CN-122007996-A - Machine tool hydraulic control intelligent compensation method and system based on machine learning

CN122007996ACN 122007996 ACN122007996 ACN 122007996ACN-122007996-A

Abstract

The invention relates to the field of hydraulic control compensation, in particular to a machine tool hydraulic control intelligent compensation method and system based on machine learning, wherein the method comprises the steps of fitting a global wear attenuation rate to obtain a decision coefficient; calculating the wear equivalent time of the grinding process to be compensated, constructing a process feature vector of the grinding process to be compensated, training a regression prediction model based on the process feature vector of the historical grinding process, inputting the process feature vector of the grinding process to be compensated into the trained regression prediction model to output a predicted value of the removal rate of the grinding process to be compensated, correcting the predicted value of the removal rate based on a decision coefficient to obtain a corrected predicted rate of the grinding process to be compensated, calculating the hydraulic pressure compensation amount of the grinding process to be compensated, and generating a control instruction according to the hydraulic pressure compensation amount to complete intelligent hydraulic compensation. By the technical scheme, the accuracy of the hydraulic compensation result can be improved.

Inventors

BAI XIANGFENG
JING WEIDONG
JING WEIFANG

Assignees

思瑞得(宁波)精密机械有限公司

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (9)

1. The intelligent compensation method for the hydraulic control of the machine tool based on machine learning is characterized by comprising the following steps: Acquiring hydraulic pressure, accumulated grinding time and initial thickness of a workpiece in any historical grinding process, fitting a global wear attenuation rate based on a historical database, and acquiring a determination coefficient in the fitting process; Calculating the abrasion equivalent time of the grinding process to be compensated based on the global abrasion attenuation rate and the accumulated grinding time of the reference process by taking the previous historical grinding process adjacent to the grinding process to be compensated as the reference process, constructing the abrasion equivalent time, the hydraulic pressure of the grinding process to be compensated and the initial thickness of the workpiece of the grinding process to be compensated as process feature vectors of the grinding process to be compensated, and obtaining the process feature vectors of any historical grinding process in the same way; training a regression prediction model based on the process feature vector of the historical grinding process, inputting the process feature vector of the grinding process to be compensated into the trained regression prediction model, and outputting a predicted value of the removal rate of the grinding process to be compensated; And calculating the hydraulic pressure compensation quantity of the grinding process to be compensated based on the correction prediction rate, the hydraulic pressure of the grinding process to be compensated and the acquired standard removal rate, and generating a control instruction according to the hydraulic pressure compensation quantity to complete hydraulic intelligent compensation.
2. The machine tool hydraulic control intelligent compensation method based on machine learning of claim 1, wherein the fitting the global wear decay rate based on the historical database while obtaining the decision coefficients in the fitting process comprises: Taking a plurality of historical grinding processes completed under standard pressure in a historical database as experimental processes, and acquiring a data pair of any experimental process, wherein the data pair comprises the true value of the accumulated grinding time and the removal rate; a regression fitting algorithm is used for all data pairs in the experimental process, so that the global wear attenuation rate is obtained; the decision coefficients of the fitted curve during the fitting process are recorded.
3. The machine tool hydraulic control intelligent compensation method based on machine learning according to claim 1, wherein the calculating the wear equivalent time of the grinding process to be compensated comprises: Calculating the product of the global wear attenuation rate and the accumulated grinding time of the reference process as a first product, calculating a negative index value of the first product, and taking the difference between a constant 1 and the negative index value as the wear equivalent time of the grinding process to be compensated.
4. The intelligent compensation method for machine tool hydraulic control based on machine learning according to claim 1, wherein the regression prediction model adopts a gaussian process regression model, a neural network, a support vector machine regression or a random forest regression model.
5. The machine-learning-based intelligent compensation method for hydraulic control of a machine tool of claim 1, wherein correcting the predicted value of the removal rate based on the decision coefficient to obtain a corrected predicted rate of the grinding process to be compensated comprises: Acquiring a predicted value of the removal rate and the corrected predicted rate of a reference process, taking a difference value of the predicted value of the removal rate and the corrected predicted rate of the reference process as a first difference value, and taking a ratio of the first difference value to the predicted value of the removal rate of the reference process as a predicted error residual of the reference process; Taking the difference value of the constant 1 and the decision coefficient as a second difference value, taking the product of the second difference value and the prediction error residual as a second product, and taking the sum of the constant 1 and the second product as a correction factor; And taking the product of the correction factor and the predicted value of the removal rate of the grinding process to be compensated as the corrected predicted rate of the grinding process to be compensated.
6. The intelligent compensation method for hydraulic control of a machine tool based on machine learning according to claim 1, wherein the calculating of the hydraulic pressure compensation amount of the grinding process to be compensated comprises: Obtaining a standard removal rate, taking a difference value between the standard removal rate and the corrected prediction rate of the grinding process to be compensated as a third difference value, and taking a ratio of the third difference value to the corrected prediction rate of the grinding process to be compensated as a rate ratio; and taking the product of the hydraulic pressure of the grinding process to be compensated and the rate ratio as the hydraulic pressure compensation quantity of the grinding process to be compensated.
7. The intelligent compensation method for hydraulic control of a machine tool based on machine learning according to claim 1, wherein the generating a control command according to the hydraulic pressure compensation amount, the performing the intelligent hydraulic compensation, comprises: Acquiring a minimum working pressure and a maximum working pressure allowed in a hydraulic system; Taking the sum of the hydraulic pressure and the hydraulic pressure compensation amount of the grinding process to be compensated as the comprehensive pressure; and limiting the comprehensive pressure between the minimum working pressure and the maximum working pressure through a limiting function, and generating a control command.
8. The intelligent compensation method for machine tool hydraulic control based on machine learning of claim 1, further comprising: and measuring the final thickness of the workpiece after the finishing of the grinding process to be compensated by using a grating thickness gauge, calculating the true value of the removal rate of the grinding process to be compensated, and updating the true value of the removal rate of the grinding process to be compensated into a historical database.
9. The intelligent compensation system for hydraulic control of a machine tool based on machine learning is characterized by comprising a processor and a memory, wherein the memory stores computer program instructions which when executed by the processor realize the intelligent compensation method for hydraulic control of a machine tool based on machine learning according to any one of claims 1-8.

