CN117351109A - Method for reconstructing section curve of shield tunnel - Google Patents
Method for reconstructing section curve of shield tunnel Download PDFInfo
- Publication number
- CN117351109A CN117351109A CN202311136149.8A CN202311136149A CN117351109A CN 117351109 A CN117351109 A CN 117351109A CN 202311136149 A CN202311136149 A CN 202311136149A CN 117351109 A CN117351109 A CN 117351109A
- Authority
- CN
- China
- Prior art keywords
- curve
- fitting
- feature vector
- point
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 239000013598 vector Substances 0.000 claims abstract description 60
- 238000013528 artificial neural network Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 32
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000003780 insertion Methods 0.000 abstract description 3
- 230000037431 insertion Effects 0.000 abstract description 3
- 238000011160 research Methods 0.000 abstract description 3
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000012937 correction Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
- G06T11/203—Drawing of straight lines or curves
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
- G06N3/0455—Auto-encoder networks; Encoder-decoder networks
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0499—Feedforward networks
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses a section curve reconstruction method of a shield tunnel, which relates to the field of tunnel interface curve research and evaluation, and comprises the following steps of S1: integrating curvature information of each interpolation point; s2: respectively carrying out curve fitting by taking each interpolation point as a starting point to obtain a plurality of fitting curve vectors; s3: transmitting each fitting curve vector into a neural network to obtain a fitting curve total feature vector; s4: coding the feature vector and transmitting the feature vector into a transducer neural network for training; s5: and finally obtaining corrected fitting point information. The invention adopts the method for reconstructing the section curve of the shield tunnel, takes curvature information of each interpolation point as a reference, sequentially fits and acquires curve information by taking each point as an initial insertion point, and transmits the curve information into a transducer neural network to extract information of each fitted curve, and integrates the curve information to obtain a method similar to a real curve. Compared with the traditional discrete curvature point fitting curve method, the method has the advantages that the reconstruction accuracy is improved by about 5%, and the method has obvious performance advantages.
Description
Technical Field
The invention relates to the field of tunnel interface curve research and evaluation, in particular to a method for reconstructing a section curve of a shield tunnel.
Background
The correction method in the current market is simpler and has low correction precision. In the research of reconstruction problems of the shield tunnel structure morphology, a plurality of scholars can monitor the stress and strain value change conditions of each interpolation point at the same time, deduce the approximate displacement condition of each interpolation point by combining the stress and strain value change, and reconstruct the corresponding fitting curve according to different optimization modes by combining the curvature information of each point. The method of the reconstruction-like curve has better fitting accuracy on the position information of each interpolation point, but often has poor performance on each fitting point among the interpolation points, and the actual position of the corresponding point cannot be fitted accurately.
The existing method is effective only for special cases, and is difficult to apply the method transplanting to the reconstruction of the tunnel section curve, and in the monitoring of the gas transmission pipeline and the water transmission pipeline with the structure morphology similar to that of a shield tunnel, a learner accurately calculates the pipeline morphological characteristics and reconstructs the accurate pipeline structure morphology through curvature monitoring on the inner wall and the outer wall of the pipeline and curvature difference of the inner wall and the outer wall. However, the method needs to uniformly arrange the monitoring units on the inner wall and the outer wall of the tunnel pipeline, and cannot meet the arrangement requirement in actual monitoring engineering.
The correction parameters are manually fine-tuned, and the correction parameters are large in workload and difficult to execute when facing large-scale problems. In the problem of fitting a curve to discrete curvature points of a tunnel section, a learner also tries to reconstruct a corresponding curve by adopting a non-uniform rational B-spline method. Although the method can solve the problem of large error of the fitted curve to a certain extent, the method also has curvature recurrence error, and the fitted curve of each interpolation point and other related control point parameters is manually fine-tuned to reconstruct the shape curve of the section structure of the real shield tunnel better.
Therefore, it is necessary to provide a method for reconstructing a cross-sectional curve of a shield tunnel to solve the above-mentioned problems.
Disclosure of Invention
The invention aims to provide a reconstruction method of a cross section curve of a shield tunnel, wherein the average error and the maximum error of obtained coordinate information of each point are minimum, and are respectively improved by 13.56% and 3.60% compared with other methods, so that the model has stronger generalization and is more suitable for solving the problem of curvature deviation.
