Foaming slurry grouting effect prediction method and system based on PSO-BP

Abstract

The invention discloses a method and system for predicting the grouting effect of foaming type grout based on PSO-BP, and belongs to the field of grouting repair engineering.The method comprises the following steps that bedrock fracture characteristics are collected to obtain fracture opening, fracture roughness and hydrodynamic pressure, meanwhile, grouting influence factor data are obtained, the grouting influence factor data comprise the grouting amount, the grout component proportion and the grouting pressure, and the grouting influence factor data comprise the grouting amount, the grout component proportion and the grouting pressure; the method comprises the following steps: dividing fracture opening, fracture roughness, hydrodynamic pressure and grouting influence factor data into a training set and a test set by using a PSO-BP neural network model, generating a grouting diffusion distance prediction model and a grouting water plugging rate prediction model, predicting the test set through the generated models, obtaining predicted values of the grouting diffusion distance and the grouting water plugging rate, and predicting the grouting diffusion distance and the grouting water plugging rate according to the predicted values of the grouting diffusion distance and the grouting water plugging rate. And according to the two groups of predicted values and measured values, calculating to obtain a mean square error, a mean absolute error and a decision coefficient corresponding to each group to evaluate the prediction effect of the slurry grouting effect. According to the method, accurate prediction of the grouting repair effect of the foaming slurry is realized through the PSO-BP neural network.

Description

Technical Field

[0001] The present invention discloses a prediction method and system for the grouting effect of foamed slurry based on PSO-BP, belonging to the technical field of underground disease grouting repair. Background Art

[0002] With the rapid progress of China's infrastructure construction, there are more and more underground projects that are aging, damaged and in need of repair during service. In underground mining and tunnel engineering, sudden disasters such as water inrush, gushing water and collapse occur from time to time. Statistics show that more than 90% of the water inrush accidents in underground projects are related to the cracks in the foundation rock mass. The groundwater scours the cracks in the bedrock, resulting in underground cavities, which in turn affects the change of the main strength of the engineering structure. In the repair of crack diseases in underground projects, grouting is one of the effective methods to improve the mechanical properties of rock masses and block groundwater.

[0003] In recent years, many new grouting materials and equipment have been put into use, greatly improving the grouting sealing effect. Among them, foamed polymer materials have become grouting materials with better comprehensive performance due to their advantages such as safety, environmental protection, fast response, high expansion rate, impermeability and durability, and are widely used in foundation reinforcement, dam leakage prevention, road maintenance and other aspects. However, in engineering practice, due to the lack of scientific theoretical and technical guidance, the grouting parameters of foamed polymer slurry are mostly determined by on-site experience, which will lead to great uncertainty in the treatment effect of engineering diseases. For example, too high grouting pressure will cause the displacement and loosening of the foundation of the underground engineering structure, while too low grouting pressure will result in the failure to reach the design strength and water blocking rate in the area to be treated; the component ratio of the foamed polymer and the grouting pressure have an impact on the diffusion distance of the slurry in the crack network, which will not only affect the reinforcement and water blocking effects in the area to be treated, but also affect the construction period and the cost of disease repair. All these seriously hinder the popularization and application of the foamed polymer grouting technology in the treatment of highway diseases and the standardization of the process flow. Summary of the Invention

[0004] The technical problem solved by the present invention is: aiming at the problem that it is difficult to accurately predict the grouting effect in the existing grouting reinforcement repair for underground engineering crack treatment, a prediction method and system for the grouting effect of foamed slurry based on PSO-BP are provided.

[0005] The present invention is realized by adopting the following technical solutions:

[0006] The present invention first discloses a prediction method for the grouting effect of foamed slurry based on PSO-BP, including the following steps:

[0007] S1. Set multiple groups of fissure grouting test groups, obtain the fissure aperture, fissure roughness, and hydrodynamic pressure of the bedrock where different grouting test groups are located, as well as the grouting influencing factor data of each grouting test group. The grouting influencing factor data includes the grouting volume, slurry component ratio, and grouting pressure. Map the numbers of each grouting test group and the grouting influencing factor data one by one to form the first data set;

[0008] S2. Obtain the measured parameters of the grouting test effect after grouting for each grouting test group. The measured parameters of the grouting test effect include the measured value of the diffusion distance of the grouting slurry and the measured value of the water blocking rate;

[0009] S3. Select a BP neural network to construct a grouting effect prediction model, and optimize the initial parameters of the BP neural network through the PSO algorithm;

[0010] S4. The predicted values of the grouting effect include the predicted value of the diffusion distance and the predicted value of the water blocking rate. Select the fissure aperture and fissure roughness of the bedrock where the grouting test group is located, combine the first data set and the measured value of the diffusion distance of the grouting slurry to construct the first training set and the first test set. Input the first training set into the grouting effect prediction model for model training, and evaluate the grouting effect prediction model through the first test set to obtain the grouting diffusion distance prediction model; Select the hydrodynamic pressure of the bedrock where the grouting test group is located, combine the first data set and the measured value of the water blocking rate to construct the second training set and the second test set. Input the second training set into the grouting effect prediction model for model training, and evaluate the grouting effect prediction model through the second test set to obtain the grouting water blocking rate prediction model;

[0011] S5. Use the grouting diffusion distance prediction model and the grouting water blocking rate prediction model to predict the actual grouting effect of the foamed slurry.

[0012] In the method for predicting the grouting effect of the foamed slurry based on PSO - BP of the present invention, further, the step S3 includes the following sub - steps:

[0013] S31. Initialize the BP neural network, determine the number of nodes in the input layer, hidden layer, and output layer of the BP neural network, and randomly initialize the weights and bias values of the BP neural network;

[0014] S32. Optimize the parameters of the BP neural network by the PSO algorithm. Take the weights and bias values of the BP neural network as the particles in the PSO algorithm, and update the velocity and position of each particle through the following formula to obtain the individual optimal position and global optimal position of the particles;

[0015]

[0016] In the formula, ω is the inertia weight, c 1 c 2 is the acceleration factor, pbest,i is the individual optimal position of particle i, g best is the global optimal position of any particle in the particle swarm, v i k and are the particle velocities of particle i at the current time k and the next time k + 1 respectively, x i k and x i k+1 are the positions of particle i at the current time k and the next time k + 1 respectively, rand 1 function, rand 2 function generates random numbers between (0, 1), and the optimal parameter combination of the BP neural network is obtained through iteration;

[0017] S33. Use the optimal parameter combination optimized by the PSO algorithm as the initial parameter for training the grouting effect prediction model.

