Method and system for rapidly predicting wind speed of tunnel face in drilling and blasting method tunnel construction

Abstract

The invention belongs to the technical field of tunnel construction, and particularly relates to a method and system for rapidly predicting the wind speed of a tunnel face in drilling and blasting method tunnel construction. The method comprises the following steps: acquiring the concentration and wind speed data of harmful gas at a tunnel face measuring point within 30 minutes of ventilation after blasting of the tunnel face; constructing a multiple linear regression model of the harmful gas concentration and the wind speed; based on correlation analysis, testing the reliability effectiveness of a control variable in the multiple linear regression model; checking whether multiple collinearity exists among independent variables in the multiple linear regression model based on multiple collinearity analysis; determining a prediction model based on the reference regression analysis result; and the wind speed value is quickly predicted based on the concentration values of CO, CO2 and SO2. The relation between the harmful gas and the wind speed is established through the concentration of the harmful gas and the wind speed data, the change of the wind speed of the tunnel face can be rapidly predicted, the frequency of the draught fan can be dynamically regulated and controlled, and high efficiency and safety of ventilation of drilling and blasting method tunnel construction are facilitated.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of tunnel construction, and particularly relates to a method and system for quickly predicting the wind speed of a tunnel construction face in a drill-and-blast method. BACKGROUND

[0002] In the process of tunnel drill-and-blast construction, a large amount of harmful gas is generated after the blasting of the tunnel face, which seriously affects the construction progress and the safety of personnel, so that after the blasting, an axial flow fan outside the tunnel is used to provide fresh air flow into the tunnel, thereby diluting the concentration of harmful gas and removing it outside the tunnel with the backflow air flow.

[0003] In the existing tunnel construction process, the fan air volume adjustment only refers to different stages of construction process, personnel comfort and environmental changes and other factors, and the fan frequency is adjusted manually without quantitatively giving the relationship between harmful gas and wind speed, resulting in the technical problem that the wind speed cannot be accurately predicted to control the fan frequency. Based on this technology, the fan frequency can be accurately and efficiently dynamically controlled to achieve the goals of energy saving, improving ventilation effect and ensuring the safety of construction personnel. SUMMARY

[0004] To solve the above problems, the present application provides a method and system for quickly predicting the wind speed of a tunnel construction face in a drill-and-blast method. By detecting the concentration of harmful gas and wind speed data during the migration of harmful gas, the present application establishes the relationship between harmful gas and wind speed, can quickly predict the change of tunnel face wind speed, and can dynamically control the fan frequency, which is helpful to the efficient and safe ventilation of drill-and-blast tunnel construction.

[0005] The technical solution of the present application is as follows: a method for quickly predicting the wind speed of a tunnel construction face in a drill-and-blast method, comprising the following steps:

[0006] S1: obtaining the concentration of harmful gas and wind speed data at the measurement point of the tunnel face within 30 minutes after the blasting of the tunnel face;

[0007] S2: constructing a multiple linear regression model of harmful gas concentration and wind speed based on the concentration of harmful gas and wind speed data;

[0008] S3: testing the reliability and effectiveness of the control variables in the multiple linear regression model based on correlation analysis;

[0009] S4: testing whether there is multiple collinearity among the independent variables in the multiple linear regression model based on multiple collinearity analysis;

[0010] S5: determining the prediction model based on the results of the benchmark regression analysis;

[0011] S6: When steps S3-S5 meet the requirements, the multiple linear regression model of harmful gas concentration and wind speed is finally determined, the relationship between multiple harmful gas concentrations and wind speed is established, and the change of the tunnel face wind speed is quickly predicted.

[0012] In the step S1, the harmful gas includes CO, CO2 and SO2.

[0013] In the step S2, the multiple linear regression model of harmful gas concentration and wind speed is specifically as follows:

[0014] Y = β0 + β1X1 + β2X2 + … + β k X k

[0015] In the formula, Y is the dependent variable, and the predicted variable wind speed; X1, …, X K are independent variables, explanatory variables, including CO, CO2 and SO2; β0 is the intercept term, the value when all independent variables are 0, and a constant term; β1, …, β K are regression coefficients of the independent variables.