In order to achieve the above purpose, the present invention provides a method for reconstructing a section curve of a shield tunnel, comprising the following steps:
s1: integrating curvature information of each interpolation point;
s2: performing curve fitting on the shield tunnel section by taking each interpolation point as an interpolation starting point to obtain a plurality of fitting curve vectors with the same number as the interpolation points;
s3: transmitting each fitting curve vector into a neural network to obtain the linear characteristic vector of the whole fitting curve and the characteristic vector of the interpolation point;
s4: performing position coding on the feature vector and transmitting the feature vector into a transducer neural network for training;
s5: and finally obtaining corrected fitting point information.
Preferably, the encoder layer is set with the relevant super parameters of the transformer neural networkNumber L enc Layer number L of decoder dec The attention multi-head quantity H and the feedforward neural network dimension d mlp Encoder network dimension d e Learning rate lambda and number of layers L of full connection layer linear Dimension d of full connection layer linear 。
Preferably, in step S2, each interpolation point is used as a starting position of curve fitting and the feature vector of the corresponding curve is marked as P 1 、P 2 、…、P n 。
Preferably, in step S3, the fitting curve feature vector information is integrated into the fitting curve total feature vector Pt.
Preferably, in step S4, the total feature vector P of the curve is fitted t The feature vector T of the fitting curve points is obtained by the incoming fully-connected network enc Will actually curve characteristic vector P r The feature vector T of the actual curve point is obtained by the incoming full-connection network dec For the characteristic vector T of the fitting curve point enc Performing position coding to obtain a feature vector T containing position information s_p T is changed by matrix s_p Conversion to a corresponding query matrix Q enc Key matrix K enc And value matrix V enc The method comprises the steps of carrying out a first treatment on the surface of the Feature vector T for actual curve points dec Performing position coding to obtain a feature vector T containing position information r_s T is changed by matrix r_p Conversion to a corresponding query matrix Q dec Key matrix K dec And value matrix V dec 。
Therefore, the method for reconstructing the section curve of the shield tunnel has the following beneficial effects:
(1) The average error and the maximum error of the coordinate information of each point obtained by the method are the smallest, and are respectively improved by 13.56% and 3.60% compared with the average error and the maximum error of other methods.
(2) The model provided by the invention has stronger generalization and is more suitable for solving the problem of curvature deviation.
(3) The invention takes curvature information of each interpolation point as a reference, and sequentially fits and acquires curve information by taking each point as an initial insertion point, and transmits the curve information into a transducer neural network to extract information of each fitted curve, and the curve information is integrated to obtain a method similar to a real curve.
(4) The algorithm performance is evaluated in the curvature calibration experiment, and the result shows that the method has obvious performance advantages compared with the traditional discrete curvature point fitting curve method, and the reconstruction accuracy is improved by about 5%.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
Fig. 1 is a flow chart of a method for reconstructing a cross-sectional curve of a shield tunnel according to the present invention.
Detailed Description
The technical scheme of the invention is further described below through the attached drawings and the embodiments.
Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs.
As used herein, the word "comprising" or "comprises" and the like means that elements preceding the word encompass the elements recited after the word, and not exclude the possibility of also encompassing other elements. The terms "inner," "outer," "upper," "lower," and the like are used for convenience in describing and simplifying the description based on the orientation or positional relationship shown in the drawings, and do not denote or imply that the devices or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the invention, but the relative positional relationship may be changed when the absolute position of the object to be described is changed accordingly. In the present invention, unless explicitly specified and limited otherwise, the term "attached" and the like should be construed broadly, and may be, for example, fixedly attached, detachably attached, or integrally formed; can be directly connected or indirectly connected through an intermediate medium, and can be communicated with the inside of two elements or the interaction relationship of the two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
As shown in fig. 1, the invention provides a method for reconstructing a section curve of a shield tunnel, which comprises the following steps: the method comprises the following steps:
setting the relevant super parameter of a transformer neural network, namely the layer number L of an encoder enc Layer number L of decoder dec The attention multi-head quantity H and the feedforward neural network dimension d mlp Encoder network dimension de, learning rate λ, full-connection layer number L linear Dimension d of full connection layer linear 。
S1: integrating curvature information of each interpolation point;
s2: performing curve fitting on the shield tunnel section by taking each interpolation point as an interpolation starting point to obtain a plurality of fitting curve vectors with the same number as the interpolation points; in step S2, each interpolation point is used as a curve fitting initial position and the corresponding curve characteristic vector is marked as P 1 、P 2 、…、P n 。
S3: transmitting each fitting curve vector into a neural network to obtain the linear characteristic vector of the whole fitting curve and the characteristic vector of the interpolation point; in step S3, the fitting curve feature vector information is integrated into a fitting curve total feature vector P t 。
S4: and performing position coding on the feature vector, and transmitting the feature vector into a transducer neural network for training to obtain a weight relation among all fitting curves and a final fitting curve. The fitted curve total feature vector P is used in step S4 t The feature vector T of the fitting curve points is obtained by the incoming fully-connected network enc Will actually curve characteristic vector P r The feature vector T of the actual curve point is obtained by the incoming full-connection network dec For the characteristic vector T of the fitting curve point enc Performing position coding to obtain a feature vector T containing position information s_p T is changed by matrix s_p Conversion to a corresponding query matrix Q enc Key matrix K enc And value matrix V enc The method comprises the steps of carrying out a first treatment on the surface of the Feature vector T for actual curve points dec Performing position coding to obtain a feature vector T containing position information r_s T is changed by matrix r_p Conversion to a corresponding query matrix Q dec Key matrix K dec And value matrix V dec 。
S5: and finally obtaining corrected fitting point information.
