CN110119568B - Method for evaluating stone-throwing effect influence factors of riprap bank protection - Google Patents
Method for evaluating stone-throwing effect influence factors of riprap bank protection Download PDFInfo
- Publication number
- CN110119568B CN110119568B CN201910383005.XA CN201910383005A CN110119568B CN 110119568 B CN110119568 B CN 110119568B CN 201910383005 A CN201910383005 A CN 201910383005A CN 110119568 B CN110119568 B CN 110119568B
- Authority
- CN
- China
- Prior art keywords
- riprap
- effect
- regression tree
- enhanced regression
- evaluating
- 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
- 230000000694 effects Effects 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000011156 evaluation Methods 0.000 claims abstract description 49
- 238000012549 training Methods 0.000 claims description 27
- 230000006870 function Effects 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 16
- 238000004088 simulation Methods 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 8
- 238000013508 migration Methods 0.000 claims description 8
- 230000005012 migration Effects 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 7
- 239000004575 stone Substances 0.000 claims description 5
- 230000008859 change Effects 0.000 abstract description 2
- 238000004451 qualitative analysis Methods 0.000 abstract description 2
- 238000004445 quantitative analysis Methods 0.000 abstract description 2
- 230000001419 dependent effect Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000009991 scouring Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000002791 soaking Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an evaluation method for influence factors of riprap effect of riprap revetment in the technical field of riprap revetment engineering, and aims to solve the technical problems that the riprap effect evaluation in the prior art is only limited to qualitative analysis, and a quantitative analysis method is lacked to accurately determine the contribution rate of each influence factor to river behavior change. The method comprises the following steps: obtaining target river reach evaluation data, wherein the target river reach evaluation data comprise a riprap effect influence factor measured value and a riprap effect evaluation index measured value; establishing an enhanced regression tree model according to the target river reach evaluation data; and evaluating the contribution rate of the riprap effect influence factors to the riprap effect according to the enhanced regression tree model.
Description
Technical Field
The invention relates to a method for evaluating influence factors of a riprap effect of riprap revetment, belonging to the technical field of riprap revetment engineering.
Background
The riprap revetment project is an underwater hidden project, and in view of underwater complex terrains, in order to ensure riprap effect, the riprap thicknesses of different areas in a construction river reach need to be determined at first. In addition, the stone throwing and throwing effect is also influenced by other factors, including: the riprap body is impacted by water flow silt to be violently moved underwater; after the riprap is finished, the riprap body is used as a part of a river bed, and is damaged and aged due to long-term soaking and scouring, so that the riprap effect is inconsistent with the engineering expectation, and bank caving and bank breaking are serious and even happen. Therefore, the method has important significance for further improving the construction mode, improving the rock-throwing utilization rate and maintaining the bank slope safety by evaluating the rock-throwing effect after the rock-throwing construction and researching the influence factors of the underwater migration of the rock-throwing body.
In the prior art, the evaluation on the riprap effect is still in a qualitative analysis stage, and a quantitative analysis method is lacked for accurately measuring the contribution rate of each influence factor to the river situation change.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a method for evaluating stone-throwing effect influence factors of a riprap revetment, which comprises the following steps: acquiring target river reach evaluation data, wherein the target river reach evaluation data comprise a riprap effect influence factor measured value and a riprap effect evaluation index measured value; establishing an enhanced regression tree model according to the target river reach evaluation data; and evaluating the contribution rate of the riprap effect influence factors to the riprap effect according to the enhanced regression tree model.
Further, the riprap effect influence factor comprises: at least any one of gradient, slope direction, flow velocity, flow direction and bed elevation; the evaluation index of the riprap effect comprises the following steps: bed elevation increment.
Further, the method for acquiring the target river reach evaluation data comprises the following steps: monitoring and obtaining target river reach evaluation data after stone throwing construction; and monitoring to obtain target river reach evaluation data after the underwater migration of the riprap body is stable.
Further, establishing an enhanced regression tree model according to the target river reach evaluation data, comprising the following steps of: establishing an enhanced regression tree process model, the enhanced regression tree process model including a loss function; extracting a training sample from the target river reach evaluation data; inputting the training sample into an enhanced regression tree process model for training so as to reduce a loss function; establishing a new enhanced regression tree process model based on the reduced loss function; and repeating the building process of the enhanced regression tree process model according to the preset learning rate and the typing times, and outputting the finally built enhanced regression tree process model as the enhanced regression tree model.
Further, the training samples account for 90% of the target river reach evaluation data, the preset learning rate is 0.1, and the preset typing times are 300.
