Underground structure prediction system and underground structure prediction method

Abstract

To provide an underground structure prediction system capable of estimating analysis conditions even when an analysis engineer is absent, and reducing delay of processes.SOLUTION: An underground structure prediction system comprises: a natural ground displacement acquisition part for acquiring natural ground displacement data of a measured result of displacement of natural ground where a tunnel is to be constructed; a stress measurement result acquisition part for acquiring stress measurement data of a measured result of stress acting on a construction along an inner peripheral surface of the tunnel; a storage part for storing physical values representing physical properties of the natural ground; an analysis part for analyzing, with an analytic model, the displacement generated in the natural ground and stress acting on timbering using a geology model representing the physical values and geological properties of the ground; and a data assimilation processing part for updating the physical values of the natural ground on the basis of the displacement and the stress obtained by analyzing a section to be analyzed by the analysis part and the natural ground displacement data and the stress measurement data obtained by measurement on the segment to be analyzed.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an underground structure prediction system and an underground structure prediction method. [Background technology]

[0002] In the construction of underground structures such as mountain tunnels and underground power plants, information-based construction is used, which allows design revisions to be made based on data from face observations and measurement work during construction. In particular, when modifying the design of mountain tunnels, for example, one of the following methods is applied: (1) standard design, (2) design under similar conditions, or (3) analytical methods. The standard design method (1) involves applying a support pattern prepared in advance for each rock mass condition according to the rock mass class. (2) The design of similar conditions is based on the actual design records of tunnels constructed in the past. The analytical method (3) is used when standard design or similar design cannot be applied, and involves design based on the displacements occurring in the ground and the stresses acting on the supports using analytical techniques. There is also a strain measurement system in which displacement sensors are installed in the supports to confirm their stability and validity, measure the strain in the supports, and predict the behavior of the ground in real time (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2022-177992 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, when trying to apply the above analysis (3), the following problems arise. For example, there is no established method for setting analysis conditions such as initial stress state, boundary conditions, and rock mass properties, and currently these are set rationally based on the experience and knowledge of analysis engineers. In addition, since the analysis involves modeling actual complex ground conditions, supports, linings, and their interactions, it is not always possible to fully reproduce actual behavior. Furthermore, evaluation of the results requires knowledge gained through experience, so highly advanced technical interpretation is required. As described above, when revising the design of a mountain tunnel using analytical techniques, there is a problem in that it takes time and effort for the analysis engineer to set the above analytical conditions and evaluate the analysis results. In addition, because it takes time to set the analysis conditions and evaluate the analysis results, if construction is suspended to make design revisions and then resumed after the results of the design revisions are obtained, taking the revisions into consideration, there will be a period of suspension of construction, which may lead to delays in the process.

[0005] The present invention has been made in consideration of the above circumstances, and its purpose is to provide an underground structure prediction system and an underground structure prediction method that can estimate analysis conditions even when an analysis engineer is not present, thereby reducing process delays. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, one aspect of the present invention is an underground structure prediction system having a ground displacement data acquisition unit that acquires ground displacement data resulting from measuring the displacement of the ground through which the tunnel is constructed by measuring the inner surface of concrete sprayed into the tunnel, a stress measurement data acquisition unit that acquires stress measurement data resulting from measuring the stress acting on a construction provided along the inner surface of the tunnel, a storage unit that stores physical property values ​​representing the physical properties of the ground, an analysis unit that performs analysis using an analytical model that analyzes the displacement occurring in the ground and the stress acting on a support structure using the physical property values ​​and a geological model that represents the geology of the ground, and a data assimilation processing unit that updates the physical property values ​​of the ground based on the displacement and the stress obtained by analyzing the section to be analyzed by the analysis unit and the ground displacement data and the stress measurement data obtained by measuring the section to be analyzed.

[0007] Moreover, one aspect of the present invention is an underground structure prediction method that causes a computer to execute a ground displacement measurement data acquisition step of acquiring ground displacement data resulting from measuring the displacement of the ground through which the tunnel is constructed by measuring the inner circumferential surface of concrete sprayed into the tunnel, a stress measurement data acquisition step of acquiring stress measurement data resulting from measuring the stress acting on a construction provided along the inner circumferential surface of the tunnel, an analysis step of performing an analysis using an analytical model that analyzes the displacement occurring in the ground and the stress acting on a support structure by using physical property values ​​that are stored in a storage unit that stores physical property values ​​that represent the physical properties of the ground and a geological model that represents the geology of the ground, and a data assimilation processing step of updating the physical property values ​​of the ground based on the displacement and the stress obtained by analyzing the section to be analyzed by the analysis step and the ground displacement data and the stress measurement data obtained by measuring the section to be analyzed. Effect of the Invention