[0018] In the method for predicting the grouting effect of foaming slurry based on PSO - BP of the present invention, further, during the iteration process of the PSO algorithm in step S32, the mean square error is used as the fitness function, and the fitness value of the BP neural network is calculated by selecting the first test set or the second test set through the following formula:

[0019]

[0020] where n in the formula is the number of test samples in the first test set or the second test set, y i is the measured value corresponding to the first test set or the second test set, is the corresponding predicted value output by the BP neural network;

[0021] Iterate until the fitness function converges to obtain the optimal parameter combination of the BP neural network.

[0022] In the method for predicting the grouting effect of foaming slurry based on PSO - BP of the present invention, further, the step S4 of setting the training set and the test set includes selecting N grouting test groups from M grouting test groups to obtain the training set, and the remaining M - N grouting test groups to obtain the test set; where,

[0023] The input variables of the first training set and the first test set include the fracture aperture, fracture roughness, grouting volume, slurry component ratio, and grouting pressure of the bedrock where the corresponding grouting test group is located, and the target output is the predicted value of the grouting slurry diffusion distance of the corresponding grouting test group;

[0024] The input variables of the second training set and the second test set include the hydrodynamic pressure, grouting volume, slurry component ratio, and grouting pressure of the bedrock where the grouting test group is located, and the target output is the predicted value of the grouting water blocking rate of the corresponding grouting test group.

[0025] In the foamed slurry grouting effect prediction method based on PSO-BP of the present invention, further, input variables are selected from the first test set and input into the grouting effect prediction model, the predicted value of the diffusion distance of grouting output by the model is compared with the measured value of the diffusion distance of the grouting slurry, and the error evaluation indexes including mean square error, mean absolute error and determination coefficient are used to evaluate the performance of the model. According to the error evaluation indexes, the parameters of the PSO algorithm are adjusted or the structure of the BP neural network model is modified, and the grouting diffusion distance prediction model is obtained through training.

[0026] In the foamed slurry grouting effect prediction method based on PSO-BP of the present invention, further, input variables are selected from the second test set and input into the grouting effect prediction model, the predicted value of the water blocking rate of grouting output by the model is compared with the measured value of the water blocking rate, and the error evaluation indexes including mean square error, mean absolute error and determination coefficient are used to evaluate the performance of the model. According to the error evaluation indexes, the parameters of the PSO algorithm are adjusted or the structure of the BP neural network model is modified, and the grouting water blocking rate prediction model is obtained through training.

[0027] In the foamed slurry grouting effect prediction method based on PSO-BP of the present invention, further, in step S4, the predicted diffusion distance value generated by the grouting diffusion distance prediction model for the first test set and the predicted water blocking rate value generated by the grouting water blocking rate prediction model for the second test set are obtained, and the mean square error, mean absolute error and determination coefficient between the predicted values and the measured parameters of the grouting test effect are calculated respectively. The comprehensive grouting performance evaluation coefficient is obtained through the following formula,

[0028]

[0029] where Y is the comprehensive grouting performance evaluation coefficient, MSE D is the mean square error between the predicted diffusion distance value and the measured value, MSE B is the mean square error between the predicted water blocking rate value and the measured value, MAE D is the mean absolute error between the predicted diffusion distance value and the measured value, MAE B is the mean absolute error between the predicted water blocking rate value and the measured value, is the determination coefficient between the predicted diffusion distance value and the measured value, is the determination coefficient between the predicted water blocking rate value and the measured value, and a, b, c, d, e, f are weight coefficients, satisfying a + b + c + d + e + f = 1.

[0030] The calibration error threshold is Y threshold , and it is compared with the comprehensive grouting performance evaluation coefficient Y,

[0031] If Y ≤ Y threshold, it is determined that the accuracy rate of the predicted value meets the accuracy requirement, and the grouting diffusion distance prediction model and the grouting water plugging rate prediction model are used as the prediction references for the grouting effect in the actual grouting process;

[0032] If Y > Y threshold , it is determined that the accuracy rate of the predicted value does not meet the accuracy requirement, and the training set test samples of the grouting diffusion distance prediction model and the grouting water plugging rate prediction model are continuously increased until the accuracy rate of the predicted value meets the accuracy requirement.

[0033] The present invention also discloses a prediction system for the grouting effect of foaming slurry based on PSO-BP, which is used to execute the above-mentioned prediction method for the grouting effect of foaming slurry of the present invention, including

[0034] A bedrock sample collection module, which obtains the fracture aperture, fracture roughness and hydrodynamic pressure of the bedrock where different grouting test groups are located;

[0035] A grouting influencing factor collection module, which obtains the grouting influencing factor data of each grouting test group. The grouting influencing factor data includes the grouting volume, the proportion of slurry components and the grouting pressure, and maps the numbers of each grouting test group and the grouting influencing factor data one by one to form a first data set;

[0036] A test data collection module, which obtains the measured parameters of the grouting test effect after grouting for each grouting test group. The measured parameters of the grouting test effect include the measured value of the grouting slurry diffusion distance and the measured value of the water plugging rate;

[0037] A grouting diffusion distance prediction model generation module, which selects a BP neural network optimized by the PSO algorithm to construct a grouting effect prediction model, selects the fracture aperture and fracture roughness of the bedrock where the grouting test group is located, combines the first data set and the measured value of the grouting slurry diffusion distance to construct a first training set and a first test set, inputs the first training set into the grouting effect prediction model for model training, and evaluates the grouting effect prediction model through the first test set to obtain the grouting diffusion distance prediction model;

[0038] A grouting water plugging rate prediction model generation module, which selects a BP neural network optimized by the PSO algorithm to construct a grouting effect prediction model, selects the hydrodynamic pressure of the bedrock where the grouting test group is located, combines the first data set and the measured value of the water plugging rate to construct a second training set and a second test set, inputs the second training set into the grouting effect prediction model for model training, and evaluates the grouting effect prediction model through the second test set to obtain the grouting water plugging rate prediction model;

[0039] A prediction error calculation module, which calculates the mean square error, mean absolute error and determination coefficient between the predicted diffusion distance and the predicted water plugging rate and the measured parameters of the grouting test effect in the test set;

[0040] The prediction performance evaluation module calculates a comprehensive grouting performance evaluation coefficient based on the mean square error, mean absolute error, and coefficient of determination generated by the prediction error calculation module, and compares it with an error threshold to evaluate the prediction effect of the slurry grouting effect.