Example 1
Feature vector T of fitting curve points enc Performing position coding to obtain a feature vector T containing position information s_p ；
for l＝1，2，…，L enc do；
T is changed by matrix s_p Conversion to a corresponding query matrix Q enc Key matrix K enc And value matrix V encl ；
Y enc ＝MultiHeadAttention(Q enc ，K enc ，V enc )；
Y enc ＝LayerNorm(T s_p +Y enc )；
Y new ＝FeedForward(Y enc )；
Y enc ＝LayerNorm(Y enc +Y new )；
end for；
Feature vector T for actual curve points dec Performing position coding to obtain a feature vector T containing position information r_s ；
for l＝1，2，…，L enc do；
T is changed by matrix r_p Conversion to a corresponding query matrix Q dec Key matrix K dec And value matrix V dec ；
Y dec ＝MaskedMultiHeadAttention(Q dec ，K dec ，V dec )；
Y dec ＝LayerNorm(T r_p +Y dec )；
Y is set to enc Decomposition into K' enc And V' enc ；
Y is set to enc Decomposition into K' enc And V' enc ；
Y′ dec ＝LayerNorm(Y′ new +Y′ dec )；
Y dec ＝FeedForward(Y dec )；
Y dec ＝LayerNorm(Y dec +Y′ new )；
endfor；
Return p=linear (Y dec )。
Therefore, the method for reconstructing the section curve of the shield tunnel is adopted, the curvature information of each interpolation point is taken as a reference, curve information is obtained by fitting each point as an initial insertion point in sequence, and the information of each fitting curve is extracted by transmitting the curve information into a transducer neural network, and the method is integrated to obtain the approximate real curve, and compared with the traditional discrete curvature point fitting curve method, the method has the advantages of improving the reconstruction precision by about 5 percent and having obvious performance.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention and not for limiting it, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that: the technical scheme of the invention can be modified or replaced by the same, and the modified technical scheme cannot deviate from the spirit and scope of the technical scheme of the invention.
Claims (5)
1. A method for reconstructing a section curve of a shield tunnel is characterized by comprising the following steps: the method comprises the following steps:
s1: integrating curvature information of each interpolation point;
s2: performing curve fitting on the shield tunnel section by taking each interpolation point as an interpolation starting point to obtain a plurality of fitting curve vectors with the same number as the interpolation points;
s3: transmitting each fitting curve vector into a neural network to obtain the linear characteristic vector of the whole fitting curve and the characteristic vector of the interpolation point;
s4: performing position coding on the feature vector and transmitting the feature vector into a transducer neural network for training;
s5: and finally obtaining corrected fitting point information.
