CN117600887A - Machine tool vibration reduction method, device, equipment and storage medium - Google Patents
Machine tool vibration reduction method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN117600887A CN117600887A CN202311374419.9A CN202311374419A CN117600887A CN 117600887 A CN117600887 A CN 117600887A CN 202311374419 A CN202311374419 A CN 202311374419A CN 117600887 A CN117600887 A CN 117600887A
- Authority
- CN
- China
- Prior art keywords
- vibration
- data
- machine tool
- sample
- vibration data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 69
- 230000009467 reduction Effects 0.000 title claims abstract description 32
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 230000008569 process Effects 0.000 claims abstract description 22
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000012549 training Methods 0.000 claims description 101
- 238000009826 distribution Methods 0.000 claims description 35
- 238000004422 calculation algorithm Methods 0.000 claims description 25
- 238000013016 damping Methods 0.000 claims description 16
- 239000006096 absorbing agent Substances 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 238000003754 machining Methods 0.000 description 22
- 238000004891 communication Methods 0.000 description 6
- 238000013459 approach Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013434 data augmentation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q11/00—Accessories fitted to machine tools for keeping tools or parts of the machine in good working condition or for cooling work; Safety devices specially combined with or arranged in, or specially adapted for use in connection with, machine tools
- B23Q11/0032—Arrangements for preventing or isolating vibrations in parts of the machine
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
- G05B19/404—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by control arrangements for compensation, e.g. for backlash, overshoot, tool offset, tool wear, temperature, machine construction errors, load, inertia
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Human Computer Interaction (AREA)
- Manufacturing & Machinery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Numerical Control (AREA)
Abstract
The invention provides a machine tool vibration reduction method, a device, equipment and a storage medium, which relate to the technical field of artificial intelligence, wherein the method comprises the following steps: obtaining vibration parameters of a target machine tool, wherein the vibration parameters comprise physical parameters of a vibrating piece of the target machine tool and processing technological parameters of a task to be processed of the target machine tool; inputting vibration parameters into a trained vibration data prediction model, and obtaining predicted vibration data output by the vibration data prediction model, wherein the predicted vibration data is a predicted value of theoretical vibration data, and the theoretical vibration data is vibration data when a target machine tool processes a task to be processed; and acquiring damper adjustment data of the target machine tool based on the predicted vibration data, wherein the damper adjustment data is used for adjusting a damper of the target machine tool. The invention can predict the vibration condition of the machine tool in advance and adjust the vibration damper, thereby being beneficial to improving the processing efficiency.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a machine tool vibration reduction method, device, equipment and storage medium.
Background
In the numerical control machining process, due to the reasons that a driving device in a machine tool rotates, a cutter is in hard contact with a workpiece and the like, the machine tool inevitably generates vibration, the vibration is more obvious under the condition of high rotating speed of a main shaft, machining errors can be caused by the vibration of the machine tool, and the service life of the machine tool can be influenced. In the prior art, vibration of a machine tool can be reduced by adopting the vibration absorber, however, various different working conditions exist in machining, the generated vibration conditions are different, prediction of vibration data of the machine tool cannot be realized in the prior art, vibration data corresponding to the current working condition can be obtained only after the normal start of machining, and the vibration absorber is adjusted by stopping after the start of machining, so that the machining efficiency is affected.
Disclosure of Invention
The invention provides a machine tool vibration reduction method, device, equipment and storage medium, which are used for solving the defect that the prediction of machine tool vibration data cannot be realized in the prior art, realizing the prediction of the machine tool vibration data before machining and improving the machining efficiency.
The invention provides a machine tool vibration reduction method, which comprises the following steps:
obtaining vibration parameters of a target machine tool, wherein the vibration parameters comprise physical parameters of a vibration piece of the target machine tool and processing technological parameters of a task to be processed of the target machine tool;
inputting the vibration parameters into a trained vibration data prediction model, and obtaining predicted vibration data output by the vibration data prediction model, wherein the predicted vibration data is a predicted value of theoretical vibration data, and the theoretical vibration data is vibration data when the target machine tool processes the task to be processed;
acquiring damper adjustment data of the target machine tool based on the predicted vibration data, the damper adjustment data being used to adjust a damper of the target machine tool;
the vibration data prediction model is obtained by training a generated countermeasure learning algorithm and a supervised learning algorithm based on a training data set, wherein the training data set comprises a plurality of groups of training data, and each group of training data comprises a sample vibration parameter and a vibration data label corresponding to the sample vibration parameter.