[0041] The present invention also discloses a computer device, comprising:

[0042] One or more processors;

[0043] A memory storing one or more computer programs;

[0044] Wherein, the processor calls the computer program to implement the above-mentioned foaming slurry grouting effect prediction method of the present invention.

[0045] The present invention combines a BP neural network based on particle swarm optimization (PSO) with the prediction of the grouting effect of foaming slurry for repairing bedrock fissures in underground engineering. According to the basic properties of the bedrock where the repaired fissures are located, foaming slurry is used for the grouting repair of fissure media. When predicting the grouting repair effect of foaming slurry, the fissure aperture, fissure roughness, and hydrodynamic pressure are selected as key input parameters, mainly based on the flow, foaming, and curing characteristics of the slurry. The fissure aperture directly affects the penetration ability and diffusion range of the slurry. If it is too large, it will be difficult for the slurry to fill densely, and if it is too small, it may affect the ability of the slurry to enter the fissure. The fissure roughness determines the adhesion and consolidation effect of the slurry in the fissure. Higher roughness helps to enhance the mechanical bite between the slurry and the bedrock and improve the stability of the repair. The hydrodynamic pressure affects the diffusion and curing behavior of the slurry under the action of groundwater. Excessive hydrodynamic pressure may cause the slurry to be washed away and difficult to consolidate, while a suitable pressure environment helps the slurry to stably fill the fissure. Therefore, the selection of these three parameters comprehensively considers the seepage characteristics, curing effect, and long-term stability of the slurry in the fissure, and can more accurately predict the final effect of the grouting repair.

[0046] According to the characteristics of the foaming slurry, the present invention simultaneously obtains the data of influencing factors affecting the grouting effect, including the grouting volume, the proportion of slurry components, and the grouting pressure. The data of fracture aperture, fracture roughness, grouting volume, proportion of slurry components, and grouting pressure of multiple groups of foamed polymer grouting test groups are divided into two parts of data sets as the first training set and the first test set. A grouting diffusion distance prediction model is generated through training with the first training set and the first test set. The diffusion distance prediction value after foamed slurry grouting is obtained by predicting the first test set with the generated grouting diffusion distance prediction model; using the same steps, a grouting water plugging rate prediction model is established based on the hydrodynamic pressure, grouting volume, proportion of slurry components, and grouting pressure of multiple groups of foamed polymer grouting test groups, and the water plugging rate prediction value after foamed slurry grouting is predicted. The mean square error, mean absolute error, and determination coefficient are calculated based on the two groups of prediction values and the measured parameters of the grouting effect of the corresponding grouting test groups. The comprehensive grouting performance evaluation coefficient is generated from the mean square error, mean absolute error, and determination coefficient of each group. Compared with the pre-set error threshold, if it does not exceed the error threshold, it is judged that the performance of the prediction model is good and meets the accuracy requirements; otherwise, it is judged that the performance of the prediction model does not meet the accuracy requirements. The present invention reasonably selects and combines grouting data parameters according to the grouting characteristics of foamed polymers, accurately predicts the grouting effect of foamed slurry, and improves the accuracy of prediction.

[0047] The basis for constructing the grouting effect prediction model in the present invention is the BP neural network model. This model is a multi-layer feedforward neural network trained according to the error backpropagation algorithm. The gradient descent method is adopted, and the mean square error between the actual output value (prediction value) and the expected output value (actual value) of the network model is used as the objective function. According to the magnitude of this error value, the weight threshold of the neural network is adjusted, and finally the error value is made small enough so that the actual output value of the network reaches the expected output value. In the initial stage of finding the optimal solution through the BP neural network in the present invention, the particle swarm optimization algorithm (PSO) is first used to accelerate the training speed. When the fitness function value does not change or the change value is less than the predefined value after multiple iterations, the search process is switched to gradient descent search according to heuristic knowledge. The PSO-BP neural network model used can accurately find the data relationship between the grouting parameter combination and the grouting effect in the prediction of the grouting effect of foamed slurry on bedrock fractures under hydrodynamic conditions in underground engineering, and make full use of the grouting characteristics of foamed slurry to improve the accuracy and stability of grouting effect prediction.

[0048] In summary, the prediction method and system for the grouting effect of foaming slurry based on PSO-BP provided by the present invention predict the grouting effect of foaming slurry by means of a particle swarm optimization BP neural network. According to the disease characteristics of the underground bedrock under the hydrodynamic conditions suitable for foaming slurry, a reasonable combination of grouting parameters is selected as the model input, so as to achieve a relatively accurate prediction of the grouting effect for the repair of foaming slurry grouting, providing a standard and objective prediction and evaluation for the pore grouting repair effect of underground engineering, which is beneficial to the subsequent construction and maintenance of underground engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic flow chart of the steps of the prediction method for the grouting effect of foaming slurry based on PSO-BP of the present invention.

[0050] Figure 2 It is a schematic structural diagram of the prediction system for the grouting effect of foaming slurry based on PSO-BP of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to specific embodiments.

[0052] Embodiment 1

[0053] As Figure 1 shown, the specific steps of the prediction method for the grouting effect of foaming slurry based on PSO-BP of the present invention are as follows.

[0054] S1. Set multiple groups of fracture grouting test groups, obtain the fracture aperture, fracture roughness and hydrodynamic pressure of the bedrock where different grouting test groups are located, as well as the grouting influence factor data of each grouting test group. The grouting influence factor data includes the grouting volume, the proportion of slurry components and the grouting pressure. Map the numbers of each grouting test group and the grouting influence factor data one by one to form a first data set.

[0055] Set multiple groups of grouting tests to ensure that the basic property parameters of each test group are not completely the same, and obtain the basic property parameters of the rock mass fractures selected for different groups of grouting tests. Here, the basic property parameters include the fracture aperture, fracture roughness, and hydrodynamic pressure. Suppose M groups of grouting test groups are set, and the number of the i-th grouting test group is designated as Si. Designate the fracture aperture in the Si-th test group as a, the fracture roughness as j, and the hydrodynamic pressure as p.