[0008] As described above, according to the present invention, even if an analysis engineer is not present, analysis conditions can be estimated, and process delays can be reduced. [Brief description of the drawings]

[0009] [Figure 1] 1 is a schematic diagram showing the configuration of an underground structure prediction system S according to one embodiment of the present invention. [Diagram 2] FIG. 1 is a diagram showing the data flow of the underground structure prediction system S. [Diagram 3] FIG. 2 is a schematic block diagram for explaining the functions of an analytical computer 20. [Figure 4] 1 is a flowchart explaining the operation of the underground structure prediction system S. [Diagram 5] FIG. 13 is a diagram illustrating an example of a probability distribution of a prediction result. [Figure 6] 1 is a flow chart showing the implementation procedure for conducting a twin experiment. [Figure 7] FIG. 1 is a diagram showing an example of an analytical model simulating the excavation of an unlined tunnel with a circular cross section. [Figure 8] 13 is a graph showing Young's modulus of the ground before and after updating by data assimilation. [Figure 9] This figure shows the correct value of the displacement in the z-axis direction of the top observation point and the ensemble prediction results. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] An underground structure prediction system S according to one embodiment of the present invention will be described below with reference to the drawings. In response to the above-mentioned problems, the underground structure prediction system S in this embodiment is a system that identifies the physical properties of the ground and predicts the displacement occurring in the ground and the stress acting on the support structure by performing excavation analysis incorporating data assimilation. Generally, calculation models are not only constructed based on mathematical models such as partial differential equations, initial and boundary conditions, and constitutive equations that are idealized and simplified for real phenomena, but also contain discretization errors and rounding errors, making it difficult to perfectly reproduce real phenomena. In contrast, the underground structure prediction system S performs data assimilation to take into account the errors in the calculation model, and can use observed data during the execution of the simulation to correct the calculation model and improve the performance of analysis and prediction. In this embodiment, the construction of a mountain tunnel is the subject of a case where the displacement occurring in the natural ground and the stress acting on the support structure are predicted as the construction proceeds. Here, the data assimilation method will be described by taking as an example a case where the ensemble Kalman filter, which is an ensemble-based data assimilation method of the sequential data assimilation method, is used.

[0011] FIG. 1 is a schematic diagram showing the configuration of an underground structure prediction system S according to one embodiment of the present invention. The underground structure prediction system S is used, for example, at a construction site where a mountain tunnel is constructed. FIG 1 illustrates the vicinity of a face K of a mountain tunnel. In a tunnel excavated through a mountain, sprayed concrete SC and steel supports SR are constructed on the inner surface except for the roadbed RB. The steel supports SR are constructed at multiple intervals in the excavation direction of the tunnel.

[0012] 《Components and roles》 The underground structure prediction system S includes a prism P, a total station TS, a strain gauge SG, a concrete stress meter SM, a measurement computer 10, a data logger DL, a router RT, and an analysis computer 20.

[0013] <Measurement of natural ground displacement> A plurality of prisms P are attached to the shotcrete SC. For example, the prisms P may be provided along the circumferential direction of the inner periphery of the tunnel. The total station TS is installed on the roadbed RB during measurement, reflects light onto a prism P, and measures distance by receiving the light reflected from the prism P. This distance measurement is performed for each prism P. The total station TS outputs the measurement results to a measurement computer 10. The measurement computer 10 measures the displacement of the ground based on the measurements obtained from the total station TS.

[0014] Stress Measurement The strain gauge SG is attached along the longitudinal direction of the steel support SR. The strain gauge SG measures the strain of the steel support SR. The concrete stress meter SM is installed between the ground and the shotcrete SC, or embedded in the shotcrete SC. For example, the concrete stress meter SM is attached to the ground and then covered with the shotcrete. The concrete stress meter SM measures the stress in the concrete. The data logger DL acquires from the strain gauges SG and stores the results measured by the concrete stress meter SM, acquires from the concrete stress meter SM and stores these stored measurement results, and outputs these stored measurement results to the measurement computer 10. This allows the measurement computer 10 to collect and store the measurement results measured by the strain gauges SG and the measurement results measured by the concrete stress meter SM. The measurement computer 10 also measures the stress acting on the steel support SR caused by the construction of the mountain tunnel based on the measurement results measured by the strain gauges SG and the measurement results measured by the concrete stress meter SM.

[0015] Data communication The router RT receives data representing the displacement of the ground measured by the measurement computer 10 and data representing the stress acting on the steel support SR from the measurement computer 10 and transmits them wirelessly to the analysis computer 20.

