CN-121766207-B - Novel wedge-shaped beam type grid dam debris flow occlusion degree prediction method

CN121766207BCN 121766207 BCN121766207 BCN 121766207BCN-121766207-B

Abstract

The invention provides a novel wedge-shaped beam type grid dam debris flow occlusion degree prediction method, which belongs to the technical field of debris flow disaster prevention and control, and comprises the following steps of obtaining basic parameters; step two, designing wedge beam type grid dam structure parameters, step three, carrying out a water tank model test to obtain measured data of the occlusion degree, step four, constructing an occlusion degree prediction model, and step five, calculating an occlusion degree prediction value of the debris flow by using the model to predict the occlusion degree. The invention takes a wedge beam type grating dam as a research object and constructs a debris flow occlusion degree prediction system based on a water tank model test and multi-factor coupling analysis.

Inventors

HU GUISHENG
SHEN WENHAO
TIAN SHUFENG
PENG SHUAISHUAI
XU ZENGQIANG

Assignees

中国科学院、水利部成都山地灾害与环境研究所

Dates

Publication Date: 20260508
Application Date: 20260302

Claims (10)

1. The novel wedge beam type grid dam debris flow occlusion degree prediction method is characterized by comprising the following steps of: The method comprises the steps of firstly, obtaining basic parameters, wherein the basic parameters are divided into debris flow characteristic parameters and channel condition parameters, the debris flow characteristic parameters comprise debris flow volume weight, debris flow solid volume concentration, primary debris flow scale and minimum granularity value of maximum block stones in the debris flow domain; step two, designing structural parameters of the wedge-shaped beam type grating dam, which comprises the following steps: Determining the angle ratio of the wedge-shaped structure, wherein the angle ratio is the ratio of the actual angle of the wedge-shaped structure of the wedge-shaped beam type grating dam to 60 degrees; Calculating the ratio of the dam body reservoir capacity to the reservoir capacity, namely calculating the ratio of the total amount of the debris flow solid matters to the reservoir capacity, namely the reservoir capacity ratio, based on the primary debris flow scale, the debris flow solid volume concentration, the average channel longitudinal slope ratio reduction, the wedge-shaped structure angle ratio, the average channel width, the dam height and the reservoir capacity; determining the opening width and the relative opening width of the dam body, namely obtaining the characteristic particle size of the debris flow basin through a pit test test of field investigation Based on characteristic particle size Setting the opening width of the dam body, and further calculating the opening width and the characteristic particle size I.e. the relative opening width; Thirdly, carrying out a water tank model test according to the basic parameters obtained in the first step and the structural parameters designed in the second step to obtain actual measurement data of the occlusion degree; Step four, constructing an occlusion degree prediction model, collecting a plurality of groups of measured occlusion degree data obtained in the step three, taking the volume weight of the debris flow, the solid volume concentration of the debris flow, the scale of the debris flow, the average channel longitudinal slope ratio drop, the wedge-shaped structure angle ratio, the relative opening width and the reservoir capacity ratio as control factors, analyzing the association relation between each control factor and the measured occlusion degree data by adopting a multiple regression analysis method, determining the function form between each control factor and the occlusion degree, and fitting to obtain the occlusion degree prediction model; And fifthly, applying the model to predict the occlusion degree, inputting the basic parameters of the debris flow channel to be predicted and the structural parameters of the corresponding wedge-shaped beam type grating dam into the occlusion degree prediction model constructed in the fourth step, and calculating to obtain the debris flow occlusion degree prediction value of the wedge-shaped beam type grating dam under the working condition.
2. The method for predicting the mud-rock flow occlusion degree of the novel wedge-shaped beam type grating dam according to claim 1, wherein when the angle ratio of the wedge-shaped structure is determined in the second step, the value range of the angle ratio is 0.75-1.25.
3. The method for predicting the mud-rock flow occlusion degree of the novel wedge-shaped beam type grating dam according to claim 1, wherein when the reservoir capacity ratio is calculated in the second step, the value range of the reservoir capacity ratio is 0-4.0, and the calculated total amount of the required mud-rock flow solid substances is obtained through the conversion of the product of the mud-rock flow scale and the mud-rock flow solid volume concentration.
4. The method for predicting mud-rock flow occlusion of a novel wedge-shaped beam type grating dam according to claim 1, wherein the characteristic particle diameter in the second step is as follows The particle size distribution curve is drawn based on soil sample screening results of a debris flow accumulation area and a circulation area, wherein the particle size value is corresponding to the accumulated content of 95% in the overall particle size distribution curve of the debris flow basin.