[0016] The regression coefficients in the linear regression model are determined by minimizing the sum of squared residuals based on the least squares method (OLS), and the specific process is as follows:

[0017] The linear regression model is assumed to be:

[0018] Y = β0 + β1X

[0019] By minimizing the sum of squared residuals, the OLS estimator

[0020]

[0021] In the formula, Y is the dependent variable; X is the independent variable; β0 is the intercept term; β1 is the regression coefficient of the independent variable; n is the sample size; X i is the independent variable; Y i is the dependent variable; is the mean of the independent variable, is the mean of the dependent variable; is the OLS estimator.

[0022] In the step S3, the specific steps of the correlation analysis test for controlling the reliability and effectiveness of the variable are as follows:

[0023] S31: Calculate the Pearson correlation coefficient between any two control variables, and the specific formula is as follows:

[0024]

[0025] In the formula, r is the Pearson correlation coefficient; Xi X is the independent variable; Y i Y is the dependent variable; X is the independent variable mean, Y is the dependent variable mean; n is the sample size;

[0026] S32: Significance test according to Pearson correlation coefficient, i.e. t test, the specific formula is:

[0027]

[0028] In the formula, t is the t test; r is the Pearson correlation coefficient; n is the sample size;

[0029] S33: Judgment of reliability and effectiveness, according to p value method:

[0030] If p < 0.01, it is significant at 1% level (highly significant, acceptable, represented by ***);

[0031] If 0.01 < p < 0.05, it is significant at 5% level (moderately significant, acceptable, represented by **);

[0032] If 0.05 < p < 0.1, it is significant at 10% level (lowly significant, acceptable, represented by *);

[0033] In the step S4, the multiple collinearity analysis test is used to judge whether there is multiple collinearity between the independent variables by calculating the variance inflation factor VIF. The specific formula of the variance inflation factor VIF is:

[0034]

[0035] In the formula, VIF j is the variance inflation factor, is the variance;

[0036] If VIF = 1, there is no collinearity;

[0037] If 1 < VIF < 5, there is low collinearity (acceptable);

[0038] If 5 ≤ VIF < 10, there is moderate collinearity (usually acceptable);

[0039] If VIF ≥ 10: there is serious collinearity (the variable needs to be removed).

[0040] A kind of drill and blast method tunnel construction face wind speed quick prediction system, including data acquisition module, model establishment module, model verification module, determine prediction model module and quick prediction module;Wherein: the parameter acquisition module is used to obtain the harmful gas concentration and wind speed variation value of face measuring point in 30min after face blasting;The model establishment module is used to build multiple linear regression model based on harmful gas concentration and wind speed;The model verification module is used to verify the reliability and effectiveness of model based on correlation analysis test module, multiple collinearity analysis;The determination prediction model module is used to obtain the prediction model based on benchmark regression analysis result;The quick prediction module is used to the wind speed quick prediction based on harmful gas concentration.

[0041] The technical effect of the application is that the multiple linear regression model is applied to the tunnel construction face wind speed quick prediction, the multiple linear regression analysis model of harmful gas concentration and wind speed is established, the relationship between harmful gas and wind speed is obtained, the change of tunnel face wind speed can be quickly predicted, and the frequency of fan can be dynamically regulated, which helps the efficient and safe ventilation of drill and blast method tunnel construction.

[0042] The following will be further described with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 It is a flowchart of the drill and blast method tunnel construction face wind speed quick prediction method of the embodiment of the application. DETAILED DESCRIPTION

[0044] Embodiment 1

[0045] As Figure 1 shown, a drill and blast method tunnel construction face wind speed quick prediction method includes the following steps:

[0046] S1: obtain the concentration of harmful gas and wind speed data at face measuring point in 30min after face blasting;

[0047] S2: build multiple linear regression model of harmful gas concentration and wind speed based on harmful gas concentration and wind speed data;

[0048] S3: test the reliable and effective degree of control variable in multiple linear regression model based on correlation analysis;

[0049] S4: test whether multiple collinearity exists between independent variables in multiple linear regression model based on multiple collinearity analysis;

[0050] S5: determine prediction model based on benchmark regression analysis result;

[0051] S6: When steps S3-S5 meet the requirements, the multiple linear regression model of harmful gas concentration and wind speed is finally determined, the relationship between multiple harmful gas concentrations and wind speed is established, and the change of the tunnel face wind speed is quickly predicted.