According to the machine tool vibration reduction method provided by the invention, the training process of the vibration data prediction model comprises the following steps:
in each training iteration, the following steps are performed:
clustering the sample vibration parameters in the training data set to obtain a plurality of subsets;
adding noise to each sample vibration parameter in a target subset, and then respectively inputting the sample vibration parameters into a data expansion model to obtain each virtual sample vibration parameter output by the data expansion model;
inputting the virtual sample vibration parameters into the vibration data prediction model to obtain virtual prediction vibration data output by the vibration data prediction model;
inputting each virtual predicted vibration data and each vibration data tag in the target subset into a discriminator respectively, and obtaining a discrimination result output by the discriminator, wherein the discrimination result reflects the probability that the data input into the discriminator is the virtual predicted vibration data;
inputting each sample vibration parameter in the target subset into the vibration data prediction model to obtain sample prediction vibration data output by the vibration data prediction model;
updating the data expansion model, the vibration data prediction model, and the arbiter based on the discrimination result, the sample predicted vibration data, and the vibration data tag.
According to the machine tool vibration reduction method provided by the invention, the updating of the data expansion model, the vibration data prediction model and the discriminator based on the discrimination result, the sample prediction vibration data and the vibration data label comprises the following steps:
acquiring distribution of a plurality of virtual prediction vibration data corresponding to the target subset as first distribution;
acquiring a distribution of the vibration data tags in the target subset as a second distribution;
determining a training loss based on the first distribution, the second distribution, the discrimination result, the sample predicted vibration data, and the vibration data tag;
updating the data expansion model, the vibration data prediction model, and the arbiter based on the training loss.
According to the machine tool vibration damping method provided by the invention, the training loss is determined based on the first distribution, the second distribution, the discrimination result, the sample predicted vibration data and the vibration data label, and the machine tool vibration damping method comprises the following steps:
acquiring a discrimination result and a real classification result corresponding to the discrimination result, and determining a first loss;
determining a second loss based on the first distribution and the second distribution;
determining a third loss based on the sample predicted vibration data and the vibration data tag;
the training loss is determined based on the first loss, the second loss, and the third loss.
According to the machine tool vibration reduction method provided by the invention, the sample vibration parameters in the training data set are clustered to obtain a plurality of subsets, and the machine tool vibration reduction method comprises the following steps:
acquiring the current training iteration number N;
if N is smaller than a preset secondary value, determining the number of the sample vibration parameters in each subset as X;
if N is not smaller than the preset sub-value, determining the number of the sample vibration parameters in each subset as Y;
wherein N, X, Y is a positive integer, and Y is greater than X.
According to the machine tool vibration reduction method provided by the invention, the method for acquiring the vibration damper adjustment data of the target machine tool based on the predicted vibration data comprises the following steps:
inputting the predicted vibration data into a trained rewarding model, and obtaining the damper adjustment data output by the rewarding model;
wherein, the training process of the reward model comprises the following steps:
inputting sample vibration data into the rewarding model, and obtaining sample vibration damper adjustment data output by the rewarding model, wherein the sample vibration data is vibration data acquired after the operation of a sample machine tool is controlled;
determining a bonus value based on the sample damper adjustment data, updating the bonus model based on the bonus value.
According to the machine tool vibration damping method provided by the invention, the method for determining the rewarding value based on the sample vibration damper adjustment data comprises the following steps:
vibration reduction data is obtained by operating the sample machine after adjusting the vibration absorber of the sample machine based on the sample vibration absorber adjustment data;
the bonus value is determined based on the vibration reduction data and the sample vibration data.
The invention also provides a machine tool vibration damper, comprising:
the parameter acquisition module is used for acquiring vibration parameters of a target machine tool, wherein the vibration parameters comprise physical parameters of a vibrating piece of the target machine tool and processing technological parameters of a task to be processed of the target machine tool;
the prediction module is used for inputting the vibration parameters into a trained vibration data prediction model, obtaining predicted vibration data output by the vibration data prediction model, wherein the predicted vibration data are predicted values of theoretical vibration data, and the theoretical vibration data are vibration data when the target machine tool processes the task to be processed;
the adjusting module is used for acquiring damper adjusting data of the target machine tool based on the predicted vibration data, wherein the damper adjusting data is used for adjusting a damper of the target machine tool;
the vibration data prediction model is obtained by training a generated countermeasure learning algorithm and a supervised learning algorithm based on a training data set, wherein the training data set comprises a plurality of groups of training data, and each group of training data comprises a sample vibration parameter and a vibration data label corresponding to the sample vibration parameter.
