KR100502648B1 - Behavior prediction method in excavation excavation - Google Patents

Abstract

The present invention relates to a method for predicting behavior during soil excavation, and more particularly, to a method for predicting behavior using fuzzy theory. The present invention has no non-uniformity in estimates and can quickly determine the ground microindex by performing accurate estimation suitable for the measured values in a short time. The object of the present invention is to provide a method for predicting the behavior during excavation, and to solve the problem, in the method for predicting the behavior during excavation, the initial design value is calculated when there is a difference between the measured behavior value and the original design value. Set the change δ that increases or decreases the ground material properties such as the earth pressure P, ground reaction coefficient Ke, ground adhesion force c, ground friction angle Ø, ground volume γ, and brace spring coefficient Ks on the membership function. s41) and the calculated behavior values when the change amount δ is plurally decreased (s48), and the ground physical properties when the calculated behavior values and the measured behavior values are roughly matched. And a brace to the spring coefficient Ks characterized in that (s51) to a hwakjeongchi.

Description

Behavior prediction method in soil excavation

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for predicting behavior during soil excavation, and more particularly to a method for predicting behavior using fuzzy theory.

Excavation work performed in the case of constructing underground structures such as hollow pit, pit, and foundation of the structure is used to prevent the impact of the excavation surface from collapsing or ground subsidence due to the progress of the excavation process. Because it is necessary to minimize and to carry out construction safely, the earthworks construction is usually performed.

In each excavation process in such a construction, the construction is progressed by sharing the imbalance of the ground caused by the excavation with the surrounding walls deeper than the excavation wall, the strut, and the excavation surface. However, the ground is indeterminate with many uneven physical properties, and even if the physical properties are sufficiently considered at the time of design, the ground may be different from the actual behavior.

Since this difference in behavior is remarkable in large depth cutting work, etc., in this kind of construction, field measurement is performed on the basis of the phenomenon measurement value obtained by performing on-site measurement. The numerical method is estimated by inverse analysis, predicts the behavior of the ground from this estimate, and is useful for construction management and safety management.

The numbers of such inverse analysis are, for example, trial and error or optimization (direct formulating, inverse formulating, Kalman filter method, etc.), and the optimization numbers are mathematically processed after the next stage of construction. The trial and error method is well used because it takes time to predict the change of the input constant.

However, such a trial and error inverse analysis has a technical problem described below.

In other words, in the trial and error inverse analysis method, when the unknown of the ground is estimated, factors such as practical experience, the sixth sense of the person in charge, and know-how based on the experience are dominant, so that individual estimates appear in the estimate and the estimate is uneven. There was a problem.

In addition, in order to estimate a plurality of unknowns, there is a problem that not only takes time, but also does not easily fit with the actual behavior. In addition, it was very difficult to calculate accurate physical property values for the whole construction area for the undefined soil quality from the results of limited measured values.

SUMMARY OF THE INVENTION The present invention has been made in view of such a conventional problem, and provides a method of predicting the behavior during excavation excavation, which is capable of performing an accurate estimation suitable for the measured value in a short time without any nonuniformity in the estimated value.

In order to achieve the above object, the present invention, when excavating the interior of the earthquake wall is measured the behavior of the earthquake wall displacement, bending moment, shear force, and the like, and based on the obtained measured behavior value, In the method of predicting the behavior during excavation of soil excavation, when there is more than a predetermined difference between the measured behavior value and the original design value, the earth pressure P set at the time of calculating the original design value, the ground reaction coefficient Ke, and the ground Changes of the soil properties such as the adhesion force C, the friction angle φ of the ground, the volume weight γ of the ground, and the membership function centered on the prop spring coefficient Ks, respectively, and the increase and decrease of the ground physical properties and the prop spring coefficient Ks on the membership function. And calculate the calculated behavior values when the change amount δ is increased or decreased a plurality of times. The following physical and bracing spring coefficient Ks was to a hwakjeongchi.

According to the method for predicting the behavior during excavation, the ground physical properties such as earth pressure P, ground reaction force Ke, ground adhesion force C, ground friction angle Ø, ground volume weight γ By setting the membership function centering on the and spring spring coefficient Ks, respectively, the original design value is not very large with the actual ground property value, and the calculated behavior value can be converged to the measured behavior value early.

In this case, the membership function can be set to the isosceles triangle function.

Best Mode for Carrying Out the Invention Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings. 1 to 7 show one embodiment of the behavior prediction method in excavation excavation according to the present invention.

Fig. 1 shows the overall flow of construction management in excavation excavation. In construction management in excavation excavation, as shown in the same drawing, first, in a certain excavation (construction) step, displacement and bending of the excavation wall Behaviors such as moment and shear force are measured (step s1).

Next, a comparison between the obtained measured behavior value A and the designed value is performed (steps S2 and S3). In this case, measured behavior A and design values are generally not very consistent. If the measured behavior A and the design value substantially coincide in step S3, construction continues.

