CN109977464B - Prediction method of piston cutting deformation based on BP neural network - Google Patents
Prediction method of piston cutting deformation based on BP neural network Download PDFInfo
- Publication number
- CN109977464B CN109977464B CN201910119528.3A CN201910119528A CN109977464B CN 109977464 B CN109977464 B CN 109977464B CN 201910119528 A CN201910119528 A CN 201910119528A CN 109977464 B CN109977464 B CN 109977464B
- Authority
- CN
- China
- Prior art keywords
- cutting
- piston
- deformation
- neural network
- machining
- 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.)
- Active
Links
- 238000005520 cutting process Methods 0.000 title claims abstract description 113
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 41
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000003754 machining Methods 0.000 claims abstract description 43
- 238000004458 analytical method Methods 0.000 claims abstract description 27
- 238000012545 processing Methods 0.000 claims abstract description 26
- 238000004088 simulation Methods 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 21
- 210000002569 neuron Anatomy 0.000 claims abstract description 17
- 230000008569 process Effects 0.000 claims abstract description 8
- 238000012360 testing method Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 6
- 238000013461 design Methods 0.000 claims description 5
- 238000012546 transfer Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 239000000463 material Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 2
- 238000002474 experimental method Methods 0.000 claims description 2
- 238000001914 filtration Methods 0.000 claims description 2
- 238000007670 refining Methods 0.000 claims description 2
- 238000009827 uniform distribution Methods 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 3
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Numerical Control (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a prediction method of piston cutting machining deformation based on a BP neural network, which carries out piston cutting machining simulation through finite element analysis, takes the predicted deformation obtained through simulation experiments as a training sample, establishes a BP neural network topology model according to the training sample, learns the BP neural network topology model according to preset training parameters, and finally predicts the piston cutting machining deformation according to the BP neural network topology model after adjusting the weight of each neuron. Compared with a simulation process prediction method, the method has the advantages that the prediction time is greatly shortened, the establishment efficiency and the prediction precision of the prediction model are improved, and a high-reliability processing prediction guide can be provided for production and processing rapidly.
Description
Technical Field
The invention relates to a method for predicting piston cutting deformation, in particular to a method for predicting piston cutting deformation based on a BP neural network.
Background
The quality of the final product of the piston as a core part of the diesel engine in the technical processing process is closely related to various technical factors in the processing process, and if the design technical parameters are unreasonable, the problem of out-of-tolerance of the size of the skirt of the piston can occur, so that the quality of the product processed by the piston is influenced, and the service performance of the diesel engine is seriously influenced.
The prediction of the piston machining deformation is generally based on simulation machining and theoretical analysis, and a deformation prediction model is established by means of finite element analysis technology, so that the deformation prediction analysis is carried out according to the prediction model. Finite element simulation analysis techniques try to analyze effective means of field quantity variation, and in recent years with advances in computer technology and advances in cutting principle research, cutting physics simulation techniques have evolved and have been applied to analysis of force loads, temperature fields, stress fields, etc. during cutting. However, cutting physical simulation calculation consumes a large amount of calculation resources, qualitative analysis is often adopted for deformation prediction analysis of different working conditions of a specific machining process, the calculation efficiency is low, and real-time prediction of each working condition through cutting simulation is often not practical.
Disclosure of Invention
The invention aims to: the invention aims to provide a prediction method of piston cutting deformation based on a BP neural network, which can accurately predict the deformation after piston cutting.
