A roadway roof displacement field advance prediction method, device and medium

Abstract

The present application relates to underground mining technical field, disclose a kind of roadway roof displacement field advance prediction method, equipment and medium, comprising: obtaining the measurement data of target roadway survey area, according to measurement data constructs the roadway numerical model of target roadway survey area;Stress simulation data is obtained by simulating and solving to roadway numerical model;Stress simulation data is used to train the model training of pre-constructed back propagation neural network, and displacement field prediction model is obtained;Displacement field prediction model is used to carry out displacement field prediction to the point data to be predicted, and displacement field prediction result is obtained.The present application can realize the nonlinear mapping from mining condition and geological parameter to displacement field, based on the inversion process of pure data driving, significantly improve the efficiency and adaptability of displacement field prediction.

Description

Technical Field

[0001] This invention relates to the field of mining technology, specifically to a method, equipment, and medium for predicting the displacement field of roadway roof. Background Technology

[0002] In coal mining and underground roadway engineering, accurate prediction of roof displacement is crucial for roadway stability evaluation, support design, and disaster prevention. In particular, roof deformation caused by the pre-support pressure during face advancement directly affects roadway safety and the effectiveness of surrounding rock control. Currently, research methods for roadway roof displacement mainly include field monitoring, numerical simulation, and empirical formulas. While these methods can monitor or predict roof displacement under specific conditions, they still have the following limitations: On-site monitoring methods are costly and have limited coverage, making it difficult to fully reflect the displacement field distribution; empirical formulas and regression methods are difficult to characterize the nonlinear dynamic process of roof displacement; numerical simulation methods rely on repeated manual parameter adjustments, which are inefficient and highly subjective; traditional neural networks are prone to getting trapped in local extrema when inverting complex displacement fields, and their generalization ability is insufficient.

[0003] Therefore, there is an urgent need for a method that can efficiently, accurately, and adaptively invert the displacement field of the roadway roof and predict the impact of advance support pressure. Summary of the Invention

[0004] The technical problem to be solved by this invention is how to efficiently and accurately invert the displacement field of the roadway roof and predict the influence of advance support pressure.

[0005] The present invention solves the above-mentioned technical problems through the following technical means: This invention provides a method for predicting the displacement field of roadway roof, characterized in that it includes: Acquire measurement data of the target roadway measurement area, and construct a roadway numerical model of the target roadway measurement area based on the measurement data; The roadway numerical model was subjected to working condition simulation to obtain stress simulation data; The pre-constructed backpropagation neural network is trained using the stress simulation data to obtain a displacement field prediction model. The displacement field prediction model is used to predict the displacement field of the data at the point to be predicted, and the displacement field prediction result is obtained.

[0006] The present invention also provides a processing device, characterized in that it includes at least one processor and at least one memory communicatively connected to the processor, wherein: the memory stores program instructions executable by the processor, and the processor can execute the above-described method for predicting the displacement field of the roadway roof by calling the program instructions.

[0007] The present invention also provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions, the computer instructions causing the computer to execute the above-described method for predicting the displacement field of the roadway roof.

[0008] The advantages of this invention are: This invention solves the problem through working condition simulation, which can consider various stress boundary forms and construct stress simulation data that comprehensively covers actual engineering conditions. By using the stress simulation data to train the backpropagation neural network model, a nonlinear mapping from mining conditions and geological parameters to the displacement field can be achieved. Based on a pure data-driven inversion process, it avoids manual intervention in traditional inversion methods and significantly improves the efficiency and adaptability of displacement field prediction. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating a method for predicting the displacement field of the roadway roof in advance, according to one embodiment of the present invention. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0011] Reference Figure 1 The diagram shown is a flowchart illustrating a method for predicting the displacement field of roadway roof according to an embodiment of the present invention. In this embodiment, the method for predicting the displacement field of roadway roof includes: S1. Obtain measurement data of the target roadway measurement area, and construct a roadway numerical model of the target roadway measurement area based on the measurement data.

[0012] In this embodiment of the invention, the measurement data consists of the coordinates of the measuring points determined by the actual displacement provided by the measurement of the target roadway measuring area, as well as the important parameters obtained by the measurement. Based on the measurement data, FLAC3D can be used to simulate and model the target roadway measuring area, and the relevant parameters can be assigned to the model.

[0013] Specifically, FLAC3D generalizes and divides the target tunnel survey area into zones based on the mechanical properties of each rock stratum and the interbedded lithology. Based on the inversion results obtained after multiple model builds, it is determined that the pre-defined measurement point locations must be within the effective range of the model during model building.

[0014] S2. Perform working condition simulation on the numerical model of the tunnel to obtain stress simulation data.