[0052] In the step S1, the harmful gas includes CO, CO2 and SO2.

[0053] In the step S2, the multiple linear regression model of harmful gas concentration and wind speed is specifically as follows:

[0054] Y = β0 + β1X1 + β2X2 + … + β k X k

[0055] In the formula, Y is the dependent variable, the predicted variable wind speed; X1, …, X K are independent variables, explanatory variables, including CO, CO2 and SO2; β0 is the intercept term, the value when all independent variables are 0, the constant term; β1, …, β K are the regression coefficients of the independent variables.

[0056] The regression coefficients in the linear regression model are determined by minimizing the sum of squared residuals based on the least squares method (OLS), and the specific process is as follows:

[0057] The linear regression model is assumed to be:

[0058] Y = β0 + β1X

[0059] By minimizing the residual square, the OLS estimator is solved

[0060] In the step S3, the specific steps of the correlation analysis test for controlling the reliability and effectiveness of the variable are as follows:

[0061] S31: Calculate the Pearson correlation coefficient between any two control variables, and the specific formula is as follows:

[0062]

[0063] In the formula, r is the Pearson correlation coefficient; X i is the independent variable; Y i is the dependent variable; is the mean of the independent variable, is the mean of the dependent variable; and n is the sample size.

[0064] S32: According to the Pearson correlation coefficient, the significance test, i.e. t test, is performed, and the specific formula is as follows:

[0065]

[0066] In the formula, t is a t test; r is a Pearson correlation coefficient; and n is a sample size.

[0067] S33: Reliability and validity judgment according to a p-value method:

[0068] If p < 0.01, significant at a 1% level (highly significant, acceptable, represented by ***);

[0069] If 0.01 < p < 0.05, significant at a 5% level (moderately significant, acceptable, represented by **);

[0070] If 0.05 < p < 0.1, significant at a 10% level (lowly significant, acceptable, represented by *).

[0071] In the step S4, a multiple collinearity analysis test is performed to determine whether there is a multiple collinearity between independent variables by calculating a variance inflation factor VIF. The specific formula of the variance inflation factor VIF is as follows:

[0072]

[0073] In the formula, VIF j is a variance inflation factor, is a variance;

[0074] If VIF = 1, there is no collinearity;

[0075] If 1 < VIF < 5, there is low collinearity (acceptable);

[0076] If 5 ≤ VIF < 10, there is moderate collinearity (generally acceptable);

[0077] If VIF ≥ 10, there is serious collinearity (the variable needs to be removed).

[0078] Example 2

[0079] A drill-and-blast tunnel construction face wind speed rapid prediction system includes a data acquisition module, a model establishment module, a model verification module, a prediction model determination module, and a rapid prediction module. The parameter acquisition module is used to acquire harmful gas concentration and wind speed change values of a face measuring point within 30 minutes after blasting of a face. The model establishment module is used to construct a multiple linear regression model based on the harmful gas concentration and the wind speed. The model verification module is used to verify reliability and validity of the model based on a correlation analysis and a multiple collinearity analysis. The prediction model determination module is used to obtain a prediction model based on a benchmark regression analysis result. The rapid prediction module is used to rapidly predict the wind speed based on the harmful gas concentration.

[0080] Example 3

[0081] According to the method for quickly predicting the wind speed of a tunnel construction face in a drill-and-blast method described in Embodiment 1, the wind speed of the construction face of the XX tunnel is quickly predicted, and the specific process is as follows:

[0082] S1: Obtain the harmful gas CO, CO2, SO2 concentrations and wind speed of the measurement point of the construction face within 30 minutes after blasting of the construction face, as shown in Table 1.

[0083] Table 1: Concentration data table of harmful gases CO, CO2, SO2 and wind speed

[0084]

[0085] S2: Based on the CO, CO2, SO2 concentrations and wind speed, a multiple linear regression model is constructed, specifically as follows:

[0086] Y = β0 + β1CO + β2CO2 + β3SO2

[0087] In the formula, β1, β2, and β3 are non-standardized regression coefficients; and β0 is a constant term.

[0088] Based on the least squares method (OLS), the regression coefficients in the linear regression model are determined by minimizing the sum of squared residuals, and the results are shown in Table 2.

