Optimization Method and System for Machining Process Parameters Based on Intelligent Numerical Control Machine Tools

Abstract

The present invention relates to the technical field of data processing, and particularly relates to an optimization method and system for machining process parameters based on intelligent numerical control machine tools. The method includes: collecting the feed speed of the numerical control machine tool at each moment, setting the initial perturbation range in the process of the simulated annealing algorithm iteratively searching for the optimal solution of the feed speed at each moment, and the reference data segment of the feed speed at each moment, dynamically adjusting the initial perturbation range according to the fluctuation degree of the reference data segment, and in each round of iterative process, constructing an objective function to evaluate the preference degree of the current solution and the new solution, so as to determine the perturbation range of the current solution, realizing the optimization of the iterative search process, obtaining the optimal machining parameters at each acquisition moment, enabling the algorithm to find the optimal feed speed at each acquisition moment faster and more accurately, and improving the optimization efficiency of the machining process parameters.

Description

Technical Field

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

[0002] As the core pillar of the equipment manufacturing industry, the technical level of numerical control machine tools has a decisive impact on the competition pattern in the industrial field. During the operation of numerical control machine tools, the key process parameter of the machine tool feed rate has traditionally relied on manual experience for setting. In actual machining practice, due to the diverse materials of different workpieces, the complex and variable machining processes, and the rich variety of tool types, the superposition of many factors makes it difficult for the feed rate value determined solely by manual experience to adapt to the dynamically changing machining requirements.

[0003] The prior art obtains the optimal solution of the feed rate of numerical control machine tools through the simulated annealing algorithm. This algorithm is based on simulating the heating and cooling process of a solid, searching for the global optimal solution in a random search space, and gradually reducing the randomness of the search by controlling the temperature to achieve parameter optimization. The metal cutting machining path planning system disclosed in the Chinese patent application document with the publication number CN119126668A includes multiple functional modules: the image acquisition module is responsible for capturing high-resolution images of the metal workpiece to be machined from multiple angles; the 3D reconstruction module reconstructs a 3D model using these images; the cutting simulation module simulates the 3D model in a virtual environment and records the stress and heat distribution during the cutting process in real time; the intelligent segmentation module divides the machining process into multiple stages based on stress and heat data; the comprehensive optimization module combines the genetic algorithm and the simulated annealing algorithm to optimize the cutting path, depth, and feed rate of each machining stage; the feedback adjustment module monitors the stress and heat changes during actual machining, and once the threshold is exceeded, it automatically adjusts the cutting parameters or pauses the machining.

[0004] However, when the above technical solution uses the simulated annealing algorithm to solve the feed rate, there is a key problem: the perturbation range of the simulated annealing algorithm is preset and fixed, and it cannot adapt to different machining task scenarios, reducing the machining accuracy and affecting the machining effect of numerical control machine tools. Specifically, it is manifested as follows: the fixed perturbation range will cause the process of the algorithm converging to the optimal solution to be extremely slow and waste a lot of time when facing scenarios with relatively simple machining tasks and obvious rules; while when dealing with tasks with high complexity and strict machining requirements, it is easy to miss potential better solutions due to insufficiently fine search. Summary of the Invention

[0005] To solve the problem that in obtaining the optimal feed rate of a numerically controlled machine tool, due to the setting of a fixed perturbation range, the traditional simulated annealing algorithm cannot adapt to different machining task scenarios, reducing machining accuracy and affecting the machining effect of the numerically controlled machine tool, the present invention proposes an optimization method based on the machining process parameters of an intelligent numerically controlled machine tool, including:

[0006] Collect the feed rates of the numerically controlled machine tool at each moment, set the initial perturbation range in the process of the simulated annealing algorithm iteratively searching for the optimal solution of the feed rate at each moment, and the reference data segment of the feed rate at each moment, analyze the fluctuation degree of the reference data segment, and dynamically adjust the initial perturbation range using the fluctuation degree to obtain the adjusted perturbation range;

[0007] In the process of iterative search, evaluate the preference degree of the current solution generated in each round and the new solution obtained by perturbing the current solution, and determine the perturbation range requirement factor of the current solution based on the difference in the preference degree of the current solution and the new solution, where the perturbation range requirement factor is used to reflect the magnitude of the adjustment requirement of the current solution;

[0008] Use the perturbation range requirement factor of the current solution generated in each round to correct the adjusted perturbation range to obtain the perturbation range of the current solution generated in each round, so as to optimize the process of the simulated annealing algorithm iteratively searching for the optimal solution of the feed rate at each moment, obtain the optimal solution of the feed rate at each moment, and optimize the machining process parameters of the numerically controlled machine tool according to the optimal solution of the feed rate at each moment.

