Numerical control machining parameter optimization method and device

Abstract

The application discloses a numerical control machining parameter optimization method and device, belongs to the numerical control machining parameter optimization technical field, and the method comprises the following steps: step 1, analyzing a part machining process, and selecting appropriate machining parameters as optimization variables; step 2, establishing a machining process multi-objective optimization function, representing the complex corresponding relationship between machining parameters and machining efficiency, machining cost and part quality; step 3, determining machining process optimization constraint conditions according to actual machining working condition requirements; step 4, establishing a machining parameter optimization model, and converting the multi-objective optimization function into a single-objective optimization function; and step 5, solving the machining parameter optimization model, minimizing the single-objective optimization function through a posterior method, and obtaining an optimal machining parameter combination; the application realizes part machining parameter optimization under complex working conditions, and can effectively improve part machining efficiency and reduce production resource waste.

Description

Technical Field

[0001] This invention relates to the field of CNC machining parameter optimization technology, and in particular to a CNC machining parameter optimization method and apparatus, which is applied to the field of CNC machining parameter optimization technology for complex aerospace parts. Background Technology

[0002] Aerospace components are characterized by a wide variety of materials, complex features, and high technical requirements, placing stringent demands on the CNC machining process. Machining parameters, as a key element of CNC machining, significantly influence part quality, machining efficiency, tool life, and production costs. Optimal machining parameters are crucial for high-quality, high-efficiency, and low-cost machining, and are of great value and significance in promoting the digitalization, automation, and industrial transformation of production, as well as achieving breakthroughs in intelligent manufacturing technologies for aerospace.

[0003] Currently, in the actual production process of complex aerospace parts, there is a lack of general solutions for guiding the selection and optimization of machining parameters. Machining parameters are often estimated based on historical experience or tool manuals, which largely depends on the skill level and experience of the personnel. Moreover, for the safety of parts and personnel, the selected machining parameters are often conservative, making it difficult to fully utilize the performance of high-end machine tools and resulting in a waste of production resources. Summary of the Invention

[0004] The purpose of this invention is to address the technical problem that, in the actual production of complex aerospace parts, there is a lack of a universal solution for guiding the selection and optimization of machining parameters. Machining parameters are often estimated based on historical experience or tool manuals, which largely depends on the skill level and experience of the personnel. Moreover, for the safety of parts and personnel, the selected machining parameters are often conservative, making it difficult to fully utilize the performance of high-end machine tools and resulting in a waste of production resources.

[0005] The purpose of this invention is to provide a method and apparatus for optimizing CNC machining parameters, which can improve machining efficiency, reduce production costs, and reduce uncertainty in experience-dependent machining modes. The CNC machining parameter optimization method provided by this invention can obtain the optimal combination of machining parameters under the current working conditions to guide production, effectively improve machining efficiency, reduce machining costs, and ensure part quality.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] A method for optimizing CNC machining parameters includes the following steps:

[0008] Step 1: Analyze the part machining process and select appropriate machining parameters as optimization variables;

[0009] Step 2: Establish a multi-objective optimization function for the processing process to characterize the complex correspondence between processing parameters and processing efficiency, processing cost, and part quality;

[0010] Step 3: Determine the optimization constraints for the processing based on the actual processing conditions.

[0011] Step 4: Establish a processing parameter optimization model and convert the multi-objective optimization function into a single-objective optimization function;

[0012] Step 5: Solve the processing parameter optimization model, minimize the single-objective optimization function using a posterior method, and obtain the optimal combination of processing parameters.

[0013] Furthermore, in step 1, appropriate machining parameters are selected as optimization variables. Specifically, the machining process of the part is analyzed, and the cutting speed v, feed rate f, depth of cut ap, and width of cut ae are selected as parameters characterizing the main motion and feed motion of the machining process, and as optimization variables X that effectively characterize the state of the machining process.

[0014] Furthermore, step 2 establishes a multi-objective optimization function for the processing process. Specifically, it determines the combination of the processing efficiency function T(X), processing cost function C(X), and processing quality function Q(X) as the multi-objective optimization function for the processing process. The objective optimization is to maximize processing efficiency while minimizing processing cost and maximizing processing quality.

[0015] Furthermore, the machining efficiency function T(X) is represented by the machining time, which includes auxiliary time t0 and cutting time t. m Tool change time t c The auxiliary time is a fixed empirical value, while the cutting time is a function of machining parameters and tool parameters; the expression for T(X) is:

[0016]

[0017] Where T is the tool life.

