Method for generating cutting amount data through artificial intelligence

Abstract

The invention provides a method for generating cutting amount data by artificial intelligence, which relates to the technical field of machining, and specifically comprises the following steps: establishing a cutting amount calculation model and a database of the cutting amount calculation model, collecting and counting original data of cutting amount, and establishing an original database; training a cutting amount calculation model by using an original database, determining values Cm of each non-numeric influence factor under different value conditions and values an of the numeric influence factors under the conditions, and storing the values Cm and the values an in the database of the cutting amount calculation model; optimizing the cutting amount calculation model, classifying the non-numeric influence factors by setting a similarity tolerance value according to the similarity between the Cm value of each non-numeric influence factor under different value conditions and the an value of the numeric influence factor under the condition, and generating substitutive representatives; and inputting a required cutting condition into the optimized cutting amount calculation model to generate cutting amount data.

Description

Technical Field

[0001] The present invention relates to the field of mechanical processing technology, and in particular to a method for generating cutting consumption data using artificial intelligence. Background Art

[0002] The cutting data used in mechanical processing usually comes from two sources, directly using existing empirical data or generating it through empirical formula calculation. This has led to two traditional cutting data storage, management and database system operation modes: a retrieval-based discrete static database that mainly stores empirical data, and a calculation-based condensed dynamic database that mainly stores empirical formulas. The former is usually not actually used except for a small number of small-variety, high-volume industries due to insufficient data volume, incomplete data, and low detection rate. The latter requires a large amount of cutting tests and theoretical research as support, which a single research institution or enterprise is basically unable to afford. Moreover, there is often a large gap between the actual working conditions of production enterprises and cutting tests and theories, which also restricts the practicality of such cutting databases.

[0003] Case-based reasoning (CBR) combines the above two modes and can generate new data based on existing empirical data. It is a new method for establishing a cutting database. In recent years, some institutions have been conducting corresponding research. This method is based on the data in the database that are most similar to the required conditions. Due to the scientific nature of the similarity judgment rules and the stability of the source data, the data reasoning calculation results are generally difficult to meet the actual application needs.

[0004] The above situation makes it impossible to effectively utilize the large amount of cutting experience data and empirical formulas collected by production enterprises and scientific research institutions at the cost of a lot of manpower and material resources. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a method for generating cutting usage data using artificial intelligence. By constructing a cutting usage data model and using empirical data for training, it can fully realize the effective reuse of a large amount of cutting experience data and empirical formulas of production enterprises and scientific research institutions.

[0006] A method for generating cutting usage data using artificial intelligence, comprising the following steps:

[0007] Step 1, establish a cutting amount calculation model and a database of the cutting amount calculation model, and store empirical formulas, cutting amounts and influencing factor names, value ranges and characteristic values;

[0008] The cutting amount calculation model formula is as follows:

[0009]

[0010] In the formula, x, a n is a variable, where x represents the cutting amount, a n Represents the numerical factor variable that affects the cutting amount x, C m , k n is a constant, corresponding to non-numerical influencing factors, C m is the specific value of the non-numerical influencing factor, k n is a numerical factor variable a n The exponential constant of

[0011] Step 2, designing an original data collection table according to the cutting data model and the cutting data model database;

[0012] The original data collection table includes cutting parameters and influencing factors thereof;

[0013] Step 3, collect and count the original data according to the original data collection table;

[0014] Step 4: Arrange the data in the original data collection table and establish an original database for training;

[0015] Step 5: Use the original database to train the cutting parameter calculation model to determine the value C of each non-numerical influencing factor under different value conditions. m And the value a of the numerical influencing factor under this condition n , stored in the database of the cutting amount calculation model; if the data is incomplete, only the valid part of the data is trained;

[0016] Step 6: Optimize the cutting amount calculation model according to the C under different value conditions of each non-numerical influencing factor. m Value and the a of the numerical influencing factors under this condition n Similarity between values, classify non-numerical influencing factors by setting similarity tolerance value, and generate substitution table;

[0017] Specifically, the classification includes classifying the values ​​of similarity tolerance within a set range into one category;

[0018] Step 7: Input the required cutting conditions into the optimized cutting parameter calculation model to generate cutting parameter data.

[0019] The beneficial effects of adopting the above technical solution are:

[0020] The present invention provides a method for generating cutting usage data by using artificial intelligence. The present invention constructs a cutting usage calculation model and uses empirical data for training. The cutting data is generated based on artificial intelligence, which can effectively and fully utilize a large amount of cutting experience data of production enterprises and scientific research institutions to the maximum extent, and obtain cutting usage data suitable for actual on-site processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is an overall flow chart of the method for generating cutting usage data using artificial intelligence in the present invention. DETAILED DESCRIPTION

[0022] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0023] A method for generating cutting usage data by artificial intelligence, such as Figure 1 As shown, the following steps are included:

[0024] Step 1, establish a cutting amount calculation model and a database of the cutting amount calculation model, and store empirical formulas, cutting amounts and influencing factor names, value ranges and characteristic values;

[0025] The cutting amount calculation model formula is as follows:

[0026]

[0027] In the formula, x, a n is a variable, where x represents the cutting amount, a n Represents the numerical factor variables that affect the cutting parameters x (such as the geometric dimensions of the workpiece and tool, the hardness of the material, other cutting parameters, etc.); C m , k n is a constant, corresponding to non-numerical influencing factors (such as material grade, etc.), C m is the specific value of the non-numerical influencing factor, k n is a numerical factor variable a n The exponential constant of

[0028] Cutting parameter data generally include but are not limited to cutting speed, feed rate, and depth of cut. Factors affecting the cutting parameter values ​​generally include but are not limited to other cutting parameters, the materials of the workpiece and tool being cut and their geometric shapes, the surface quality requirements of the workpiece being cut, the type of cutting fluid, the rated speed of the equipment, production efficiency and cost targets, etc.