[0009] In the above technical solution, the initial perturbation range provides a starting exploration space for the search of the simulated annealing algorithm and is the basis for the algorithm to start optimizing. The reference data segment reflects the state of the feed rate with the help of historical data. A large fluctuation degree means complex machining conditions and requires a larger perturbation range to explore a wider solution space. If the fluctuation is small, it indicates that the machining state is relatively stable, and the perturbation range can be reduced to accelerate the algorithm convergence. By dynamically adjusting the perturbation range, the simulated annealing algorithm can adapt to different machining scenarios, maintain sufficient search ability under complex working conditions, avoid missing the optimal solution, and thus improve the overall machining efficiency and quality. Furthermore, in the process of iterative search for the optimal solution, the adjusted perturbation range based on the fluctuation degree is combined with the perturbation range requirement factor determined based on the preference degree of the current solution and the new solution, enabling the algorithm to dynamically adjust the perturbation range according to the quality of the current solution and the new solution. This search strategy can find the optimal solution at each moment faster and more accurately. Optimize the process parameters of the numerically controlled machine tool during machining based on the optimal solution at each moment to keep it in the best operating state and ensure the machining effect of the numerically controlled machine tool.

[0010] Preferably, the perturbation range of the current solution generated in each round is determined based on the following method:

[0011] Multiply the adjusted perturbation range in the process of the simulated annealing algorithm iteratively searching for the optimal solution of the feed rate at each moment by the perturbation range demand factor of the current solution generated in each round to correct the adjusted perturbation range, and use the multiplied value as the perturbation range of the current solution generated in each round.

[0012] The above technical solution combines the adjusted perturbation range with the perturbation range demand factor of the current solution. In different machining scenarios and iteration rounds, the algorithm can flexibly adjust the perturbation range according to the actual situation.

[0013] Preferably, the method for evaluating the preference degree of the current solution generated in each round and the new solution obtained by perturbing the current solution is as follows:

[0014] Construct an objective function for reflecting the preference degree of the current solution / new solution:

[0015]

[0016] Wherein, is the objective function value of the current solution / new solution. Both the current solution / new solution are the feed rates, and the objective function value has a negative correlation with the preference degree. is the energy consumption per unit time corresponding to the current solution / new solution. is the production capacity per unit time corresponding to the current solution / new solution.

[0017] The above technical solution constructs a specific objective function, takes into account the energy consumption per unit time and the production capacity, can intuitively reflect the advantages and disadvantages of the current solution and the new solution, provides a quantitative basis for deciding whether to accept the new solution. Since the objective function value has a negative correlation with the preference degree, it is possible to clearly compare the quality of different solutions during the iteration process, guide the algorithm to search in a more optimal direction, and help improve the optimization efficiency.

[0018] Preferably, the perturbation range demand factor of the current solution satisfies the following relational expression:

[0019]

[0020] In the formula, is the perturbation range demand factor of the current solution generated in the th round of iteration in the process of iteratively searching for the optimal solution of the feed rate at the th moment. is the normalization function. and are respectively the objective function values of the current solution and the new solution generated in the th round of iteration. In the process of iteratively searching for the optimal solution of the feed rate at the -th moment, the temperature value after the -th round of iteration, is the temperature threshold set in the process of iteratively searching for the optimal solution of the feed rate at the -th acquisition moment.

[0021] By incorporating the difference in the objective function values of the current solution and the new solution generated in the previous round, as well as factors such as the temperature value, the above technical solution can accurately reflect the magnitude of the adjustment requirement of the current solution.

[0022] Preferably, the degree of fluctuation of the reference data segment is determined by the following method:

[0023] Among all the feed rates included in the reference data segment of the feed rate at each moment, if there is a numerical difference between a certain feed rate and the previous feed rate, mark the feed rate and obtain the numerical change amount of the marked feed rate;

[0024] Calculate the degree of fluctuation: ; where is the degree of fluctuation of the reference data segment of the feed rate at the -th moment, is the normalization function, is the number of feed rates in the reference data segment, is the -th moment, is the total number of all marked feed rates in the reference data segment of the feed rate at the -th moment,

[0025] The above technical solution not only considers the numerical change amount of the feed rate, but also considers the proportion of the number of feed rates with numerical changes. This way of comprehensively considering multiple factors avoids the one-sidedness brought by single-factor evaluation and more comprehensively and accurately quantifies the fluctuation characteristics of the reference data segment.

[0026] Preferably, the method for dynamically adjusting the initial perturbation range by using the degree of fluctuation to obtain the adjusted perturbation range is:

[0027] Multiply the initial perturbation range in the process of iteratively searching for the optimal solution of the feed rate at each moment by the degree of fluctuation of the reference data segment of the feed rate at each moment to dynamically adjust the initial perturbation range, and use the multiplied value as the adjusted perturbation range.