[0056] As set in this embodiment, 16 groups of grouting test samples are used to establish a grouting effect prediction model, and the basic property parameters of the 16 test groups are recorded as shown in Table 1.

[0057] Table 1 Basic properties of rock mass fractures used in grouting tests

[0058]

[0059]

[0060] In addition to the basic properties of the fracture medium itself, the influencing factors on the grouting effect of the foaming slurry also include the influence of the grouting volume, the influence of the slurry component ratio, and the influence of the grouting pressure. According to the grouting repair processes of 16 groups of grouting test groups, the data of the influencing factors of grouting for each grouting test group are obtained. The data of the influencing factors of grouting include the grouting volume, the slurry component ratio, and the grouting pressure, and each grouting test group and the data of the influencing factors of grouting are mapped one by one to form the first data set.

[0061] The grouting volume is an important factor determining the filling effect of bedrock fractures. A reasonable grouting volume can ensure that the slurry is evenly distributed in the fracture network, fill the target area, and form a dense consolidated body. If the grouting volume is insufficient, it may lead to incomplete filling of the fractures and reduce the plugging effect; while if the grouting volume is too large, it will not only cause material waste, but may also cause the slurry to overflow to non-target areas, and even lead to fracture expansion or disturbance to the surrounding structure of the bedrock.

[0062] The foaming polymer slurry generally adopts a two-component slurry configured with a main agent and a foaming agent. The two-component ratio of the grouting slurry directly determines the foaming effect, curing speed, and consolidated body performance of the slurry. When the slurry component ratio is appropriate, the slurry can fully foam and fill the fractures, forming a dense and uniform consolidated structure, improving the compressive strength and durability. If the proportion of the main agent is too high and the foaming is insufficient, it will lead to poor filling effect; while if the proportion of the foaming agent is too high, it may cause excessive foaming, forming a loose structure and reducing the strength of the consolidated body. Therefore, the two-component ratio should be precisely controlled according to the target fracture conditions.

[0063] The grouting pressure is a key parameter affecting the slurry diffusion range and plugging efficiency. An appropriate grouting pressure can push the slurry to diffuse to the deep areas of the fractures and achieve a uniform plugging effect; however, too low a pressure will limit the slurry diffusion range and cause incomplete filling of the deep fractures; while too high a pressure may cause fracture expansion or material overflow, resulting in poor reinforcement effect in the target area and material waste.

[0064] For the 16 groups of grouting test groups set in this embodiment, the grouting volume, the slurry component ratio, and the grouting pressure in each test group are recorded to generate the first data set, as shown in Table 2.

[0065] Table 2 The First Data Set

[0066]

[0067]

[0068] The data in the first data set is used to combine with the basic property data of bedrock fissures collected by the grouting test groups to complete the establishment of the input variables of the grouting effect prediction model.

[0069] S2. Obtain the measured parameters of the grouting test effect after grouting for each grouting test group. The measured parameters of the grouting test effect include the measured value of the diffusion distance of the grouting slurry and the measured value of the water blocking rate.

[0070] The specific test steps for collecting the measured parameters of the grouting test effect of the grouting test groups are as follows.

[0071] (1) Bedrock fissure investigation and borehole layout. Obtain the basic properties of bedrock fissures through geological surveys, ultrasonic waves, radar detection, depth cameras, etc. According to the fissure distribution and grouting design requirements, reasonably arrange the boreholes, determine the borehole positions, spacing, and depth to ensure that the grouting can cover the main areas of the fissures.

[0072] (2) Install grouting equipment, including a grouting pump, high-pressure pipeline, grouting hole connection device, and pressure monitoring equipment. Check whether the equipment is operating normally to ensure that the pressure data can be recorded in real time during the construction process.

[0073] (3) Prepare the two-component foamed polymer material according to the design requirements to ensure accurate proportioning. After preparing the slurry, conduct small-scale tests in the laboratory or on-site to measure parameters such as the foaming ratio, viscosity, and curing time of the slurry, and verify whether it meets the diffusion requirements. Taking the isocyanate foaming material as an example, with the polyol ratio fixed at 1, prepare polymer slurries with different foaming ratios by adjusting the isocyanate ratio, and conduct compressive strength analysis on the consolidated bodies formed under different proportions. Finally, select the best proportion with the highest compressive strength. The proportion of the slurry components is a technical parameter that those skilled in the art can obtain according to the conventional proportion for different foaming materials used in grouting repair.

[0074] (4) Grouting implementation. Grout, inject the slurry into the fissures. At the initial stage of grouting, gradually increase the pressure from low pressure to the design pressure to prevent the slurry from overflowing or the fissures from expanding; real-time monitoring, use the pressure sensors and flow meters installed at different positions of the fissures to monitor the slurry diffusion range and pressure changes in real time, and stop grouting when the designed grouting volume is reached; mark the diffusion position, observe the position where the slurry reaches by laying markers (such as pigments or tracers) at the fissure outlet or seepage points; data recording, use technologies such as ultrasonic waves or geological radar to measure the maximum distance of slurry diffusion as the measured value of the diffusion range.

[0075] (5) Water blocking rate measurement. After the grouting is completed and cured, measure the change in the water blocking rate of the fissures through groundwater flow.

[0076] (6) Combine the measured values of the slurry diffusion distance and the water blocking rate, and evaluate the overall effect of the grouting construction through the subsequent grouting effect prediction model.

[0077] In this embodiment, the training set and the test set of the grouting effect prediction model are selected from the first data set and the measured parameters of the grouting test effect. Among the M groups of grouting test groups, N groups of grouting test groups are selected for training to obtain the training set, and the remaining M - N groups of grouting test groups are used as the test set.

[0078] S3. Select a BP neural network to construct a grouting effect prediction model, and optimize the initial parameters of the BP neural network through the particle swarm optimization algorithm (PSO).