[0029] Experimental research has shown that for any cutting amount and various influencing factors, within its effective numerical range, an empirical formula of the following form can be established:

[0030]

[0031] In the formula, x, a n is a variable, x represents a cutting value, a nIt indicates the specific value of a numerical factor that affects the cutting parameters (such as the geometric dimensions of the workpiece and tool, the hardness of the material, other cutting parameters, etc.). m , k n is a constant corresponding to each non-numerical influencing factor (such as material grade, etc.), C m With a specific value, k n is a numerical factor variable a n The index of .

[0032] Through regression analysis of the cutting parameters under various influencing factors, it can be concluded that m 、a n value, thereby establishing an empirical formula to complete the calculation of cutting parameters under specified conditions.

[0033] Step 2, designing an original data collection table according to the cutting data model and the cutting data model database;

[0034] The original data collection table includes cutting parameters and influencing factors thereof.

[0035] Step 3: Collect and count the original data according to the original data collection table; the original data can come from the production site, cutting test, data manual and other technical information. The original data can be incomplete data that lacks some influencing factors;

[0036] Step 4: Arrange the data in the original data collection table and establish an original database for training;

[0037] Step 5: Use the original database to train the cutting parameter calculation model to determine the value C of each non-numerical influencing factor under different value conditions. m And the value a of the numerical influencing factor under this condition n , stored in the database of the cutting amount calculation model; if the data is incomplete, only the valid part of the data is trained;

[0038] Step 6: Optimize the cutting amount calculation model according to the C under different value conditions of each non-numerical influencing factor. m Value and the a of the numerical influencing factors under this condition n Similarity between values, classify non-numerical influencing factors by setting similarity tolerance value, and generate substitution table;

[0039] Specifically, the classification includes classifying the values ​​of similarity tolerance within a set range into one category, and data of the same category can replace each other;

[0040] Step 7: Input the required cutting conditions into the optimized cutting parameter calculation model to generate cutting parameter data.

[0041] Taking the generation of cutting parameters for turning the outer arc surface as an example, the non-numerical factors affecting the cutting speed are workpiece material, tool material, workpiece geometric characteristics, and tool geometric characteristics. The numerical factors are feed rate, cutting depth, tool rake angle, and workpiece outer arc radius. An empirical formula for cutting speed can be established:

[0042]

[0043] Where C1, C2, C3, and C4 are workpiece material, tool material, workpiece geometric characteristics, and tool geometric characteristics constants, respectively; a1, a2, a3, and a4 are feed rate, cutting depth, tool rake angle, and workpiece outer arc radius variables, respectively; k1, k2, k3, and k4 are feed rate, cutting depth, tool rake angle, and workpiece outer arc radius exponential constants, respectively.

[0044] By using 3128 pieces of original data to perform regression calculation on the above formula, the values ​​of constants C1, C2, C3, C4, k1, k2, k3, and k4 are determined.

[0045] If you need to calculate the cutting speed under the following conditions in production: ordinary turning, workpiece material TC4, tool material HX, circular blade rake angle 6°, feed rate 0.12, cutting depth 0.5, outer arc radius 32, find the following formula in the model database:

[0046]

[0047] Substituting the above variable values, we get:

[0048] V c =24.07×0.12 -0.4 0.5 -0.13 6 0.2 32 -0.03 =79.3(m / min)

[0049] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) to form a technical solution.

Claims

1. A method for generating cutting usage data using artificial intelligence, characterized in that: The following steps are involved: Step 1, establish a cutting amount calculation model and a database of the cutting amount calculation model, and store empirical formulas, cutting amounts and influencing factor names, value ranges and characteristic values; Step 2, designing an original data collection table according to the cutting data model and the cutting data model database; Step 3, collect and count the original data according to the original data collection table; Step 4: Arrange the data in the original data collection table and establish an original database for training; Step 5: Use the original database to train the cutting parameter calculation model to determine the value C of each non-numerical influencing factor under different value conditions. m And the value a of the numerical influencing factor under this condition n , stored in the database of the cutting amount calculation model; if the data is incomplete, only the valid part of the data is trained; Step 6: Optimize the cutting amount calculation model according to the C under different value conditions of each non-numerical influencing factor. m Value and the a of the numerical influencing factors under this condition n Similarity between values, classify non-numerical influencing factors by setting similarity tolerance value, and generate substitution table; Step 7: Input the required cutting conditions into the optimized cutting parameter calculation model to generate cutting parameter data.

2. The method for generating cutting usage data by artificial intelligence according to claim 1, characterized in that: The cutting amount calculation model formula described in step 1 is as follows: In the formula, x, a n is a variable, where x represents the cutting amount, a n Represents the numerical factor variable that affects the cutting amount x, C m , k n is a constant, corresponding to non-numerical influencing factors, C m is the specific value of the non-numerical influencing factor, k n is a numerical factor variable a n The exponential constant.

3. The method for generating cutting usage data by artificial intelligence according to claim 1, characterized in that: The original data collection table described in step 2 includes cutting parameters and their influencing factors.

4. The method for generating cutting usage data by artificial intelligence according to claim 1, characterized in that: The classification in step 6 specifically includes classifying the values ​​of similarity tolerance within a set range into one category.

Images