Further, the method for evaluating the contribution rate of the riprap effect influence factors to the riprap effect according to the enhanced regression tree model comprises the following steps: checking whether the enhanced regression tree model has reliability; if the enhanced regression tree model has reliability, inputting the measured value of the riprap effect influence factor into the enhanced regression tree model to obtain the contribution rate of the riprap effect influence factor to the riprap effect evaluation index; and if the enhanced regression tree model has no reliability, reestablishing the enhanced regression tree model according to the adjusted learning rate and the classification times.
Further, checking whether the enhanced regression tree model has reliability comprises the following steps: extracting a check sample from the target river reach evaluation data, wherein the check sample is residual data of the target river reach evaluation data after a training sample is extracted; inputting the check sample into an enhanced regression tree model, and outputting a riprap effect evaluation index simulation value; and judging whether the enhanced regression tree model has reliability according to the riprap effect evaluation index simulation value and the riprap effect evaluation index measured value.
Further, the enhanced regression tree model has reliable judgment criteria, including: the root mean square error of the riprap effect evaluation index analog value is less than or equal to 0.5; the standardized average error of the simulation value of the riprap effect evaluation index is less than 50 percent; the goodness of fit of the riprap effect evaluation index simulation value is more than 0.8.
Compared with the prior art, the invention has the following beneficial effects: the relative contribution rate among the influence factors influencing the underwater migration result of the riprap body can be obtained, and guidance is provided for numerical simulation; the accuracy of the riprap migration result is high by using the model for training, the collected topographic data is predicted before riprap construction by using the model as a basis, riprap amount is reduced at a place with serious siltation, riprap is enlarged in a region with serious scouring, and a scientific basis is provided for improving riprap efficiency.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a diagram illustrating predicted value and actual value data according to an embodiment of the present invention;
fig. 3 is a diagram illustrating the contribution ratio of the impact factor obtained by calculation in the embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
As shown in fig. 1, which is a flow chart of the present invention, the method includes the following steps:
acquiring topographic data of a riverbed of a target river reach, and respectively extracting actual measurement data of riprap effect influence factors and riprap effect evaluation indexes according to different flow rate monitoring sites;
when related data are extracted according to different flow rate monitoring sites, all the data are in the same coordinate system;
data for simulating the riprap revetment engineering effect are at least collected in two different time periods, namely topographic data monitored immediately after riprap construction and topographic data monitored after riprap underwater migration is stable, river reach collected twice are the same, and monitoring points for extracting topographic information are the same;
the target river reach riverbed terrain data comprises: gradient, slope direction, flow velocity, flow direction and bed elevation data, wherein the terrain data are riprap effect influence factors (namely independent variables); and selecting the elevation increment of the bottom bed from the riprap construction to the underwater migration stability of the riprap body as a riprap effect evaluation index (namely a dependent variable).
And step two, establishing an enhanced regression tree model, randomly extracting 90% of target river reach riverbed topographic data as a training sample for training the model, and always ensuring that a new model is established on the basis of the reduction of the loss function of the original model by using an integrated learning method. In the learning process, the loss function is continuously reduced, and the model is continuously improved. The preset learning rate is 0.1, the typing times are 300, namely the partial derivative of each loss function reduction is controlled to be 0.1, the training is stopped after 300 times of operation, and the output result is the common result of 300 times of training.
The enhanced regression tree is the sum of a plurality of iteratively trained regression trees, and comprises two parts: the first part is the training error, the second part is the sum of the complexity of each tree, and the formula is as follows:
wherein Obj (θ) is an enhanced regression tree function, l is a loss function, Ω is a regular function, n is the number of trees, T is the number of leaves per tree, y is i For the measured value of the dependent variable of each group of the training samples in the ith training,
for the predicted value of the dependent variable of each group of data of training samples in the ith training, f j And dividing results for each group of data independent variables of the training samples in the jth training.
(1) The single regression tree model is:
in the formula, x i F is the set space of the regression tree. The method aims to distribute input data to leaf nodes according to attributes, wherein each leaf node corresponds to a real number;
(2) When the enhanced regression tree function is trained, a new function is added on the basis of the original enhanced regression tree function every time, so that the target function is reduced as much as possible, namely:
in the formula, obj (t) To enhance the regression tree function after adding the new function,
for the t training time, the predicted value of the dependent variable of each group of data of the training sample, f t Dividing results for each group of data independent variables of training samples before the t-th training;
(3) Complexity of the tree: the complexity includes the number of nodes in one tree and the modulo square of the output score on each tree leaf node, that is:
in the formula, omega j Is the vector of the leaf.
The loss function in the learning process continuously descends, the model is continuously improved, and for preventing the model from being over-fitted, the learning rate is set, namely:
where v is the learning rate.