[0018] Furthermore, the processing cost function C(X) includes the processing cost c. m Tool cost c t The processing cost is determined by the processing time, and the tool cost is determined by the number of tool changes; the expression for C(X) is:

[0019] C(X) = c m +c t .

[0020] Furthermore, the machining quality function Q(X) includes the surface roughness of the part, which is determined by measuring the surface roughness of the part after machining; the expression for Q(X) is:

[0021]

[0022] Among them, f z feed per tooth, r e The radius of the blade tip arc.

[0023] Furthermore, step 3 determines the optimization constraints of the machining process, specifically including: the cutting speed v and feed rate f are within the maximum and minimum ranges allowed by the machine tool usage specifications, respectively; the machine tool power is within the maximum and minimum ranges allowed by the usage specifications, respectively; the depth of cut ap and width of cut ae are within the maximum and minimum ranges allowed by the tool usage specifications, respectively; and the surface roughness of the part is within the range allowed by the part design drawing.

[0024] Furthermore, step 4 establishes a machining parameter optimization model. Specifically, the input X of the model consists of four optimization variables: cutting speed v, feed rate f, depth of cut ap, and cutting width ae. The machining efficiency function T(X), machining cost function C(X), and machining quality function Q(X) are all objective functions related to X. Through attention weighting and a neural network model, the three optimization objectives T(X), C(X), and Q(X) are transformed into a unified objective function F(X).

[0025] Attention weighting assigns attention weights w1, w2, and w3 to T(X), C(X), and Q(X), respectively, along with a bias b. The weights and biases are determined based on the relative influence of the processing parameters. The inputs to the neural network model are: T(X), C(X), Q(X), w1, w2, w3, and b. The output is the optimization objective function value. The single-objective optimization function model for the processing parameters is expressed as:

[0026] F(X) = f(T(X), C(X), Q(X))

[0027] Furthermore, in step 5, the processing parameter optimization model is solved. The objective function is minimized by using a posterior method based on the development and evolution of the population. Selection, crossover, and mutation operations are used to maintain the diversity and convergence of the solution set. Multiple solutions are output in each iteration, and the optimal solution is selected to obtain the optimal combination of processing parameters.

[0028] A CNC machining parameter optimization device, applied in the method, includes:

[0029] The selection module is used to select appropriate processing parameters as optimization variables;

[0030] A module is established to create an optimization objective function that characterizes the complex relationships between processing parameters and processing efficiency, production costs, and part quality.

[0031] The constraint module is used to determine the optimization constraints of the machining process based on the actual machining conditions.

[0032] The modeling module converts multi-objective optimization functions into single-objective optimization functions;

[0033] The solution module minimizes the objective function using a posterior method to obtain the optimal combination of processing parameters.

[0034] Beneficial effects of the present invention

[0035] (1) The present invention can obtain the optimal machining parameters for CNC machining and improve the machining efficiency of parts under complex working conditions;

[0036] (2) It helps to reduce production costs and reduce uncertainty in experience-dependent processing models;

[0037] (3) It is applicable not only to the optimization of processing parameters, but also to optimization problems in other complex environments. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the method principle of the present invention. Detailed Implementation

[0039] To more intuitively and clearly illustrate the structural principles and working methods in the embodiments of the present invention, the following description will be based on the accompanying drawings. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0040] Example 1:

[0041] A method for optimizing CNC machining parameters is provided, including:

[0042] Step 1: Analyze the part machining process and select appropriate machining parameters as optimization variables;

[0043] Step 2: Establish a multi-objective optimization function for the processing process to characterize the complex correspondence between processing parameters and processing efficiency, processing cost, and part quality;

[0044] Step 3: Determine the optimization constraints for the processing based on the actual processing conditions.

[0045] Step 4: Establish a processing parameter optimization model and convert the multi-objective optimization function into a single-objective optimization function;

[0046] Step 5: Solve the processing parameter optimization model, minimize the single-objective optimization function using a posterior method, and obtain the optimal combination of processing parameters.

[0047] In step 1, appropriate machining parameters are selected as optimization variables. Specifically, the machining process of the part is analyzed, and the cutting speed v, feed rate f, depth of cut ap, and width of cut ae are selected as parameters characterizing the main motion and feed motion of the machining process, and as optimization variables X that effectively characterize the state of the machining process.

[0048] Step 2 establishes a multi-objective optimization function for the processing process. Specifically, it involves determining a combination of the processing efficiency function T(X), the processing cost function C(X), and the processing quality function Q(X) as the multi-objective optimization function for the processing process. The objective optimization is to maximize processing efficiency while minimizing processing cost and maximizing processing quality.