The technical scheme is as follows: a prediction method of piston cutting deformation based on BP neural network comprises the following steps:
(1) Taking the piston stress distribution obtained by the piston rough machining finite element analysis as an initial condition of the piston finish machining finite element analysis, and simultaneously considering a workpiece tool to add clamping constraint of a piston part into a piston finish machining finite element simulation model;
(2) Working condition design is carried out aiming at cutting machining parameters related to rough machining of the piston, cutting working conditions are grouped according to a principle of control variables, and corresponding variation trends of the cutting parameters can be accurately distinguished by different working condition groups;
(3) Solving the distribution result of the cutting stress and strain of each group of pistons by using a finite element analysis method to obtain a machining deformation simulation result under the corresponding working condition;
(4) Performing a piston cutting experiment according to a designed cutting working condition, measuring an experimental value of the deformation of a piston skirt under the corresponding working condition, checking the change rule of cutting deformation along with cutting parameters by comparing and analyzing the experimental value with an imitation value, and correcting a corresponding finite element analysis model;
(5) Taking the predicted object deformation obtained by finite element analysis as a training sample, determining the neuron numbers of an input layer, an hidden layer and an output layer in the BP neural network topology model based on the piston cutting deformation and an input-output mapping relation, and establishing the BP neural network topology model based on the piston cutting deformation;
(6) Performing BP neural network topology model learning based on piston cutting processing deformation according to a preset training sample, adjusting weights of all neurons in the model according to learning conditions, and performing reliability verification on the BP neural network topology model with the weights adjusted;
(7) And calculating a piston deformation value in the piston cutting process by using a BP neural network topology model based on the piston cutting deformation.
The beneficial effects are that: (1) Compared with a finite element simulation process prediction method, the method has the advantages that the prediction time is greatly shortened, the establishment efficiency of a prediction model is improved, and a processing prediction guide with higher reliability can be provided for production and processing rapidly; (2) The relative error value of the predicted value and the actual value is not more than 10%, and the invention has higher reliability.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a flow chart of a method of the present invention for a BP neural network topology model based on piston cutting process variation;
FIG. 3 is a schematic view of a piston long and short shaft deformation in an embodiment of the invention;
FIG. 4 is a graph showing the variation of the error curve of the training result of the present invention.
Detailed Description
The mechanism for generating the cutting processing deformation of the piston is that a part blank containing initial residual stress gradually evolves an internal stress field under the comprehensive influence of factors such as clamping force load, cutting heat load, constraint condition change caused by a large amount of material removal and the like, and after the processing is completed and the clamping is removed, the internal stress field further evolves and causes the part to deform, and finally, the self-balancing state of the internal stress is achieved. To predict and analyze the machining deformation of the piston, the evolution process of the internal stress field under the external effect needs to be tracked, and the data processing is performed by an efficient operation means, so that the final machining deformation is obtained.
As shown in fig. 1, in this example, the machining deformation prediction is performed for the machining process of rough boring of the forged aluminum piston of the PA6 type marine diesel engine, and the specific steps are as follows:
(1) And taking the piston stress distribution obtained by the piston rough machining finite element analysis as an initial condition of the piston finish machining finite element analysis, and simultaneously considering a workpiece tool, wherein the clamping constraint of the piston part is added into a piston finish machining finite element simulation model, and the workpiece tool can be clamping force load and the like.
(2) Working condition design is carried out aiming at cutting machining parameters related to a rough boring machining process, cutting depth, cutting speed and feeding quantity are selected as influence analysis factors of piston machining deformation according to a principle of control variables, and each machining parameter adopts three horizontal parameters of high, medium and low to obtain 9 groups of different test working conditions.
(3) The stress strain distribution result of the marine diesel engine piston cutting processing is obtained by utilizing a finite element analysis method, and the processing deformation simulation result under the corresponding working condition is obtained, wherein the specific steps are as follows:
31. geometric model and grid division: and (3) establishing a tool model by using three-dimensional modeling software such as Unigraphics, introducing process simulation finite element software such as DEFORM-3D, meshing and setting tool and workpiece material properties.
32. And sequentially applying different cutting speeds, cutting depths and feeding amounts in the working condition table, calculating cutting force and cutting heat under corresponding working conditions through corresponding software such as DEFORM-3D, deriving the curves of the cutting force and the cutting heat, introducing the curves into mathematical software such as Matlab for data filtering, removing abnormal data which seriously deviate from the average level, and obtaining steady-state values of the cutting force and the cutting temperature under each working condition.