[0089] Table 2: Determination of regression coefficients

[0090] Type Wind speed CO 0.012032 *** ]]> (4.278) <![CDATA[CO2]]> 0.0013188 *** ]]> (29.546) SO2 -0.00606478 *** ]]> (-8.276) _cons 0.3522586 *** ]]> (15.086) N 282.000 r2_a 0.877 F 667.760

[0091] * p < 0.1, ** p < 0.05, *** p < 0.01

[0092] As can be seen from Table 2, the R-square value of the model after adjustment is 0.877, indicating that the relationship between the CO, CO2, SO2 concentrations and wind speed is significant. When the model is subjected to F test, it is found that the model passes the F test (F = 667.760, p = 0.000 < 0.05), which means that the model is meaningful. r2_a is the correlation coefficient of the multiple linear regression analysis model, representing the explanatory power of the independent variables (each influencing parameter) on the dependent variable (wind speed), and is between 0 and 1. The closer to 1, the better the fitting, and N is the sample size, and _cons is the constant term.

[0093] Therefore, the multiple linear regression model is:

[0094] Y = 0.3522586 + 0.012032CO + 0.0013188CO2 - 0.00606478SO2

[0095] S3: Based on the correlation analysis test, the reliability and effectiveness of the control variables are tested.

[0096] S31: Calculate the Pearson correlation coefficient between any two control variables, the specific formula is:

[0097]

[0098] In the formula, r is the Pearson correlation coefficient; X i is the independent variable; Y i is the dependent variable; is the independent variable mean, is the dependent variable mean; n is the sample size;

[0099] S32: According to the Pearson correlation coefficient, the significance test is t test, the specific formula is:

[0100]

[0101] In the formula, t is the t test; r is the Pearson correlation coefficient; n is the sample size;

[0102] S33: The judgment of reliable effectiveness, the results are shown in Table 3, according to the p value method:

[0103] If p < 0.01, it is significant at 1% level (highly significant, acceptable, represented as ***);

[0104] If 0.01 < p < 0.05, it is significant at 5% level (moderate significant, acceptable, represented as **);

[0105] If 0.05 < p < 0.1, it is significant at 10% level (low significant, acceptable, represented as *).

[0106] Table 3 Correlation analysis parameter table

[0107] Type Wind speed CO CO2 SO2 Wind speed 1 CO 0.398 *** ]] 1 CO2 0.920 *** ]]> 0.469 *** ]] 1 SO2 0.690 *** ]] 0.718 *** ]]> 0.840 *** ]] 1

[0108] In Table 3, ***, **, * represent significant at 1%, 5%, 10% level respectively, according to Table 2, each parameter is significant at 1% level, which indicates that there is a significant positive correlation.

[0109] S3: Based on multiple collinearity analysis, whether there is multiple collinearity between variables;

[0110] Through calculating the variance inflation factor VIF to judge whether there is multiple collinearity between independent variables, the results are shown in Table 4, and the specific formula of variance inflation factor VIF is:

[0111]

[0112] In the formula, VIF j is the variance inflation factor, is the variance;

[0113] Table 4 Multicollinearity test table

[0114] Type VIF 1 / VIF SO2 6.250 0.160 CO2 3.890 0.257 CO 2.360 0.424 Mean VIF 4.170

[0115] The results in Table 4 show that the average VIF values ​​are all less than 10, indicating that there is no serious collinearity problem between the variables and subsequent regression analysis can be performed.

[0116] S4: Based on the results of the baseline regression analysis, a multivariate linear regression prediction model for harmful gas concentration and wind speed is finally determined, and the relationship between the concentration of multiple harmful gases and wind speed is established. Specifically:

[0117] Y=0.3522586+0.012032CO+0.0013188CO2-0.00606478SO2

[0118] This allows for rapid prediction of wind speed changes at the tunnel face.

[0119] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with this technical field within the technical scope disclosed by the present invention should be covered by the scope of protection of the present invention.

