CN113579849A - Digital twinning control method and system for weak rigidity drilling - Google Patents
Digital twinning control method and system for weak rigidity drilling Download PDFInfo
- Publication number
- CN113579849A CN113579849A CN202110987837.XA CN202110987837A CN113579849A CN 113579849 A CN113579849 A CN 113579849A CN 202110987837 A CN202110987837 A CN 202110987837A CN 113579849 A CN113579849 A CN 113579849A
- Authority
- CN
- China
- Prior art keywords
- burr
- processing
- parameters
- parameter
- decision information
- 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
Images
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
- B23Q15/00—Automatic control or regulation of feed movement, cutting velocity or position of tool or work
- B23Q15/007—Automatic control or regulation of feed movement, cutting velocity or position of tool or work while the tool acts upon the workpiece
- B23Q15/013—Control or regulation of feed movement
- B23Q15/02—Control or regulation of feed movement according to the instantaneous size and the required size of the workpiece acted upon
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a digital twinning control method and system for weak rigidity drilling. The method comprises the following steps: collecting multidimensional time sequence data in the processing process; segmenting the time sequence data according to the processing progress; performing time-frequency domain analysis on the segmented data, and extracting a characteristic value; inputting the characteristic value into a mechanism model to obtain a burr state parameter; predicting the burr height according to the burr state parameters, the current processing technological parameters and the residual processing progress; constructing a decision information base; the decision information base comprises optimal processing technological parameters for inhibiting burrs in different processing scenes; and when the predicted burr height exceeds the burr height threshold, selecting the optimal processing process parameter from the decision information base for feedback. The invention utilizes the virtual model of digital twin and the real-time virtual-real interaction characteristic to realize the burr control in the field of weak rigidity drilling.
Description
Technical Field
The invention relates to the technical field of machining, in particular to a digital twinning control method and system for weak rigidity drilling.
Background
The robot drilling is applied to scenes such as aircraft skin, has large flexibility and large operation range, and can realize large-space drilling processing which is difficult to realize by traditional drilling equipment. However, the robot belongs to a typical weak rigid processing device, and the robot is subjected to exciting force to generate large vibration during processing. Instability in the machining conditions leads to a reduction in the quality of the machining, which is typical of a less rigid drilling system as the exit burr. Burrs generated under the weak rigidity drilling condition have the characteristics of large height and complex shape, and can cause the defects of microcracks, stress concentration and the like on a workpiece, so that the service life of the workpiece is greatly reduced. Through the research of documents, the existing research mostly reduces the negative influence caused by large amplitude flutter by optimizing the processing technological parameters (mainly main shaft rotating speed, feed amount and the like), designing special processing equipment (such as a step drill and the like) or using a special auxiliary processing mode (such as vibration drilling and the like), thereby inhibiting burr generation and optimizing the processing quality. However, this method belongs to open-loop control, and cannot make decision adjustment in time for the abnormality that has occurred in the machining process. Therefore, a closed-loop system needs to be constructed to realize process monitoring and timely feedback adjustment of the abnormality. However, the weak rigidity drilling processing has complex coupling influence factors such as multi-degree-of-freedom vibration, multi-directional cutting force and the like, and the specific change of the processing process is difficult to accurately solve through the traditional univariate or multivariate analysis. Therefore, the closed-loop control system established by the traditional method is difficult to realize accurate monitoring and effective control of the machining process,
disclosure of Invention
The invention aims to provide a digital twinning control method and system for weak rigidity drilling, which utilize a virtual model of digital twinning and real-time virtual-real interaction characteristics to realize burr control in the field of weak rigidity drilling.
In order to achieve the purpose, the invention provides the following scheme:
a digital twinning control method facing weak-rigidity drilling comprises the following steps:
collecting multidimensional time sequence data in the processing process;
segmenting the time sequence data according to the processing progress;
performing time-frequency domain analysis on the segmented data, and extracting a characteristic value;
inputting the characteristic value into a mechanism model to obtain a burr state parameter;
predicting the burr height according to the burr state parameters, the current processing technological parameters and the residual processing progress;
constructing a decision information base; the decision information base comprises optimal processing technological parameters for inhibiting burrs in different processing scenes;
and when the predicted burr height exceeds the burr height threshold, selecting the optimal processing process parameter from the decision information base for feedback.
Optionally, the machining process parameters include spindle speed and feed.
