Optimization method based on intelligent numerical control machine tool machining process parameters

Abstract

The invention relates to the field of data processing, in particular to an optimization method based on machining process parameters of an intelligent numerical control machine tool, which comprises the following steps: acquiring the process parameters of the intelligent numerical control machine tool; obtaining an improved filter order; and inputting the process parameters into a filtering module, carrying out notch processing by using an improved filter order to obtain filtered process parameters, and inputting the filtered process parameters into a parameter optimization module to obtain optimized process parameters. According to the method, the order of the filter is dynamically adjusted by calculating the drifting degree of the dominant frequency of the technological parameters and the periodicity existence possibility, so that interference and noise caused by mechanical resonance can be eliminated or weakened when the parameters are filtered by using the improved order of the filter subsequently, and the filtering precision and the parameter optimization effect are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and more specifically, to an optimization method based on machining process parameters of intelligent numerical control machine tools. Background Art

[0002] In the process of CNC machine tool processing, the selection of process parameters such as cutting parameters (cutting speed, feed rate, cutting depth, etc.) and tool parameters (tool radius, etc.) directly affects the processing efficiency. By optimizing these parameters, the processing speed can be significantly increased, the processing cycle can be shortened, and production efficiency can be improved. For example, under the premise of ensuring processing quality, appropriately increasing the cutting speed and feed rate can significantly reduce the processing time.

[0003] Prior art, such as the patent application document with publication number CN114758064A, discloses a method for optimizing machine tool cutting process parameters based on digital twins, which includes: realizing multi-dimensional expression of the characteristics, functions, and behaviors of the actual cutting process of the machine tool through three-dimensional modeling software and a virtual platform; reducing the complexity of the underlying code through text data transmission in JSON format, collecting the process data of the machine tool in the cutting process in real time, and analyzing the influence weight of the process data on the cutting force through grey correlation theory; calling the encapsulated digital twin cutting force prediction model based on the unity3D visual mapping method, and performing real-time call analysis on the dynamic changes of the cutting force; generating an optimization plan for the cutting force fluctuation in real time based on the influencing factor weights and the digital twin prediction model.

[0004] However, the above-mentioned machine tool cutting process parameter optimization method does not take into account the influence of the mechanical resonance of the machine tool. Due to mechanical resonance, the machine tool produces continuous and regular vibrations. This vibration will interfere with the relative movement between the tool and the workpiece, making it impossible for the cutting process to proceed stably according to the predetermined trajectory and parameters, thereby affecting the processing accuracy and surface quality. Summary of the invention

[0005] In order to solve the technical problem that the optimization effect of the above-mentioned process parameters is reduced, the present invention provides the following technical solution.

[0006] The optimization method based on intelligent CNC machine tool processing parameters includes: Obtain the process parameters of intelligent CNC machine tools; Get the improved filter order; Input the process parameters into a filtering module, apply an improved filter order to perform notch processing to obtain filtered process parameters, and input the filtered process parameters into a parameter optimization module to obtain optimized process parameters; Among them, the improved filter order satisfies the relationship: ; In the formula, is the improved filter order, is the drift degree of the main frequency of the process parameters, The main frequency of the process parameters may be periodic. As parameters, is an exponential function with the natural constant e as base, is the ceiling function.

[0007] The present invention analyzes the degree of drift of the main frequency of the process parameters and the possibility of periodicity of the main frequency, so that a certain frequency range can be reserved during the design of the filter to cope with slight changes in the main frequency, and these periodic factors can be optimized to reduce the noise caused by periodic interference. The filtered process parameters can more accurately reflect the processing requirements of the machine tool, thereby guiding the machine tool to perform more precise processing.

[0008] Preferably, Fourier transform is performed on the process parameters to convert the time domain signal into a frequency domain signal, and the main frequency is extracted from the frequency domain signal.

[0009] The present invention takes into account that in the actual processing process, the main frequency of the process parameters will be affected by various influencing factors, resulting in drift and periodic changes. By calculating the drift degree of the main frequency and the possibility of periodicity, the filter order is dynamically adjusted, so that when the improved filter order is used to filter the parameters subsequently, the interference and noise caused by mechanical resonance can be eliminated or weakened, thereby significantly improving the filtering accuracy and parameter optimization effect.