33. And performing grid division and tool constraint adding on the forged aluminum piston model, and applying the cutting force and the cutting heat obtained by calculation according to each group to the finite element model of the piston based on ANSYS finite element analysis software to obtain the deformation of the skirt portion of the piston.
(4) According to the control variable principle in the test working conditions, the actual cutting test is sequentially carried out for the 9 working conditions, and the cutting result is measured by using a test tool such as a three-coordinate measuring machine, so that the cutting deformation under each working condition, including the radial dimensions of the piston skirt in the major axis and minor axis directions, is calculated. And comparing the test machining data with the finite element simulation data, analyzing the finite element analysis result and the test machining result, and verifying the change rule of the cutting machining deformation along with the cutting machining parameters. If the test data and the finite element simulation data generate great deviation, the finite element simulation model needs to be adjusted, and the finite element simulation model in the cutting process is ensured to have higher reliability by checking whether the processing constraint conditions are consistent with the test working conditions, refining the grid of the finite element model, adjusting the application modes of cutting force and cutting heat and the like, so that the generation of samples in the next step is ensured to have higher reference degree.
(5) Taking the deformation of the predicted object obtained by finite element analysis as a training sample, designing a marine diesel engine cutting orthogonal test according to the requirements of a BP neural network prediction model on the compatibility, the ergodic performance and the compactness of the training sample, and taking the cutting influence factors into consideration, wherein three factors five levels (the three factors comprise cutting speed, cutting depth and feeding amount; the five levels refer to the deformation of the cutting piston length and the short axis under each working condition combination under the reasonable three-factor processing influence factors, wherein each influence factor is uniformly divided into five level values (for example, the cutting speed five levels are 100/200/300/400/500 m/min) and the deformation of the cutting piston length and the short axis under each working condition combination is calculated based on the finite element simulation model after adjustment and optimization;
the number of neurons of an input layer, an hidden layer and an output layer in the BP neural network topology model based on the piston cutting deformation and the input-output mapping relation are determined, and the structure of the BP neural network topology model is different due to different parameters to be predicted. As shown in fig. 2, three elements of cutting in cutting processing are used as an input layer of a BP neural network topology model, namely, the input layer has 3 neurons, and an input vector is x= [ X1, X2, X3], wherein X1 is a cutting depth, X2 is a cutting speed, and X3 is a feeding amount; the deformation of the piston skirt generated after the cutting processing is used as an output layer of the BP neural network topological model, namely, the output layer is provided with 2 neurons, the output vector is Y= [ [ delta ] x and [ delta ] Y ], wherein [ delta ] x is the accumulated deformation of the piston in the long axis direction, [ delta ] Y is the accumulated deformation of the piston in the short axis direction, and the hidden layer is provided with 5 neurons.
(6) And learning the topology model of the BP neural network according to a preset training sample. The method specifically comprises the following steps:
61. because the neural network has strong adaptability, firstly, input variables are normalized, so that the importance of each variable from the beginning is equivalent, and the ownership of the network is within a small range, thereby reducing the difficulty of network training. When the input and output are linearly normalized to [0.1,0.9 ]]The performance of the network is better during the interval, so can be adoptedSample normalization formula for BP neural network topology model training, wherein x is max And x min Respectively, x is the maximum value and the minimum value of a certain dimension of the input/output vector i And->Is the value before and after normalization.
62. The Levebberg-Marquardt algorithm is selected as a training algorithm of the BP neural network based on the piston cutting deformation to adjust the weight of each neuron in the BP neural network topology model. The basic iterative formula of the algorithm is:wherein g h =g/2, g is the error function E versus the weight vector +.>Is a gradient of (2); j is an error function, jacobian matrix I for differentiating the weight vector is an identity matrix, and lambda is a certain non-negative number capable of being adaptively adjusted; the existence of lambda and I can well treat the pathological matrix J T J, stability of the algorithm is better.