[0028] By multiplying the degree of fluctuation by the initial disturbance range and dynamically adjusting the disturbance range according to the fluctuation characteristics of the actual processing data, the above technical solution is beneficial to avoiding falling into local optimal solutions while ensuring the search accuracy, and improving the efficiency of finding the global optimal feed rate.

[0029] Preferably, the method for setting the reference data segment of the feed rate at each moment is as follows:

[0030] The feed rates at the moments immediately preceding the feed rate at each moment form the reference data segment of the feed rate at that moment, where

[0031] is a preset value.

[0032] Preferably, a method for optimizing the machining process parameters of a numerically controlled machine tool according to the optimal solution of the feed rate at each moment is as follows:

[0033] Set a fixed time interval. During the machining process, every time a time interval elapses, obtain the optimal machining parameters of the feed rate at the last moment of the time interval, and set the feed rate of the numerically controlled machine tool according to the optimal machining parameters.

[0034] Preferably, another method for optimizing the machining process parameters of a numerically controlled machine tool according to the optimal solution of the feed rate at each moment is as follows:

[0035] The present invention also provides an optimization system for machining process parameters based on an intelligent numerically controlled machine tool. The optimization system for the process parameters includes a memory and a processor. A computer program is stored on the memory, and the processor executes the computer program to implement the steps of the optimization method.

[0036] The present invention has the following effects:

[0037] During the process of the simulated annealing algorithm iteratively searching for the optimal solution of each feed rate, the present invention first dynamically adjusts the initial disturbance range of the simulated annealing algorithm, and then constructs an objective function to further evaluate the preference degree of the current solution and the new solution generated in each round of the iterative process to determine the disturbance range of the current solution. This series of operations enables the simulated annealing algorithm to better adapt to different machining task scenarios, avoid excessive search and accelerate convergence in simple tasks, enhance the search ability and not miss potential optimal solutions in complex tasks, effectively improve the optimization effect of the machining parameters of the numerically controlled machine tool, improve the machining accuracy and overall machining quality, and enhance the adaptability and stability of the numerically controlled machine tool in various machining processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] 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 drawings, several embodiments of the present invention are shown in an exemplary rather than restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0039] Figure 1 is a schematic flowchart of the method of the present invention. Specific Embodiments

[0040] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0041] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0042] Referring to Figure 1 , the optimization method of machining process parameters based on intelligent numerical control machine tools provided by the present invention includes steps S1 - S4:

[0043] S1: Collect the feed rate of the numerical control machine tool.

[0044] In one embodiment, a high-precision speed sensor is used to collect data on the feed rate of the numerical control machine tool, and the collection frequency is 10 times per second to ensure that rich and representative feed rate data is obtained, providing a data basis for subsequent optimization of process parameters.

[0045] S2: Set the reference data segment of the feed rate at each moment, and adjust the initial disturbance range of the feed rate by using the fluctuation degree of the reference data segment.

[0046] The simulated annealing algorithm is derived from the simulation of the physical annealing process. At high temperatures, particles have high energy and can move freely within a large range. As the temperature decreases, the particles gradually stabilize in the state with the lowest energy. When solving problems, the algorithm randomly searches within the neighborhood of the current solution and accepts new solutions with a certain probability to avoid falling into local optimal solutions.

[0047] In the simulated annealing algorithm, the perturbation range is a core parameter. It defines the boundary of the neighborhood that can be explored starting from the current solution, thereby determining the size of the neighborhood of the current solution. The neighborhood is the solution space around the current solution and is the local space for searching the optimal solution. The optimal solution may lie within this neighborhood. The simulated annealing algorithm is based on the current solution and searches for possible new solutions within the neighborhood according to the perturbation range. The larger the perturbation range, the greater the change in the current solution, and the greater the difference between the new solution and the current solution; conversely, it is smaller. The new solution may be better or worse than the current solution. The simulated annealing algorithm will accept the new solution with a certain probability. Even if the new solution is worse, there is still a chance of being accepted.

[0048] In the current scenario, the traditional simulated annealing algorithm sets a fixed and identical initial perturbation range for all feed rates , with the unit of , and this value is an empirical value. This initial perturbation range provides an initial exploration space scale for the search of the algorithm. The parameters in the simulated annealing algorithm also include: the initial temperature value is set to 100 degrees Celsius, the temperature threshold is set to 50 degrees Celsius, the cooling coefficient is 0.97, and the initial solution (the current solution in the first round of iteration) when iteratively searching for the optimal solution of the feed rate at each moment is set to 500, with the unit of , and these values are all empirical values.