Claims

1. A method for rapidly predicting wind speed at the tunnel face during drill-and-blast tunnel construction, characterized by: It includes the following steps: S1: Obtain the concentration and wind speed data of harmful gases at the measuring points of the tunnel face within 30 minutes after the tunnel face blasting. S2: Based on the concentration and wind speed data of harmful gases, construct a multiple linear regression model of harmful gas concentration and wind speed. S3: Based on correlation analysis, test the reliability and effectiveness of control variables in the multiple linear regression model. S4: Based on multicollinearity analysis, test whether there is multicollinearity between independent variables in the multiple linear regression model. S5: Based on the results of benchmark regression analysis, determine the prediction model. S6: When steps S3 - S5 meet the requirements, finally determine the multiple linear regression model of harmful gas concentration and wind speed, establish the relationship between various harmful gas concentrations and wind speeds, and quickly predict the change of wind speed at the tunnel face.

2. The method for rapid prediction of wind speed at the tunnel face during drill and blast tunnel construction according to claim 1, characterized in that: In step S1, the harmful gases include CO, CO2, and SO2.

3. The method for rapid prediction of wind speed at the tunnel face during drill and blast tunnel construction according to claim 1, characterized in that: In step S2, the multiple linear regression model of harmful gas concentration and wind speed is specifically: Y=β0+β1X1+β2X2+…+β k X k Where Y is the dependent variable and wind speed is the predicted variable; X1,…,X K are independent variables and explanatory variables, including CO, CO2, and SO2; β0 is the intercept term, which takes value when all independent variables are 0, and is a constant term; β1,…,β K is the regression coefficient of the independent variable.

4. The method for rapid prediction of wind speed at the tunnel face during drill and blast tunnel construction according to claim 3, wherein: Based on the least squares method (OLS), determine the regression coefficients in the linear regression model by minimizing the sum of squared residuals. The specific process is as follows: Assume the linear regression model is: Y = β0 + β1X By minimizing the square of the residual, the OLS estimator is obtained In the formula, Y is the dependent variable; X is the independent variable; β0 is the intercept term; β1 is the regression coefficient of the independent variable; n is the sample size; X i is the independent variable; Y i is the dependent variable; is the mean of the independent variable, is the mean of the dependent variable; is the OLS estimator.

5. The method for rapid prediction of wind speed at the tunnel face during drill and blast tunnel construction according to claim 1, characterized in that: In step S3, the specific steps for correlation analysis to test the reliability and effectiveness of control variables are: S31: Calculate the Pearson correlation coefficient between any two control variables. The specific formula is: Where r is the Pearson correlation coefficient; X i is the independent variable; Y i is the dependent variable; is the mean of the independent variable, is the mean of the dependent variable; n is the sample size; S32: Conduct a significance test, that is, a t - test, based on the Pearson correlation coefficient. The specific formula is: In the formula, t is the t - test; r is the Pearson correlation coefficient; n is the sample size. S33: Judgment of reliability and effectiveness. According to the p - value method: If p < 0.01, it is significant at the 1% level (highly significant, acceptable, represented by ***). If 0.01 < p < 0.05, it is significant at the 5% level (moderately significant, acceptable, represented by **). If 0.05 < p < 0.1, it is significant at the 10% level (lowly significant, acceptable, represented by *).

6. The method for rapid prediction of wind speed at the tunnel face during drill and blast tunnel construction according to claim 1, characterized in that: In step S4, judge whether there is multicollinearity between independent variables through calculating the variance inflation factor VIF. The specific formula of the variance inflation factor VIF is: Where, VIF j is the variance inflation factor, is the variance; If VIF = 1, there is no collinearity. If 1 < VIF < 5, there is low - level collinearity (acceptable). If 5 ≤ VIF < 10, there is moderate collinearity (usually acceptable). If VIF ≥ 10: there is severe collinearity (variables need to be removed).

7. A rapid wind speed prediction system for tunnel construction using the drill-and-blast method, characterized by: It includes a data acquisition module, a model establishment module, a model test module, a prediction model determination module, and a rapid prediction module. Among them: the parameter acquisition module is used to obtain the concentration and wind speed change values of harmful gases at the measuring points of the tunnel face within 30 minutes after the tunnel face blasting; the model establishment module is used to construct a multiple linear regression model based on the concentration and wind speed of harmful gases; the model test module is used to test the reliability and effectiveness of the model based on correlation analysis and multicollinearity analysis; the prediction model determination module is used to obtain the prediction model based on the results of benchmark regression analysis; the rapid prediction module is used to quickly predict based on the wind speed of harmful gas concentration.

Images