The invention also provides electronic equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the machine tool vibration reduction method when executing the computer program.
The invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which when executed by a processor implements a machine tool vibration reduction method as described in any one of the above.
According to the machine tool vibration reduction method, device, equipment and storage medium, vibration parameters (including physical parameters of a vibrating piece of a target machine tool and machining process parameters of a task to be machined of the target machine tool) are input into the vibration data prediction model, predicted vibration data of the target machine tool when the target machine tool is machined of the task to be machined, which is output by the model, is obtained by training a training data of a plurality of groups of vibration data labels corresponding to the sample vibration parameters and the sample vibration parameters through generation of an anti-learning algorithm, in the training process based on generation of the anti-learning algorithm and the supervised learning algorithm, the vibration data prediction model can learn the association between the vibration parameters and the vibration data of the target machine tool, the capacity of predicting the vibration data based on the input vibration parameters is achieved, after the predicted vibration data output by the vibration data prediction model is obtained, damper adjustment data of the target machine tool are obtained based on the predicted vibration data, and a damper of the target machine tool is adjusted based on the damper adjustment data. The invention can predict the vibration data of the machine tool before the machining starts and adjust the vibration damper of the machine tool based on the prediction result, thereby improving the machining efficiency.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a machine tool vibration damping method provided by the invention;
FIG. 2 is a schematic diagram of the machine tool vibration damping device provided by the invention;
fig. 3 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The machine tool vibration damping method, apparatus, device and storage medium of the present invention are described below with reference to fig. 1-3.
As shown in fig. 1, the machine tool vibration damping method provided by the invention comprises the following steps:
s100, obtaining vibration parameters of a target machine tool, wherein the vibration parameters comprise physical parameters of a vibration piece of the target machine tool and processing technological parameters of a task to be processed of the target machine tool;
s200, inputting vibration parameters into a trained vibration data prediction model, and obtaining predicted vibration data output by the vibration data prediction model, wherein the predicted vibration data is a predicted value of theoretical vibration data, and the theoretical vibration data is vibration data when a target machine tool processes a task to be processed;
s300, acquiring damper adjustment data of the target machine tool based on the predicted vibration data, wherein the damper adjustment data is used for adjusting a damper of the target machine tool.
The vibration data prediction model is obtained by training a generated countermeasure learning algorithm and a supervised learning algorithm based on a training data set, wherein the training data set comprises a plurality of groups of training data, and each group of training data comprises a sample vibration parameter and a vibration data label corresponding to the sample vibration parameter.
According to the machine tool vibration reduction method, vibration parameters (including physical parameters of a vibrating piece of a target machine tool and machining process parameters of a task to be machined of the target machine tool) are input into a vibration data prediction model, predicted vibration data of the target machine tool, which is output by the model, when the task to be machined is machined of the target machine tool is machined, is obtained, wherein the vibration data prediction model is obtained by training by adopting a generated countermeasure learning algorithm, in the training process of generating the countermeasure learning algorithm and the supervised learning algorithm, the vibration data prediction model can learn the association between the vibration parameters and the vibration data of the target machine tool, the capability of predicting the vibration data based on the input vibration parameters is achieved, after the predicted vibration data output by the vibration data prediction model is obtained, damper adjustment data of the target machine tool are obtained based on the predicted vibration data, and the damper of the target machine tool is adjusted based on the damper adjustment data. The invention can predict the vibration data of the machine tool before the machining starts and adjust the vibration damper of the machine tool based on the prediction result, thereby improving the machining efficiency.
Specifically, the vibrating element of the target machine tool is a member that affects the machining effect of the target machine tool when vibrating, for example, a machining tool, a spindle, or the like of the target machine tool. Physical parameters of the vibrating member include the material, size, weight, etc. of the vibrating member. The target machine tool is used for processing tasks which are not started by the target machine tool, and the technical scheme provided by the application predicts the vibration condition of the target machine tool when the target machine tool executes the task to be processed before the task to be processed of the target machine tool starts to obtain predicted vibration data. The processing technological parameters of the task to be processed of the target machine tool comprise the material of a cutting object (namely a processed workpiece) in the task to be processed, the spindle rotating speed of the target machine tool and the like.