[0049] In the initial stage of the simulated annealing algorithm, based on this initial perturbation range, a certain degree of random perturbation is applied to each feed rate to generate new solutions. By comparing the quality of the new solutions and the current solution, and accepting the new solution according to a certain probability and other mechanisms, it gradually searches and attempts to find the optimal solution of the feed rate at each moment. If the initial perturbation range is set too large, it may cause the algorithm to be too aggressive in the initial stage of the search, skipping some potential high-quality solution regions; while if it is set too small, the algorithm may get stuck in a long-term search in the local area, difficult to jump out of the local optimal solution and unable to fully explore the entire solution space.

[0050] Therefore, this solution first obtains the fluctuation degree of the reference data segment (equivalent to the neighborhood of the feed rate) of the feed rate at each moment, and uses this fluctuation degree to dynamically adjust the initial perturbation range in the process of the simulated annealing algorithm iteratively searching for the optimal solution of the feed rate at this moment, obtaining the adjusted perturbation range. And further, in the process of iteratively searching for the optimal solution of the feed rate at the acquisition moment, accurately obtain the perturbation range demand factor of the current solution generated in each round, and use the perturbation range demand factor of the current solution to perform a secondary correction on the adjusted perturbation range to obtain the perturbation range of the current solution generated in each round, so as to optimize the process of the simulated annealing algorithm iteratively searching for the optimal solution of the feed rate at each moment.

[0051] In one embodiment, the method for setting the reference data segment of the feed rate at each moment is as follows:

[0052] The feed rates at the moments immediately preceding the feed rate at each moment form the reference data segment of the feed rate at the acquisition moment. , this value is an empirical value. If the number of feed rates before a certain moment is less than 100, then the feed rate 60 seconds in advance can be used to ensure that there is enough data for constructing the reference data segment.

[0053] After obtaining the reference data segment of the feed rate at each moment, it is necessary to analyze the degree of fluctuation of the reference data segment. The more frequent and obvious the numerical changes within a reference data segment are, the greater the degree of fluctuation indicates.

[0054] In one embodiment, the method for obtaining the degree of fluctuation of the reference data segment of the feed rate at each moment includes:

[0055] For the reference data segment of the feed rate at the th acquisition moment, the following judgment is made among all the feed rates included in this reference data segment:

[0056] If there is a numerical difference between a certain feed rate and the previous feed rate, then mark this feed rate and obtain the numerical change amount of the marked feed rate.

[0057] For example, if all the feed rates included in this reference data segment are:

[0058] 10, 12, 12, 13. It can be seen that the first 12 and 13 have numerical differences from their respective previous feed rates. Mark the first 12 and 13. The numerical change amount of the first 12 is , and the numerical change amount of 13 is .

[0059] Calculate the degree of fluctuation based on the above judgment results:

[0060]

[0061] In the formula, is the degree of fluctuation of the reference data segment of the feed rate at the th moment, is the number of feed rates in the reference data segment, is the total number of all marked feed rates in the reference data segment of the feed rate at the th acquisition moment, is the variance of the numerical change amounts of all marked feed rates in the reference data segment of the feed rate at the th acquisition moment. is a normalization function.

[0062] In this formula, The larger it is, it indicates that in the reference data segment of the feed rate at the

[0063] In this formula, The larger it is, it shows that in the reference data segment of the feed rate at the

[0064] After obtaining the fluctuation degree of the reference data segment of the feed rate at each moment, it is necessary to adaptively adjust the initial disturbance range of the feed rate at this moment according to this fluctuation degree.

[0065] In one embodiment, the disturbance range of the feed rate at each moment is adaptively adjusted according to the following formula:

[0066]

[0067] In the formula, is the adjusted disturbance range of the feed rate at the th moment, which is the preliminary adaptive result of the initial disturbance range of the feed rate at the th moment, is the initial disturbance range uniformly set by the simulated annealing algorithm for the feed rates at all moments, represents the fluctuation degree of the reference data segment of the feed rate at the

[0068] If the fluctuation degree in the reference data segment of the feed rate at a certain moment is larger, it means that in the process of obtaining the optimal solution of the feed rate at this moment by using the simulated annealing algorithm, the feed rates included in the reference data segment of this feed rate are more discrete. In order to maintain the flexibility of the search under large data fluctuations, avoid falling into local extrema and more easily find the global optimal solution, it is necessary to set a larger disturbance range for the feed rate at this moment. Therefore, this formula is designed so that when the th moment has a larger fluctuation degree of the reference data segment of the feed rate, the th moment has a larger adjusted disturbance range of the feed rate.