2. The method for reconstructing the section curve of the shield tunnel according to claim 1, wherein the method comprises the following steps: setting relevant super parameters of a transformer neural network: encoder layer number L enc Layer number L of decoder dec The attention multi-head quantity H and the feedforward neural network dimension d mlp Encoder network dimension d e Learning rate lambda and number of layers L of full connection layer linear Dimension d of full connection layer linear 。
3. The method for reconstructing the section curve of the shield tunnel according to claim 2, wherein the method comprises the following steps: in step S2, each interpolation point is used as a curve fitting initial position and the corresponding curve characteristic vector is marked as P 1 、P 2 、…、P n 。
4. A method of reconstructing a cross-sectional profile of a shield tunnel according to claim 3, wherein: in step S3, the fitting curve feature vector information is integrated into a fitting curve total feature vector P t 。
5. The method for reconstructing the section curve of the shield tunnel according to claim 4, wherein the method comprises the following steps: the fitted curve total feature vector P is used in step S4 t The feature vector T of the fitting curve points is obtained by the incoming fully-connected network enc Will actually curve characteristic vector P r The feature vector T of the actual curve point is obtained by the incoming full-connection network dec For the characteristic vector T of the fitting curve point enc Performing position coding to obtain a feature vector T containing position information s_p T is changed by matrix s_p Conversion to a corresponding query matrix Q enc Key matrix K enc And value matrix V enc The method comprises the steps of carrying out a first treatment on the surface of the Feature vector T for actual curve points dec Performing position coding to obtain a feature vector T containing position information r_s， T is changed by matrix r_p Conversion to a corresponding query matrix Q dec Key matrix K dec And value matrix V dec 。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311136149.8A CN117351109B (en) | 2023-09-05 | 2023-09-05 | Method for reconstructing section curve of shield tunnel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311136149.8A CN117351109B (en) | 2023-09-05 | 2023-09-05 | Method for reconstructing section curve of shield tunnel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117351109A true CN117351109A (en) | 2024-01-05 |
| CN117351109B CN117351109B (en) | 2024-06-07 |
Family
ID=89362113
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311136149.8A Active CN117351109B (en) | 2023-09-05 | 2023-09-05 | Method for reconstructing section curve of shield tunnel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117351109B (en) |
Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5237649A (en) * | 1989-04-20 | 1993-08-17 | Kabushiki Kaisha Toshiba | Method and system for acquiring interpolation points from straight short vectors representing figure in curve fitting |
| JP2000148223A (en) * | 1998-11-06 | 2000-05-26 | Yaskawa Electric Corp | Curve interpolation acceleration and deceleration control method |
| US20100220099A1 (en) * | 2007-10-30 | 2010-09-02 | National University Corporation Yokohama National University | Interpolation processing method and interpolation processor |
| US20190095739A1 (en) * | 2017-09-27 | 2019-03-28 | Harbin Institute Of Technology | Adaptive Auto Meter Detection Method based on Character Segmentation and Cascade Classifier |
| CN109840930A (en) * | 2019-01-30 | 2019-06-04 | 厦门大学 | Curve constructing method based on the control of geometrical characteristic point |
| CN110163799A (en) * | 2019-05-05 | 2019-08-23 | 杭州电子科技大学上虞科学与工程研究院有限公司 | A kind of super-resolution point cloud generation method based on deep learning |
| CN110244658A (en) * | 2019-06-04 | 2019-09-17 | 浙江工业大学 | A method of cutting interpolation position precision is improved based on improved BP |
| WO2020098708A1 (en) * | 2018-11-14 | 2020-05-22 | 北京市商汤科技开发有限公司 | Lane line detection method and apparatus, driving control method and apparatus, and electronic device |
| AU2020103105A4 (en) * | 2020-10-29 | 2021-01-07 | Nanjing Forestry University | A Method for Segmenting Point Cloud Data on Shield Tunnel Surface |
| CN112562086A (en) * | 2021-02-22 | 2021-03-26 | 中国空气动力研究与发展中心低速空气动力研究所 | Ice-shaped model curved surface reconstruction method |