[0049] The machining efficiency function T(X) is represented by the machining time, which includes auxiliary time t0 and cutting time t. m Tool change time t c The auxiliary time is a fixed empirical value, while the cutting time is a function of machining parameters and tool parameters; the expression for T(X) is:

[0050]

[0051] Where T is the tool life.

[0052] The processing cost function C(X) includes the processing cost c. m Tool cost c t The processing cost is determined by the processing time, and the tool cost is determined by the number of tool changes; the expression for C(X) is:

[0053] C(X) = c m +c t .

[0054] The machining quality function Q(X) includes the surface roughness of the part, which is determined by measuring the surface roughness of the part after machining; the expression for Q(X) is:

[0055]

[0056] Among them, f z feed per tooth, r e The radius of the blade tip arc.

[0057] Step 3 determines the optimization constraints of the machining process, specifically including: the cutting speed v and feed rate f are within the maximum and minimum ranges allowed by the machine tool usage specifications, respectively; the machine tool power is within the maximum and minimum ranges allowed by the usage specifications, respectively; the depth of cut ap and width of cut ae are within the maximum and minimum ranges allowed by the tool usage specifications, respectively; and the surface roughness of the part is within the range allowed by the part design drawing.

[0058] Specifically, step 4, establishing the machining parameter optimization model, involves the model input X consisting of four optimization variables: cutting speed v, feed rate f, depth of cut ap, and cutting width ae. The machining efficiency function T, machining cost function C, and machining quality function Q are all objective functions related to X. Through attention weighting and a neural network model, the three optimization objectives T(X), C(X), and Q(X) are transformed into a unified objective function F(X).

[0059] In this model, attention weighting involves assigning attention weights w1, w2, and w3 to T(X), C(X), and Q(X), respectively, along with a bias b. The weights and biases are determined based on the relative influence of the processing parameters. The inputs to the neural network model are: T(X), C(X), Q(X), w1, w2, w3, and b. The output is the optimization objective function value. The single-objective optimization function model for the processing parameters can be implicitly represented as:

[0060] F(X) = f(T(X), C(X), Q(X))

[0061] In step 5, the processing parameter optimization model is solved. The objective function is minimized by using a posterior method based on the development and evolution of the population. The diversity and convergence of the solution set are maintained by using selection, crossover, and mutation operations. Multiple solutions are output in each iteration, and the optimal solution is selected to obtain the optimal combination of processing parameters.

[0062] Example 2:

[0063] This embodiment provides a method for optimizing CNC machining parameters, which mainly includes the following:

[0064] Step 1: Analyze the part machining process and select appropriate machining parameters as optimization variables;

[0065] Step 2: Establish a multi-objective optimization function for the processing process to characterize the complex correspondence between processing parameters and processing efficiency, processing cost, and part quality;

[0066] Step 3: Determine the optimization constraints for the processing based on the actual processing conditions.

[0067] Step 4: Establish a processing parameter optimization model and convert the multi-objective optimization function into a single-objective optimization function;

[0068] Step 5: Solve the processing parameter optimization model, minimize the single-objective optimization function using a posterior method, and obtain the optimal combination of processing parameters.

[0069] The present invention analyzes the machining process of a part, selecting cutting speed v, feed rate f, depth of cut ap, and width of cut ae as parameters characterizing the main motion and feed motion of the machining process, and as optimization variables X to effectively characterize the state of the machining process. The cutting speed v is related to the tool diameter d and the machine tool speed n. The feed rate f is related to the feed per tooth f0. z It is related to the number of teeth z of the cutting tool. The optimization variable can be designed as a tensor X.

[0070] X = [v, f, ap, ae]

[0071]

[0072] f = f z z

[0073] The combination of the processing efficiency function T(X), processing cost function C(X), and processing quality function Q(X) described in this invention is used as a multi-objective optimization function for the processing process. The objective optimization is to maximize processing efficiency while minimizing processing cost and maximizing processing quality.