[0069] S3: When iteratively searching for the optimal solution of each feed rate, further correct the perturbed range after adjusting the feed rate to optimize the process of iteratively searching for the optimal solution of the feed rate at each moment by the simulated annealing algorithm.

[0070] Although the previous steps obtained the preliminary adaptive results of the initial perturbed range of the feed rate at each moment, that is, the perturbed range after adjusting the feed rate. However, since the process of obtaining the optimal solution of the feed rate at each moment by the simulated annealing algorithm is an iterative search process, in each round of iteration, the perturbed range determines the size of the range for generating a new solution from the current solution (the size of the solution search space). If the perturbed range after adjusting the feed rate is used as the standard during the iterative search process, it is impossible to perform dynamic adaptation according to the actual situation during the iteration process, and it is very likely that the perturbed range is too large or too small, affecting the search efficiency of the optimal solution. That is to say, the initially obtained perturbed range after adjusting the feed rate at each moment does not accurately match the optimal perturbed range required for each round of iteration.

[0071] Based on the principle of the simulated annealing algorithm, when iteratively searching for the optimal solution of the feed rate at each moment, a current solution will be generated in each round of iteration. Based on the current solution, a perturbation operation is performed on it according to the perturbed range, and a new solution will be generated. In this process, the optimal perturbed range of the current solution generated in each round of iteration is not fixed, but is usually closely related to two factors:

[0072] One is the difference in the objective function values corresponding to the current solution and the new solution generated in the previous round of iteration. Because the simulated annealing algorithm simulates the physical annealing process and gradually searches for the optimal solution through iteration, the difference in the objective function values of the current solution and the new solution generated in each round of iteration reflects whether the current search direction is approaching the optimal solution and the degree of approach. If the difference is large, it indicates that the current perturbed range may make the search span larger, and in subsequent iterations, the perturbed range may need to be adjusted appropriately according to the situation to improve the search efficiency. On the contrary, if the difference is small, it means that the search may have approached a relatively optimal region, and at this time, the perturbed range needs to be finely adjusted to avoid missing the optimal solution.

[0073] The other is the temperature value corresponding to each round of iteration. The temperature value in the simulated annealing algorithm is a key control parameter, which determines the probability that the algorithm accepts a worse solution. As the iteration progresses, the temperature gradually decreases, and the algorithm is more and more inclined to accept better solutions. In the high-temperature stage, a larger perturbed range is allowed to explore in a wider solution space; in the low-temperature stage, in order to converge to the optimal solution, the perturbed range should be reduced accordingly. Therefore, the temperature value directly affects the optimal perturbed range of the current solution in each round of iteration.

[0074] Therefore, in this step, when iteratively searching for the optimal solution at each feed rate, by deeply analyzing the difference between the objective function value of the new solution generated in each round and the objective function value of the current solution generated in each round, as well as the temperature value after each round of iteration, the perturbation range requirement factor of the current solution generated in each round is determined to reflect the magnitude of the adjustment requirement of the current solution. Then, the perturbation range requirement factor of the current solution is used to further correct the perturbation range after the adjustment of the feed rate, so as to obtain the perturbation range applicable to the current solution generated in each round, in order to optimize the process of iteratively searching for the optimal solution of the feed rate at each moment by the simulated annealing algorithm.

[0075] In one embodiment, during the iterative search process, an objective function is constructed to evaluate the preference degree of the current solution generated in each round and the new solution obtained by perturbing the current solution.

[0076] The constructed objective function is:

[0077]

[0078] In this formula, is the objective function value of the current solution / new solution, and the objective function value has a negative correlation with the preference degree.

[0079] In this formula, is the energy consumption per unit time corresponding to the current solution / new solution. Specifically, for a certain feed rate (whether it is the feed rate corresponding to the current solution or the feed rate corresponding to the new solution obtained by perturbing the current solution), when processing a workpiece with the same cutting area as that processed by the feed rate at the th acquisition moment, the energy consumed within one minute can be accurately obtained from a large amount of past historical data by means of big data technology. The larger the value of , the larger the objective function value

[0080] This indicates that in the process of screening the optimal solution, the higher the feed rate, the more difficult it is to become an ideal optimal solution, that is, the lower the preference degree. In this formula, is the production capacity per unit time corresponding to the current solution / new solution. Specifically, for a certain feed rate (whether it is the feed rate corresponding to the current solution or the feed rate corresponding to the new solution obtained by perturbing the current solution), when processing a workpiece with the same cutting area as that processed by the feed rate at the th acquisition moment, the length of the workpiece that can be processed within one minute. This value can also be extracted from historical data using big data technology. The higher the feed rate value, the higher its priority during the screening process and the more likely it is to become the optimal solution ultimately sought.