And inputting the vibration parameters into a trained vibration data prediction model, wherein the vibration data prediction model is trained and then provided with vibration conditions when a target machine tool processes a task to be processed under the state without a vibration damper based on the input vibration parameters, and outputting predicted vibration data. Specifically, the predicted vibration data includes the amplitude and vibration frequency of each vibrating member of the target machine tool.
The training process of the vibration data prediction model is described in detail below. The training process of the vibration data prediction model consists of a plurality of training iterations, in each of which the following steps are performed:
clustering sample vibration parameters in the training data set to obtain a plurality of subsets;
noise is added to each sample vibration parameter in the target subset, and then the sample vibration parameters are respectively input into the data expansion model, so that each virtual sample vibration parameter output by the data expansion model is obtained;
inputting the virtual sample vibration parameters into a vibration data prediction model, and obtaining virtual prediction vibration data output by the vibration data prediction model;
inputting each virtual predicted vibration data and each vibration data tag in the target subset into a discriminator respectively, and obtaining a discrimination result output by the discriminator, wherein the discrimination result reflects the probability that the data input into the discriminator is the virtual predicted vibration data;
inputting each sample vibration parameter in the target subset into a vibration data prediction model to obtain sample prediction vibration data output by the vibration data prediction model;
and updating the data expansion model, the vibration data prediction model and the discriminator based on the discrimination result, the sample vibration data prediction data and the vibration data tag.
The training method based on supervised learning requires a large amount of labeling data, but the conventional general database includes a small amount of sample vibration parameters and vibration data labels corresponding to the sample vibration parameters, and the acquisition and labeling of the data require relatively high cost, that is, the quantity of training data in the training data set is small. In the method provided by the invention, the vibration data prediction model is trained by adopting the combination of a generation countermeasure learning algorithm and a supervised learning algorithm. Specifically, the training data set is expanded by adopting the generated countermeasure learning algorithm, the generated countermeasure network model (a sample expansion model and a discriminator) and the vibration data prediction model are trained based on the expanded data and the labeled data, and parameters are updated, so that the model can learn by using more data and learning modes, and the accuracy of the data output by the model obtained by training is improved.
For a plurality of sample vibration parameters which are similar to each other, the corresponding vibration data of the sample vibration parameters should have consistent distribution, the invention utilizes the point to firstly cluster the sample vibration parameters in the training data set, the similar sample vibration parameters can be clustered into one type by the clustering, and the sample vibration parameters belonging to one type and the vibration data labels corresponding to the sample vibration parameters form a subset. In each training iteration, a subset is determined as the target subset, which may be repeated during the training process. The random noise is added to the sample vibration parameters in the target subset and then the sample vibration parameters are input into the data expansion model, and each virtual sample vibration parameter output by the data expansion model is obtained, that is, the data expansion model generates data similar to the sample vibration parameters based on the sample vibration parameters, and the generated data are not really collected data and are called virtual sample vibration parameters. In the process of generating the virtual sample vibration parameters by the data expansion model, constraint can be added, and the virtual sample vibration parameters generated by the data expansion model are constrained to have certain similarity with each sample vibration parameter of the target subset. Specifically, the distance between the virtual sample vibration parameter generated by constraint and each sample vibration parameter of the target subset is not greater than the distance between any two sample vibration parameters in the target subset.
In the method provided by the invention, the purpose is to enable the data output by the vibration data prediction model to be more accurate, that is, the data output by the vibration data prediction model is expected to be more approximate to the true value. Therefore, in the method provided by the invention, the virtual sample vibration parameters output by the data expansion model are not judged by adopting a traditional training mode of the countermeasure generation network, but are input into the vibration data prediction model after the virtual sample vibration parameters are generated, and the virtual prediction vibration data output by the vibration data prediction model based on the virtual sample vibration parameters are judged by using the discriminator. Specifically, after the virtual sample vibration parameter is generated, the virtual sample vibration parameter and the real sample vibration parameter are each input to a discriminator, which discriminates the probability that the input data is the real sample vibration parameter. Through training, the virtual sample vibration parameters generated by the data expansion model can be more in line with the actual rules, and the data output by the vibration data prediction model is more approximate to the actual values.