[0074] Specifically, the processing efficiency function of this invention is represented by processing time, which includes auxiliary time such as clamping preparation t0 and cutting time t. m Tool change time t c The auxiliary time t0 is a fixed empirical value (const). The cutting time t... m It is the time required for material removal, which is related to the tool diameter d, workpiece machining length L, cutting speed v, and feed per tooth f. z The tool tooth count z is related to the tool change time t. c With cutting time t m And it is related to tool life T. Tool life can be expressed by Taylor expansion as a function of the tool life coefficient c. t Cutting speed v, feed per tooth f z The depth of cut (ap) and width of cut (ae) are related, while x, y, z, and w represent the influencing factors of the corresponding parameters, determined according to the tool manual. The machining time function T(X) can be expressed as:

[0075]

[0076] t0 = const

[0077]

[0078] t c =c T v x f z y ae z ap w

[0079] Specifically, the processing cost function of this invention includes processing cost c. m Tool cost c t Processing costs are determined by processing time t. m It is determined that k1 is the corresponding empirical scaling factor. Tool cost is determined by the number of tool changes, and k2 is the corresponding empirical scaling factor. The machining cost function C(X) can be expressed as...

[0080] C(X) = c m +c t

[0081] c m =k1t m

[0082]

[0083] Specifically, the machining quality function of this invention mainly includes the surface roughness of the part and the feed per tooth f. z The radius of the blade tip arc r e This is relevant and can be determined by measuring the surface of the machined part. The machining quality function Q(X) can be expressed as:

[0084]

[0085] The determination of machining process optimization constraints described in this invention includes: the cutting speed v is within the maximum allowable value specified in the machine tool usage specifications. max minimum v min Within the specified range; the feed rate f is within the maximum allowable range specified in the machine tool operating specifications. max minimum f min Within the specified range; the machine tool power p is the maximum p allowed by the usage specifications. max Minimum p min Within the range; depth of cut ap is within the maximum allowable ap specified in the tool usage specifications. max Minimum AP min Within the range; the cutting width ae is within the maximum ae allowed by the tool usage specifications. max The smallest ae min Within the specified range; the surface roughness Ra of the part is within the Ra allowed by the part design drawing. max Within the range. The process optimization constraints can be expressed as:

[0086] v min ≤v≤v max

[0087] f min ≤f≤f max

[0088] p min ≤p≤p max

[0089] ap min ≤ap≤ap max

[0090] ae min ≤ae≤ae m1x

[0091] 0≤Ra≤Ra max

[0092] The machining parameter optimization model described in this invention takes four optimization variables, X, as input: cutting speed v, feed rate f, depth of cut ap, and width of cut ae. Machining efficiency function T, machining cost function C, and machining quality function Q are all objective functions related to X. Through attention weighting and a neural network model, the three optimization objectives T(X), C(X), and Q(X) are transformed into a unified objective function F(X). w1, w2, and w3 are the attention weights of T(X), C(X), and Q(X), respectively, and b is the bias. The weights and biases are set to 0.6, 0.2, 0.2, and 0.4, respectively. The input to the neural network model is T(X), C(X), Q(X), w1, w2, w3, and b. The output is the value of the optimization objective function. The single-objective optimization function model for machining parameters can be implicitly represented as F(X).

[0093] F(X) = f(T(X), C(X), Q(X))

[0094] The processing parameter optimization model described in this invention minimizes the objective function based on the evolution of the population using a posterior method. It maintains the diversity and convergence of the solution set by employing selection, crossover, and mutation operations. In each iteration, multiple solutions are output, from which the optimal solution is selected to obtain the optimal combination of processing parameters.

[0095] The main algorithm is as follows:

[0096] (1) Create an initial population, i.e., randomly generate optimization variables X that satisfy the constraints, such as... Figure 1 (x1, x2, x3, x4);

[0097] (2) Calculate the population fitness, i.e., processing time T(X), and select the individual with the shortest processing time, such as... Figure 1 The fitness of the population is (1,2,2,6), and the optimal individual is the individual with a fitness of 1.

[0098] (3) Determine whether the termination condition is met, that is, the number of iterations reached when the optimal fitness and average fitness of the population are stable. This can be set to 200. If it is met, output the optimal individual, which is also the optimal combination of processing parameters. If it is not met, continue.

[0099] (4) Perform selection, crossover, and mutation operations, and jump to (2).

[0100] Example 3:

[0101] This embodiment provides a CNC machining parameter optimization device, including:

[0102] The selection module is used to select appropriate processing parameters as optimization variables;

[0103] A module is established to create an optimization objective function that characterizes the complex relationships between processing parameters and processing efficiency, production costs, and part quality.

[0104] The constraint module is used to determine the optimization constraints of the machining process based on the actual machining conditions.

[0105] The modeling module converts multi-objective optimization functions into single-objective optimization functions.

[0106] The solution module minimizes the objective function using a posterior method to obtain the optimal combination of processing parameters.