[0081] In one embodiment, when iteratively searching for the optimal solution of each feed rate, the method for obtaining the perturbation range demand factor of the current solution generated in each round is as follows:

[0082] During the process of iteratively searching for the optimal solution of the feed rate at the th acquisition moment, if the difference between the objective function value of the current solution and the objective function value of the new solution generated in the th round of iteration is greater, it indicates that the preference degree of the new solution generated in the th round of iteration is higher, which means that the new solution obtained in the th round of iteration is closer to the optimal solution. At this time, there is no need to conduct a large-scale search anymore, so in the subsequent process, it is necessary to continue the search within a relatively small range, and then the perturbation range demand factor of the new solution obtained in the th round of iteration is smaller. On the contrary, if the difference between the objective function value of the current solution and the objective function value of the new solution generated in the th round of iteration is small, it indicates that the new solution generated in the th round of iteration is not significantly better than the current solution, so in the subsequent process, it is necessary to continue the search within a relatively large range, and then the perturbation range demand factor of the new solution obtained in the th round of iteration is larger.

[0083] It is calculated according to the following relational expression:

[0084]

[0085] In the formula, is the perturbation range demand factor of the current solution generated in the th round of iteration during the process of using the simulated annealing algorithm to iteratively search for the optimal solution of the feed rate at the th moment, is the normalization function, and are respectively the objective function values of the current solution and the new solution generated in the th round of iteration, is the temperature value after the th round of iteration during the process of using the simulated annealing algorithm to iteratively search for the optimal solution of the feed rate at the th moment, is the temperature threshold set during the process of using the simulated annealing algorithm to iteratively search for the optimal solution of the feed rate at the th moment.

[0086] In the formula, during the process of iteratively searching for the In the process of obtaining the optimal solution of the feed rate at a collection moment, in the current solution generated in the round of iteration and the difference in the objective function values of the new solution respectively. The larger this value is, the greater the preference degree of the new solution obtained through random perturbation during the round of iteration, indicating that the new solution obtained through random perturbation during the round of iteration is closer to the optimal solution. To output the optimal solution more accurately, during the round of iteration, a relatively small perturbation range requirement factor is set for the current solution generated in the

[0087] Specifically: The difference between the objective function value of and the objective function value of

[0088] is large, including two cases: is very large, is very small, that is, the objective function value of the current solution generated in the round of iteration is large, and the objective function value of the new solution is small. Since the objective function value and the preference degree are negatively correlated, it indicates that during the round of iteration, the preference degree of the generated current solution is very low, and the preference degree of the new solution obtained after perturbing the current solution is very high. At this time, it shows that the new solution is closer to the optimal solution. Then, during the round of iteration, a smaller perturbation range should be set, that is, the perturbation range requirement factor of the current solution generated in the round of iteration is smaller.

[0089] At this time is a relatively large positive number. In the formula, a negative correlation relationship between is constructed through and , and The larger the value of , the greater the difference between the objective function value of and the objective function value of , and the perturbation range requirement factor of the current solution generated in the

[0090] The second case is is very small, is very large, that is, the objective function value of the current solution generated in the round of iteration is small, and the objective function value of the new solution is large. Since the objective function value and the preference degree are negatively correlated, it indicates that during the The preference level of the current solution generated by the round iteration is very high, while the preference level of the new solution obtained after perturbing the current solution is very low, indicating that the farther the new solution is from the optimal solution, then at the round of iteration, a larger perturbation range needs to be set, that is, at the round of iteration, the perturbation range requirement factor of the current solution generated is larger.

[0091] At this time is a very small negative number. In the formula, by a positive correlation relationship between and is constructed, and the smaller this negative number is, it indicates that the difference between the objective function value of and the objective function value of is larger. At the same time, the larger the value of the perturbation range requirement factor of the current solution generated at the round of iteration is.

[0092] In this formula, represents the magnitude of the temperature value after the round of iteration relative to the temperature threshold. The smaller this value is, it indicates that the round of iteration is closer to the last round of iteration. At this time, the simulated annealing algorithm tends to perform a more detailed local search around the current solution, reducing the scope of global search, and thus gradually converging to an optimal solution. Therefore, at the round of iteration, the requirement factor for the perturbation range of the current solution generated will be smaller.

[0093] In summary, the above solution enables the simulated annealing algorithm to dynamically adjust the perturbation range of the new solution generated in the current round according to the iteration result of the previous round and the temperature state of the current round, so as to improve the accuracy and efficiency of the algorithm's iterative search for the optimal solution and avoid the problems of falling into local optimality or improper search range.