Further, as described above, the vibration data corresponding to the plurality of vibration parameters should have a uniform distribution, although there is a difference therebetween. That is, if vibration parameters in the same subset, the distribution of their corresponding vibration data is uniform. Based on this, the data expansion model, the vibration data prediction model, and the discriminator are updated based on the discrimination result, the sample prediction vibration data, and the vibration data tag, and include:
acquiring distribution of a plurality of virtual prediction vibration data corresponding to the target subset as first distribution;
acquiring the distribution of the vibration data tags in the target subset as a second distribution;
determining a training loss based on the first distribution, the second distribution, the discrimination result, the sample predicted vibration data, and the vibration data tag;
updating the data augmentation model, the vibration data prediction model, and the arbiter based on the training loss.
Specifically, determining the training loss based on the first distribution, the second distribution, the discrimination result, the sample predicted vibration data, and the predicted vibration data tag includes:
acquiring a discrimination result and a real classification result corresponding to the discrimination result, and determining a first loss;
determining a second loss based on the first distribution and the second distribution;
determining a third loss based on the sample predicted vibration data and the vibration data tag;
training losses are determined based on the first loss, the second loss, and the third loss.
The first loss and the second loss reflect the degree of the approach of the data generated by the data expansion model to the real data and the degree of the approach of the vibration prediction data output by the vibration data prediction model to the real data, and the third loss reflects the degree of the approach of the vibration prediction data output by the vibration data pre-storage model to the real data. Determining a training loss based on the first loss, the second loss, and the third loss, and updating the data expansion model, the arbiter, and the vibration data prediction model based on the training loss may enable the vibration data prediction model to output more accurate predicted vibration data.
Further, in order to improve training efficiency, in the method provided by the invention, the parameters of sample vibration in the training data set are clustered to obtain a plurality of subsets, including:
acquiring the current training iteration number N;
if N is smaller than the preset secondary value, determining that the number of the sample vibration parameters in each subset is X;
if N is not smaller than the preset secondary value, determining the number of the sample vibration parameters in each subset as Y;
wherein N, X, Y is a positive integer, and Y is greater than X.
Specifically, in the initial training stage, the model does not have high performance, and the data distribution obtained according to more data is more accurate, so in the method provided by the invention, the number of the sample vibration parameters in each subset is smaller in the initial training stage, that is, when the current training iteration number N is smaller than a preset number, and the number of the sample vibration parameters in each subset is larger in the later training stage, that is, when the current training iteration number N is not smaller than the preset number. Thus, in the later stage of training, the number of virtual predictive vibration data used for generating the first distribution is higher than that in the earlier stage of training, that is, the more backward, the higher the accuracy of the generated distribution, so that more detailed feedback information can be provided for model training in the later stage of model training, and in the earlier stage of training, the more coarse-grained feedback information can be used to reduce the calculation amount.
After the parameters of the model are converged, training is finished. The trained vibration data prediction model is used for outputting predicted vibration data based on the input vibration parameters.
After obtaining the predicted vibration data output by the trained vibration data pre-storage model, obtaining the damper adjustment data of the target machine tool based on the predicted vibration data, wherein the method comprises the following steps:
inputting the predicted vibration data into the trained rewarding model, and obtaining the damper adjustment data output by the rewarding model;
the training process of the reward model comprises the following steps:
inputting sample vibration data into a reward model, and obtaining sample vibration damper adjustment data output by the reward model, wherein the sample vibration data is vibration data acquired after the operation of a sample machine tool is controlled;
a bonus value is determined based on the sample damper adjustment data, and a bonus model is updated based on the bonus value.
Determining a bonus value based on sample damper adjustment data, comprising:
vibration reduction data is obtained by operating the sample machine after adjusting the vibration absorber of the sample machine based on the sample vibration absorber adjustment data;
a prize value is determined based on the vibration damping data and the sample vibration data.