And step three, outputting a simulation result check model.
Inputting independent variables of the remaining 10% of data in the trained model to simulate the bed elevation increment, verifying the bed elevation increment and the measured value, calculating the root mean square error, the normalized average error and the goodness of fit of the model training data, and verifying the reliability of the model. Wherein,
root mean square error
Normalized mean error
Goodness of fit
If RMES is less than or equal to 0.5m, NME is less than 50 percent, R 2 If the model is more than 0.8, the model reaches the expectation, and the fourth step is carried out; otherwise, adjusting parameters to enable the model accuracy to meet the requirements. When parameters are adjusted, the learning rate and the typing times (namely the number of trees) of the model need to be adjusted at the same time, and when the learning rate is increased, the typing times are reduced; and when the learning rate is reduced, the typing times are increased.
As shown in fig. 2, the root mean square error of the predicted value and the true value of the data in the embodiment of the present invention is 0.35m, the normalized average error is 11.88%, the goodness of fit reaches 0.91, and the model has reliability.
Step four, evaluating the contribution rate of factors influencing the underwater migration of the riprap body: and calculating the contribution rates of different influence factors to the riprap effect through model operation, evaluating the importance of the influence of each factor on the engineering effect of riprap revetment implementation, and outputting the influence effect of the slope, the slope direction, the flow speed, the flow direction and the bottom bed elevation on the bottom bed elevation increment. Fig. 3 shows the influence factor contribution rate obtained by calculation in the embodiment of the present invention.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (7)
1. A method for evaluating stone throwing effect influence factors of stone throwing and bank protection is characterized by comprising the following steps:
acquiring target river reach evaluation data, wherein the target river reach evaluation data comprise a riprap effect influence factor measured value and a riprap effect evaluation index measured value;
establishing an enhanced regression tree model according to the target river reach evaluation data;
evaluating the contribution rate of the riprap effect influence factors to the riprap effect according to the enhanced regression tree model;
establishing an enhanced regression tree model according to the target river reach evaluation data, comprising the following steps of:
establishing an enhanced regression tree process model, the enhanced regression tree process model including a loss function;
extracting a training sample from the target river reach evaluation data;
inputting the training sample into an enhanced regression tree process model for training so as to reduce a loss function;
establishing a new enhanced regression tree process model based on the reduced loss function;
and repeating the process of establishing the enhanced regression tree process model according to the preset learning rate and the typing times, and outputting the finally established enhanced regression tree process model as the enhanced regression tree model.
2. The method for evaluating the riprap effect influence factor of riprap revetment according to claim 1, wherein the riprap effect influence factor comprises: at least any one of gradient, slope direction, flow velocity, flow direction and bed elevation;
the evaluation index of the riprap effect comprises the following steps: bed elevation increment.
3. The method for evaluating the riprap effect influence factor of riprap revetment according to claim 1, wherein the step of obtaining target river reach evaluation data comprises the steps of:
monitoring and obtaining target river reach evaluation data after stone throwing construction;
and monitoring to obtain target river reach evaluation data after the underwater migration of the riprap body is stable.
4. The method for evaluating the riprap effect influence factor of the riprap revetment according to claim 1, wherein the training samples account for 90% of the target river reach evaluation data, the preset learning rate is 0.1, and the preset typing times are 300.
5. The method for evaluating the riprap effect influence factor for riprap revetment according to claim 1, wherein the method for evaluating the contribution rate of the riprap effect influence factor to the riprap effect according to the enhanced regression tree model comprises the following steps:
checking whether the enhanced regression tree model has reliability;
if the enhanced regression tree model has reliability, inputting the measured value of the riprap effect influence factor into the enhanced regression tree model to obtain the contribution rate of the riprap effect influence factor to the riprap effect evaluation index;
and if the enhanced regression tree model has no reliability, reestablishing the enhanced regression tree model according to the adjusted learning rate and the typing times.
6. The method for evaluating the influence factors of the riprap and riprap effect of claim 5, wherein checking whether the enhanced regression tree model has reliability comprises the following steps:
extracting a check sample from the target river reach evaluation data, wherein the check sample is residual data of the target river reach evaluation data after a training sample is extracted;
inputting the check sample into the enhanced regression tree model, and outputting a simulation value of the riprap effect evaluation index;
and judging whether the enhanced regression tree model has reliability according to the riprap effect evaluation index simulation value and the riprap effect evaluation index measured value.