| US20210133365A1 (en) * | 2017-11-30 | 2021-05-06 | Siemens Industry Software Limited | A method of operating a cad system model for modelling an article to be manufactured |
| CN113112568A (en) * | 2021-06-08 | 2021-07-13 | 中国空气动力研究与发展中心低速空气动力研究所 | Two-dimensional ice shape simplification method based on arc fitting |
| US20220121952A1 (en) * | 2019-02-18 | 2022-04-21 | Iruiz Contracting Limited | Non-homogeneous machine learning architecture for approximating an overall function and training method thereof |
| CN114463515A (en) * | 2022-02-11 | 2022-05-10 | 中煤科工集团南京设计研究院有限公司 | Shield tunnel section extraction method based on ground laser scanning point cloud |
| CN114545863A (en) * | 2022-03-07 | 2022-05-27 | 中南大学 | Track smoothing method for numerical control machining based on B spline curve fitting |
| CN114972640A (en) * | 2022-05-20 | 2022-08-30 | 天津理工大学 | Cable three-dimensional reconstruction and parameter calculation method based on point cloud |
| CN115018157A (en) * | 2022-06-06 | 2022-09-06 | 中南大学 | Surface subsidence prediction method based on EMD-CASSA-ELM |
| US20220390920A1 (en) * | 2021-06-03 | 2022-12-08 | Chongqing University | Method for constructing body-in-white spot welding deformation prediction model based on graph convolutional network |
| CN115713026A (en) * | 2022-09-23 | 2023-02-24 | 武汉大学 | Time series data prediction model construction method, device, equipment and readable storage medium |
| CN115730710A (en) * | 2022-11-11 | 2023-03-03 | 广东工业大学 | Attention mechanism-based electric automobile daily charging demand curve prediction method |
| CN116484635A (en) * | 2023-05-06 | 2023-07-25 | 江苏科技大学 | Sinusoidal arc fitting local fairing method based on improved arc fitting algorithm |
-
2023
- 2023-09-05 CN CN202311136149.8A patent/CN117351109B/en active Active
Patent Citations (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5237649A (en) * | 1989-04-20 | 1993-08-17 | Kabushiki Kaisha Toshiba | Method and system for acquiring interpolation points from straight short vectors representing figure in curve fitting |
| JP2000148223A (en) * | 1998-11-06 | 2000-05-26 | Yaskawa Electric Corp | Curve interpolation acceleration and deceleration control method |
| US20100220099A1 (en) * | 2007-10-30 | 2010-09-02 | National University Corporation Yokohama National University | Interpolation processing method and interpolation processor |
| US20190095739A1 (en) * | 2017-09-27 | 2019-03-28 | Harbin Institute Of Technology | Adaptive Auto Meter Detection Method based on Character Segmentation and Cascade Classifier |
| US20210133365A1 (en) * | 2017-11-30 | 2021-05-06 | Siemens Industry Software Limited | A method of operating a cad system model for modelling an article to be manufactured |
| WO2020098708A1 (en) * | 2018-11-14 | 2020-05-22 | 北京市商汤科技开发有限公司 | Lane line detection method and apparatus, driving control method and apparatus, and electronic device |
| CN109840930A (en) * | 2019-01-30 | 2019-06-04 | 厦门大学 | Curve constructing method based on the control of geometrical characteristic point |
| US20220121952A1 (en) * | 2019-02-18 | 2022-04-21 | Iruiz Contracting Limited | Non-homogeneous machine learning architecture for approximating an overall function and training method thereof |
| CN110163799A (en) * | 2019-05-05 | 2019-08-23 | 杭州电子科技大学上虞科学与工程研究院有限公司 | A kind of super-resolution point cloud generation method based on deep learning |
| CN110244658A (en) * | 2019-06-04 | 2019-09-17 | 浙江工业大学 | A method of cutting interpolation position precision is improved based on improved BP |
| AU2020103105A4 (en) * | 2020-10-29 | 2021-01-07 | Nanjing Forestry University | A Method for Segmenting Point Cloud Data on Shield Tunnel Surface |
| CN112562086A (en) * | 2021-02-22 | 2021-03-26 | 中国空气动力研究与发展中心低速空气动力研究所 | Ice-shaped model curved surface reconstruction method |
| US20220390920A1 (en) * | 2021-06-03 | 2022-12-08 | Chongqing University | Method for constructing body-in-white spot welding deformation prediction model based on graph convolutional network |
| CN113112568A (en) * | 2021-06-08 | 2021-07-13 | 中国空气动力研究与发展中心低速空气动力研究所 | Two-dimensional ice shape simplification method based on arc fitting |
| CN114463515A (en) * | 2022-02-11 | 2022-05-10 | 中煤科工集团南京设计研究院有限公司 | Shield tunnel section extraction method based on ground laser scanning point cloud |