On the other hand, when it is determined in step s3 that the measured behavior value A and the design value do not substantially coincide (when a predetermined difference is recognized between both members), the behavior prediction method of the present embodiment is executed in step s4. Details of this behavior prediction method are shown below in FIG. 2.

In the behavior prediction method of step s4, when the soil physical properties such as earth pressure P, ground reaction coefficient Ke, ground adhesion force c, ground friction angle Ø, ground volume weight γ, and brace spring coefficient Ks are determined (step s5), The deformation prediction (pure analysis) of the soil in the next excavation step is performed (step s6).

In addition, when the stress degree which generate | occur | produces in the soil wall is investigated from the deformation | transformation prediction obtained by this analysis (step s7), and it is determined in step s8 that safety of the soil wall can be ensured, construction continues. .

On the other hand, when it is determined in step s8 that the safety of the antifouling wall cannot be secured, a countermeasure is performed (step s9), and construction continues.

The detailed procedure of the behavior prediction method performed in step s4 is shown in FIG. When the procedure shown in the drawing is started, first, at step s40, the ground physical properties such as earth pressure P, ground reaction force Ke, ground adhesion force c, ground friction angle Ø, ground volume weight γ, and brace spring coefficient Ks For each, the membership function based on fuzzy theory is set.

This membership function is set for each existing soil layer along the depth direction of the ground. An example of the membership function set in Fig. 3A is shown. The membership function shown in the drawing shows the soil physical properties and brace spring coefficients such as earth pressure P, ground reaction force Ke, ground adhesive force c, ground friction angle Ø, ground volume weight γ, etc. It is a function of an isosceles triangle centered on Ks.

The function of this isosceles triangle is the minimum and maximum at the position where the accuracy (ambiguity) of the design value is reduced by half. In addition, since the maximum and minimum values may not be absolutely higher or lower than the friction angle Ø of the cohesive C ground of the ground, for example, these absolute values may be set to the maximum or minimum.

More specifically, with reference to FIG. 3 (b), for example, the membership function Gn for the earth pressure P is set for each soil layer by the equation shown below. Unknown values other than earth pressure P are set in the same manner.

<Example of membership function (for P)>

Gn = {(Pn-P _min ) / (P _o -P _min )} × 0.5 + 0.5 (P _min ≤P _m <P _o )

Gn = {(Pn-P _o ) / (P _max -P _o )} × 0.5 + 0.5 (P _o ≤P _n <P _max )

Where P _o , P _min , and P _max are the original design values, minimum and maximum values of the earth pressure, respectively (minimum and maximum values are set by the designer based on engineering judgment).

The shape of the membership function is not limited to that shown in Fig. 3A, and may be, for example, a normal distribution function.

Subsequently, in step s41, the ground physical property values (soil pressure P, ground reaction coefficient Ke, ground adhesion force c, ground friction angle Ø, ground volume weight γ) and brace spring coefficient Ks are increased and decreased on the membership function described above. Is set.

The change amount δ in this case sets the change amount δ of the accuracy (grade) G for the design of the membership function. For example, the change amount δ is set to a value of 0.1 to 0.05 for the accuracy of the design value. Next, in step s42, the repetition number n of the calculation of the calculation behavior value B is set.

In the following step s43, the ground physical properties set at the time of calculating the original design value (soil pressure P, ground reaction coefficient Ke, ground adhesion force c, ground friction angle Ø, ground volume weight γ) and the support spring coefficient Ks For each soil level, the calculated behavior value Bo is obtained.

In step S44, the measured action value A measured in step S1 and the calculation behavior value Bo are compared for each floor. If it is determined in step S44 that the measured behavior value A is larger than the calculated behavior value Bo, the unknown value (earth pressure P, the reaction force coefficient of the ground, the adhesive force c of the ground, the friction angle of the ground is increased so that the calculated behavior value Bo becomes large in the step S45. Ø, ground material properties such as ground weight γ and bottom-to-spring coefficient Ks) are increased or decreased by the change amount δ on the membership function.

Specifically, in the case of earth pressure P in Fig. 3 (b), it is changed by δ in the direction toward P _max .

On the other hand, if it is determined in step S44 that the measured behavior value A is smaller than the calculated behavior value Bo, the unknown value (earth pressure P, the reaction coefficient Ke of the ground, the friction angle of the ground Ø, the ground of the ground so that the calculated behavior value Bo becomes small in step S46). The ground physical properties such as the volume weight r and the brace spring coefficient Ks) are increased or decreased by the change amount δ on the membership function.

For example, in the case of earth pressure P in Fig. 3 (b), it is changed by δ in the direction toward P _min .

In the next step s47, the unknown properties (the earth pressure P, the ground reaction coefficient Ke, the ground adhesion force c, the ground friction angle Ø, the ground volume weight γ, etc.) increased or decreased by the amount of change δ in steps s45 and s46, and the brace spring coefficient Ks) is calculated.

For example, the earth pressure P is calculated by the following equation in accordance with the determination in step S45 or step S46.