[0010] Preferably, the process of obtaining the drift degree of the main frequency of the process parameter includes: Decomposing the main frequency into multiple components, calculating the degree of change of each component and the possibility of noise components of each component; The noise component possibility of each component is used as a weight, and a weighted mean of the degree of change of all components is calculated and used as the drift degree of the main frequency.

[0011] Decomposing the main frequency into multiple components can analyze the change of the main frequency in more detail. Each component may represent the characteristics of the main frequency in different frequency bands or different time periods. Calculating the degree of change of each component can quantify the fluctuation of the main frequency in different aspects, which helps to identify the main sources and trends of the main frequency drift.

[0012] Preferably, the degree of change of each component satisfies the relationship: ; In the formula, For the The degree of change of the For the The first The value corresponding to each time point, For the The first The value corresponding to each time point, is the total number of components decomposed from the main frequency, For the The total number of time points in a component.

[0013] By comparing the degree of change between different components, the component with the most significant change can be identified, and the filter design can be optimized based on the change characteristics of the component.

[0014] Preferably, the noise component of each component may satisfy the relationship: ; In the formula, For the The noise component probability of the components, For the The variance of the values ​​corresponding to all time points in a component, For the The component and The Pearson correlation coefficient of the components, The total number of components decomposed into the main frequency.

[0015] In the main frequency decomposition process, by calculating the noise component likelihood of each component, it is possible to identify which of the main frequency components are more likely to be contaminated by noise. For components with a high probability of being noise components, stricter filters can be applied to reduce the impact of noise.

[0016] Preferably, the possibility that the main frequency of the process parameter has periodicity satisfies the relationship: ; In the formula, The main frequency has periodicity. For the The maximum point and the The first-order difference value of the distance between the maximum points, The total number of maximum points in the main frequency.

[0017] By analyzing the periodicity of the main frequency, regular fluctuations in the process parameters can be identified, and then these fluctuations can be removed through notch processing, thereby improving the stability and accuracy of the processing.

[0018] Preferably, the degree of change of each component satisfies the relationship: ; In the formula, For the The degree of change of the For the The first The value corresponding to each time point, For the The first The value corresponding to each time point, For the The total number of time points in a component.

[0019] Preferably, the optimization adopts a genetic algorithm or a particle swarm optimization algorithm.

[0020] The beneficial effects of the present invention are: The improved filter order is calculated based on the degree of main frequency drift of process parameters, the possibility of periodicity of main frequency and specific parameters, realizing adaptive optimization of the filter. This adaptive optimization can be flexibly adjusted according to the characteristics of different process parameters, improving the filtering effect and further improving the processing accuracy.

[0021] By filtering the process parameters, noise and interference can be effectively removed, and the stability and reliability of the signal can be improved. At the same time, by optimizing the filtered process parameters through the parameter optimization module, more accurate processing parameters can be obtained, thereby improving processing accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 It is a method flow chart of steps S1 to S3 in the optimization method of machining process parameters based on intelligent CNC machine tools in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0024] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0025] The application scenario of the present invention is: firstly, the parameter data is processed by using a notch filter, and then the processed parameter data is further optimized.

[0026] Reference Figure 1The optimization method based on the processing parameters of intelligent CNC machine tools includes steps S1 to S3, which are as follows: S1: Obtain the process parameters of the intelligent CNC machine tool.

[0027] In the background database of intelligent CNC machine tools, historical data within the past month of the optimization parameters required for the intelligent CNC machine tool processing technology are obtained.

[0028] Specifically, the process parameters obtained include cutting parameters (cutting speed, feed speed, etc.) and tool parameters (tool radius, etc.), and the acquired data is cleaned to remove invalid or abnormal data.

[0029] Among them, any parameter such as cutting speed is taken as an example for analysis.

[0030] S2: Get the improved filter order.

[0031] During the machining process, the resonant frequency of the machine tool will change dynamically due to the influence of many factors. For example, during the cutting process, the cutting force will change with the change of cutting parameters (such as cutting speed, feed rate, cutting depth, etc.). The change of cutting force will cause the rigidity and dynamic response of the machine tool system to change, thus affecting the resonant frequency.

[0032] The dynamic change of the machine tool resonant frequency may cause vibration and deformation during the machining process, thus affecting the machining accuracy. It will also cause the cutting parameters to be constantly adjusted, thus reducing the machining efficiency.

[0033] In the embodiment of the present invention, a notch filter is used during the machining process of a machine tool to eliminate or weaken the influence of the mechanical resonance of the machine tool on the machining parameters. By reducing the interference of the resonance frequency on the machining process, the machining accuracy and efficiency can be improved.