[0079] Specifically, it includes the following sub - steps:

[0080] S31. Initialize the BP neural network, determine the number of nodes in the input layer, hidden layer and output layer of the BP neural network. The number of input layer nodes is equal to the number of input features (such as fracture aperture, roughness, grouting volume, etc. in the training set). The number of hidden layer nodes is set according to experience or experiments. The number of output layer nodes represents the diffusion distance prediction value, initially 1. Randomly initialize the weights and bias values of the BP neural network.

[0081] The BP neural network algorithm uses the gradient descent method, takes the mean square error between the actual output value (predicted value) and the expected output value (actual value) of the network model as the objective function, and adjusts the weight threshold of the neural network according to the size of this error value. Finally, make the error value small enough so that the actual output value of the network reaches the expected output value. For a three - layer BP neural network, let X=(x 1 ,x 2 ,……x m ) T be the input vector, where m is the number of features of the input vector; W=(w 1 ,w 2 ,…,w m ) represents the weight vector between the input layer and the hidden layer; θ=(θ 1 ,θ 2 ,…,θ q ) T represents the bias vector, where q is the number of neurons in the hidden layer; Φ(x) is the transfer function of the hidden layer, and this function uses the Sigmoid function; α=(α 1 ,α 2 ,…,α L ) T is the weight vector from the hidden layer to the output layer, where L is the number of output neurons; O=(O 1 ,O 2 ,…,O L ) is the output vector, is the transfer function of the output layer, and this function uses the Softmax function.

[0082] S32. Optimize the parameters of the BP neural network using the PSO algorithm. Take the weights and bias values of the BP neural network as the particles in the PSO algorithm, calculate the fitness of each particle, update the individual optimal position and global optimal position of each particle, and thus update the position and velocity of the particles. Repeat the iteration until the fitness function converges or reaches the preset number of iterations. Through the above process, the PSO algorithm continuously adjusts the weights and biases of the neural network, and finally obtains an optimized BP neural network model.

[0083] The PSO algorithm optimizes the parameters of the BP neural network. Take the weights and bias values of the BP neural network as the particles in the PSO algorithm, and update the velocity and position of each particle through the following formula to obtain the individual optimal position and global optimal position of the particle.

[0084]

[0085] In the formula, ω is the inertia weight, c 1 , c 2 is the acceleration factor, p best,i is the individual optimal position of particle i, g best is the global optimal position of any particle in the particle swarm, v i k , are the particle velocities of particle i at the current time k and the next time k + 1 respectively, x i k , x i k+1 are the positions of particle i at the current time k and the next time k + 1 respectively, rand 1 function, rand 2 function generate random numbers between (0, 1), and the optimal parameter combination of the BP neural network is obtained through iteration.

[0086] During the iteration process of the PSO algorithm, use the mean square error as the fitness function, and select the first test set or the second test set to calculate the fitness value of the BP neural network through the following formula:

[0087]

[0088] where n is the number of test samples in the test set, y i is the measured value corresponding to the test set, is the corresponding predicted value output by the BP neural network.

[0089] Iterate until the fitness function converges to obtain the optimal parameter combination of the BP neural network.

[0090] S33. Use the optimal parameter combination optimized by the PSO algorithm as the initial parameters for training the grouting effect prediction model.

[0091] S4. The predicted values of the grouting effect include the predicted value of the diffusion distance and the predicted value of the water blocking rate. The optimized grouting effect prediction model in step S3 is used to predict the diffusion distance and the water blocking rate after the foaming grout is grouted, and the comprehensive grouting effect prediction of the grouting slurry is obtained.

[0092] The fracture aperture and fracture roughness of the bedrock where the grouting test group is located are selected, combined with the first data set and the measured value of the grouting slurry diffusion distance, to construct the first training set and the first test set. The grouting slurry diffusion distance in the first training set and the first test set is predicted by the grouting diffusion distance prediction model to generate the predicted value of the diffusion distance. Among them, the first training set is input into the grouting effect prediction model for model training, and the grouting effect prediction model is evaluated through the first test set to obtain the grouting diffusion distance prediction model.

[0093] In this embodiment, the sample data of 16 groups of grouting test groups are randomly divided into a training set and a test set. The data set is randomly divided into the first training set and the first test set according to a ratio of 3:1. Among them, the sample data of 12 groups of grouting test groups are used as the first training set, and the sample data of the other 4 groups of grouting test groups are used as the first test set. The input variables of the first training set and the first test set include the fracture aperture, fracture roughness, grouting volume, slurry component ratio, and grouting pressure of the bedrock where the grouting test group is located, and the target output corresponds to the predicted value of the grouting slurry diffusion distance of the grouting test group. After obtaining the data of the first training set, the data needs to be cleaned first, the data integrity is checked, and the outliers and missing values are removed to ensure the data quality. Then all features are normalized to the range of [0,1] or [-1,1] and then input into the BP neural network model to avoid the influence of the scale difference between features on model training.

[0094] The parameters optimized by the particle swarm optimization algorithm (PSO) are used as the initial values of the grouting effect prediction model, and the BP neural network of the grouting effect prediction model is further trained through the first training set to improve the prediction accuracy. The learning rate is dynamically adjusted during the training process to avoid overfitting.

[0095] The input variables are selected from the first training set and input into the grouting effect prediction model, and the predicted value of the diffusion distance of the grouting output by the model is compared with the measured value of the grouting slurry diffusion distance, and the following evaluation indexes are used to evaluate the model prediction error.

[0096] The mean square error MSE measures the square deviation between the measured value of the grouting slurry diffusion distance and the predicted value of the diffusion distance, as shown in the following formula:

[0097]

[0098] The mean absolute error MAE measures the absolute deviation between the measured value of the grouting slurry diffusion distance and the predicted value of the diffusion distance, as shown in the following formula:

[0099]

[0100] Coefficient of determination R 2 , to evaluate the goodness of fit of the grouting diffusion distance prediction model to the grouting slurry diffusion distance, as shown in the following formula:

[0101]

[0102] In the above formula, n is the number of test samples in the test set, here referring to the number of samples in the first test set, which is 12, and y i is the measured value corresponding to the test set, here referring to the measured value of the grouting slurry diffusion distance, is the corresponding predicted value output by the BP neural network, here referring to the predicted value of the diffusion distance.