63. Selecting hyperbolic tangent sigmoid function as transfer function of neural network training hidden layer neuron, selecting logarithmic sigmoid function as transfer function of output layer neuron, using 25 groups of training samples obtained by finite element analysis as input, setting network target error as 0.001 when the neural network is trained, and utilizing Matlab tool to make BP neural network training.
(7) And carrying out deformation prediction analysis and reliability verification according to the trained BP neural network topology model. In order to verify the prediction capability of the network, 6 test samples are selected and input to verify the prediction result of the network, the change of the error curve of the training result is shown as a graph in fig. 4, wherein the prediction value is the prediction feedback performed by the trained network topology model according to the input parameters, and finally, the relative error value predicted by the prediction model of the invention is not more than 10%, which indicates that the established prediction model has higher reliability.
Claims (4)
1. The method for predicting the deformation of the piston cutting machining based on the BP neural network is characterized by comprising the following steps of:
(1) Taking the piston stress distribution obtained by piston rough machining finite element analysis as an initial condition of piston finish machining finite element analysis, and simultaneously considering a workpiece tool to add clamping constraint of a piston part into a piston finish machining finite element simulation model;
(2) Working condition design is carried out aiming at cutting machining parameters related to rough machining of the piston, cutting working conditions are grouped according to a principle of control variables, and corresponding variation trends of the cutting parameters can be accurately distinguished by different working condition groups; working condition design is carried out aiming at cutting machining parameters related to a rough boring machining process, cutting depth, cutting speed and feeding quantity are selected as influence analysis factors of piston machining deformation according to a principle of control variables, and each machining parameter adopts three horizontal parameters of high, medium and low;
(3) Utilizing finite element analysis to solve the distribution result of the cutting machining stress and strain of each group of pistons, and obtaining a machining deformation simulation result under corresponding working conditions;
(4) Performing a piston cutting experiment according to a designed cutting working condition, measuring an experimental value of the deformation of a piston skirt under the corresponding working condition, checking the change rule of cutting deformation along with cutting parameters by comparing and analyzing the experimental value with an imitation value, and correcting a corresponding finite element analysis model;
(5) Taking the predicted object deformation obtained by finite element analysis as a training sample, determining the neuron numbers of an input layer, an hidden layer and an output layer in the BP neural network topology model based on the piston cutting deformation and an input-output mapping relation, and establishing the BP neural network topology model based on the piston cutting deformation;
(6) Performing BP neural network topology model learning based on piston cutting processing deformation according to a preset training sample, adjusting weights of all neurons in the model according to learning conditions, and performing reliability verification on the BP neural network topology model with the weights adjusted;
(7) Calculating a piston deformation value in the piston cutting process by using a BP neural network topology model based on the piston cutting deformation;
step (3) further comprises the steps of:
31. establishing a cutter model by using modeling software, importing process simulation finite element software to divide grids, and setting cutter and workpiece material properties;
32. sequentially applying cutting speed, cutting depth and feeding amount to each working condition group, calculating cutting force and cutting heat under corresponding working conditions through process simulation finite element software, deriving a data curve of the cutting force and the cutting heat, filtering the derived data, removing abnormal data which seriously deviate from an average level, and obtaining steady state values of the cutting force and the cutting heat under each working condition;
33. performing grid division on the piston model by utilizing finite element analysis software, adding corresponding workpiece tool constraint, and adding each group of calculated cutting force and cutting heat data into the finite element model of the piston to obtain the corresponding deformation size of the skirt part of the piston;
step (4) further comprises the steps of:
41. according to the principle of control variables, cutting machining tests are carried out aiming at each working condition group, and the cutting machining deformation of the piston is measured, wherein the cutting machining deformation of the piston comprises radial dimensions in the major axis and minor axis directions of a piston skirt;
42. analyzing the finite element analysis result and the test processing result by comparing the test processing data and the finite element simulation data, and verifying the change rule of the cutting processing deformation along with the cutting processing parameters; and the reliability of the finite element simulation model in the cutting process is ensured by checking whether the processing constraint conditions are consistent with the test working conditions, refining the finite element model grid and adjusting the application modes of cutting force and cutting heat.