[0034] First, the process parameters are Fourier transformed to convert the time domain signal into a frequency domain signal, and the main frequency is extracted from the frequency domain signal, that is, the frequency component that occupies a dominant position or has the highest energy in the frequency domain signal.

[0035] Secondly, when the main frequency of the process parameters changes or drifts, that is, the frequency drift of the main frequency, the frequency drift causes the filter to be unable to accurately locate the main frequency of the process parameters, thus affecting the filtering effect. In order to cope with frequency drift, the filter needs to be able to adapt to these changes to continue to filter effectively and keep tracking the target frequency. When the main frequency of the process parameters drifts significantly, a simple low-order filter may not be able to cope with it, so it is necessary to adjust the order of the filter to improve accuracy and response speed.

[0036] In addition, the frequency components in the frequency signal will change periodically over time. If the signal frequency changes continuously within a certain period, the filter design must be able to dynamically adjust to keep up with these periodic changes. When the frequency changes very frequently (that is, the period is very short), the signal may contain more complex frequency components. In order to process these complex frequency components, a higher filter order may be required. High-order filters have better frequency selectivity and faster response speed, so they are more suitable for processing signals with frequent frequency changes.

[0037] In summary, by adjusting the order of the filter according to the frequency drift degree of the main frequency and the periodicity of the frequency change, the performance of the filter can be significantly improved, making it more effective in filtering out interference signals related to the mechanical resonance of the machine tool.

[0038] The calculation process of the frequency drift degree of the main frequency is as follows: Perform empirical mode decomposition (EMD) on the main frequency to obtain a series of IMF components; Calculate the degree of change of each component, that is, satisfy the relationship:

[0039] In the formula, For the The degree of change of the For the The first The value corresponding to each time point, For the The first The value corresponding to each time point, is the total number of components decomposed from the main frequency, For the The total number of time points in a component. The larger the value, the The more drastic the change of a component relative to other components, the more The components contain more noise or abnormal information. By calculating the degree of change of each component, it is possible to identify which components may contain noise, so as to perform corresponding filtering operations in signal processing.

[0040] In one embodiment, another method is provided to calculate the degree of change of a component by considering the change of the component itself between different time points, that is, satisfying the relationship:

[0041] In the formula, For the The degree of change of the For the The first The value corresponding to each time point, For the The first The value corresponding to each time point, For the The total number of time points in a component.

[0042] Calculate the noise component possibility of each component, that is, satisfy the relationship:

[0043] In the formula, For the The noise component probability of the components, For the The variance of the values ​​corresponding to all time points in a component, For the The component and The Pearson correlation coefficient of the components, The total number of components decomposed into the main frequency.

[0044] For components, whose variance represents the average of the squares of the deviations of the component data points from their mean. The larger the variance, the more dispersed the data points of the component, that is, the higher the randomness of the component. If the Pearson correlation coefficient between two different components is close to 1 or -1, it means that there is a very light linear correlation between the two components. If it is close to 0, it means that there is almost no linear correlation between the two components.

[0045] It comprehensively considers the randomness of the component (measured by the variance) and its correlation with other components (measured by the Pearson correlation coefficient). In practical applications, this indicator can be used to identify and remove the noise component in the signal to obtain a purer signal.

[0046] After calculating the degree of change of each component and the possibility of the noise component of each component, the possibility of the noise component of each component is used as the weight, and the weighted mean of the degree of change of all components is calculated and used as the degree of drift of the main frequency. That is, the degree of drift of the main frequency satisfies the relationship:

[0047] In the formula, is the drift degree of the main frequency, For the The degree of change of the For the The noise component probability of the components, The total number of components decomposed into the main frequency.

[0048] The calculation process of the possibility of periodic changes in the main frequency is as follows: Obtain all the maximum points in the main frequency and the position of each maximum point in the main frequency, and calculate the distance between adjacent maximum points (this distance represents the interval between two adjacent frequency components in the frequency domain).

[0049] Furthermore, the first-order difference of the distances between adjacent maximum points is performed, that is, the change between adjacent distances is calculated, and the results of the first-order difference are accumulated and calculated (the accumulated sum reflects the overall change trend of the distances between adjacent maximum points).