[0103] Compare the measured value of the grouting slurry diffusion distance with the predicted value of the diffusion distance, and verify the model performance through charts (such as scatter plots or line charts). If the model performance meets the requirements, the training ends, and a grouting diffusion distance prediction model for predicting the first test set is obtained through training; if the model performance does not meet the requirements, it can be optimized by increasing the sample size of the first training set, adjusting the parameters of the PSO algorithm (such as the number of particles, inertia weight), modifying the BP network structure (such as the number of hidden layer nodes or layers), etc.

[0104] Taking the adjustment of the PSO algorithm parameters as an example, the search range of the particles can be controlled by adjusting the inertia weight in the PSO, and the degree to which the particles approach the individual optimum and the global optimum can be determined by adjusting the learning factor.

[0105] In the inertia weight ω adjustment strategy: a larger inertia weight ω is beneficial for global search, while a smaller inertia weight ω is beneficial for local search. Usually, a linear decreasing strategy of the following formula is adopted,

[0106]

[0107] where ω max and ω mix are the maximum and minimum inertia weights, T is the total number of iterations, t is the current number of iterations, and the empirical value is usually taken in [0.4, 0.9].

[0108] c 1 , c 2 is the acceleration factor in the PSO algorithm, which determines the degree to which the particles approach the individual optimum and the global optimum. In the adjustment strategy: increasing the acceleration factor c 1 makes the particles more inclined to local search, and increasing the acceleration factor c 2 makes the particles more inclined to global search. The commonly used empirical value is c 1 = c 2 = 2, and it can also be dynamically adjusted through the following formula:

[0109]

[0110] Make the global search stronger in the early stage and the local search more refined in the later stage. T is the total number of iterations, and t is the current number of iterations.

[0111] Select the hydrodynamic pressure of the bedrock where the grouting test group is located, combine it with the first data set and the measured water-blocking rate to construct the second training set and the second test set. Predict the grouting water-blocking rate in the second training set and the second test set through the grouting water-blocking rate prediction model to generate the predicted water-blocking rate value. Input the second training set into the grouting effect prediction model for model training, and evaluate the grouting effect prediction model through the second test set to obtain the grouting water-blocking rate prediction model.

[0112] The division of the second training set and the second test set is the same as that of the first training set and the first test set, except for the selection of input variables. The input variables of the second training set and the second test set include the hydrodynamic pressure of the bedrock where the grouting test group is located, the grouting volume, the proportion of slurry components, and the grouting pressure, and the target output corresponds to the predicted water-blocking rate value of the grouting test group.

[0113] Select input variables from the second training set and input them into the grouting effect prediction model. Output the predicted water-blocking rate value of the model for grouting, compare it with the measured water-blocking rate value, use error evaluation indicators including mean square error, mean absolute error, and coefficient of determination to evaluate the model performance, adjust the parameters of the PSO algorithm or modify the structure of the BP neural network model according to the error evaluation indicators, and train to obtain the grouting water-blocking rate prediction model. The specific training process is the same as that of the grouting diffusion distance prediction model and will not be elaborated here.

[0114] The above has evaluated the grouting diffusion distance prediction model and the grouting water-blocking rate prediction model through the first test set and the second test set respectively. This embodiment also evaluates the comprehensive prediction effect of the grouting effect of the slurry through the mean square error, mean absolute error, and coefficient of determination between the predicted diffusion distance value, the predicted water-blocking rate value and the measured parameters of the grouting test effect. Obtain the mean square error, mean absolute error, and coefficient of determination between the predicted diffusion distance value made by the grouting diffusion distance prediction model for the first test set and the predicted water-blocking rate value made by the grouting water-blocking rate prediction model for the second test value respectively with the measured parameters of the grouting test effect, and calculate the comprehensive grouting performance evaluation coefficient through the following formula,

[0115]

[0116] Among them, Y is the comprehensive grouting performance evaluation coefficient, which is used to measure the overall performance of the prediction model for the grouting effect; MSE D is the mean square error between the predicted diffusion distance value and the measured value, MSE Bis the mean square error between the predicted value and the measured value of the water plugging rate, measuring the squared deviation between the predicted value and the actual value, MAE D is the mean absolute error between the predicted value and the measured value of the diffusion distance, MAE B is the mean absolute error between the predicted value and the measured value of the water plugging rate, measuring the average deviation between the predicted value and the actual value is the coefficient of determination between the predicted value and the measured value of the diffusion distance is the coefficient of determination between the predicted value and the measured value of the water plugging rate, measuring the explanatory ability of the model for the variation of the two. a, b, c, d, e, f are weight coefficients used to adjust the contribution degree of different evaluation indexes, satisfying a + b + c + d + e + f = 1

[0117] The calibration error threshold is Y threshold , usually, the error threshold of the grouting project can be set as follows: the error threshold of the diffusion distance is ±5% to ±15%, and for projects with higher precision requirements, it can be set as ±5%; the error threshold of the water plugging rate is ±5% to ±10%, and for key projects, it can be set as ±5%

[0118] Compare the comprehensive grouting performance evaluation coefficient Y with the error threshold Y threshol , if Y ≤ Y threshold , it is judged that the accuracy rate of the predicted value meets the precision requirements, and the grouting diffusion distance prediction model and the grouting water plugging rate prediction model are used as the prediction references for the grouting effect in the actual grouting process; if Y > Y threshold , it is judged that the accuracy rate of the predicted value does not meet the precision requirements, and continue to increase the training set test samples of the grouting diffusion distance prediction model and the grouting water plugging rate prediction model until the accuracy rate of the predicted value meets the precision requirements

[0119] S5. Use the grouting diffusion distance prediction model and the grouting water plugging rate prediction model to predict the actual grouting effect of the foaming slurry

[0120] When actually predicting the grouting effect of the foaming slurry, select the fracture aperture, fracture roughness, grouting volume, slurry component ratio, and grouting pressure of the bedrock where the grouting repair is located and input them into the grouting diffusion distance prediction model to output the predicted value of the diffusion distance of the injected foaming slurry; select the hydrodynamic pressure, grouting volume, slurry component ratio, and grouting pressure of the bedrock where the grouting repair is located and input them into the grouting water plugging rate prediction model to output the predicted value of the grouting water plugging rate of the injected foaming slurry, and summarize the predicted value of the slurry diffusion distance and the predicted value of the grouting water plugging rate to obtain the prediction of the actual grouting effect of the foaming slurry

[0121] Example 2

[0122] See Figure 2, the figure discloses the PSO-BP-based foamed grout injection effect prediction system of the present invention, which is used to execute the foamed grout injection effect prediction method in Embodiment 1, and specifically includes a bedrock sample collection module, a grouting influencing factor collection module, a test data collection module, a grouting diffusion distance prediction model generation module, a grouting water plugging rate prediction model generation module, a prediction error calculation module, and a prediction performance evaluation module.