[0094] In one embodiment, the perturbation range of the current solution generated in each round is determined based on the following method:

[0095] In the process of obtaining the optimal solution of the feed rate at each moment, the smaller the perturbation range requirement factor of the new solution generated in a certain round of iteration is, it indicates that the fluctuation amplitude of the reference data segment of the feed rate at this moment is smaller, and it can also reflect that the preference level of the new solution generated in this round of iteration is larger, and the temperature value corresponding to this round of iteration is lower. Then, a smaller perturbation range needs to be set for the new solution generated in this round of iteration to ensure a more accurate output of the optimal solution.

[0096] According to the above logic, the calculation formula for the perturbation range of the new solution generated in each round of iteration is:

[0097]

[0098] In the formula, represents the perturbation range of the current solution generated in the -th iteration during the process of searching for the optimal solution of the feed rate at the -th moment of iteration, is the perturbation range after adjustment of the feed rate at the -th acquisition moment, is the perturbation range requirement factor of the current solution generated in the -th iteration during the process of searching for the optimal solution of the feed rate at the -th acquisition moment by using the simulated annealing algorithm.

[0099] In this formula, since the perturbation range requirement factor is determined based on factors such as the difference in the objective function values of the current solution and the new solution, the perturbation range requirement factor can reflect the magnitude of the adjustment requirement of the current solution. In this way, in different machining scenarios and iteration rounds, the algorithm can flexibly adjust the perturbation range according to the actual situation. When the machining process is relatively complex and there may be a large deviation between the current solution and the optimal solution, a larger perturbation range requirement factor will prompt the perturbation range to increase appropriately, enabling the algorithm to have a larger search space to find a better solution and ensuring that the algorithm does not fall into a local optimum; while in the case where the machining is relatively stable and the current solution is already close to the optimal solution, a smaller perturbation range requirement factor will cause the perturbation range to shrink, enabling the algorithm to perform a more refined search when approaching the optimal solution, accelerating the algorithm convergence speed, improving the efficiency of finding the global optimal solution, and enhancing the dynamic adaptability of the simulated annealing algorithm to different machining task scenarios.

[0100] S4: Obtain the optimal solution of the feed rate at each moment according to the optimized iterative search process, and optimize the machining process parameters of the numerical control machine tool according to the optimal solution of the feed rate at each moment.

[0101] The process of iterative search for the feed rate at each moment by the simulated annealing algorithm is optimized through step S2 - step S3. When the iterative search ends, the optimal machining parameters at each moment are obtained.

[0102] In one embodiment, the method for optimizing the machining process parameters of the numerical control machine tool is:

[0103] According to the characteristics of the machining process and the performance of the machine tool, a fixed time interval is set, with an empirical value of 5 minutes. During the machining process, every 5 minutes, obtain the optimal solution of the feed rate at the last moment of this 5 - minute period according to step S2 - step S3, and set the feed rate of the numerical control machine tool according to this optimal solution to achieve the optimization of the machining process parameters of the numerical control machine tool.

[0104] In one embodiment, the method for optimizing the machining process parameters of a numerically controlled machine tool is as follows:

[0105] Determine the adjustment interval according to the progress of the machining task. Suppose the entire machining task is to machine a workpiece with a length of . It can be set that every time of the machining length is completed, a progress is reached. Obtain the optimal machining parameters at the corresponding moment when this progress is reached according to steps S2 - S3, and set the feed rate of the numerically controlled machine tool according to the optimal machining parameters to achieve the optimization of the machining process parameters of the numerically controlled machine tool.

[0106] By providing the most suitable feed rate for the machine tool at different machining stages in this way, it is ensured that the machine tool can maintain the best operating state throughout the machining process, avoiding adjusting the machining parameters too frequently, and at the same time, according to different situations in the machining process, timely providing the optimal feed rate for the numerically controlled machine tool to ensure the optimization of the machining effect and the performance of the machine tool.

[0107] In addition, the present invention also provides an optimization system for the machining process parameters of an intelligent numerically controlled machine tool. The intelligent teaching system includes a memory and a processor. A computer program is stored on the memory, and the processor executes the computer program to implement the steps of the optimization method.

[0108] In the description of this specification, the meanings of "a plurality of" and "several" are at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0109] Although this specification has shown and described multiple embodiments of the present invention, it is obvious to those skilled in the art that such embodiments are provided only by way of example. Those skilled in the art will think of many changes, alterations and alternative ways without departing from the spirit and idea of the present invention. It should be understood that various alternative solutions to the embodiments of the present invention described herein can be adopted in the process of practicing the present invention.