Optionally, the predicting the burr height according to the burr state parameter, the current processing process parameter and the remaining processing progress specifically includes:
inputting the burr state parameters, the current processing technological parameters and the residual processing progress into a gated cyclic neural network, and predicting burr state parameters in the residual processing progress;
and inputting the predicted burr state parameters into the full-connection layer to predict the burr height.
Optionally, the constructing a decision information base specifically includes:
predicting burr heights under different scenes according to process parameters, burr state parameters and processing progress under different processing scenes;
and according to the predicted burr heights under different processing scenes, performing self-adaptive calculation by adopting a genetic algorithm, determining the optimal processing technological parameters of the different processing scenes, and forming a decision information base.
Optionally, selecting an optimal processing process parameter from the decision information base for feedback, specifically including:
selecting a plurality of decision libraries which are closest from the decision information libraries by adopting an Euclidean distance algorithm according to the burr state parameters, the current processing technological parameters and the residual processing progress respectively;
weighting the burr state parameters, the current processing technological parameters and the residual processing progress;
and selecting the optimal processing technological parameter from a plurality of decision libraries by adopting a neighbor model according to the weighted burr state parameter, the current processing technological parameter and the residual processing progress.
The invention also provides a weak rigidity drilling oriented digital twin control system, which comprises:
the data acquisition module is used for acquiring multi-dimensional time sequence data in the processing process;
the segmentation module is used for segmenting the time sequence data according to the processing progress;
the characteristic extraction module is used for carrying out time-frequency domain analysis on the segmented data and extracting a characteristic value;
the burr state parameter determining module is used for inputting the characteristic value into a mechanism model to obtain a burr state parameter;
the burr height prediction module is used for predicting the burr height according to the burr state parameters, the current processing process parameters and the residual processing progress;
the decision information base building module is used for building a decision information base; the decision information base comprises optimal processing technological parameters for inhibiting burrs in different processing scenes;
and the optimal processing parameter selection module is used for selecting the optimal processing parameter from the decision information base for feedback when the predicted height of the burr exceeds the burr height threshold.
Optionally, the machining process parameters include spindle speed and feed.
Optionally, the burr height prediction module specifically includes:
the burr state parameter prediction unit in the residual processing progress is used for inputting the burr state parameter, the current processing technological parameter and the residual processing progress into a gated cyclic neural network and predicting the burr state parameter in the residual processing progress;
and the burr height prediction unit is used for inputting the predicted burr state parameters into the full-connection layer to predict the burr height.
Optionally, the decision information base building module specifically includes:
the burr height prediction unit under different scenes is used for predicting burr heights under different scenes according to process parameters, burr state parameters and processing progress under different processing scenes;
and the decision information base construction unit is used for carrying out self-adaptive calculation by adopting a genetic algorithm according to the predicted burr heights in different processing scenes, determining the optimal processing technological parameters in different processing scenes and forming a decision information base.
Optionally, the optimal processing parameter selection module specifically includes:
a decision base selection unit, configured to select a plurality of decision bases that are closest to each other from the decision information bases by using an euclidean distance algorithm according to the burr state parameter, the current processing technology parameter, and the remaining processing progress, respectively;
the weighting unit is used for weighting the burr state parameters, the current processing technological parameters and the residual processing progress;
and the optimal processing technological parameter selection unit is used for selecting the optimal processing technological parameters from the multiple decision libraries by adopting a neighbor model according to the weighted burr state parameters, the current processing technological parameters and the residual processing progress.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
1) based on a mechanism model, the invention realizes the real-time monitoring of the weak rigidity drilling process by a gated recurrent neural network (GRU) to achieve the rapid identification of the abnormal processing state;
2) the method realizes the decision and feedback of the abnormity based on a Genetic Algorithm (GA) and a neighbor algorithm (KNN);
3) the method is based on the real-time performance and the virtual-real interaction of the digital twin, realizes the optimal control of the machining state, and reduces the burr height of weak-rigidity drilling.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow chart of a digital twinning control method for weak stiffness drilling according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the basic components of a digital twin control system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a glitch status monitoring algorithm according to an embodiment of the present invention;
FIG. 4 is a flow chart of the operation of a digital twin control system according to an embodiment of the present invention;
fig. 5 is a schematic diagram illustrating search of decision information according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a digital twinning control method and system for weak rigidity drilling, which utilize a virtual model of digital twinning and real-time virtual-real interaction characteristics to realize burr control in the field of weak rigidity drilling.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in FIG. 1, the invention provides a digital twinning control method facing weak-rigidity drilling, which comprises the following steps:
step 101: and collecting multidimensional time sequence data in the processing process.