[0123] The bedrock sample collection module obtains the fracture aperture, fracture roughness, and hydrodynamic pressure of the bedrock where different grouting test groups are located;

[0124] The grouting influencing factor collection module obtains the grouting influencing factor data of each grouting test group. The grouting influencing factor data includes the grouting volume, slurry component ratio, and grouting pressure, and maps the numbers of each grouting test group and the grouting influencing factor data one by one to form a first data set;

[0125] The test data collection module obtains the measured parameters of the grouting test effect after grouting for each grouting test group. The measured parameters of the grouting test effect include the measured value of the grouting slurry diffusion distance and the measured value of the water plugging rate;

[0126] The grouting diffusion distance prediction model generation module selects a BP neural network optimized by the PSO algorithm to construct a grouting effect prediction model, selects the fracture aperture and fracture roughness of the bedrock where the grouting test group is located, combines the first data set and the measured value of the grouting slurry diffusion distance to construct a first training set and a first test set, inputs the first training set into the grouting effect prediction model for model training, and evaluates the grouting effect prediction model through the first test set to obtain a grouting diffusion distance prediction model.

[0127] The grouting water plugging rate prediction model generation module selects a BP neural network optimized by the PSO algorithm to construct a grouting effect prediction model, selects the hydrodynamic pressure of the bedrock where the grouting test group is located, combines the first data set and the measured value of the water plugging rate to construct a second training set and a second test set, inputs the second training set into the grouting effect prediction model for model training, and evaluates the grouting effect prediction model through the second test set to obtain a grouting water plugging rate prediction model.

[0128] The prediction error calculation module calculates the mean square error, mean absolute error, and coefficient of determination between the predicted values of the diffusion distance and the water plugging rate and the measured parameters of the grouting test effect in the test set;

[0129] The prediction performance evaluation module calculates a comprehensive grouting performance evaluation coefficient based on the mean square error, mean absolute error, and coefficient of determination generated by the prediction error calculation module, and compares it with the error threshold to evaluate the prediction effect of the grouting effect of the slurry.

[0130] Embodiment 3

[0131] The present invention also provides a computer device, including: one or more processors and a memory storing one or more computer programs; wherein, the processor calls the computer program to implement the steps of the foam slurry grouting effect prediction method based on PSO-BP described in Embodiment 1, and the specific implementation process can refer to the description of Embodiment 1.

[0132] For the specific implementation process of each step, please refer to the description of the foregoing method.

[0133] It should be understood that in the embodiments of the present invention, the so-called processor may be a central processing unit (CPU), and this processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or this processor may also be any conventional processor, etc. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A part of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0134] In this article, the orientation or positional relationship indicated by the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", "vertical", "horizontal", etc. is the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and is only for the convenience of clearly expressing the technical solution and description, and therefore cannot be construed as a limitation to the present invention.

[0135] In this article, the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, and in addition to including the listed elements, it may also include other elements not specifically listed.

[0136] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. The prediction method of foaming slurry grouting effect based on PSO-BP is characterized by The steps include: S1, setting up multiple groups of fissure grouting test groups, obtaining the fissure opening, fissure roughness and dynamic water pressure of the bedrock where different grouting test groups are located, and the grouting influencing factor data of each grouting test group, wherein the grouting influencing factor data includes the grouting amount, the slurry component ratio and the grouting pressure, and mapping the numbers of each grouting test group and the grouting influencing factor data one by one to form a first data set; S2, obtaining the measured parameters of the grouting test effect after grouting of each grouting test group, wherein the measured parameters of the grouting test effect include the measured value of the grouting slurry diffusion distance and the measured value of the water blocking rate; S3, select BP neural network to build a grouting effect prediction model, and optimize the initialization parameters of BP neural network through PSO algorithm; S4. The predicted values ​​of the grouting effect include the predicted values ​​of the diffusion distance and the predicted values ​​of the water blocking rate. The fissure opening and the fissure roughness of the bedrock where the grouting test group is located are selected in combination with the first data set and the measured value of the diffusion distance of the grouting slurry to construct a first training set and a first test set. The first training set is input into the grouting effect prediction model for model training. The grouting effect prediction model is evaluated through the first test set to obtain a grouting diffusion distance prediction model. The dynamic water pressure of the bedrock where the grouting test group is located is selected in combination with the first data set and the measured value of the water blocking rate to construct a second training set and a second test set. The second training set is input into the grouting effect prediction model for model training. The grouting effect prediction model is evaluated through the second test set to obtain a grouting water blocking rate prediction model. S5. Use the grouting diffusion distance prediction model and the grouting water blocking rate prediction model to predict the actual grouting effect of the foaming slurry.

2. The method for predicting the effect of foaming slurry grouting based on PSO-BP according to claim 1, characterized in that The step S3 includes the following sub-steps: S31, BP neural network initialization, determining the node number parameters of the BP neural network input layer, hidden layer and output layer, and randomly initializing the weight and bias value of the BP neural network; S32, PSO algorithm optimizes the parameters of BP neural network, takes the weight and bias value of BP neural network as the particles in PSO algorithm, and updates the speed and position of each particle by the following formula to obtain the individual optimal position and global optimal position of the particle; Where ω is the inertia weight, c1 and c2 are acceleration factors, and p best,i is the individual optimal position of particle i, g best is the global optimal position of any particle in the particle swarm, v i k , are the particle velocities of particle i at the current time k and the next time k+1, respectively, i k 、x i k+1 are the positions of particle i at the current time k and the next time k+1 respectively. The rand1 function and rand2 function generate random numbers between (0,1). After iteration, the optimal parameter combination of the BP neural network is obtained. S33. The optimal parameter combination optimized by the PSO algorithm is used as the initial parameters for training the grouting effect prediction model.