5. The method for predicting the debris flow occlusion degree of the novel wedge-shaped beam type grating dam according to claim 1, wherein when a water tank model test is carried out in the third step, a test device comprising a hopper, a water tank, a wedge-shaped beam type grating dam model and a tailing pond is built, the blocking area of an overflow port and the total area of an overflow section before the debris flow in the water tank stops moving completely are recorded immediately, and the ratio of the blocking area to the total area of the overflow port is calculated to serve as actual measurement data of the occlusion degree.
6. The novel wedge-shaped beam type grid dam debris flow occlusion degree prediction method according to claim 1 is characterized in that before multiple regression analysis is adopted in the fourth step, dimensionless treatment is carried out on control factors, so that the control factors are in the same order of magnitude, and accuracy of association relation analysis is improved.
7. The method for predicting mud-rock flow occlusion of a novel wedge-shaped beam type grating dam according to claim 1, wherein when determining the opening width of the dam body in the second step, the opening width is based on the characteristic particle size Is set at 1.62 times.
8. The method for predicting the degree of occlusion of mud-rock flow of a novel wedge-shaped beam type grating dam according to claim 1, wherein the number of the groups of the measured data of the degree of occlusion collected in the fourth step is not less than 70, after a model for predicting the degree of occlusion is obtained by fitting, the fitting goodness of the model is verified by calculating the error between the model predicted value and the measured value, and the fitting goodness of the model is ensured ≥0.7。
9. The novel wedge beam type grid dam debris flow occlusion degree prediction method is characterized by comprising the following steps of carrying out fusion modeling by adopting three algorithms of multiple linear regression, support vector machine regression and random forest regression when the occlusion degree prediction model is built in the fourth step, wherein the three algorithms comprise firstly carrying out characteristic importance analysis on all control factors by using the random forest regression model, screening out core control factors with obvious influence on the occlusion degree, feeding back importance weights of the core control factors to the multiple linear regression model as characteristic weight coefficients when the multiple linear regression model builds linear association, taking initial occlusion degree prediction values output by the multiple linear regression model as additional input characteristics into training samples of the support vector machine regression model, enabling a nonlinear fitting result of the support vector machine regression model to be reversely transferred to the random forest regression model when the occlusion degree is matched with the nonlinear association of the control factors, adjusting the integration weights of decision trees in the random forest regression model, enabling the random forest regression model to more focus on correction of nonlinear residual errors, and realizing accurate linear fitting of the three algorithms and the nonlinear association.
10. The method for predicting the mud-rock flow occlusion degree of the novel wedge-shaped beam type grating dam according to claim 9, wherein when the occlusion degree prediction model is constructed in the fourth step, model optimization is realized through layered collaboration and cross verification of three algorithms, namely multiple linear regression, support vector machine regression and random forest regression, and the specific operations comprise: Dividing the measured data of the occlusion degree and control factors into a training set and a verification set, training a multiple linear regression model by using the training set data to obtain a linear prediction basic model of the occlusion degree, and simultaneously calculating a prediction residual error of the model on the verification set; Secondly, taking the predicted residual as a new predicted target, training a support vector machine regression model by combining control factors, and fitting a nonlinear association which cannot be captured by a linear model to obtain a residual correction model; Thirdly, performing full-dimensional fitting on the control factors and the actually measured occlusion degree by using a random forest regression model to obtain an integrated learning prediction model, and evaluating the suitability of the linear basic model and the residual error correction model by using an out-of-bag data verification result of the random forest regression model; Fourthly, according to the verification result of the random forest regression model, adjusting the characteristic coefficient of the multiple linear regression model and the kernel function parameter of the support vector machine regression model, and then superposing the adjusted linear prediction result and the nonlinear residual error correction result to obtain a preliminary fusion prediction value; And fifthly, carrying out weighted fusion on the preliminary fusion predicted value and the predicted value of the random forest regression model, dynamically distributing weights according to the predicted precision of the three models on the verification set, and finally forming the occlusion degree predicted model considering the linear rule, the nonlinear characteristic and the integration stability.