Step 102: and segmenting the time sequence data according to the processing progress.
Step 103: and carrying out time-frequency domain analysis on the segmented data, and extracting a characteristic value.
Step 104: and inputting the characteristic value into a mechanism model to obtain a burr state parameter.
Step 105: and predicting the burr height according to the burr state parameters, the current machining process parameters and the residual machining progress.
Step 106: constructing a decision information base; the decision information base comprises optimal processing technological parameters for inhibiting burrs in different processing scenes.
Step 107: and when the predicted burr height exceeds the burr height threshold, selecting the optimal processing process parameter from the decision information base for feedback.
As an optional embodiment, step 105 specifically includes:
step 1051: and inputting the burr state parameters, the current processing technological parameters and the residual processing progress into a gated cyclic neural network, and predicting the burr state parameters in the residual processing progress.
Step 1052: and inputting the predicted burr state parameters into the full-connection layer to predict the burr height.
As an optional embodiment, step 106 specifically includes:
step 1061: and predicting the burr heights under different scenes according to the process parameters, the burr state parameters and the machining progress under different machining scenes.
Step 1062: and according to the predicted burr heights under different processing scenes, performing self-adaptive calculation by adopting a genetic algorithm, determining the optimal processing technological parameters of the different processing scenes, and forming a decision information base.
As an optional embodiment, step 107 specifically includes:
step 1071: selecting a plurality of decision bases (GA) which are closest from the decision information bases by adopting an Euclidean distance algorithm according to the burr state parameters, the current processing technological parameters and the residual processing progress respectivelyi)。
Step 1072: and weighting the burr state parameters, the current processing technological parameters and the residual processing progress.
Step 1073: and selecting the optimal processing technological parameter from a plurality of decision libraries by adopting a neighbor model according to the weighted burr state parameter, the current processing technological parameter and the residual processing progress.
The following detailed description of the principles and processes of the present invention refers to specific embodiments:
as shown in FIG. 2, the invention establishes a digital twin model (a physical model and a virtual model) and a model interaction module (a virtual model sensing physical model: a data processing module, a virtual model controlling physical model: a decision control module):
the solid model comprises processing equipment and a workpiece entity, and realizes a drilling processing process;
the virtual model is used for establishing mapping of the entity model in a digital space, wherein the mapping comprises mapping of model physical attributes (geometry, materials and the like) and mechanism rules (a force field model, a vibration model and the like), and the monitoring process of the entity model is realized through simulation of drilling behaviors;
the data processing module is used for collecting multi-dimensional data (such as force, vibration, heat and the like) in the processing process of the entity model, transmitting the multi-dimensional data through a uniform interface, denoising the original data by utilizing wavelet transformation, and identifying the characteristics by utilizing mechanism rules;
the decision control module performs feedback control on abnormal machining phenomena obtained through virtual model simulation (simulate), and comprises two parts: generation of information bases (offline) and search of decision information (online).
The digital twin control system is implemented as follows:
and a data processing stage, which comprises two stages of data acquisition and preprocessing and time-frequency domain feature extraction (the data processing stage is used for converting acquired signals into available data of a virtual model). The data processing module performs equal-proportion segmentation or self-adaptive segmentation on the acquired time series data according to the processing progress, and extracts shallow characteristic values (such as mean value, amplitude and the like) in each time period through time-frequency domain analysis.
In the state monitoring stage, the characteristic values (time-frequency domain characteristic values, such as the mean value F of each time segment and the amplitude ratio A extracted by frequency domain analysis) obtained by data processing are usedtEtc.) input mechanism model H to obtain the spur state parameters. Then, the obtained burr state parameters H, current processing process parameters (feed amount f, spindle rotation speed v), residual processing progress t and other information are transmitted to a gated recurrent neural network (GRU) of a core to perform time sequence prediction simulation of the burr state parameters, and the process is as shown in fig. 3 (a mechanism model integrates priori knowledge in the system, so that the system can eliminate interference information at the bottom layer, and the training speed of a virtual model behavior rule model is increased). Because the workpiece state of the final processing stage has great influence on burr generation, the invention finally constructs a full-connection layer in the GRU network, and transmits the burr state parameter set (Hi) of the final stage obtained by GRU simulation as input into full-connection to carry out burr height under the current processing state
And (4) predicting. And establishing a burr height threshold according to the machining process requirement, and comparing the burr height threshold with a simulation result of the system to acquire abnormal information.