The sample machine tool is a machine tool consistent with the target machine tool. The bonus model is for outputting damper adjustment data adapted to optimize vibration data based on the input vibration data. In the training process of the rewarding model, a sample machine tool is actually operated, vibration data of the sample machine tool are collected and used as sample vibration data, the sample vibration data are input into the rewarding model, sample damper adjusting data output by the rewarding model are obtained, then a damper of the sample machine tool is adjusted based on the sample damper adjusting data, then the sample machine tool is continuously moved based on the fact that vibration parameters are unchanged when the sample vibration data are collected, and vibration reduction data are collected. It can be seen that the difference between the vibration damping data and the sample vibration data reflects the accuracy of the damper adjustment data.
Specifically, determining the bonus value based on the vibration damping data and the sample vibration data includes:
acquiring vibration reduction data and vibration frequency change values and amplitude change values of each vibrating piece in sample vibration data;
and carrying out weighted summation on the variation value of the vibration frequency and the variation value of the vibration amplitude of each vibrating piece according to the weight of the vibrating piece to obtain a reward value, wherein the corresponding weight of the vibrating piece is determined based on the influence degree of the vibrating piece on the machining precision during vibration.
The weight of each vibrating member may be determined in advance based on the degree of influence of each vibrating member on the machining accuracy, specifically, the degree of influence of the vibrating member on the machining accuracy may be determined based on whether the vibrating member is in direct contact with the machined workpiece or not, for example, the degree of influence of the tool on the machining accuracy is higher than the degree of influence of the spindle on the machining accuracy, or the degree of influence of the vibrating member on the machining accuracy may be determined based on the position and effect of the vibrating member in the machine tool, by a professional according to empirical knowledge.
The machine tool vibration damping device provided by the invention is described below, and the machine tool vibration damping device and the machine tool vibration damping method described above can be correspondingly referred to each other.
As shown in fig. 2, the machine tool vibration damping device provided by the present invention includes:
the parameter obtaining module 210 is configured to obtain vibration parameters of the target machine tool, where the vibration parameters include physical parameters of a vibrating member of the target machine tool and processing parameters of a task to be processed of the target machine tool.
The prediction module 220 is configured to input the vibration parameter to a trained vibration data prediction model, obtain predicted vibration data output by the vibration data prediction model, where the predicted vibration data is a predicted value of theoretical vibration data, and the theoretical vibration data is vibration data when the target machine tool processes a task to be processed.
An adjustment module 230 for obtaining damper adjustment data of the target machine tool based on the predicted vibration data, the damper adjustment data being used to adjust a damper of the target machine tool.
The vibration data prediction model is obtained by training a generated countermeasure learning algorithm and a supervised learning algorithm based on a training data set, wherein the training data set comprises a plurality of groups of training data, and each group of training data comprises a sample vibration parameter and a vibration data label corresponding to the sample vibration parameter.
Fig. 3 illustrates a physical schematic diagram of an electronic device, as shown in fig. 3, where the electronic device may include: processor 310, communication interface (Communications Interface) 320, memory 330 and communication bus 340, wherein processor 310, communication interface 320, memory 330 accomplish communication with each other through communication bus 340. Processor 310 may invoke logic instructions in memory 330 to perform a machine tool vibration reduction method comprising: obtaining vibration parameters of a target machine tool, wherein the vibration parameters comprise physical parameters of a vibrating piece of the target machine tool and processing technological parameters of a task to be processed of the target machine tool; inputting vibration parameters into a trained vibration data prediction model, and obtaining predicted vibration data output by the vibration data prediction model, wherein the predicted vibration data is a predicted value of theoretical vibration data, and the theoretical vibration data is vibration data when a target machine tool processes a task to be processed; acquiring damper adjustment data of the target machine tool based on the predicted vibration data, the damper adjustment data being used to adjust a damper of the target machine tool; the vibration data prediction model is obtained by training a generated countermeasure learning algorithm and a supervised learning algorithm based on a training data set, wherein the training data set comprises a plurality of groups of training data, and each group of training data comprises a sample vibration parameter and a vibration data label corresponding to the sample vibration parameter.