Claims

1. An optimization method based on processing parameters of intelligent CNC machine tools, characterized in that: include: Collecting the feed speed of the CNC machine tool at each moment, setting the initial disturbance range of the simulated annealing algorithm in the process of iteratively searching for the optimal solution of the feed speed at each moment, and the reference data segment of the feed speed at each moment, analyzing the fluctuation degree of the reference data segment, dynamically adjusting the initial disturbance range using the fluctuation degree, and obtaining the adjusted disturbance range; In the process of iterative search, the preferred degree of the current solution generated in each round and the new solution obtained by perturbing the current solution are evaluated, and the disturbance range requirement factor of the current solution is determined based on the difference in preferred degree between the current solution and the new solution. The disturbance range requirement factor is used to reflect the size of the adjustment requirement of the current solution; The disturbance range requirement factor of the current solution generated in each round is used to correct the adjusted disturbance range, and the disturbance range of the current solution generated in each round is obtained, so as to optimize the process of iterative search for the optimal solution of the feed speed at each moment by the simulated annealing algorithm, obtain the optimal solution of the feed speed at each moment, and optimize the machining process parameters of the CNC machine tool according to the optimal solution of the feed speed at each moment.

2. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 1 is characterized in that: The perturbation range of the current solution generated in each round is determined based on the following method: The disturbance range adjusted by the simulated annealing algorithm in the process of iteratively searching for the optimal solution of the feed speed at each moment is multiplied by the disturbance range requirement factor of the current solution generated in each round to realize the correction of the adjusted disturbance range, and the multiplied value is used as the disturbance range of the current solution generated in each round.

3. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 1 is characterized in that: The method for evaluating the optimality of the current solution generated in each round and the new solution obtained by perturbing the current solution is: Construct an objective function to reflect the preference of the current solution / new solution: ； in, is the objective function value of the current solution / new solution. Both the current solution / new solution are feed speed, and the objective function value is negatively correlated with the optimization degree. is the energy consumption per unit time corresponding to the current solution / new solution, is the unit time capacity corresponding to the current solution / new solution.

4. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 3 is characterized in that: The disturbance range requirement factor of the current solution satisfies the following relationship: ； In the formula, It is an iterative search In the process of finding the optimal solution for the feed speed at the moment, The perturbation range requirement factor of the current solution generated by the round iteration, is the normalization function, and Respectively The objective function values ​​of the current solution and the new solution generated by the round iteration, To iterate the search In the process of finding the optimal solution for the feed speed at the moment, The temperature value after the round iteration, Iterate to search for The temperature threshold is set during the process of finding the optimal solution for the feed speed at each acquisition moment.

5. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 1 is characterized in that: The degree of fluctuation of the reference data segment is determined in the following way: Among all the feed speeds included in the reference data segment of the feed speed at each moment, if a certain feed speed has a numerical difference with a previous feed speed, the feed speed is marked, and the numerical change of the marked feed speed is obtained; Calculate the volatility: ; In the formula, For the The fluctuation degree of the reference data segment of the feed speed at each moment, is the normalization function, is the number of feed rates within the reference data segment, For the The total number of all marked feed speeds in the reference data segment of the feed speed at the moment, For the The variance of the numerical changes of all marked feed speeds in the reference data segment of the feed speed at a moment.

6. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 5 is characterized in that: The method of dynamically adjusting the initial disturbance range by using the fluctuation degree to obtain the adjusted disturbance range is: The initial disturbance range in the process of iteratively searching for the optimal solution of the feed speed at each moment by the simulated annealing algorithm is multiplied by the fluctuation degree of the reference data segment of the feed speed at that moment to achieve dynamic adjustment of the initial disturbance range, and the multiplied value is used as the adjusted disturbance range.

7. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 6 is characterized in that: The method for setting the reference data segment of the feed speed at each moment is: The feed rate immediately before each moment The feed speed at each moment constitutes the reference data segment of the feed speed at that moment. is the default value.

8. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 1 is characterized in that: One method to optimize the machining process parameters of CNC machine tools according to the optimal solution of feed speed at each moment is: A fixed time interval is set. During the processing, every time a time interval passes, the optimal processing parameters of the feed speed at the last moment of the time interval are obtained, and the feed speed of the CNC machine tool is set according to the optimal processing parameters.

9. The optimization method based on processing parameters of intelligent CNC machine tools according to claim 1 is characterized in that: Another method to optimize the machining process parameters of CNC machine tools according to the optimal solution of feed speed at each moment is: The progress of multiple processing tasks is set. During the processing, each time a progress is reached, the optimal processing parameters of the feed speed at the corresponding moment of the progress are obtained, and the feed speed of the CNC machine tool is set according to the optimal processing parameters.

10. An optimization system based on intelligent CNC machine tool processing parameters, characterized in that: The process parameter optimization system comprises a memory and a processor, wherein a computer program is stored in the memory, and the processor executes the computer program to implement the steps of the optimization method according to any one of claims 1 to 9.

Images