Description

Machine tool hydraulic control intelligent compensation method and system based on machine learning Technical Field The present invention relates to the field of hydraulic control compensation. In particular to a machine tool hydraulic control intelligent compensation method and system based on machine learning. Background In precision machining equipment, a hydraulic system is used as a key execution link for adjusting the contact pressure of the grinding disc and a workpiece, and the consistency of the removal rate and the final thickness is directly determined. However, as the accumulated use time of the polishing disc increases, the passivation of abrasive particles and the abrasion of the bonding agent may cause nonlinear drift of the process state, and it is difficult to maintain the stability of the processing process if only fixed process parameter settings are relied on. In the existing machine learning method, if the accumulated time is directly used as the characteristic input, the characteristic is often based on the stability assumption, and the quick degradation at the initial stage of wear and the non-stable characteristic which tends to be gentle at the later stage cannot be distinguished, so that systematic prediction deviation is generated. In the prior art, the accumulated grinding time is generally directly taken as a characteristic to be input into a prediction model, or a mapping relation between technological parameters and removal rate is established in a pure data driving mode. Such methods are based mostly on the assumption of stationarity that by default the amount of process state change caused by the same time interval is constant. However, in the double-end-face grinding process, the abrasion of the grinding disc presents remarkable non-stable degradation characteristics, the abrasion rate is high in the initial stage and the later stage is gentle under the influence of coupling of abrasive particle passivation and abrasion of a bonding agent. If the original time is directly input into the model, the model cannot distinguish physical differences of different abrasion stages, and the early rapid degradation and the later slow abrasion are mistakenly regarded as equivalent states, so that systematic prediction deviation is generated, and the hydraulic compensation result is inaccurate. Disclosure of Invention In order to solve the above-described technical problems, the present invention provides the following aspects. According to the hydraulic pressure control intelligent compensation method for the machine tool based on machine learning, hydraulic pressure, accumulated grinding time and initial workpiece thickness of any historical grinding process are obtained, global abrasion attenuation rate is fitted based on a historical database, a determination coefficient in the fitting process is obtained at the same time, a previous historical grinding process adjacent to the grinding process to be compensated is used as a reference process, abrasion equivalent time of the grinding process to be compensated is calculated based on the global abrasion attenuation rate and the accumulated grinding time of the reference process, the abrasion equivalent time, the hydraulic pressure of the grinding process to be compensated and the initial workpiece thickness of the grinding process to be compensated are built into technological feature vectors of the grinding process to be compensated, the technological feature vectors of any historical grinding process are obtained in the same way, a regression prediction model is trained based on the technological feature vectors of the historical grinding process, a predicted value of a removal rate of the grinding process to be compensated is output, the predicted value of the removal rate is corrected based on the determination coefficient, the predicted value of the removal rate to be compensated is obtained, the corrected prediction rate of the grinding process to be compensated is calculated, the hydraulic pressure control hydraulic pressure command is calculated according to the hydraulic pressure command of the hydraulic pressure compensation intelligent compensation. Preferably, the fitting of the global wear attenuation rate based on the historical database and the obtaining of the decision coefficients in the fitting process simultaneously comprise the steps of taking a plurality of historical grinding processes completed under standard pressure in the historical database as experimental processes, obtaining data pairs of any experimental process, wherein the data pairs comprise the true values of the accumulated grinding time and the removal rate, obtaining the global wear attenuation rate by using a regression fitting algorithm on the data pairs of all the experimental processes, and recording the decision coefficients of a fitting curve in the fitting process. Preferably, the calculating the wear equivalent time of the grinding process to