Further, the logic instructions in the memory 330 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform a machine tool vibration reduction method provided by the above methods, the method comprising: obtaining vibration parameters of a target machine tool, wherein the vibration parameters comprise physical parameters of a vibrating piece of the target machine tool and processing technological parameters of a task to be processed of the target machine tool; inputting vibration parameters into a trained vibration data prediction model, and obtaining predicted vibration data output by the vibration data prediction model, wherein the predicted vibration data is a predicted value of theoretical vibration data, and the theoretical vibration data is vibration data when a target machine tool processes a task to be processed; acquiring damper adjustment data of the target machine tool based on the predicted vibration data, the damper adjustment data being used to adjust a damper of the target machine tool; the vibration data prediction model is obtained by training a generated countermeasure learning algorithm and a supervised learning algorithm based on a training data set, wherein the training data set comprises a plurality of groups of training data, and each group of training data comprises a sample vibration parameter and a vibration data label corresponding to the sample vibration parameter.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A method of damping vibration in a machine tool, comprising:
obtaining vibration parameters of a target machine tool, wherein the vibration parameters comprise physical parameters of a vibration piece of the target machine tool and processing technological parameters of a task to be processed of the target machine tool;
inputting the vibration parameters into a trained vibration data prediction model, and obtaining predicted vibration data output by the vibration data prediction model, wherein the predicted vibration data is a predicted value of theoretical vibration data, and the theoretical vibration data is vibration data when the target machine tool processes the task to be processed;
acquiring damper adjustment data of the target machine tool based on the predicted vibration data, the damper adjustment data being used to adjust a damper of the target machine tool;
the vibration data prediction model is obtained by training a generated countermeasure learning algorithm and a supervised learning algorithm based on a training data set, wherein the training data set comprises a plurality of groups of training data, and each group of training data comprises a sample vibration parameter and a vibration data label corresponding to the sample vibration parameter.
2. The machine tool vibration reduction method according to claim 1, wherein the training process of the vibration data prediction model includes:
in each training iteration, the following steps are performed:
clustering the sample vibration parameters in the training data set to obtain a plurality of subsets;
adding noise to each sample vibration parameter in a target subset, and then respectively inputting the sample vibration parameters into a data expansion model to obtain each virtual sample vibration parameter output by the data expansion model;
inputting the virtual sample vibration parameters into the vibration data prediction model to obtain virtual prediction vibration data output by the vibration data prediction model;
inputting each virtual predicted vibration data and each vibration data tag in the target subset into a discriminator respectively, and obtaining a discrimination result output by the discriminator, wherein the discrimination result reflects the probability that the data input into the discriminator is the virtual predicted vibration data;
inputting each sample vibration parameter in the target subset into the vibration data prediction model to obtain sample prediction vibration data output by the vibration data prediction model;
updating the data expansion model, the vibration data prediction model, and the arbiter based on the discrimination result, the sample predicted vibration data, and the vibration data tag.
3. The machine tool vibration reduction method according to claim 2, wherein the updating the data expansion model, the vibration data prediction model, and the discriminator based on the discrimination result, the sample predicted vibration data, and the vibration data tag includes:
acquiring distribution of a plurality of virtual prediction vibration data corresponding to the target subset as first distribution;
acquiring a distribution of the vibration data tags in the target subset as a second distribution;
determining a training loss based on the first distribution, the second distribution, the discrimination result, the sample predicted vibration data, and the vibration data tag;
updating the data expansion model, the vibration data prediction model, and the arbiter based on the training loss.
4. A machine tool vibration reduction method according to claim 3, wherein the determining a training loss based on the first distribution, the second distribution, the discrimination result, the sample predicted vibration data, and the vibration data tag comprises:
acquiring a discrimination result and a real classification result corresponding to the discrimination result, and determining a first loss;
determining a second loss based on the first distribution and the second distribution;
determining a third loss based on the sample predicted vibration data and the vibration data tag;
the training loss is determined based on the first loss, the second loss, and the third loss.
5. The machine tool vibration reduction method according to claim 2, wherein the clustering of the sample vibration parameters in the training dataset results in a plurality of subsets, comprising:
acquiring the current training iteration number N;
if N is smaller than a preset secondary value, determining the number of the sample vibration parameters in each subset as X;
if N is not smaller than the preset sub-value, determining the number of the sample vibration parameters in each subset as Y;
wherein N, X, Y is a positive integer, and Y is greater than X.
6. The machine tool vibration reduction method according to claim 1, wherein the acquiring damper adjustment data of the target machine tool based on the predicted vibration data includes:
inputting the predicted vibration data into a trained rewarding model, and obtaining the damper adjustment data output by the rewarding model;
wherein, the training process of the reward model comprises the following steps:
inputting sample vibration data into the rewarding model, and obtaining sample vibration damper adjustment data output by the rewarding model, wherein the sample vibration data is vibration data acquired after the operation of a sample machine tool is controlled;
determining a bonus value based on the sample damper adjustment data, updating the bonus model based on the bonus value.