P _{n + 1} = {(G _n ± δ-0.5) /0.5} × (P _o -P _min ) + P _min (P _min ≤P _n <P _O )

δ-0.5)/0.5}×(P_max-P_o)+P_o (P_o≤P_n<P_max)P _{n + 1} = {(G _n

δ-0.5) /0.5} × (P _max -P _o ) + P _o (P _o ≤P _n <P _max )

The above expression indicates the case of the (n + 1) th operation.

In step s48, the calculation behavior value B1 is calculated based on the obtained unknown value. In subsequent step s49, it is determined whether or not the calculated behavior value B has been performed by the set number of repetitions n. If not, the process returns to step s44 if no n times has been performed, and if it is executed n times, the process proceeds to step s50.

In step s50, the measured behavior value A is compared with the calculated behavior value Bn repeatedly calculated n times, and if they are approximately identical, the unknown value calculated in step S47 in the following step s51 (earth pressure P, ground). The procedure is concluded by determining the ground property values and the brace spring coefficient Ks such as the reaction force Ke of the ground, the adhesion force of the ground, the friction angle Ø of the ground, and the volume weight γ of the ground.

If it is determined in step s50 that the actual behavior value A and the calculated behavior value Bn do not substantially coincide, the process returns to step s41 to set a new change amount δ and the number of repetitions, and the processing procedure is continued.

4-7 are calculation results which displayed the calculated behavior value obtained by the said procedure with an actual value. In the graph shown in Fig. 4, the repetition count n is 1, i.e., based on the original design value. As can be seen from the figure, the deformation of the soil wall deviates significantly from the calculated behavior value and the measured behavior value.

4 to 7, in the graph of deformation, 0 is the measured behavior value, and in the same figure, the calculated behavior value is indicated by the solid line. In the graph shown in FIG. 5, the iteration count n is 2, and in the graph shown in FIG. 6, the iteration count n is 10.

When the repeat count n shown in FIG. 5 is 2, the degree of agreement between the calculated behavior value and the measured behavior value is not sufficient. However, when the repeat count n becomes 5 as shown in FIG. It can be seen that the calculated behavior and the measured behavior agree well.

The degree of agreement between the calculated behavior value and the measured behavior value is almost different from the case of 5 even if the repetition number n is 10, and it can be seen that the unknown procedure is set at about 5 repetition times n.

However, according to the above-described behavior prediction method for soil excavation, earth pressure P, ground residual coefficient Ke, ground adhesive force c, ground friction angle Ø, ground volume weight γ, etc. set at the time of initial calculation are calculated. By setting up the membership function centered on the ground property value and the brace spring coefficient of Ks, respectively, the initial design value is not much different from the actual ground property value, and it is possible to settle the calculated behavior value to the actual behavior value by the repeated operation number of several times. have.

In this case, the determination of the soil properties such as earth pressure P, soil reaction coefficient Ke, soil adhesion c, soil friction angle Ø, ground volume weight γ, and support spring coefficient Ks is governed by the sixth sense and experience of the site concerned. It is not possible to determine the exact unknowns in a short time and accurately.

As described in detail in the above embodiments, according to the method for predicting the behavior of soil excavation in accordance with the present invention, it is possible to accurately and quickly determine the unclear ground microindex, so that the analysis result is easy for construction management and safety management of soil excavation. Can be reflected.

1 is a flow chart showing the overall flow of construction management in excavation excavation to which the behavior prediction method according to the present invention is applied

2 is a flowchart showing an example of the procedure of the behavior prediction method according to the present invention.

FIG. 3A is an explanatory diagram showing an example of a membership function set by the behavior prediction method shown in FIG.

3 (b) is an explanatory diagram showing a specific example of FIG. 3 (a).

FIG. 4 is a graph of the results obtained in the first iteration count of the behavior prediction method shown in FIG.

FIG. 5 is a graph of the results obtained in the second iteration operation count of the behavior prediction method shown in FIG.

6 is a graph of the results obtained in the fifth iteration of the operation prediction method of FIG.

FIG. 7 is a graph of the results obtained in the 10th iteration operation count of the behavior prediction method shown in FIG.

Claims (2)

When excavating the inside of the excavation wall, the behavior of the excavation excavation method that predicts the future behavior based on the measured behavior value obtained by measuring the displacement of the excavation wall, the bending moment, the shear force, and the like. To If there is more than a predetermined difference between the measured behavior value and the original design value, The membership function centered on the soil physical properties such as earth pressure P, ground reaction coefficient Ke, ground adhesion force c, ground friction angle Ø, ground volume weight γ, and brace spring coefficient Ks set at the time of calculating the original design value, Setting each one, Setting a change amount δ which increases and decreases the ground property value and the brace spring coefficient Ks on the membership function; Calculating a calculated behavior value when the change amount δ is increased by a plurality of times; Setting the ground property value and the brace spring coefficient Ks when the calculated behavior value and the measured behavior value approximately coincide with each other; Behavior prediction method during excavation, characterized in that consisting of. The method for predicting behavior during excavation of earth excavation according to claim 1, wherein the membership function is an isosceles triangle function.

Images