3. The method for predicting the grouting effect of foaming slurry based on PSO-BP according to claim 2 is characterized in that: In the PSO algorithm iteration process of step S32, the mean square error is used as the fitness function, and the first test set or the second test set is selected to calculate the fitness value of the BP neural network by the following formula: Where n is the number of test samples in the first test set or the second test set, y i is the measured value corresponding to the first test set or the second test set, is the corresponding prediction value output by the BP neural network; Iterate until the fitness function converges to obtain the optimal parameter combination of the BP neural network.

4. The method for predicting the grouting effect of foaming slurry based on PSO-BP according to claim 1, characterized in that: The step S4 of setting the training set and the test set includes selecting N groups of grouting test groups from the M groups of grouting test groups to obtain the training set, and the remaining MN groups of grouting test groups to obtain the test set; wherein, The input variables of the first training set and the first test set include the fracture aperture, fracture roughness, grouting volume, slurry component ratio, and grouting pressure of the bedrock corresponding to the grouting test group, and the target output is the predicted value of the grouting slurry diffusion distance of the corresponding grouting test group; The input variables of the second training set and the second test set include the dynamic water pressure of the bedrock where the grouting test group is located, the grouting volume, the slurry component ratio, and the grouting pressure, and the target output corresponds to the predicted value of the grouting water blocking rate of the grouting test group.

5. The method for predicting the effect of foaming slurry grouting based on PSO-BP according to claim 4 is characterized in that: Input variables are selected from the first test set and input into the grouting effect prediction model. The output model predicts the diffusion distance of the grouting and compares it with the measured value of the diffusion distance of the grouting slurry. The model performance is evaluated using error evaluation indicators including mean square error, mean absolute error and determination coefficient. The parameters of the PSO algorithm are adjusted or the BP neural network model structure is modified according to the error evaluation indicators to obtain the grouting diffusion distance prediction model through training.

6. The method for predicting the grouting effect of foaming slurry based on PSO-BP according to claim 4 is characterized in that: Input variables are selected from the second test set and input into the grouting effect prediction model. The output model predicts the water blocking rate of grouting and compares it with the measured value of the water blocking rate. The model performance is evaluated using error evaluation indicators including mean square error, mean absolute error and determination coefficient. The parameters of the PSO algorithm are adjusted or the BP neural network model structure is modified according to the error evaluation indicators to obtain the grouting water blocking rate prediction model through training.

7. The method for predicting the grouting effect of foaming slurry based on PSO-BP according to claim 1, characterized in that: In step S4, the diffusion distance prediction value generated by the grouting diffusion distance prediction model for the first test set and the water plugging rate prediction value generated by the grouting water plugging rate prediction model for the second test set are obtained, and the mean square error, mean absolute error and determination coefficient between the measured parameters of the grouting test effect are calculated respectively, and the comprehensive grouting performance evaluation coefficient is obtained by the following formula: Among them, Y is the comprehensive grouting performance evaluation coefficient, MSE D MSE is the mean square error between the predicted diffusion distance and the measured diffusion distance. B is the mean square error between the predicted value and the measured value of the water plugging rate, MAE D is the mean absolute error between the predicted diffusion distance and the measured diffusion distance, MAE B is the average absolute error between the predicted value and the measured value of the water plugging rate, is the determination coefficient between the predicted and measured diffusion distances, is the determination coefficient of the predicted value and the measured value of the water plugging rate, a, b, c, d, e, f are weight coefficients, and a+b+c+d+e+f=1. The calibration error threshold is Y threshold , compared with the comprehensive grouting performance evaluation coefficient Y, If Y≤Y threshold , then the accuracy of the predicted value is judged to meet the accuracy requirements, and the grouting diffusion distance prediction model and the grouting water blocking rate prediction model are used as the prediction reference of the grouting effect in the actual grouting process; If Y>Y threshold , it is judged that the accuracy of the predicted value does not meet the precision requirement, and the training set test samples of the grouting diffusion distance prediction model and the grouting water blocking rate prediction model are continued to be increased until the accuracy of the predicted value meets the precision requirement.

8. A foaming slurry grouting effect prediction system based on PSO-BP, used to execute the method according to any one of claims 1 to 7, characterized in that: include The bedrock sample collection module obtains the fracture aperture, fracture roughness and dynamic water pressure of the bedrock in different grouting test groups; A grouting influencing factor acquisition module is used to obtain the grouting influencing factor data of each grouting test group, wherein the grouting influencing factor data includes the grouting amount, the slurry component ratio and the grouting pressure, and each grouting test group number and the grouting influencing factor data are mapped one by one to form a first data set; The test data acquisition module obtains the measured parameters of the grouting test effect after grouting of each grouting test group, wherein the measured parameters of the grouting test effect include the measured value of the grouting slurry diffusion distance and the measured value of the water blocking rate; The grouting diffusion distance prediction model generation module selects the BP neural network optimized by the PSO algorithm to construct the grouting effect prediction model, selects the fracture opening and fracture roughness of the bedrock where the grouting test group is located, combines the first data set and the measured value of the grouting slurry diffusion distance to construct the first training set and the first test set, inputs the first training set into the grouting effect prediction model for model training, and evaluates the grouting effect prediction model through the first test set to obtain the grouting diffusion distance prediction model; The grouting water plugging rate prediction model generation module selects the BP neural network optimized by the PSO algorithm to construct the grouting effect prediction model, selects the dynamic water pressure of the bedrock where the grouting test group is located, combines the first data set and the measured value of the water plugging rate to construct the second training set and the second test set, inputs the second training set into the grouting effect prediction model for model training, and evaluates the grouting effect prediction model through the second test set to obtain the grouting water plugging rate prediction model; The prediction error calculation module calculates the mean square error, mean absolute error and determination coefficient between the actual measured parameters of the grouting test effect in the test set according to the predicted values ​​of the diffusion distance and the water plugging rate; The prediction performance evaluation module calculates the comprehensive grouting performance evaluation coefficient based on the mean square error, mean absolute error and determination coefficient generated by the prediction error calculation module, and compares it with the error threshold to evaluate the prediction effect of the grouting effect.

9. A computer device, characterized in that: include: one or more processors; a memory storing one or more computer programs; The processor calls the computer program to implement: The method for predicting the grouting effect of foaming slurry based on PSO-BP as described in any one of claims 1 to 7.

Images