| CN114545863A (en) * | 2022-03-07 | 2022-05-27 | 中南大学 | Track smoothing method for numerical control machining based on B spline curve fitting |
| CN114972640A (en) * | 2022-05-20 | 2022-08-30 | 天津理工大学 | Cable three-dimensional reconstruction and parameter calculation method based on point cloud |
| CN115018157A (en) * | 2022-06-06 | 2022-09-06 | 中南大学 | Surface subsidence prediction method based on EMD-CASSA-ELM |
| CN115713026A (en) * | 2022-09-23 | 2023-02-24 | 武汉大学 | Time series data prediction model construction method, device, equipment and readable storage medium |
| CN115730710A (en) * | 2022-11-11 | 2023-03-03 | 广东工业大学 | Attention mechanism-based electric automobile daily charging demand curve prediction method |
| CN116484635A (en) * | 2023-05-06 | 2023-07-25 | 江苏科技大学 | Sinusoidal arc fitting local fairing method based on improved arc fitting algorithm |
Non-Patent Citations (8)
| Title |
|---|
| 刘冠兰;徐亚明;柏文峰;: "地铁隧道断面变形分析方法研究", 测绘通报, no. 02, 25 February 2016 (2016-02-25) * |
| 叶耀东;: "B样条在盾构隧道沉降曲线拟合中的应用", 地下空间与工程学报, no. 02, 15 April 2016 (2016-04-15) * |
| 吴粼;高新闻;陈鸿;: "NURBS曲线在隧道断面拟合中的应用研究", 水利与建筑工程学报, no. 02, 15 April 2016 (2016-04-15) * |
| 周红梅;王燕铭;刘志刚;卢秉恒;: "基于最少控制点的非均匀有理B样条曲线拟合", 西安交通大学学报, no. 01, 10 January 2008 (2008-01-10) * |
| 张旭;侯茂盛;刘智超;刘涛;李丽娟;: "基于光纤布拉格光栅传感器的板壳结构曲面重构算法", 激光与光电子学进展, no. 09, 10 May 2020 (2020-05-10) * |
| 徐进;柯映林;曲巍崴;: "基于特征点自动识别的B样条曲线逼近技术", 机械工程学报, no. 11, 15 November 2009 (2009-11-15) * |
| 朱晓锦;张合生;谢春宁;樊红朝;: "一种基于曲率信息的太空帆板空间曲面拟合算法分析", 系统仿真学报, no. 11, 5 June 2007 (2007-06-05) * |
| 石屹;金登男;: "基于贝塞尔曲线拟合的心电信号模式分类方法", 计算机工程与设计, no. 04, 16 April 2013 (2013-04-16) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117351109B (en) | 2024-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106198909B (en) | A kind of aquaculture water quality prediction technique based on deep learning | |
| CN109784480A (en) | A kind of power system state estimation method based on convolutional neural networks | |
| CN110689171A (en) | Turbine health state prediction method based on E-LSTM | |
| CN116070799B (en) | Photovoltaic power generation amount prediction system and method based on attention and deep learning | |
| CN108920812A (en) | A kind of machining surface roughness prediction technique | |
| CN113988449B (en) | Wind power prediction method based on transducer model | |
| CN110707763A (en) | AC/DC power distribution network load prediction method based on ensemble learning | |
| CN114897277B (en) | Burst-type landslide displacement prediction method based on LSTM | |
| CN110398219B (en) | Joint arm type coordinate measuring machine parameter calibration method based on hybrid optimization algorithm | |
| CN103294647B (en) | Embedded head-position difficult labor dimension reduction method is kept based on orthogonal tensor neighbour | |
| CN117351109B (en) | Method for reconstructing section curve of shield tunnel | |
| CN116629126A (en) | Soft measurement modeling method based on dynamic multi-head attention mechanism | |
| CN108599170B (en) | Tidal current obtaining method suitable for alternating current-direct current system | |
| CN109711662A (en) | A kind of power grid robust state estimation method based on multisource data fusion | |
| CN115204509A (en) | Method for predicting number of inpatients in respiratory system | |
| CN116483036B (en) | Transformer-based self-encoder soft measurement modeling method | |
| CN113377744B (en) | Reconstruction method and device for structural anomaly monitoring data with environmental temperature correlation | |
| CN107680679B (en) | Big data driven student aerobic capacity grouping method | |
| CN110909492B (en) | Sewage treatment process soft measurement method based on extreme gradient lifting algorithm | |
| CN110109904B (en) | Environment-friendly big data oriented water quality soft measurement method | |
| CN111458031A (en) | Non-contact remote measurement method for measuring and calculating body temperature | |
| CN114626012A (en) | GNSS sequence prediction method and system of multi-scale attention mechanism | |
| CN113006774B (en) | Online graph neural network prediction method for oil pressure peak in fracturing construction | |
| CN114186771A (en) | Hybrid regularization random configuration network industrial process operation index estimation method | |
| CN114241553A (en) | Method for migrating facial expressions to faces of virtual characters |
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 |