2. The method for predicting the deformation of the piston cutting process based on the BP neural network according to claim 1, wherein the step (1) further comprises: the cutting parameters are cutting depth, cutting speed and feeding amount, and the machining parameters are set to different parameter levels according to the uniform distribution condition of the cutting parameters, so that different groups of test working conditions are obtained.
3. The method for predicting the deformation of the piston cutting process based on the BP neural network according to claim 1, wherein the method comprises the following steps: step (5) further comprises the input layer comprising three neurons of depth of cut, cutting speed and feed; the output layer comprises two neurons, namely the accumulated deformation of the piston skirt part in the long axis direction and the accumulated deformation of the piston in the short axis direction, which are generated after the cutting processing is completed.
4. A method for predicting the deformation of a piston cutting process based on a BP neural network according to any one of claims 1 to 3, characterized by: the step (6) further comprises the steps of carrying out normalization processing on the input variables, selecting a Levebberg-Marquardt algorithm as a training algorithm of the BP neural network based on piston cutting processing deformation for adjusting the weight of each neuron, selecting a hyperbolic tangent sigmoid function as a transfer function of neurons of an hidden layer, selecting a logarithmic sigmoid function as a transfer function of neurons of an output layer, setting a network target error to be 0.001 during neural network training, and carrying out BP neural network training based on the piston cutting processing deformation by utilizing Matlab software.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910119528.3A CN109977464B (en) | 2019-02-18 | 2019-02-18 | Prediction method of piston cutting deformation based on BP neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910119528.3A CN109977464B (en) | 2019-02-18 | 2019-02-18 | Prediction method of piston cutting deformation based on BP neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109977464A CN109977464A (en) | 2019-07-05 |
| CN109977464B true CN109977464B (en) | 2023-11-24 |
Family
ID=67077024
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910119528.3A Active CN109977464B (en) | 2019-02-18 | 2019-02-18 | Prediction method of piston cutting deformation based on BP neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109977464B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7432201B2 (en) * | 2019-12-19 | 2024-02-16 | 株式会社ブリヂストン | Estimation device, estimation method, program, and learning model generation device |
| CN111037365B (en) * | 2019-12-26 | 2021-08-20 | 大连理工大学 | Cutter state monitoring data set enhancing method based on generative countermeasure network |
| CN112528955B (en) * | 2020-12-25 | 2022-05-13 | 华中科技大学 | High-frequency element machining size precision prediction method and system |
| CN112668227B (en) * | 2020-12-31 | 2024-02-02 | 华中科技大学 | Thin-wall part cutter relieving deformation error prediction model establishment method and application thereof |
| CN112906136B (en) * | 2021-03-16 | 2022-07-22 | 上海交通大学 | Method and system for predicting laser thermoforming deformation of hull plate |
| CN113283138B (en) * | 2021-05-25 | 2024-03-26 | 大连理工大学 | Deep-learning-based dynamic response analysis method for deep-sea culture platform |
| CN114154240A (en) * | 2021-11-26 | 2022-03-08 | 重庆长安汽车股份有限公司 | Front cover coating overturning deformation prediction method and device and storage medium |
| CN115630558B (en) * | 2022-12-21 | 2023-04-14 | 南京航空航天大学 | Method for predicting assembly deformation of composite material component |
| CN116992778B (en) * | 2023-08-30 | 2024-04-26 | 西北工业大学 | Milling residual stress deformation prediction method, storage medium and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160059A (en) * | 2015-07-11 | 2015-12-16 | 西安工业大学 | BP and GA based blade machining cutting quantity optimization selection method |
| CN106484960A (en) * | 2016-09-20 | 2017-03-08 | 江苏科技大学 | A kind of suppression diesel engine closes the process parameter optimizing method of keyhole system machining deformation |