7. The machine tool vibration reduction method according to claim 6, wherein the determining a prize value based on the sample damper adjustment data comprises:
vibration reduction data is obtained by operating the sample machine after adjusting the vibration absorber of the sample machine based on the sample vibration absorber adjustment data;
the bonus value is determined based on the vibration reduction data and the sample vibration data.
8. A vibration damping device for a machine tool, comprising:
the parameter acquisition module is used for acquiring vibration parameters of a target machine tool, wherein the vibration parameters comprise physical parameters of a vibrating piece of the target machine tool and processing technological parameters of a task to be processed of the target machine tool;
the prediction module is used for inputting the vibration parameters into a trained vibration data prediction model, obtaining predicted vibration data output by the vibration data prediction model, wherein the predicted vibration data are predicted values of theoretical vibration data, and the theoretical vibration data are vibration data when the target machine tool processes the task to be processed;
the adjusting module is used for acquiring damper adjusting data of the target machine tool based on the predicted vibration data, wherein the damper adjusting data is used for adjusting a damper of the target machine tool;
the vibration data prediction model is obtained by training a generated countermeasure learning algorithm and a supervised learning algorithm based on a training data set, wherein the training data set comprises a plurality of groups of training data, and each group of training data comprises a sample vibration parameter and a vibration data label corresponding to the sample vibration parameter.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the machine tool vibration reduction method according to any one of claims 1 to 7 when executing the computer program.
10. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a machine tool vibration reduction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311374419.9A CN117600887A (en) | 2023-10-20 | 2023-10-20 | Machine tool vibration reduction method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311374419.9A CN117600887A (en) | 2023-10-20 | 2023-10-20 | Machine tool vibration reduction method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117600887A true CN117600887A (en) | 2024-02-27 |
Family
ID=89946948
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311374419.9A Pending CN117600887A (en) | 2023-10-20 | 2023-10-20 | Machine tool vibration reduction method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117600887A (en) |
-
2023
- 2023-10-20 CN CN202311374419.9A patent/CN117600887A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Predicting tool wear with multi-sensor data using deep belief networks | |
| Kan et al. | A review on prognostic techniques for non-stationary and non-linear rotating systems | |
| US20160335548A1 (en) | Methods and apparatus for predicting fault occurrence in mechanical systems and electrical systems | |
| US20190213475A1 (en) | Reducing machine-learning model complexity while maintaining accuracy to improve processing speed | |
| WO2008156595A1 (en) | Hybrid method for simulation optimization | |
| CN114662386A (en) | Bearing fault diagnosis method and system | |
| KR20200125029A (en) | Method and apparatus for regression analysis | |
| CN114660993B (en) | Numerical control machine tool fault prediction method based on multi-source heterogeneous data feature dimension reduction | |
| JP2020123270A (en) | Arithmetic unit | |
| CN116047934A (en) | Real-time simulation method and system for unmanned aerial vehicle cluster and electronic equipment | |
| CN116226676B (en) | Machine tool fault prediction model generation method suitable for extreme environment and related equipment | |
| CN117600887A (en) | Machine tool vibration reduction method, device, equipment and storage medium | |
| CN111400964B (en) | Fault occurrence time prediction method and device | |
| CN112419098B (en) | Power grid safety and stability simulation sample screening and expanding method based on safety information entropy | |
| CN111667186B (en) | Method and device for determining the energy consumption of a machine for production | |
| CN113348473B (en) | Management artificial intelligence module installation foundation | |
| Dudek | Tournament searching method to feature selection problem | |
| US20230110981A1 (en) | Method and system for a controller | |
| Zeng et al. | Prediction of equipment maintenance using optimized support vector machine | |
| CN117826618B (en) | Adaptive control method and system based on cold rolling mill control system | |
| CN111077769A (en) | Method for controlling or regulating a technical system | |
| CN116068962B (en) | Process route planning method based on numerical control machine tool and related equipment | |
| Rezaeipanah et al. | Meta-heuristic approach based on genetic and greedy algorithms to solve flexible job-shop scheduling problem | |
| EP4060546A1 (en) | Method and system for controlling a production plant to manufacture a product | |
| EP3916496A1 (en) | An industrial process model generation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |