after data processing in the tap model and corresponding generated performance parameters including cutting force F, cutting temperature T and tool wear rate μ through orthogonal test; S3: performing weight distribution on the performance parameters generated by the geometric parameters after data processing; S4: optimizing the geometric parameters after data processing and the performance parameters after weight distribution through a multi-island genetic algorithm; encoding the geometric parameters after data processing into gene data, randomly generating chromosome individuals and forming n2 populations; substituting the gene data carried by the chromosome individuals into the fitting equation to obtain corresponding performance parameters; calculating the fitness of the obtained performance parameters; performing traditional genetic operations on the chromosome individuals in each population according to the fitness; randomly selecting e chromosome individuals for population migration to generate n2 new populations; selecting elite individuals in the new populations, the elite individuals being the chromosome individuals with the maximum fitness; S5: judging whether the selected elite individuals meet the termination condition, if not, repeating S4 for gene iteration; if yes, decoding and performing reverse normalization processing on the selected elite individuals, and outputting the parameters after reverse normalization processing as the optimal geometric parameters of the tap; S6: outputting the parameters after reverse normalization processing as the optimal geometric parameters of the tap; The value range of the geometric parameters of the model of the tap to be optimized in S1 is: ， Where, γ pmin For γ p The minimum value in the range, γ pmax For γ p The maximum value in the range, α pmin For α p The minimum value in the range, α pmax For α p The maximum value in the range, N min Let N be the minimum value in the range of N. max The maximum value in the range of N; The fitting equation of the geometric parameters after data processing and corresponding generated performance parameters in S2 is: ， where g1 (γ p , α p , N) is a fitting equation corresponding to the cutting force F, g2 (γ p , α p , N) is a fitting equation corresponding to the tool wear rate μ, and g3 (γ p , α p , N) is a fitting equation corresponding to the cutting temperature T.

2. The tap parameter optimization method of claim 1, wherein: The data processing method of the geometric parameters in S1 is normalization processing in dimensionless form, and the normalization processing formula is: ， Wherein, x represents any kind of geometric parameter to be normalized, x new represents the result obtained by normalization, x max represents the maximum value of the value range of x, x min represents the minimum value of the value range, wherein x new The value range of x is [0, 1], and the precision is set to 10 -3 .

3. The tap parameter optimization method of claim 1, wherein: The weight distribution method in S3 is realized through a weight matrix, and the weight matrix is: B=[A1 A2 A3], wherein A1 is the weight distribution of cutting force, A2 is the weight distribution of cutting temperature, and A3 is the weight distribution of tool wear rate, and A1+A2+A3=1.

4. The tap parameter optimization method of claim 3, wherein: The performance index function generated by the weight matrix is: W=A1F+A2T+A3μ.

5. The tap parameter optimization method of claim 1, wherein: The coding mode of the gene data in S4 is to encode the tap front angle γ p , the tap rear angle α p , and the number of chip taper teeth N as X1, X2, X3, so that each chromosome individual is {X1, X2, X3}, wherein the values of X1, X2 and X3 are between [0, 1], and the precision is set to 10 -3 , the population number is n2, and the number of chromosome individuals in each population is n1.

6. The tap parameter optimization method of claim 4, wherein: The fitness in S4 is calculated through a fitness function, and the fitness function is: ， wherein R is fitness, and W is a performance index function.

7. The tap parameter optimization method of claim 3, wherein: The traditional genetic operations in S4 include gene selection, gene crossover and gene mutation, the gene selection is realized in a roulette selection manner, the probability of a chromosome individual in a population being selected is proportional to the corresponding fitness function of the chromosome individual, and the selection formula is: ， where P i is the probability of the ith gene being selected, R i is the fitness of the ith chromosome individual; Each generation of chromosome individuals in each population is sorted in reverse order by fitness: The sequence of chromosome individuals is Let chromosome individuals G1 and G2, G3 and and are crossed according to a gene crossover probability, and the formula of the gene crossover probability is: ， where P c max represents an upper bound of the gene crossover probability, P c min represents a lower bound of the gene crossover probability, R max represents a maximum value of the population fitness, R avg represents an average value of the population fitness, R c max represents a maximum value of the fitness in the parent participating in the crossover, c0takes 9.903438, The gene crossover is realized in a two-point crossover manner, crossover points are set between X1 and X2 and X2 and X3, and the parts of the two chromosome individuals between the two set crossover points are exchanged; The gene mutation is realized in a basic bit mutation manner, and the gene mutation probability formula is: ， where P m max represents the upper bound of the probability of gene mutation, P m min represents the lower bound of the probability of gene mutation, R max represents the maximum value of the fitness of the population, R avg represents the average value of the fitness of the population, R m represents the fitness of the mutated chromosome individual, and c0is 9.903438.

8. The tap parameter optimization method of claim 1, wherein: The termination condition in S5 is that when the fitness of the elite individual and the population fitness no longer increases, or the number of gene iterations t reaches a preset iteration number T, the elite individual is decoded and reverse normalized, and the parameters after reverse normalization processing are output as the optimal geometric parameters of the tap.

Images