The decision control process includes two parts: the off-line information base generation stage is used for generating a plurality of feedback strategies, and the on-line decision feedback stage is used for adjusting the processing state in real time.
And a generation stage of a decision information base, wherein the established virtual model is used as a simulation platform, the main shaft rotating speed v, the feeding amount f, the residual processing progress t and the current burr state parameter H are used as input parameters of a virtual model behavior rule (mechanism model H + gated recurrent neural network GRU (H, f, v, t)), the adaptability score (burr height value) of the current process parameter is calculated through simulation, the simulation scores of all sample parameters are sequenced, and a sample with a higher score (good processing process parameter in a certain time segment) is selected and coded by a coder of a variable-score self-coder. The exchange of partial fragments and the re-random generation (crossover and mutation) of partial fragments for each generated set of coding sequences through a genetic algorithm generate filial samples and the filial samples are generated through decoding of a decoder of a variational self-encoder. In order to generate a better sample, the sub-sample is continuously simulated, and through continuous iteration, optimal machining process parameters of the inhibition burr under each scene can be generated. Meanwhile, because the online decision adjustment needs to be limited within a certain range, the system constructs a final decision information base consisting of a plurality of strategy groups by taking the spindle rotating speed, the feeding amount, the residual processing progress and the burr state parameters as division standards.
Genetic algorithms are typically heuristic algorithms that can be used for single-target optimization processes, but they run at a slow speed and are therefore not suitable for use in control systems with relatively high real-time performance. The present invention utilizes genetic algorithms to generate spur control decisions prior to operation of a digital twin system. The key of the genetic algorithm is the selection process, and different selection processes are provided for different scenes. The method takes the established digital twin virtual model as a target environment of a genetic algorithm, utilizes the burr height prediction value obtained by virtual model simulation to calculate the fitness of randomly generated processing technological parameters, residual processing progress and burr state parameters, then carries out one iteration through selection, intersection and variation, and screens out the optimal processing technological parameters capable of inhibiting burr generation through the set iteration times. However, since the system is intended to adjust the state during the machining process, the machining process parameters cannot be adjusted at will, and therefore, the system needs to be limited to a certain adjustment range. Aiming at the problem, the invention selects to establish a sub-decision library under different conditions.
In the decision feedback stage, the decision feedback stage searches the optimal strategy according to the identified abnormal information, namely the identified abnormal informationAnd adjusting the processing technological parameters according to the processing state. Meanwhile, in order to prevent the processing state from generating unknown excitation caused by the sudden change of the processing technological parameters, the parameter adjustment in the processing process is limited in a certain area. The specific process is as follows: as shown in fig. 5, the closest strategy group (i.e., the euclidean distance between the strategy group and the input burr state parameter value is the minimum within the range of the feed amount, the spindle rotation speed and the residual machining progress) is selected according to the feed amount, the spindle rotation speed, the residual machining progress and the burr state parameter in turn according to the importance, and then a certain weight difference (specifically, increasing K is given to four variables of the spindle rotation speed, the feed amount, the residual machining progress and the current state parameter) is given to the spindle rotation speed, the spindle rotation speed and the residual machining progressiThe weight value of the system is used for strengthening the rotating speed or the feeding amount of the main shaft, so that the change range of strengthening elements is reduced as much as possible by the searched strategy, the probability of excitation generation of the system is reduced, and the weighted value is used as the input of a neighbor model (KNN) to search and select the optimal machining process parameter (f)new,vnew) In order to improve the effectiveness of feedback, the system transmits the fed-back new parameters to a virtual model for simulation (predicting the burr state parameters and the final burr height value under the new parameters), if the simulation result shows that the feedback strategy can realize the burr suppression, the feedback strategy is converted into a controller adjusting code (Control function) for Control, and if the simulation result shows that the feedback strategy can realize the burr suppression, the feedback strategy adjusts the weight and searches the strategy again.
The digital twin control system operation mode established by the present invention is shown in fig. 4. Setting initial processing technological parameters, and drilling a processing entity by using a weak rigidity drilling system; the method comprises the following steps that a drilling system is internally provided with a sensor and externally provided with a sensor, dynamic signals related to the drilling process are collected, signal preprocessing and characteristic extraction are carried out at an edge end, and processed target parameters (such as cutting force characteristic values and the like) are obtained; after receiving the target parameters, the virtual model performs virtual simulation according to the established behavior rules, so as to realize mapping of a real machining state and monitor burr state parameters in the drilling process; the decision control module matches the process state model and the local optimization model in the model base according to the monitored abnormal information, utilizes the matched model to perform local optimization in the corresponding decision information base, outputs the optimal decision information, and realizes the real-time adjustment of the processing process parameters by controlling the related equipment.
Based on the weak rigidity drilling-oriented digital twin control method provided by the invention, the invention also provides a weak rigidity drilling-oriented digital twin control system, which comprises the following steps:
the data acquisition module is used for acquiring multi-dimensional time sequence data in the processing process;
the segmentation module is used for segmenting the time sequence data according to the processing progress;
the characteristic extraction module is used for carrying out time-frequency domain analysis on the segmented data and extracting a characteristic value;
the burr state parameter determining module is used for inputting the characteristic value into a mechanism model to obtain a burr state parameter;
the burr height prediction module is used for predicting the burr height according to the burr state parameters, the current processing process parameters and the residual processing progress;
the decision information base building module is used for building a decision information base; the decision information base comprises optimal processing technological parameters for inhibiting burrs in different processing scenes;
and the optimal processing parameter selection module is used for selecting the optimal processing parameter from the decision information base for feedback when the predicted height of the burr exceeds the burr height threshold.
The burr height prediction module specifically includes:
the burr state parameter prediction unit in the residual processing progress is used for inputting the burr state parameter, the current processing technological parameter and the residual processing progress into a gated cyclic neural network and predicting the burr state parameter in the residual processing progress;
and the burr height prediction unit is used for inputting the predicted burr state parameters into the full-connection layer to predict the burr height.
The decision information base building module specifically comprises:
the burr height prediction unit under different scenes is used for predicting burr heights under different scenes according to process parameters, burr state parameters and processing progress under different processing scenes;
and the decision information base construction unit is used for carrying out self-adaptive calculation by adopting a genetic algorithm according to the predicted burr heights in different processing scenes, determining the optimal processing technological parameters in different processing scenes and forming a decision information base.
The optimal processing technological parameter selection module specifically comprises:
a decision base selection unit, configured to select a plurality of decision bases that are closest to each other from the decision information bases by using an euclidean distance algorithm according to the burr state parameter, the current processing technology parameter, and the remaining processing progress, respectively;
the weighting unit is used for weighting the burr state parameters, the current processing technological parameters and the residual processing progress;
and the optimal processing technological parameter selection unit is used for selecting the optimal processing technological parameters from the multiple decision libraries by adopting a neighbor model according to the weighted burr state parameters, the current processing technological parameters and the residual processing progress.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (10)
1. A digital twinning control method for weak rigidity drilling is characterized by comprising the following steps:
collecting multidimensional time sequence data in the processing process;
segmenting the time sequence data according to the processing progress;
performing time-frequency domain analysis on the segmented data, and extracting a characteristic value;
inputting the characteristic value into a mechanism model to obtain a burr state parameter;
predicting the burr height according to the burr state parameters, the current processing technological parameters and the residual processing progress;
constructing a decision information base; the decision information base comprises optimal processing technological parameters for inhibiting burrs in different processing scenes;
and when the predicted burr height exceeds the burr height threshold, selecting the optimal processing process parameter from the decision information base for feedback.
2. The digital twinning control method for weak rigidity drilling as claimed in claim 1, wherein the machining process parameters include main shaft rotation speed and feed amount.
3. The weak-rigidity drilling oriented digital twinning control method as claimed in claim 1, wherein the predicting of burr height according to the burr state parameter, current machining process parameter and remaining machining progress specifically comprises:
inputting the burr state parameters, the current processing technological parameters and the residual processing progress into a gated cyclic neural network, and predicting burr state parameters in the residual processing progress;
and inputting the predicted burr state parameters into the full-connection layer to predict the burr height.
4. The weak-rigidity drilling oriented digital twinning control method as claimed in claim 1, wherein the constructing of the decision information base specifically comprises:
predicting burr heights under different scenes according to process parameters, burr state parameters and processing progress under different processing scenes;
and according to the predicted burr heights under different processing scenes, performing self-adaptive calculation by adopting a genetic algorithm, determining the optimal processing technological parameters of the different processing scenes, and forming a decision information base.