[0050] Furthermore, if the cumulative sum is close to zero or shows periodic fluctuations, then it is considered that the main frequency is likely to be periodic. This is because periodic signals are usually manifested as a series of equally spaced frequency components in the frequency domain, and these components form a regular distribution of maximum points in the spectrum.

[0051] Then the possibility of periodicity of the main frequency satisfies the relationship:

[0052]

[0053] In the formula, The main frequency has periodicity. For the The maximum point and the The first-order difference value of the distance between the maximum points, is the total number of maximum points in the main frequency; For the The maximum point and the The distance between the maximum points, For the The value of the maximum point, For the The value of the maximum point, For the The location of the maximum point, For the The location of the maximum point, It is an exponential function with the natural constant e as its base.

[0054] in, The smaller the value of means, the smaller the distance between adjacent maximum points is, that is, these maximum points are more evenly distributed to some extent. In signal processing or data analysis, if the distribution of maximum points is more regular, the probability that the main frequency they represent exists is higher.

[0055] Finally, the order of the filter is improved according to the drift degree of the main frequency and the possibility of periodicity of the main frequency.

[0056] Specifically, the higher the drift of the main frequency, the higher the required filter order is to increase the filter response; the higher the possibility of periodicity of the main frequency, the higher the required filter order is to ensure that the filter can keep up with such periodic changes.

[0057] Then the improved filter order satisfies the relationship:

[0058] In the formula, is the improved filter order, is the drift degree of the main frequency of the process parameters, The main frequency of the process parameters may be periodic. As parameters, is an exponential function with the natural constant e as base, is the ceiling rounding function. .

[0059] S3: inputting the process parameters into a filtering module, applying an improved filter order to perform notch processing to obtain filtered process parameters, and inputting the filtered process parameters into a parameter optimization module to obtain optimized process parameters.

[0060] The process parameters to be optimized, such as cutting speed, are input into the filter module of the built-in notch filter. According to the filter order improved by S2, the notch filter will filter the input parameters to remove noise or interference components.

[0061] The filtered parameters will be input into the parameter optimization module, which has built-in optimization algorithms (such as genetic algorithm, particle swarm optimization algorithm, etc.) to further optimize the parameters so that the performance indicators of the processing technology reach the optimal or near-optimal level.

[0062] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0063] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. An optimization method based on processing parameters of intelligent CNC machine tools, characterized in that: include: Obtain the process parameters of intelligent CNC machine tools; Get the improved filter order; Input the process parameters into a filtering module, apply an improved filter order to perform notch processing to obtain filtered process parameters, and input the filtered process parameters into a parameter optimization module to obtain optimized process parameters; Among them, the improved filter order satisfies the relationship: ; In the formula, is the improved filter order, is the drift degree of the main frequency of the process parameters, The main frequency of the process parameters may be periodic. As parameters, is an exponential function with the natural constant e as base, is the ceiling function.

2. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 1 is characterized in that: Perform Fourier transform on the process parameters, convert the time domain signal into the frequency domain signal, and extract the main frequency from the frequency domain signal.

3. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 2 is characterized in that: The process of obtaining the drift degree of the main frequency of the process parameters includes: Decomposing the main frequency into multiple components, calculating the degree of change of each component and the possibility of noise components of each component; The noise component possibility of each component is used as a weight, and a weighted mean of the degree of change of all components is calculated and used as the drift degree of the main frequency.

4. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 3 is characterized in that: The degree of change of each component satisfies the relationship: ; In the formula, For the The degree of change of the For the The first The value corresponding to each time point, For the The first The value corresponding to each time point, is the total number of components decomposed from the main frequency, For the The total number of time points in a component.

5. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 4 is characterized in that: The noise component of each component is likely to satisfy the relationship: ; In the formula, For the The noise component probability of the components, For the The variance of the values ​​corresponding to all time points in a component, For the The component and The Pearson correlation coefficient of the components, The total number of components decomposed into the main frequency.

6. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 3 is characterized in that: The possibility that the main frequency of the process parameters is periodic satisfies the relationship: ; In the formula, The main frequency has periodicity. For the The maximum point and the The first-order difference value of the distance between the maximum points, The total number of maximum points in the main frequency.

7. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 3 is characterized in that: The degree of change of each component satisfies the relationship: ; In the formula, For the The degree of change of the For the The first The value corresponding to each time point, For the The first The value corresponding to each time point, For the The total number of time points in a component.

8. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 1 is characterized in that: The optimization adopts a genetic algorithm or a particle swarm optimization algorithm.

Images