7. The method for evaluating the riprap effect influence factor of riprap revetment according to claim 6, wherein the enhanced regression tree model has a reliable judgment criterion comprising:
the root mean square error of the riprap effect evaluation index analog value is less than or equal to 0.5;
the standardized average error of the simulation value of the riprap effect evaluation index is less than 50 percent;
the goodness of fit of the riprap effect evaluation index simulation value is more than 0.8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910383005.XA CN110119568B (en) | 2019-05-09 | 2019-05-09 | Method for evaluating stone-throwing effect influence factors of riprap bank protection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910383005.XA CN110119568B (en) | 2019-05-09 | 2019-05-09 | Method for evaluating stone-throwing effect influence factors of riprap bank protection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110119568A CN110119568A (en) | 2019-08-13 |
| CN110119568B true CN110119568B (en) | 2022-10-14 |
Family
ID=67522016
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910383005.XA Active CN110119568B (en) | 2019-05-09 | 2019-05-09 | Method for evaluating stone-throwing effect influence factors of riprap bank protection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110119568B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118036901B (en) * | 2024-04-11 | 2024-06-28 | 广东河海工程咨询有限公司 | Ecological embankment protection bank safety precaution system based on intelligent monitoring |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107608938A (en) * | 2017-08-08 | 2018-01-19 | 安徽师范大学 | The factor screening method towards two-value classification of tree algorithm is returned based on enhancing |
| CN109359738A (en) * | 2018-10-19 | 2019-02-19 | 西南交通大学 | A kind of Landslide hazard appraisal procedure based on QPSO-BP neural network |
-
2019
- 2019-05-09 CN CN201910383005.XA patent/CN110119568B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107608938A (en) * | 2017-08-08 | 2018-01-19 | 安徽师范大学 | The factor screening method towards two-value classification of tree algorithm is returned based on enhancing |
| CN109359738A (en) * | 2018-10-19 | 2019-02-19 | 西南交通大学 | A kind of Landslide hazard appraisal procedure based on QPSO-BP neural network |
Non-Patent Citations (2)
| Title |
|---|
| 基于增强回归树的农田总氮流失强度省际差异及其影响因素分析;王迪等;《生态与农村环境学报》;20171231;第33卷(第12期);全文 * |
| 波浪作用下抛石潜堤的动态稳定数值模拟;胡宝琳等;《工程力学》;20140228;第31卷(第2期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110119568A (en) | 2019-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112417573B (en) | GA-LSSVM and NSGA-II shield tunneling multi-objective optimization method based on existing tunnel construction | |
| US11709979B1 (en) | Bridge damage identification method considering uncertainty | |
| CN107315862A (en) | A kind of method for setting up open-cut foundation ditch engineering investigation and analog parameter relation | |
| KR102155101B1 (en) | System and method for predicting ground layer information, and a recording medium having computer readable program for executing the method | |
| CN107291667A (en) | A kind of interwell communication degree determines method and system | |
| CN105678417A (en) | Prediction method and device for tunnel face water inflow of construction tunnel | |
| CN110489844A (en) | One kind being suitable for the uneven large deformation grade prediction technique of soft rock tunnel | |
| CN103984996B (en) | Optimization prediction method for taboo searching algorithm and genetic algorithm of algal bloom mechanism time varying model | |
| CN113158315A (en) | Rock-soil body parameter three-dimensional non-stationary condition random field modeling method based on static cone penetration data | |
| CN110119568B (en) | Method for evaluating stone-throwing effect influence factors of riprap bank protection | |
| CN117786794A (en) | Shield tunneling existing tunnel deformation optimization control method and system | |
| CN106568647A (en) | Nerve network-based concrete strength predication method | |
| CN115481565A (en) | Earth pressure balance shield tunneling parameter prediction method based on LSTM and ant colony algorithm | |
| CN110717270B (en) | Oil and gas reservoir simulation method based on data | |
| CN114818548A (en) | Aquifer parameter field inversion method based on convolution generated confrontation network | |
| CN108959721B (en) | Method for calculating reliability of long-span bridge member under action of windmill load | |
| CN113569487B (en) | BP neural network-based method for predicting step blasting throwing effect | |
| CN110826714A (en) | Dynamic regulation and control method for rocky foundation pit blasting parameters | |
| CN113158297B (en) | Hydraulic arc steel gate parameter load identification method | |
| CN112396312B (en) | Open caisson sinking construction automatic decision-making method and system | |
| CN106874627B (en) | Detection method for detecting construction quality and working state of mine anchor rod | |
| CN108763164A (en) | Evaluation method for coal and gas outburst inversion similarity | |
| CN116911148A (en) | Method and system for evaluating earthquake damage of sedimentary basin building group | |
| CN115809598A (en) | Slope deformation prediction and control value correction method based on safety coefficient | |
| CN113032996B (en) | Water channeling channel identification method for hypotonic fractured reservoir horizontal well |
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 |