5. The weak-rigidity drilling oriented digital twinning control method as claimed in claim 1, wherein the step of selecting the optimal machining process parameters from the decision information base for feedback specifically comprises:
selecting a plurality of decision libraries which are closest from the decision information libraries by adopting an Euclidean distance algorithm according to the burr state parameters, the current processing technological parameters and the residual processing progress respectively;
weighting the burr state parameters, the current processing technological parameters and the residual processing progress;
and selecting the optimal processing technological parameter from a plurality of decision libraries by adopting a neighbor model according to the weighted burr state parameter, the current processing technological parameter and the residual processing progress.
6. A digital twinning control system for weak rigid drilling, comprising:
the data acquisition module is used for acquiring multi-dimensional time sequence data in the processing process;
the segmentation module is used for segmenting the time sequence data according to the processing progress;
the characteristic extraction module is used for carrying out time-frequency domain analysis on the segmented data and extracting a characteristic value;
the burr state parameter determining module is used for inputting the characteristic value into a mechanism model to obtain a burr state parameter;
the burr height prediction module is used for predicting the burr height according to the burr state parameters, the current processing process parameters and the residual processing progress;
the decision information base building module is used for building a decision information base; the decision information base comprises optimal processing technological parameters for inhibiting burrs in different processing scenes;
and the optimal processing parameter selection module is used for selecting the optimal processing parameter from the decision information base for feedback when the predicted height of the burr exceeds the burr height threshold.
7. The digital twinning control system for weak rigid drilling as claimed in claim 6, wherein the machining process parameters include spindle rotation speed and feed amount.
8. The digital twinning control system for weak rigid drilling as claimed in claim 6, wherein the burr height prediction module specifically comprises:
the burr state parameter prediction unit in the residual processing progress is used for inputting the burr state parameter, the current processing technological parameter and the residual processing progress into a gated cyclic neural network and predicting the burr state parameter in the residual processing progress;
and the burr height prediction unit is used for inputting the predicted burr state parameters into the full-connection layer to predict the burr height.
9. The weak-rigidity-drilling-oriented digital twinning control system of claim 6, wherein the decision information base construction module specifically comprises:
the burr height prediction unit under different scenes is used for predicting burr heights under different scenes according to process parameters, burr state parameters and processing progress under different processing scenes;
and the decision information base construction unit is used for carrying out self-adaptive calculation by adopting a genetic algorithm according to the predicted burr heights in different processing scenes, determining the optimal processing technological parameters in different processing scenes and forming a decision information base.
10. The digital twinning control system for weak rigid drilling as claimed in claim 6, wherein the optimal machining process parameter selection module specifically comprises:
a decision base selection unit, configured to select a plurality of decision bases that are closest to each other from the decision information bases by using an euclidean distance algorithm according to the burr state parameter, the current processing technology parameter, and the remaining processing progress, respectively;
the weighting unit is used for weighting the burr state parameters, the current processing technological parameters and the residual processing progress;
and the optimal processing technological parameter selection unit is used for selecting the optimal processing technological parameters from the multiple decision libraries by adopting a neighbor model according to the weighted burr state parameters, the current processing technological parameters and the residual processing progress.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110987837.XA CN113579849A (en) | 2021-08-26 | 2021-08-26 | Digital twinning control method and system for weak rigidity drilling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110987837.XA CN113579849A (en) | 2021-08-26 | 2021-08-26 | Digital twinning control method and system for weak rigidity drilling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113579849A true CN113579849A (en) | 2021-11-02 |
Family