[0107] In summary, the method and apparatus provided by this invention effectively improve processing efficiency, reduce production costs, and reduce uncertainty in experience-dependent processing modes. The CNC machining parameter optimization method provided by this invention can obtain the optimal combination of machining parameters under the current working conditions to guide production, effectively improve processing efficiency, reduce processing costs, and ensure part quality.

[0108] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing CNC machining parameters, characterized in that, Includes the following steps: Step 1: Analyze the part machining process and select appropriate machining parameters as optimization variables; Step 2: Establish a multi-objective optimization function for the processing process to characterize the complex correspondence between processing parameters and processing efficiency, processing cost, and part quality; Step 3: Determine the optimization constraints for the processing based on the actual processing conditions. Step 4: Establish a processing parameter optimization model and convert the multi-objective optimization function into a single-objective optimization function; Step 5: Solve the processing parameter optimization model, minimize the single-objective optimization function using a posterior method, and obtain the optimal combination of processing parameters.

2. The method according to claim 1, characterized in that: Step 1: Select appropriate machining parameters as optimization variables. Specifically, analyze the part machining process and select cutting speed v, feed rate f, depth of cut ap, and width of cut ae as parameters characterizing the main motion and feed motion of the machining process, and as optimization variables X that effectively characterize the state of the machining process.

3. The method according to claim 2, characterized in that: Step 2 establishes a multi-objective optimization function for the processing process. Specifically, it determines the combination of the processing efficiency function T(X), processing cost function C(X), and processing quality function Q(X) as the multi-objective optimization function for the processing process. The objective optimization is to maximize processing efficiency while minimizing processing cost and maximizing processing quality.

4. The method according to claim 3, characterized in that: The machining efficiency function T(X) is represented by the machining time, which includes auxiliary time t0 and cutting time t. m Tool change time t c The auxiliary time is a fixed empirical value, while the cutting time is a function of machining parameters and tool parameters; the expression for T(X) is: Where T is the tool life.

5. The method according to claim 4, characterized in that: The processing cost function C(X) includes the processing cost c. m Tool cost c t The processing cost is determined by the processing time, and the tool cost is determined by the number of tool changes; the expression for C(X) is: C(X)=c m +c t 。 6. The method according to claim 5, characterized in that: The machining quality function Q(X) includes the surface roughness of the part, which is determined by measuring the surface roughness of the part after machining; the expression for Q(X) is: Among them, f z feed per tooth, r e The radius of the blade tip arc.

7. The method according to claim 6, characterized in that: Step 3 determines the optimization constraints of the machining process, specifically including: the cutting speed v and feed rate f are within the maximum and minimum ranges allowed by the machine tool usage specifications, respectively; the machine tool power is within the maximum and minimum ranges allowed by the usage specifications, respectively; the depth of cut ap and width of cut ae are within the maximum and minimum ranges allowed by the tool usage specifications, respectively; and the surface roughness of the part is within the range allowed by the part design drawing.

8. The method according to claim 7, characterized in that: Step 4 establishes the machining parameter optimization model. Specifically, the input X of the model consists of four optimization variables: cutting speed v, feed rate f, depth of cut ap, and cutting width ae. The machining efficiency function T(X), machining cost function C(X), and machining quality function Q(X) are all objective functions related to X. Through attention weighting and a neural network model, the three optimization objectives T(X), C(X), and Q(X) are transformed into a unified objective function F(X). Attention weighting assigns attention weights w1, w2, and w3 to T(X), C(X), and Q(X), respectively, along with a bias b. The weights and biases are determined based on the relative influence of the processing parameters. The inputs to the neural network model are: T(X), C(X), Q(X), w1, w2, w3, and b. The output is the optimization objective function value. The single-objective optimization function model for the processing parameters is expressed as: F(X)=f(T(X), C(X), Q(X)).

9. The method according to claim 8, characterized in that: Step 5 solves the processing parameter optimization model. The objective function is minimized by using a posterior method based on the development and evolution of the population. Selection, crossover, and mutation operations are used to maintain the diversity and convergence of the solution set. Multiple solutions are output in each iteration, and the optimal solution is selected to obtain the optimal combination of processing parameters.

10. A CNC machining parameter optimization device, applied in the method as described in claim 9, characterized in that, include: The selection module is used to select appropriate processing parameters as optimization variables; A module is established to create an optimization objective function that characterizes the complex relationships between processing parameters and processing efficiency, production costs, and part quality. The constraint module is used to determine the optimization constraints of the machining process based on the actual machining conditions. The modeling module converts multi-objective optimization functions into single-objective optimization functions; The solution module minimizes the objective function using a posterior method to obtain the optimal combination of processing parameters.

Images