Description

Novel wedge-shaped beam type grid dam debris flow occlusion degree prediction method Technical Field The invention relates to the technical field of debris flow disaster prevention and control, in particular to a novel wedge-shaped beam type grid dam debris flow occlusion prediction method. Background Debris flow is a typical sudden geological disaster in mountain areas and constitutes a serious threat to towns, traffic lines, hydraulic engineering and the like. In current mud-rock flow prevention and cure, traditional protective structure has apparent defect: The physical dam has the limitations of poor water permeability, easy rapid loss of functions due to sediment accumulation, frequent manual dredging, weak impact resistance, easy damage due to long-term scouring by debris flow, and even possible conversion into secondary disaster sources, and is difficult to adapt to long-term disaster prevention requirements of mountain areas. The traditional permeable type sand blocking dam has the defects that the design depends on experience, key factors (such as debris flow scale, dam body structure and adaptability of debris flow motion trail) influencing the blocking effect are considered insufficiently, so that the design subjectivity is strong, the blocking degree and critical blocking judging model is imperfect, the blocking coarse and fine functions are easily lost due to blocking of an overflow section, the structure optimization focuses on stability and anti-flushing performance, the regulation and control capability is not improved from the angles of diversion and energy dissipation of the debris flow, and the engineering applicability and the protection effect are limited. Along with the promotion of the important engineering construction in mountain areas, the requirements on the accuracy and the adaptability of the debris flow protection structure are remarkably improved, and a technical method capable of scientifically predicting the blocking degree of the permeable check dam and combining structural optimization and functional reliability is needed. Disclosure of Invention The invention provides a novel wedge-shaped beam type grid dam debris flow occlusion degree prediction method, which is used for constructing a debris flow occlusion degree prediction system based on a water tank model test and multi-factor coupling analysis by taking a wedge-shaped beam type grid dam as a research object. The method for predicting the blocking degree of the mud-rock flow of the novel wedge-shaped beam type grating dam comprises the following steps of firstly, obtaining basic parameters, wherein the basic parameters are divided into mud-rock flow characteristic parameters and channel condition parameters, and the mud-rock flow characteristic parameters comprise mud-rock flow volume weight, mud-rock flow solid volume concentration, primary mud-rock flow scale and minimum granularity value of maximum block stones in the mud-rock flow region; the channel condition parameters comprise channel average width of a cross section of a wedge-shaped beam type grating dam of a debris flow basin, average channel longitudinal slope ratio reduction in the cross section range of the wedge-shaped beam type grating dam, step two, design of wedge-shaped beam type grating dam structure parameters, wherein the angle ratio is determined to be the ratio of the actual angle of the wedge-shaped beam type grating dam to 60 DEG, calculation of the dam body reservoir capacity to reservoir capacity ratio, calculation of the reservoir capacity of the wedge-shaped beam type grating dam based on the channel average width, the preliminarily set dam height, the wedge-shaped structure angle ratio, the minimum granularity value of the maximum block stone and the average channel longitudinal slope ratio reduction, calculation of the ratio of the total amount of debris flow solid matters to the reservoir capacity, namely the reservoir capacity ratio, and determination of the opening width of a dam body and the relative opening width, namely the characteristic particle size of the debris flow basin is obtained through a pit test test of field steps based on the size of the debris flow, the volume concentration of the debris flow, the average channel longitudinal slope ratio reduction, the average channel longitudinal slope ratio and the average channel longitudinal slope ratio, the wedge-shaped structure angle ratio, the channel average angle and the channel average angle of the damBased on characteristic particle sizeSetting the opening width of the dam body, and further calculating the opening width and the characteristic particle sizeThe method comprises the steps of obtaining a corresponding opening width, carrying out a water tank model test according to the basic parameters obtained in the first step and the structural parameters designed in the second step to obtain measured data of the blocking degree, building a blocking degree prediction mode