ID=78239427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110987837.XA Pending CN113579849A (en) | 2021-08-26 | 2021-08-26 | Digital twinning control method and system for weak rigidity drilling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113579849A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114161422A (en) * | 2021-12-20 | 2022-03-11 | 东华大学 | Method for predicting height of burrs at outlet of stainless steel plate drilled by robot |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104932421A (en) * | 2015-06-19 | 2015-09-23 | 华中科技大学 | Numerical control machine work process CPS modeling method based on instruction domain analysis |
| CN109492881A (en) * | 2018-10-19 | 2019-03-19 | 江苏科技大学 | Based on the twin mechanical processing technique dynamic evaluation method of number |
| CN110389534A (en) * | 2019-07-01 | 2019-10-29 | 东华大学 | A kind of Multi-axis motion control virtual experimental system based on the twin technology of number |
| CN110900307A (en) * | 2019-11-22 | 2020-03-24 | 北京航空航天大学 | Numerical control machine tool cutter monitoring system driven by digital twin |
| US20200282504A1 (en) * | 2017-08-07 | 2020-09-10 | Franz Haimer Maschinenbau Kg | Creating a digital twin in a processing centre |
| CN111968004A (en) * | 2020-08-07 | 2020-11-20 | 东华大学 | High-precision product assembling and adjusting integrated system based on digital twins |
-
2021
- 2021-08-26 CN CN202110987837.XA patent/CN113579849A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104932421A (en) * | 2015-06-19 | 2015-09-23 | 华中科技大学 | Numerical control machine work process CPS modeling method based on instruction domain analysis |
| US20200282504A1 (en) * | 2017-08-07 | 2020-09-10 | Franz Haimer Maschinenbau Kg | Creating a digital twin in a processing centre |
| CN109492881A (en) * | 2018-10-19 | 2019-03-19 | 江苏科技大学 | Based on the twin mechanical processing technique dynamic evaluation method of number |
| CN110389534A (en) * | 2019-07-01 | 2019-10-29 | 东华大学 | A kind of Multi-axis motion control virtual experimental system based on the twin technology of number |
| CN110900307A (en) * | 2019-11-22 | 2020-03-24 | 北京航空航天大学 | Numerical control machine tool cutter monitoring system driven by digital twin |
| CN111968004A (en) * | 2020-08-07 | 2020-11-20 | 东华大学 | High-precision product assembling and adjusting integrated system based on digital twins |
Non-Patent Citations (1)
| Title |
|---|
| 许敏俊等: "数字孪生驱动下的弱刚性钻削毛刺控制", 《计算机集成制造系统》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114161422A (en) * | 2021-12-20 | 2022-03-11 | 东华大学 | Method for predicting height of burrs at outlet of stainless steel plate drilled by robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhou et al. | Tool wear condition monitoring in milling process based on current sensors | |
| CN110070060B (en) | Fault diagnosis method for bearing equipment | |
| CN109102005B (en) | Small sample deep learning method based on shallow model knowledge migration | |
| Du Preez et al. | Machine learning in cutting processes as enabler for smart sustainable manufacturing | |
| CN114619292B (en) | Milling cutter wear monitoring method based on fusion of wavelet denoising and attention mechanism with GRU network | |
| CN113664612A (en) | Numerical control machine tool milling cutter abrasion real-time monitoring method based on deep convolutional neural network | |
| CN108197647B (en) | Rapid clustering method for automobile starter endurance test data | |
| CN113458873B (en) | Method for predicting wear loss and residual life of cutter | |
| CN109034076A (en) | A kind of automatic clustering method and automatic cluster system of mechanical fault signals | |
| CN112207631B (en) | Method for generating tool detection model, method, system, device and medium for detecting tool detection model | |
| CN113579849A (en) | Digital twinning control method and system for weak rigidity drilling | |
| CN115587290A (en) | Aero-engine fault diagnosis method based on variational self-coding generation countermeasure network | |
| Song et al. | Intelligent diagnosis method for machinery by sequential auto-reorganization of histogram | |
| CN114757087A (en) | Tool wear prediction method based on dynamic principal component analysis and LSTM | |
| CN117477993A (en) | Brushless direct current motor rotating speed control method and system | |
| CN115781136B (en) | Intelligent recognition and optimization feedback method for welding heat input abnormality | |
| CN113523904A (en) | Cutter wear detection method | |
| CN116401545A (en) | Multimode fusion type turbine runout analysis method | |
| CN115730255A (en) | Motor fault diagnosis and analysis method based on transfer learning and multi-source information fusion | |
| CN116628418A (en) | Vulnerable part failure prediction method based on sensor signals and deep migration learning | |
| CN116111906A (en) | Special motor with hydraulic brake for turning and milling and control method thereof | |
| CN110543908B (en) | Control chart pattern recognition method based on dynamic observation window | |
| CN112183642A (en) | Method and system for detecting coal consumption of cement firing based on random forest model | |
| CN118409608B (en) | Rotation control method and system of servo high-speed machine | |
| CN117341261B (en) | Intelligent control method and system for servo direct-drive screw press |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211102 |
|
| RJ01 | Rejection of invention patent application after publication |