[0015] In this embodiment of the invention, the working condition simulation solution involves reasonably setting the rock strata mechanical parameters and boundary constraints, constructing a roof displacement field database containing multiple working conditions, simulating the roof deformation response under different mining and support conditions, and solving the stress using FLAC3D to finally obtain stress simulation data under multiple working conditions.

[0016] Specifically, the step of performing a working condition simulation solution on the roadway numerical model to obtain stress simulation data includes: The roadway numerical model is simulated using preset simulated working conditions loads to obtain a simulated numerical model. The stress is solved by the simulated numerical model to obtain stress simulation data.

[0017] Specifically, the simulated working load involves setting different boundary load conditions on the roadway data model to simulate the roof deformation response under different mining and support conditions, and setting multiple different boundary stress forms for the boundary load.

[0018] For example, based on the distribution of measuring points, it's possible to determine whether the working conditions are decreasing or increasing in the x, y, and z directions. Using FLAC3D, uniform loads, triangular and trapezoidal loads along the x-direction, and trapezoidal loads along the y-direction are applied to the z-plane of the mining area model, along with normal stress. On the x-plane, uniform loads and trapezoidal loads along the y and z directions are applied, along with normal stress. On the y-plane, uniform loads and trapezoidal loads along the x and z directions are applied, along with normal stress. For the shearing motion in the xy, xz, and yz planes, shear stress is applied to each side with a shear boundary displacement of 600 mm. The displacement loads are controlled by three types of velocity boundary conditions: uniform, triangular, and trapezoidal, with a boundary velocity of 3e-4 and a running step of 2000 steps. Therefore, 27 working conditions can be obtained. Stress is solved using FLAC3D, ultimately yielding all stress simulation data for these 27 working conditions.

[0019] S3. Using the stress simulation data, train the pre-constructed backpropagation neural network to obtain the displacement field prediction model.

[0020] In this embodiment of the invention, the pre-constructed backpropagation neural network is a BP neural network model, consisting of an input layer, hidden layers, and an output layer. The hidden layer has the following node count determined through trial and error, and the activation function is tansig, expressed as: Output layer: Top plate displacement value, activation function is purelin. The expression for purelin is: .

[0021] Specifically, the step of training a pre-constructed backpropagation neural network using the stress simulation data to obtain a displacement field prediction model includes: The stress simulation data is normalized to obtain normalized data; The normalized data is predicted using the backpropagation neural network to obtain the displacement field prediction result; The mean square error of the backpropagation neural network is calculated based on the displacement field prediction results. The model parameters of the backpropagation neural network are optimized based on the mean square error to obtain the displacement field prediction model.

[0022] In detail, the function for calculating the mean square error can be expressed as:

[0023] Furthermore, the normalized data can be divided into training and testing samples in a ratio of 4:1. The training samples can be used to optimize the model parameters, and the testing samples can be used to verify the generalization ability of the displacement field prediction model.

[0024] S4. Use the displacement field prediction model to predict the displacement field of the point data to be predicted, and obtain the displacement field prediction result.

[0025] In this embodiment of the invention, the data of the points to be predicted are imported into MATLAB, and the displacement field prediction model can be used to predict the displacement field, while also predicting the influence of the advance support pressure.

[0026] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting the displacement field of roadway roof, characterized in that, include: Acquire measurement data of the target roadway measurement area, and construct a roadway numerical model of the target roadway measurement area based on the measurement data; The roadway numerical model was subjected to working condition simulation to obtain stress simulation data; The pre-constructed backpropagation neural network is trained using the stress simulation data to obtain a displacement field prediction model. The displacement field prediction model is used to predict the displacement field of the data at the point to be predicted, and the displacement field prediction result is obtained.

2. The method for predicting the displacement field of the roadway roof as described in claim 1, characterized in that, The process of performing a working condition simulation on the numerical model of the roadway to obtain stress simulation data includes: The roadway numerical model is simulated using preset simulated working conditions loads to obtain a simulated numerical model. The stress is solved by the simulated numerical model to obtain stress simulation data.

3. The method for predicting the displacement field of the roadway roof as described in claim 1, characterized in that, The step of training a pre-constructed backpropagation neural network using the stress simulation data to obtain a displacement field prediction model includes: The stress simulation data is normalized to obtain normalized data; The normalized data is predicted using the backpropagation neural network to obtain the displacement field prediction result; The mean square error of the backpropagation neural network is calculated based on the displacement field prediction results. The model parameters of the backpropagation neural network are optimized based on the mean square error to obtain the displacement field prediction model.

4. A processing device, characterized in that, The method includes at least one processor and at least one memory communicatively connected to the processor, wherein the memory stores program instructions executable by the processor, and the processor can execute the method as described in any one of claims 1 to 3 by invoking the program instructions.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause the computer to perform the method as described in any one of claims 1 to 3.

Images