[0016] Data Analysis The analysis computer 20 is installed in an office located near the construction site where the mountain tunnel is constructed. An excavation analysis program and a data assimilation program are installed in the analysis computer 20, and the analysis computer 20 performs excavation analysis processing, data assimilation processing using the prediction results obtained by the excavation analysis and measurement data, and predictive analysis processing for performing predictive analysis. A display device is connected to the analysis computer 20, which displays the results of various processes.

[0017] Data flow of this system FIG. 2 is a diagram showing the data flow of the underground structure prediction system S. The underground structure prediction system S is broadly composed of a measurement data collection unit 1, an analysis unit 2, and a prediction result display unit 3. The measurement data collection unit 1 includes various measuring instruments (prism P, total station TS, strain gauge SG, concrete stress meter SM, etc.) installed at the mountain tunnel construction site, and devices (data logger DL, router RT) used to transfer measurement data to a PC in the office. The analysis unit 2 acquires the measurement data from the measurement data collection unit 1, performs analysis processing, prediction processing, etc., and outputs the analysis results and prediction results to the prediction result display unit 3. The prediction result display unit 3 is, for example, a display device, and displays various data. The analysis unit 2 and the prediction result display unit 3 are realized by an analysis computer 20.

[0018] <Analysis computer> FIG. 3 is a schematic block diagram illustrating the functions of analytical computer 20. The analytical computer 20 has a communication unit 201 , a memory unit 202 , a natural ground displacement data acquisition unit 203 , a stress measurement data acquisition unit 204 , an analysis unit 205 , a data assimilation processing unit 206 , a control unit 207 , and an output unit 208 . The communication unit 201 is wirelessly connected to the router RT and communicates with the router RT. The storage unit 202 stores various data, for example, various measurement data acquired from the measurement computer 10. The storage unit 202 also stores physical property values ​​of a geological model that represents the physical properties of the natural ground in which the tunnel is to be constructed. The memory unit 202 is configured with a storage medium, for example, a hard disk drive (HDD), a flash memory, an electrically erasable programmable read only memory (EEPROM), a random access read / write memory (RAM), a read only memory (ROM), or any combination of these storage media. The storage unit 202 may be, for example, a non-volatile memory.

[0019] The ground displacement data acquisition unit 203 acquires ground displacement data, which is the result of measuring the displacement of the ground where the tunnel is constructed by measuring the inner surface of the concrete sprayed into the tunnel, from the measurement computer 10 via the router RT and the communication unit 201. The stress measurement data acquisition unit 204 acquires stress measurement data, which is the result of measuring the stress acting on construction materials (e.g., sprayed concrete SC, steel supports SR) installed along the inner surface of the tunnel, from the measurement computer 10 via the router RT and the communication unit 201.

[0020] The analysis unit 205 uses the physical property values ​​and a geological model that represents the geology of the natural ground, and performs analysis with an analysis model that analyzes the displacement occurring in the natural ground and the stress acting on the support. The analytical model is a model that simulates the excavation of a mountain tunnel, and is a model for calculating the displacement of the nodes of the analytical model that correspond to the positions of the observation points where observation data is obtained, the stress acting on the elements, etc. For example, the analytical model is a model in which an analytical mesh is applied to a geological model, and various parameters are assigned to each rock mass in the geological model. The geological model is, for example, a three-dimensional geological model that represents the geological structure of the natural ground when excavating a tunnel in a mountain as a construction target. The geological model may be created using the results of a geological survey of the natural ground where the tunnel as a construction target is to be excavated. The three-dimensional geological model includes, for example, models that represent a plurality of geological strata, and physical property values ​​(various parameters) are assigned to elements corresponding to each strata. The various parameters include physical property values ​​such as Young's modulus, Poisson's ratio, density, adhesion, and internal friction angle of the natural ground. Such physical property values ​​are stored in the storage unit 202.

[0021] The analysis unit 205 can perform analysis on sections that have already been constructed, but also predicts the displacement and stress in a target section that is the section to be predicted by analyzing a target section that will be excavated in the future. When making predictions, the analysis unit 205 performs predictions starting from the most recent construction section (for example, the second section described later) using physical property values ​​updated by the data assimilation processing unit 206 described later. The analysis unit 205 obtains, as analysis results, a probability distribution of a predicted displacement value obtained by predicting the displacement, and a probability distribution of a predicted stress value obtained by predicting the stress.

[0022] The data assimilation processing unit 206 updates the physical properties of the ground based on the displacement and stress obtained by analysis of the section to be analyzed by the analysis unit 205, and the ground displacement data and stress measurement data obtained by measuring the section to be analyzed. In addition, the data assimilation processing unit 206 updates the physical property values ​​of the ground stored in the memory unit 202 by performing data assimilation processing using a state vector generated based on the analysis results of the first section and the analysis results of the second section, and an observation vector generated based on the ground displacement data and stress measurement data obtained by measuring the first section and the ground displacement data and stress measurement data obtained by measuring the second section. The first section is a section constructed by carrying out a first excavation in the natural ground. The second section is a section constructed by carrying out a second excavation from the first section. Although the first section and the second section are described as being continuous sections, they do not necessarily have to be continuous sections as long as it is possible to obtain the probability distribution of the prediction results described later.

[0023] The control unit 207 controls each part of the analytical computer 20 .

[0024] The output unit 208 outputs information that enables the magnitude relationship between the displacement and stress predicted for the target section and the displacement reference value and stress reference value to be understood. The output unit 208 displays on the display screen an image in which a graph representing the probability distribution of the predicted displacement values ​​is supplemented with a figure representing the displacement reference value, or an image in which a graph representing the probability distribution of the predicted stress values ​​is supplemented with a figure representing the stress reference value.

[0025] The communication unit 201, ground displacement data acquisition unit 203, stress measurement data acquisition unit 204, analysis unit 205, data assimilation processing unit 206, control unit 207, and output unit 208 of the analysis computer 20 may be composed of a processing device such as a CPU (central processing unit) or a dedicated electronic circuit.

[0026] 《Implementation Procedure》 FIG. 4 is a flowchart explaining the operation of the underground structure prediction system S. (Processing step S1) The analysis unit 205 of the analysis computer 20 performs an excavation analysis assuming excavation of construction steps 0 to n. One construction step is a process of excavating one section from the tunnel entrance toward the extension direction and constructing shotcrete SC and steel supports SR. When the analysis unit 205 uses the ensemble Kalman filter as a data assimilation method, it performs the excavation analysis using multiple prediction models (ensembles) with varying parameters, taking into account the uncertainty of the parameters handled in the excavation analysis. The parameters mentioned here correspond to physical property values ​​such as Young's modulus, Poisson's ratio, density, adhesion, and internal friction angle of the natural ground.

[0027] (Processing step S2) For each section, excavation of the ground for a mountain tunnel and construction of support structures such as steel shoring and sprayed concrete will be carried out.

[0028] (Processing step S3) The displacement of the ground and the stress acting on the support are measured in construction steps n-1 and n. A measurement computer 10, a total station TS, and a prism P are used to measure the displacement of the ground. A measurement computer 10, a strain gauge SG, a concrete stress meter SM, and a data logger DL are used to measure the stress acting on the support. The strain gauge SG measures the strain of the steel support SR and outputs the measurement results to the data logger DL. The concrete stress meter SM measures the stress of the shotcrete. The measurement data of the displacement and the stress are stored in the measurement computer 10 and then transferred to the analysis computer 20 via the router RT.

[0029] (Processing step S4) When the analysis computer 20 receives the measurement data from the measurement computer 10, the data assimilation processing unit 206 generates a state vector from the results of the excavation analysis of construction steps n-1, n, and generates an observation vector from the measurement data of construction steps n-1, n. The state vector is composed of the physical properties of the ground, the increments of displacement of the nodes of the analytical model corresponding to the positions of the observation points that occurred from construction step n-1 to n, and the increments of stress of the elements. There are as many state vectors as there are ensembles. The observation vector is composed of the increment in ground displacement and the increment in support stress at the observation point that occurred from construction step n-1 to n. Then, the data assimilation processor 206 performs data assimilation of the state vector and the observation vector, and updates the physical property values ​​of the natural ground for each state vector. For example, it calculates the Kalman gain from the variance-covariance of the observation data and the variance-covariance calculated from the ensemble state vector, and updates the physical property values ​​of the geological model that constitutes the ensemble state vector.

[0030] (Processing step Q1) The control unit 207 of the analysis computer 20 determines whether construction of the mountain tunnel has been completed or not, and if construction has been completed (processing step Q1-YES), terminates the processing, and if construction has not been completed (processing step Q1-NO), transitions to processing step S5.

[0031] (Processing step S5) The analysis unit 205 of the analysis computer 20 assigns the updated physical properties of the ground to the pre-excavation analysis model, performs excavation analysis from construction steps 0 to n+1, and predicts the displacement of the ground and the stress acting on the support structure that will occur as construction progresses in the section in construction step n+1. Here, for example, construction has been completed up to the section corresponding to construction step n (e.g., the second section), and the construction step starting from this section, in which the mountain is excavated and shotcrete SC and steel support SR are constructed, is construction step n+1. The analysis unit 205 predicts the displacement of the ground and the stress acting on the steel support SR that will occur when construction proceeds with the section to be excavated.

[0032] Here, the numerical analysis is performed taking into account the uncertainty of the parameters, so a probability distribution is obtained as the prediction result. FIG. 5 is a diagram showing an example of a probability distribution of a prediction result. When the prediction result is obtained by the analysis unit 205, the output unit 208 outputs the prediction result by displaying it on the display screen of the analysis computer 20. For example, the output unit 208 displays a probability distribution as a prediction result on the display screen as shown in FIG. 5. The output unit 208 displays a probability distribution on a graph for each of the predicted value of the natural ground displacement, the predicted value of the stress acting on the steel support, and the predicted value of the stress acting on the sprayed concrete. More specifically, for the predicted value of the natural ground displacement, the horizontal axis of the graph represents the distance excavated from the entrance, and the vertical axis represents the predicted value of the displacement. In addition, the vertical axis displays a displacement standard value (control standard value). The displacement standard value is a value input in advance by an analysis engineer in the construction of a mountain tunnel. The horizontal axis displays a probability distribution on a graph at a position on the graph indicating the distance from the entrance according to the current position of the face. In this figure, it is shown that each value indicated by the probability distribution is distributed within a range that does not exceed the displacement reference value. In addition, the minimum and maximum predicted values ​​of the ground displacement in the section from the tunnel entrance to the face position are output. Here, at the face position, the probability distribution falls between the minimum and maximum predicted values ​​of the ground displacement. In this way, since the prediction results are displayed on the screen, they are easy to interpret; even if an analysis engineer is not present, the prediction results can be interpreted using the displacement reference value, probability distribution, minimum and maximum predicted values, etc.

[0033] The predicted values ​​of stress acting on steel supports are roughly the same as the predicted values ​​of ground displacement, except that the vertical axis represents the predicted values ​​of stress. In this diagram, the probability distribution, which is the predicted value of stress acting on steel supports, exceeds the stress standard value (control standard value) set in advance by the analysis engineer. In this case, the person in charge (analysis engineer, designer, etc.) who checks the displayed contents can consider the need to modify the tunnel design.

[0034] The predicted values ​​of stress acting on the shotcrete are roughly the same as the predicted values ​​of natural ground displacement, except that the vertical axis represents the predicted values ​​of stress acting on the shotcrete. In this figure, some of the probability distributions, which are the predicted values ​​of stress acting on the shotcrete, exceed the stress standard value (control standard value) set in advance by the analysis engineer. In this case, the person in charge (analysis engineer, designer, etc.) who checks the displayed contents can consider the need to modify the tunnel design.

[0035] In addition, the predicted ground displacement, the predicted stress acting on the steel supports, and the predicted stress acting on the sprayed concrete are displayed side by side on the screen, so that the relationship between these three predicted results can be taken into account when examining the need to modify the tunnel design.

[0036] (Processing step Q2) After processing step S5, the control unit 207 of the analysis computer 20 judges whether or not the design and construction method need to be reviewed based on the results of the predictive analysis. For example, the control unit 207 makes a judgment based on whether the ground displacement and shoring stress indicated by the prediction results of an arbitrary number of ensembles exceed the control reference value, whether the average value of the ensemble prediction results exceeds the control reference value, etc. If the ground displacement or shoring stress indicated by the ensemble prediction results exceeds the control reference value, or if the average value of the ensemble prediction results exceeds the control reference value, the control unit 207 judges that the design and construction method need to be reviewed (processing step Q2-YES), and shifts the processing to processing step S6. On the other hand, if the ground displacement or shoring stress indicated by the ensemble prediction results does not exceed the control reference value, or if the average value of the ensemble prediction results does not exceed the control reference value, the control unit 207 judges that the design and construction method need not be reviewed (processing step Q2-NO), and shifts the processing to processing step S2.

[0037] (Processing step S6) The analysis section 205 of the analysis computer 20 uses the physical properties of the natural ground updated by data assimilation to perform predictive analysis when the support pattern and construction method are changed.

[0038] (Processing step S7) The analyst modifies the design and construction method based on the results of the predictive analysis executed in processing step S6. For example, the analyst considers and judges the content of the changes based on criteria such as whether the ground displacement indicated by the prediction results of an arbitrary number of ensembles exceeds the control standard value, whether the support stress indicated by the prediction results of an arbitrary number of ensembles exceeds the control standard value, whether the average value of the ensemble prediction results exceeds the control standard value, whether the design is economical, and whether the construction method is economical. After the design and construction method are reviewed by the analysis engineer, the control unit 207 shifts the process to processing step S2. Then, construction is carried out according to the design and construction method after the design and construction method have been changed.

[0039] Thereafter, if the data assimilation processing unit 206 determines in processing step Q2 that no review of the design and construction methods is necessary (if no review is required), for subsequent data assimilation, it uses the observation vector created from the measurement data obtained in processing step S3 and the state vector created from the analysis results obtained in processing step S5. In addition, if the data assimilation processing unit 206 determines in processing step Q2 that the design and construction methods need to be revised (if a revision has been made), then for subsequent data assimilation it uses the observation vector created from the measurement data obtained in processing step S3 and the state vector created from the analysis results obtained in processing step S6.

[0040] <effect> According to the above-mentioned embodiment, data such as ground displacement and support stress observed on-site and data on ground displacement and support stress predicted by direct analysis under a plurality of different conditions can be assimilated to update and estimate the physical properties of the ground in the analysis model used in the direct analysis. This makes it possible to obtain a probability distribution of the predicted results by performing numerical analysis taking into account the uncertainty of parameters, making it easier to interpret the predicted results and allowing the predicted results to be judged even when the engineer in charge of analysis is not present.

[0041] In addition, when performing numerical analysis to change the design and construction methods in the construction of mountain tunnels, (a) the analysis engineers need time and effort to set the analysis conditions and evaluate the analysis results, and (b) if construction is interrupted to modify the design or construction method, it leads to delays in the process. In response to this problem, the above-mentioned underground structure prediction system S can provide the following effects. Even if an analyst is not present, data assimilation makes it possible to estimate analytical conditions (physical properties of the ground, initial stress, etc.). Furthermore, since a probability distribution is obtained as the analysis result, the analysis result can be easily evaluated even if an analysis engineer is not present. In addition, the analytical model can be updated using statistical mathematical techniques, which can improve the prediction performance of ground behavior, reduce the time and effort required to set analytical conditions, and enable evaluation of analysis results, thereby shortening the period during which construction is suspended.

[0042] In the above-described embodiment, an example was given of data assimilation using both the displacement of the ground and the stress acting on the support structure as measurement data, but either one of the measurement data may be used for data assimilation. In addition, in the above-mentioned embodiment, the underground structure prediction system S is described as being targeted at the construction of mountain tunnels, but the underground structure is not limited to mountain tunnels, and can also be applied to the construction of large-scale underground cavities such as underground power plants.

[0043] Furthermore, according to the above-described embodiment, an example has been described in which the displacement of the ground and the increment in stress acting on the support from construction steps n-1 to n are used for data assimilation, but this is not limited to this, and increments from any construction step may be used for data assimilation, and cumulative displacement and stress since the start of measurement may be used for data assimilation.

[0044] Basis for invention A twin experiment was conducted to simulate the construction of a tunnel by unlined excavation. In this twin experiment, it was confirmed that the Young's modulus of the ground could be identified by the excavation analysis combined with data assimilation, and the prediction performance of the ground displacement that occurs during construction was improved. The twin experiment here refers to an experimental method in which the results of calculations using an analytical model with correct conditions set are considered as measurement data, and the measurement data is sequentially assimilated into an analytical model with conditions different from the correct answer. By conducting a twin experiment, true values ​​that would normally be unknown can be found, so it can be used to verify whether or not it is possible to identify the correct conditions through data assimilation depending on the information obtained.

[0045] Procedure for twin experiments FIG. 6 is a flow chart showing the implementation procedure for conducting a twin study. (Processing step S11) The analysis unit 205 performed an excavation analysis assuming excavation of construction steps 0 to n using an ensemble that is an analytical model of the Young's modulus of the natural ground different from the correct answer. (Processing step S12) The analysis unit 205 performed an excavation analysis assuming excavation of construction steps 0 to n, using an analytical model of the correct Young's modulus of the natural ground.

[0046] (Processing step S13) In processing step S13, the data assimilation processor 206 generated a state vector from the results of the excavation analysis of construction steps n-1 and n, and generated an observation vector from the measurement data of construction steps n-1 and n. The state vector is composed of the physical properties of the ground and the increments of displacement of the nodes of the analytical model corresponding to the positions of the observation points that occurred from construction steps n-1 to n. There are as many state vectors as there are ensembles. The observation vector is composed of the increments of displacement of the ground at the observation points that occurred from construction steps n-1 to n. Next, the data assimilation processor 206 updated the Young's modulus of the natural ground of the state vector by data assimilation of the state vector and the observation vector.

[0047] (Processing step Q11) After processing step S13, a determination is made as to whether construction of the tunnel in the section that is the subject of the twin experiment has been completed. If construction has been completed (processing step Q11-YES), the processing is terminated. If construction has not been completed (processing step Q11-NO), the processing proceeds to processing step S14.

[0048] (Processing step S14) The Young's modulus of the ground updated by the data assimilation process in processing step S13 was applied to the analysis model before excavation, and excavation analysis was performed from construction steps 0 to n+1 to predict the ground displacement caused by excavation. After performing process step S14, the process proceeds to process step S12, where the analytical model of the correct Young's modulus of the natural ground is used to excavate the section corresponding to construction step n+1, and then the process proceeds to process step S13. In the data assimilation in process step S13, the observation vector created from the analysis results obtained in process step S12 and the state vector created from the analysis results obtained in process step S14 were used.

[0049] Analysis model A twin experiment was conducted using an analytical model simulating the excavation of an unlined tunnel with a circular cross section, as shown in Figure 7 . The commercial software FLAC3D was used for the numerical analysis. The ground was assumed to be homogeneous, and the constitutive law was linear elasticity. The tunnel diameter D was 10m, and the analysis domain was set to 1 / 4 of the area in consideration of symmetry. The analysis domain was 25m in the tunnel transverse direction (x), 30m in the tunnel longitudinal direction (y), and 25m in the vertical direction (z), with the analysis model having approximately 800 nodes and 700 elements.

[0050] The correct Young's modulus E of the natural ground was set to 50 MPa. The possible range of E in the prediction model (ensemble) was set to 47.25≦E≦57.75 based on the statistics in Reference 1 below, and uniform random numbers were generated and applied to the analysis model. Reference 1: Umetaro Yamaguchi, Relationship between uniaxial compressive strength and Young's modulus of rocks, Zairyo, Vol. 17, No. 181, pp. 902-907, 1968

[0051] For both the correct model and the predictive model, displacement observation points were set up at the top and side walls of the tunnel cross section every 6 m from TD (Tunnel Distance) 0 m, and relative displacements were calculated from the absolute displacements in the three components of the x, y, and z axes at each observation point before and after 2 m of excavation, and used for data assimilation. In this twin experiment, data assimilation was performed every time the face was excavated by 2 m, the Young's modulus of the ground was updated, and the ground displacement that would occur when an unexcavated area was excavated was predicted.

[0052] Results of the twin experiment Figure 8 is a graph showing the Young's modulus of the ground before (prior distribution) and after (posterior distribution) updating by data assimilation, and Figure 9 shows the correct value of the displacement in the z-axis direction at the observation point at the top of the tunnel with a TD (Tunnel Distance) of 0 m and the ensemble prediction results.

[0053] The vertical axis of Figure 8 is the Young's modulus of the rock mass, and the horizontal axis is the TD of the face. To improve the visibility of the symbols, the symbols are arranged so that the prior distribution and the posterior distribution do not overlap. The vertical axis of Figure 9 is the z-axis displacement of the observation point at the top of the tunnel at TD0m, and the horizontal axis is the TD of the tunnel face. The graph on the right side of Figure 9 is the same scale as the graph on the left, and shows the frequency distribution of the z-axis displacement when the tunnel face is located at TD14m.

[0054] As shown in Figure 8, with data assimilation, the posterior distribution of E becomes less variable relative to the prior distribution as the face position advances. Also, in Figure 9, the variation in the displacement prediction results after data assimilation becomes smaller compared to the displacement prediction results before data assimilation. These results confirmed that in twin experiments using simulated observational values, data assimilation improved the prediction performance of ground displacements that occur when excavating untouched areas.

[0055] Comparison with other technologies 1 Reference 2 below reports a method of back analysis used during tunnel construction, but this is a method for estimating the deformation coefficient and stress ratio of the ground from the displacement of the ground during tunnel construction, and it is not possible to obtain each piece of information independently. Reference 2: Shunsuke Sakurai and Kunifumi Takeuchi, "Back analysis method of displacement measurement results during tunnel excavation," Proceedings of the Japan Society of Civil Engineers, 1983 On the other hand, according to this embodiment, the physical properties of the ground can be estimated from the displacement of the ground during tunnel construction.

[0056] <Comparison with other technologies 2> Examples of prior art that utilize ground displacement measurement in measurement work include a method for predicting the final displacement amount based on the displacement speed obtained from displacement measurement in mountain tunnel construction (Patent Publication No. 2021-188994), a method and device for predicting the deformation suppression effect of countermeasure work (Patent Publication No. 2020-200648), and a method and system for setting parameters to be used in creep analysis from the time-series changes in borehole diameter displacement and performing creep analysis (Patent Publication No. 2020-66843). These technologies are deterministic methods that predict the future from the measured values ​​obtained, and the reliability of the prediction results cannot be presented. On the other hand, according to the present embodiment, the system utilizes data assimilation, which is a probabilistic method, and handles uncertainties in the physical properties of the ground, geological structures, analytical models, etc., and therefore has the advantage that reliable prediction results are output, making it easy to evaluate the prediction results.

[0057] The functions of the analysis computer 20 in the above-mentioned embodiment may be realized by recording a program for realizing the functions of the analysis computer 20 in a computer-readable recording medium, reading the program recorded in the recording medium into a computer system, and executing the program. The term "computer system" as used herein includes hardware such as an OS and peripheral devices. The term "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into a computer system. The term "computer-readable recording medium" may also include a medium that dynamically holds a program for a short period of time, such as a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, and a medium that holds a program for a certain period of time, such as a volatile memory inside a computer system that is a server or client in such a case. The above-mentioned program may be a program for realizing a part of the above-mentioned functions, or may be a program that can realize the above-mentioned functions in combination with a program already recorded in the computer system, or may be a program that is realized using a programmable logic device such as an FPGA (Field Programmable Gate Array).

[0058] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included. [Explanation of symbols]

[0059] 1...measurement data collection unit, 2...analysis unit, 3...prediction result display unit, 10...measurement computer, 20...analysis computer, 201...communication unit, 202...storage unit, 203...ground displacement data acquisition unit, 204...stress measurement data acquisition unit, 205...analysis unit, 206...data assimilation processing unit, 207...control unit, 208...output unit, S...underground structure prediction system

Claims

1. a ground displacement data acquisition unit that acquires ground displacement data obtained by measuring an inner circumferential surface of the concrete sprayed into the tunnel to measure the displacement of the ground in which the tunnel is constructed; a stress measurement data acquisition unit that acquires stress measurement data that is a result of measuring stress acting on a construction provided along an inner circumferential surface of the tunnel; A storage unit that stores physical property values ​​representing physical properties of the natural ground; an analysis unit that performs analysis using an analysis model that uses the physical property values ​​and a geological model that represents the geology of the natural ground to analyze a displacement occurring in the natural ground and a stress acting on a support; a data assimilation processing unit that updates physical property values ​​of the natural ground based on the displacement and the stress obtained by analyzing the section to be analyzed by the analysis unit and the natural ground displacement data and the stress measurement data obtained by measuring the section to be analyzed. An underground structure prediction system having the above-mentioned function.

2. A first section constructed by performing a first excavation on the natural ground, and a second section constructed by performing a second excavation from the first section, The data assimilation processing unit includes: a state vector generated based on an analysis result of the first section and an analysis result of the second section; an observation vector generated based on the natural ground displacement data and the stress measurement data obtained by measuring the first section and the natural ground displacement data and the stress measurement data obtained by measuring the second section; By using this data assimilation process, the physical properties of the ground are updated. The underground structure prediction system according to claim 1 .

3. The analysis unit is Using the updated physical property values, an analysis is performed on a target section to be excavated from the second section as a starting point, thereby predicting displacement and stress in the target section; The underground structure prediction system comprises: An output unit that outputs information that enables the magnitude relationship between the displacement and stress predicted for the target section and the displacement reference value and the stress reference value to be understood. The underground structure prediction system according to claim 1 or claim 2, further comprising:

4. The analysis unit is As the analysis results, a probability distribution of a predicted displacement value and a probability distribution of a predicted stress value are obtained. The underground structure prediction system according to claim 3 .

5. The output unit is An image in which a figure representing the displacement reference value is added to a graph representing a probability distribution of the displacement prediction value, or an image in which a figure representing the stress reference value is added to a graph representing a probability distribution of the stress prediction value is displayed on a display screen. The underground structure prediction system according to claim 4 .

6. On the computer, a ground displacement measurement data acquisition step of acquiring ground displacement data which is a result of measuring the displacement of the ground in which the tunnel is constructed by measuring an inner peripheral surface of the concrete sprayed into the tunnel; a stress measurement data acquisition step of acquiring stress measurement data which is a result of measuring stress acting on a construction provided along an inner circumferential surface of the tunnel; an analysis step of performing an analysis using an analysis model that analyzes a displacement occurring in the ground and a stress acting on a support, using the physical property values ​​stored in a storage unit that stores the physical property values ​​representing the physical properties of the ground and a geological model that represents the geology of the ground; a data assimilation processing step for updating physical properties of the natural ground based on the displacement and the stress obtained by analyzing the analysis target section in the analysis step and the natural ground displacement data and the stress measurement data obtained by measuring the analysis target section; A method for predicting underground structures.

Images