| CN106570563A (en) * | 2015-10-13 | 2017-04-19 | 中国石油天然气股份有限公司 | Deformation prediction method and apparatus based on Kalman filtering and BP neural network |
| CN108182325A (en) * | 2017-12-29 | 2018-06-19 | 清华大学深圳研究生院 | A kind of thin-walled workpiece machining Deformation Prediction analysis method |
| CN108959833A (en) * | 2018-09-26 | 2018-12-07 | 北京工业大学 | Tool wear prediction technique based on improved BP neural network |
-
2019
- 2019-02-18 CN CN201910119528.3A patent/CN109977464B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160059A (en) * | 2015-07-11 | 2015-12-16 | 西安工业大学 | BP and GA based blade machining cutting quantity optimization selection method |
| CN106570563A (en) * | 2015-10-13 | 2017-04-19 | 中国石油天然气股份有限公司 | Deformation prediction method and apparatus based on Kalman filtering and BP neural network |
| CN106484960A (en) * | 2016-09-20 | 2017-03-08 | 江苏科技大学 | A kind of suppression diesel engine closes the process parameter optimizing method of keyhole system machining deformation |
| CN108182325A (en) * | 2017-12-29 | 2018-06-19 | 清华大学深圳研究生院 | A kind of thin-walled workpiece machining Deformation Prediction analysis method |
| CN108959833A (en) * | 2018-09-26 | 2018-12-07 | 北京工业大学 | Tool wear prediction technique based on improved BP neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109977464A (en) | 2019-07-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109977464B (en) | Prediction method of piston cutting deformation based on BP neural network | |
| CN110609531B (en) | Workshop scheduling method based on digital twin | |
| CN113176761B (en) | Quality prediction and technological parameter optimization method for multi-processing characteristic sheet part | |
| CN108875213B (en) | Reactor core thermal hydraulic multi-scale analysis method | |
| CN106842922B (en) | Numerical control machining error optimization method | |
| CN108062427A (en) | The method that gradient rate controlling based on numerical computations reduces turbine disk forging residual stress | |
| CN111881574B (en) | Wind turbine generator set key component reliability modeling method based on distribution function optimization | |
| CN115015318B (en) | Macro-micro analysis method and platform for hot forging Quan Liucheng of large-scale component | |
| CN109885900A (en) | A kind of Cutting parameters optimization method based on gray analysis | |
| CN112149714A (en) | Method for determining energy efficiency characteristic index reference value of coal-electric unit based on data mining | |
| CN116520772A (en) | Sheet part numerical control machining quality prediction method based on multitasking transfer learning | |
| CN107391890B (en) | Prediction and optimal control method for oil casing threaded joint machining chatter defect | |
| CN113657029B (en) | Efficient approximate optimization method for heterogeneous data driven aircraft | |
| CN111948977B (en) | Multi-objective optimization method and system for stainless steel processing | |
| CN112861433A (en) | Product low-carbon design method based on multi-level integrated framework | |
| CN115270542A (en) | Method and device for identifying J-C constitutive parameters for titanium alloy cutting | |
| Gao et al. | A 6-sigma robust optimization method for stamping forming of automobile covering parts based on residual error and radial basis interpolation | |
| CN115470583A (en) | Method for obtaining optimal machining parameters of cantilever part based on numerical simulation | |
| CN113673166A (en) | Digital twin model working condition self-adaption method and system for machining quality prediction | |
| CN117435308B (en) | Modelica model simulation method and system based on parallel computing algorithm | |
| CN110781615A (en) | CAE simulation accuracy evaluation method | |
| CN116595883B (en) | Real-time online system state correction method for numerical reactor | |
| CN110096742B (en) | Pre-forging forming die design method based on prediction model and particle swarm optimization | |
| Dai et al. | Research on establishing prediction model for aerospace aluminum alloy milling force with the help of RBF neural network | |
| CN116305425A (en) | Cloud theory-based high earth-rock dam static parameter rapid inversion identification method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |