Dynamic prediction method for shear strength of ionic rare earth tailing soil

Abstract

The invention discloses a dynamic prediction method for the shear strength of ionic type rare earth tailing soil, and relates to the technical field of dynamic prediction of the shear strength of the ionic type rare earth tailing soil. Comprising the following steps: collecting soil bulk density data, tailing soil shear strength data, real-time rainfall data, soil real-time moisture evaporation data and underground water level data, and establishing a soil bulk density three-dimensional diagram and an underground water level curve diagram; the method comprises the following steps: constructing a real-time dynamic model of soil moisture and rainfall data and soil real-time moisture evaporation data and a dynamic prediction model of the shear strength and the stability coefficient of the tailing soil based on the soil moisture; according to the method disclosed by the invention, by integrating a sensor technology, a multiple regression algorithm, a random forest algorithm, a polynomial regression model construction technology of statistics and a big data analysis technology, the predicted shear strength becomes a more accurate index under the same condition; and the accuracy in the dynamic prediction process of the shear strength of the ionic rare earth tailing soil is enhanced.

Description

Technical Field

[0001] The invention relates to the technical field of dynamic prediction of the shear strength of ionic rare earth tailings soil, in particular to a dynamic prediction method for the shear strength of ionic rare earth tailings soil. Background Art

[0002] For ionic rare earth tailings soil, shear strength is an important parameter to measure its stability. Traditional shear strength assessment methods are mostly based on static experiments, which are difficult to accurately reflect the behavior under dynamic conditions in the actual environment. With the advancement of science and technology, comprehensive analysis methods combining field monitoring data, laboratory test results and numerical simulations have gradually become a research hotspot. By using sensor networks to collect key parameters such as humidity, temperature, and pore water pressure of tailings soil in real time, and using machine learning algorithms or physical models for data analysis, dynamic prediction of shear strength change trends can be achieved, which not only helps to improve the safety management level of tailings ponds, but also provides a scientific basis for the management and resource utilization of tailings soil. Although the existing dynamic prediction methods for the shear strength of ionic rare earth tailings soil have made great progress, there are still some problems that need to be optimized. The main influencing factor of the shear strength of ionic rare earth tailings soil is soil moisture. Soil cohesion and soil internal friction angle are two key parameters of the shear strength of ionic rare earth tailings soil. Soil moisture affects the prediction of the shear strength of ionic rare earth tailings soil by changing soil cohesion and soil internal friction angle. In the process of predicting the shear strength of ionic rare earth tailings soil, the soil meteorological factors that affect soil moisture are mainly rainfall and evaporation, and soil moisture will affect soil shear strength. When the soil moisture is too high, the shear strength of the tailings soil is significantly reduced, which may lead to inaccurate prediction results; the height of the groundwater level is not easy to be observed by the naked eye, but too high a groundwater level will change the soil properties and change the shear strength of the tailings soil, resulting in a large error in the prediction result. Summary of the invention

[0003] In order to solve the above problems, the present invention provides a method for dynamically predicting the shear strength of ionic rare earth tailings soil, which comprises the following steps: S1. Use a soil bulk density measuring instrument to obtain soil bulk density; S2. Use a direct shear instrument to obtain the shear force and displacement data of ionic rare earth tailings soil; use a tipping bucket rain gauge to obtain a pulse electrical signal carrying real-time rainfall data in the tailings soil shear strength prediction area; use a micro evaporator to obtain real-time soil water evaporation data; use an ultrasonic sensor to obtain groundwater level data in the tailings soil shear strength prediction area; S3. Use ArcGIS to create a three-dimensional map of soil bulk density; S4. Use tailings soil shear force and displacement data to obtain shear strength data; convert pulse electrical signals carrying real-time rainfall data into digital signals and input them into a computer to obtain real-time rainfall data; pre-process shear strength data, real-time rainfall data and real-time soil moisture evaporation data, and use a multivariate regression algorithm to obtain a real-time dynamic model of soil moisture and rainfall data, and real-time soil moisture evaporation data; obtain a dynamic prediction model of tailings soil shear strength and stability coefficient based on soil moisture through a random forest algorithm; S5. Using a statistical polynomial regression model, determine the groundwater level threshold, draw a two-dimensional coordinate graph, and obtain a groundwater level curve graph; S6. Analyze the soil bulk density three-dimensional graph, shear strength data, real-time rainfall data, real-time soil water evaporation data and groundwater level curve, and intervene in the shear strength prediction process based on the analysis results.

[0004] Preferably, the process of obtaining soil bulk density in S1 includes: The predicted area is selected, and the soil bulk density meter is based on the knife ring method. The original soil sample is collected and dried to constant weight. The volume of the knife ring and the drying weight of the soil sample are measured to calculate the soil bulk density.

[0005] Preferably, the process of obtaining the shear force and displacement data of the ionic rare earth tailings soil in S2 includes: The direct shear instrument simulates the forces received by the soil in actual scenarios by applying horizontal and vertical forces on the soil samples. The ionic rare earth tailings soil samples collected from the test area are placed in the shear box of the direct shear instrument, and a predetermined vertical load is applied. The vertical load simulates the gravity or other vertical stress above the soil layer. While the vertical load remains unchanged, the horizontal shear force is applied at a constant rate. During the shearing process, the direct shear instrument monitors the changes in shear force and displacement in real time to obtain the shear force and displacement data of the tailings soil.

[0006] Preferably, the process of obtaining the pulse electrical signal carrying real-time rainfall data in the tailings soil shear strength prediction area in S2 includes: The tipping bucket rain gauge has built-in rain receiver, water diversion funnel, tipping bucket, sensor and signal recorder. Rainwater enters the tipping bucket rain gauge through the rain receiver, flows through the water diversion funnel and enters the tipping bucket. The preset value of rainfall for the tipping bucket to flip is set to 0.1mm. When the rainfall inside the tipping bucket reaches the preset value, it loses balance due to gravity and flips. All the rainwater in the tipping bucket is discharged and reset to the initial state. The flipping action of the tipping bucket is connected to the sensor through a mechanical transmission device. When the tipping bucket flips, the sensor is triggered to generate a pulse electrical signal. The larger the number of pulse electrical signals, the greater the number of times the tipping bucket flips and the greater the rainfall. The signal recorder then records the number of times the tipping bucket flips to obtain a pulse electrical signal carrying real-time rainfall data.

[0007] Preferably, the process of obtaining real-time soil water evaporation data in S2 includes: The micro-evaporator consists of a soil column, an outer barrel and a weighing device. Soil samples are collected in real time and filled into the soil column. The soil column is then placed in the outer barrel and the weighing device is installed to obtain the initial mass of the soil column. The micro-evaporator is placed in the test area to ensure that the sample soil is consistent with the evaporation environment of the test area. The mass of the soil column after evaporation is measured by the weighing device, and the mass of the soil column after evaporation is subtracted from the initial mass of the soil column to obtain the real-time water evaporation data of the tailings soil.

[0008] Preferably, the process of obtaining groundwater level data in S2 includes: The ultrasonic sensor has a built-in ultrasonic transmitter and an ultrasonic receiver. The ultrasonic transmitter consists of a transmitter and a transducer. The transmitter generates an electrical signal, and the transducer consists of a piezoelectric chip. The piezoelectric chip generates mechanical vibration when it receives voltage excitation, and converts electrical energy into ultrasonic pulses. The ultrasonic transmitter transmits the pulse to the top of the groundwater body to be tested. When the ultrasonic pulse contacts the groundwater surface, based on the principle that sound waves will be reflected between different media, the ultrasonic receiver receives the reflected ultrasonic pulse and measures the round-trip time of the ultrasonic wave. The distance between the sensor and the groundwater surface is obtained by calculation, and then the distance from the sensor to the ground is subtracted to obtain the groundwater level data.

[0009] Preferably, the process of establishing the three-dimensional map of soil bulk density by using ArcGIS in S3 includes: The soil bulk density data were cleaned and imported into ArcGIS 3D modeling software. A 3D network was created in the software, and the soil bulk density value was assigned to each node of the grid. The software visualization function was used to generate a 3D graph of the soil bulk density distribution.

[0010] Preferably, the process of obtaining the dynamic model of soil moisture and real-time rainfall data, soil real-time moisture evaporation data, and the dynamic prediction model of tailings soil shear strength and stability coefficient based on soil moisture in S4 includes: B1. Establish a two-dimensional coordinate system, with the horizontal axis representing the position and the vertical axis representing the shear force. Map the shear force and its displacement data to the two-dimensional coordinate system and draw a shear force and displacement relationship curve. If the curve has a peak, the shear force corresponding to the peak is the shear strength. If the curve has no peak, when the shear force corresponding to the shear displacement reaches stability, the shear force is the shear strength. B2. Amplify and filter the collected pulse electrical signals carrying real-time rainfall data, convert the analog electrical signals into digital signals through an analog-to-digital converter and input them into a computer for data decoding, data calibration and data storage, and obtain real-time rainfall data; B3. Clean and standardize the soil shear strength data, real-time rainfall data, and real-time soil water evaporation data; B4. Using the multiple regression algorithm, set the dependent variable as soil moisture content, the independent variables as real-time rainfall and real-time soil moisture evaporation, use the statistical software SPSS to input data, select the least squares method for model fitting, obtain the SPSS output model, and check the consistency of the SPSS output model with the assumptions of linear regression; B5. When checking the inconsistency between the assumptions of the SPSS model and the linear regression, compare the difference between the model prediction value and the actual observation value, take measures to transform variables, delete abnormal data and use weighted least squares method to optimize the SPSS output model, and obtain the dynamic model of soil moisture and real-time rainfall data and soil real-time moisture evaporation data; B6. Using the dynamic model of soil moisture, real-time rainfall data and real-time soil moisture evaporation data, the real-time soil moisture data is calculated based on the fact that the real-time soil moisture data is equal to the real-time rainfall data minus the real-time soil moisture evaporation data; B7. Use the random forest algorithm to set random forest parameters. The data set consists of real-time soil moisture data and soil shear strength data. The data set is divided into training set, validation set and test set. Samples are extracted from the training set to train the random forest model. The validation set is input into the random forest model. The model fit is adjusted through parameters. The test set is then input into the adjusted random forest model to evaluate the model performance and obtain a dynamic prediction model for tailings soil shear strength and its stability coefficient based on soil moisture.

[0011] Preferably, the process of obtaining the groundwater level curve map in S5 includes: C1. Clean and standardize the groundwater level data and tailings soil shear strength data; C2. Importing the preprocessed data into SPSS, using the statistical software SPSS to fit the polynomial regression model, starting from a quadratic polynomial to fit the nonlinear relationship between the groundwater level data and the shear strength, and increasing the order of the polynomial until a polynomial regression model that fits the nonlinear relationship between the groundwater level data and the shear strength is obtained; C3. The polynomial regression model of the nonlinear relationship between groundwater level data and shear strength includes the statistical results of regression coefficient, R-square value, F statistic and P value. The fitting effect of the model is evaluated according to the statistical results. When the fitting is good and the P value is less than 0.05, the safety threshold of tailings soil shear strength is determined by using the regression results. The corresponding groundwater level data is obtained by regression equation calculation, and the obtained groundwater level data is set as the threshold that does not affect the shear strength prediction result. C4. Establish a two-dimensional coordinate system, where the horizontal axis is time and the vertical axis is the groundwater level value. Map the groundwater level data and the groundwater level threshold into the two-dimensional coordinate system to obtain a groundwater level curve.

[0012] Preferably, the process of analyzing the soil bulk density three-dimensional graph, shear strength data, real-time rainfall data, real-time soil water evaporation data and groundwater level curve graph in S6, and intervening in the shear strength prediction process through the analysis results includes: D1. Analyze the three-dimensional graph of soil bulk density. The larger the soil bulk density in the horizontal direction, the predicted regional climate is dry, the terrain is steep, and the vegetation coverage rate is low. The smaller the soil bulk density in the horizontal direction, the predicted regional climate is humid, the terrain is flat, and the vegetation coverage rate is high. The larger the soil bulk density in the vertical direction, the greater the soil depth. The smaller the soil bulk density in the vertical direction, the soil is affected by human disturbance and natural disasters, and the soil depth shows a decreasing trend. D2. Analyze the real-time rainfall data and the real-time soil moisture evaporation data, and combine the dynamic model of soil moisture with the real-time rainfall data and the real-time soil moisture evaporation data to obtain the relationship and dynamic change law between the real-time soil moisture and the real-time rainfall and the real-time soil moisture evaporation; D3. Through the dynamic prediction model of shear strength and its stability coefficient based on soil moisture, real-time soil moisture data is input, and the model outputs the corresponding soil shear strength data. Among them, the stability coefficient is the degree to which the soil shear strength remains stable with the change of soil moisture. By analyzing the fluctuation degree of soil shear strength with the change of soil moisture, the standard deviation of soil shear strength under different soil moisture conditions is calculated as the stability coefficient. The smaller the standard deviation, the smaller the fluctuation of soil shear strength with the change of moisture, and the higher the stability coefficient; the larger the standard deviation, the greater the fluctuation of soil shear strength with the change of moisture, and the lower the stability coefficient; D4. Analyze the groundwater level curve. If the groundwater level data does not exceed the groundwater level threshold, the groundwater level value is normal and does not affect the prediction of tailings soil shear strength. If it exceeds the groundwater level threshold, the groundwater level value is too high, affecting the prediction of tailings soil shear strength. D5. Based on the above analysis results, the shear strength prediction process is intervened.

[0013] The beneficial effects of the present invention are as follows: Compared with the traditional method for dynamically predicting the shear strength of tailings soil, the method of dynamically predicting the shear strength of tailings soil in the present invention is closely integrated with modern information technology, and achieves real-time and comprehensive monitoring of the dynamic prediction process of the shear strength of ionic rare earth tailings soil by integrating sensor technology, multivariate regression algorithm, random forest algorithm, statistical polynomial regression model construction technology and big data analysis technology. The method can accurately capture rainfall data, real-time soil water evaporation data, groundwater level data and tailings soil shear strength data, and construct a dynamic model of soil meteorological factors such as soil moisture and rainfall data, and real-time soil water evaporation data. The dynamic prediction model of shear strength and its stability coefficient based on soil moisture is used to monitor the influence of soil moisture on the prediction results of soil shear strength, report abnormalities in time, and intervene and handle them by the client, thus solving the problem of soil moisture affecting the prediction results of soil shear strength; through the groundwater level curve and its safety threshold, the groundwater level data is monitored in real time, and the abnormal signal is transmitted to the client, thus solving the problem of interference of the groundwater level on the prediction of soil shear strength, ensuring that the method in the present invention can refine the dynamic prediction standard of tailings soil shear strength within a more accurate range, so that the predicted shear strength becomes a more accurate indicator under the same conditions. The development and application of this method have significantly enhanced the accuracy of the dynamic prediction process of ionic rare earth tailings soil shear strength. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of a method for dynamically predicting the shear strength of ionic rare earth tailings soil. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] like Figure 1 As shown, the present invention provides a technical solution: a method for dynamically predicting the shear strength of ionic rare earth tailings soil, which comprises the following steps: S1. Use a soil bulk density measuring instrument to obtain soil bulk density; S2. Use a direct shear instrument to obtain the shear force and displacement data of ionic rare earth tailings soil; use a tipping bucket rain gauge to obtain a pulse electrical signal carrying real-time rainfall data in the tailings soil shear strength prediction area; use a micro evaporator to obtain real-time soil water evaporation data; use an ultrasonic sensor to obtain groundwater level data in the tailings soil shear strength prediction area; S3. Use ArcGIS to create a three-dimensional map of soil bulk density; S4. Use tailings soil shear force and displacement data to obtain shear strength data; convert pulse electrical signals carrying real-time rainfall data into digital signals and input them into a computer to obtain real-time rainfall data; pre-process shear strength data, real-time rainfall data and real-time soil moisture evaporation data, and use a multivariate regression algorithm to obtain a real-time dynamic model of soil moisture and rainfall data, and real-time soil moisture evaporation data; obtain a dynamic prediction model of tailings soil shear strength and stability coefficient based on soil moisture through a random forest algorithm; S5. Using a statistical polynomial regression model, determine the groundwater level threshold, draw a two-dimensional coordinate graph, and obtain a groundwater level curve graph; S6. Analyze the soil bulk density three-dimensional graph, shear strength data, real-time rainfall data, real-time soil water evaporation data and groundwater level curve, and intervene in the shear strength prediction process based on the analysis results.

[0017] The process of obtaining soil bulk density in S1 includes: The predicted area is selected, and the soil bulk density meter is based on the knife ring method. The original soil sample is collected and dried to constant weight. The volume of the knife ring and the drying weight of the soil sample are measured to calculate the soil bulk density.

[0018] The process of obtaining the shear force and displacement data of ionic rare earth tailings soil in S2 includes: The direct shear instrument simulates the forces received by the soil in actual scenarios by applying horizontal and vertical forces on the soil samples. The ionic rare earth tailings soil samples collected from the test area are placed in the shear box of the direct shear instrument, and a predetermined vertical load is applied. The vertical load simulates the gravity or other vertical stress above the soil layer. While the vertical load remains unchanged, the horizontal shear force is applied at a constant rate. During the shearing process, the direct shear instrument monitors the changes in shear force and displacement in real time to obtain the shear force and displacement data of the tailings soil.

[0019] The process of obtaining the pulse electric signal carrying real-time rainfall data in the tailings soil shear strength prediction area in S2 includes: The tipping bucket rain gauge has built-in rain receiver, water diversion funnel, tipping bucket, sensor and signal recorder. Rainwater enters the tipping bucket rain gauge through the rain receiver, flows through the water diversion funnel and enters the tipping bucket. The preset value of rainfall for the tipping bucket to flip is set to 0.1mm. When the rainfall inside the tipping bucket reaches the preset value, it loses balance due to gravity and flips. All the rainwater in the tipping bucket is discharged and reset to the initial state. The flipping action of the tipping bucket is connected to the sensor through a mechanical transmission device. When the tipping bucket flips, the sensor is triggered to generate a pulse electrical signal. The larger the number of pulse electrical signals, the greater the number of times the tipping bucket flips and the greater the rainfall. The signal recorder then records the number of times the tipping bucket flips to obtain a pulse electrical signal carrying real-time rainfall data.

[0020] The process of obtaining real-time soil water evaporation data in S2 includes: The micro-evaporator consists of a soil column, an outer barrel and a weighing device. Soil samples are collected in real time and filled into the soil column. The soil column is then placed in the outer barrel and the weighing device is installed to obtain the initial mass of the soil column. The micro-evaporator is placed in the test area to ensure that the sample soil is consistent with the evaporation environment of the test area. The mass of the soil column after evaporation is measured by the weighing device, and the mass of the soil column after evaporation is subtracted from the initial mass of the soil column to obtain the real-time water evaporation data of the tailings soil.

[0021] The process of obtaining groundwater level data in S2 includes: The ultrasonic sensor has a built-in ultrasonic transmitter and an ultrasonic receiver. The ultrasonic transmitter consists of a transmitter and a transducer. The transmitter generates an electrical signal, and the transducer consists of a piezoelectric chip. The piezoelectric chip generates mechanical vibration when it receives voltage excitation, and converts electrical energy into ultrasonic pulses. The ultrasonic transmitter transmits the pulse to the top of the groundwater body to be tested. When the ultrasonic pulse contacts the groundwater surface, based on the principle that sound waves will reflect between different media, the ultrasonic receiver receives the reflected ultrasonic pulse and measures the round-trip time of the ultrasonic wave. The distance between the sensor and the groundwater surface is calculated through the s=vt formula, and then the distance from the sensor to the ground is subtracted to obtain the groundwater level data.

[0022] The process of using ArcGIS in S3 to create a three-dimensional soil bulk density map includes: The soil bulk density data were cleaned and imported into ArcGIS 3D modeling software. A 3D network was created in the software, and the soil bulk density value was assigned to each node of the grid. The software visualization function was used to generate a 3D graph of the soil bulk density distribution.

[0023] The process of obtaining soil moisture and real-time rainfall data, the dynamic model of soil real-time moisture evaporation data and the dynamic prediction model of tailings soil shear strength and stability coefficient based on soil moisture in S4 includes: B1. Establish a two-dimensional coordinate system, with the horizontal axis representing the position and the vertical axis representing the shear force. Map the shear force and its displacement data to the two-dimensional coordinate system and draw a shear force and displacement relationship curve. If the curve has a peak, the shear force corresponding to the peak is the shear strength. If the curve has no peak, when the shear force corresponding to the shear displacement reaches stability, the shear force is the shear strength. B2. Amplify and filter the collected pulse electrical signals carrying real-time rainfall data, convert the analog electrical signals into digital signals through an analog-to-digital converter and input them into a computer for data decoding, data calibration and data storage, and obtain real-time rainfall data; B3. Clean and standardize the soil shear strength data, real-time rainfall data, and real-time soil water evaporation data; B4. Using the multiple regression algorithm, set the dependent variable as soil moisture content, the independent variables as real-time rainfall and real-time soil moisture evaporation, use the statistical software SPSS to input data, select the least squares method for model fitting, obtain the SPSS output model, and check the consistency of the SPSS output model with the assumptions of linear regression; B5. When checking the inconsistency between the assumptions of the SPSS model and the linear regression, compare the difference between the model prediction value and the actual observation value, take measures to transform variables, delete abnormal data and use weighted least squares method to optimize the SPSS output model, and obtain the dynamic model of soil moisture and real-time rainfall data and soil real-time moisture evaporation data; B6. Using the dynamic model of soil moisture, real-time rainfall data and real-time soil moisture evaporation data, the real-time soil moisture data is calculated based on the fact that the real-time soil moisture data is equal to the real-time rainfall data minus the real-time soil moisture evaporation data; B7. Use the random forest algorithm to set random forest parameters. The data set consists of real-time soil moisture data and soil shear strength data. The data set is divided into training set, validation set and test set. Samples are extracted from the training set to train the random forest model. The validation set is input into the random forest model. The model fit is adjusted through parameters. The test set is then input into the adjusted random forest model to evaluate the model performance and obtain a dynamic prediction model for tailings soil shear strength and its stability coefficient based on soil moisture.

[0024] The process of obtaining the groundwater level curve map in S5 includes: C1. Clean and standardize the groundwater level data and tailings soil shear strength data; C2. Importing the preprocessed data into SPSS, using the statistical software SPSS to fit the polynomial regression model, starting from a quadratic polynomial to fit the nonlinear relationship between the groundwater level data and the shear strength, and increasing the order of the polynomial until a polynomial regression model that fits the nonlinear relationship between the groundwater level data and the shear strength is obtained; C3. The polynomial regression model of the nonlinear relationship between groundwater level data and shear strength includes the statistical results of regression coefficient, R-square value, F statistic and P value. The fitting effect of the model is evaluated according to the statistical results. When the fitting is good and the P value is less than 0.05, the safety threshold of tailings soil shear strength is determined by using the regression results. The corresponding groundwater level data is obtained by regression equation calculation, and the obtained groundwater level data is set as the threshold that does not affect the shear strength prediction result. C4. Establish a two-dimensional coordinate system, where the horizontal axis is time and the vertical axis is the groundwater level value. Map the groundwater level data and the groundwater level threshold into the two-dimensional coordinate system to obtain a groundwater level curve.

[0025] In S6, the three-dimensional soil bulk density graph, shear strength data, real-time rainfall data, real-time soil water evaporation data, and groundwater level curve are analyzed, and the process of intervening in the shear strength prediction process through the analysis results includes: D1. Analyze the three-dimensional graph of soil bulk density. The larger the soil bulk density in the horizontal direction, the predicted regional climate is dry, the terrain is steep, and the vegetation coverage rate is low. The smaller the soil bulk density in the horizontal direction, the predicted regional climate is humid, the terrain is flat, and the vegetation coverage rate is high. The larger the soil bulk density in the vertical direction, the greater the soil depth. The smaller the soil bulk density in the vertical direction, the soil is affected by human disturbance and natural disasters, and the soil depth shows a decreasing trend. D2. Analyze the real-time rainfall data and the real-time soil moisture evaporation data, and combine the dynamic model of soil moisture with the real-time rainfall data and the real-time soil moisture evaporation data to obtain the relationship and dynamic change law between the real-time soil moisture and the real-time rainfall and the real-time soil moisture evaporation; D3. Through the dynamic prediction model of shear strength and its stability coefficient based on soil moisture, real-time soil moisture data is input, and the model outputs the corresponding soil shear strength data. Among them, the stability coefficient is the degree to which the soil shear strength remains stable with the change of soil moisture. By analyzing the fluctuation degree of soil shear strength with the change of soil moisture, the standard deviation of soil shear strength under different soil moisture conditions is calculated as the stability coefficient. The smaller the standard deviation, the smaller the fluctuation of soil shear strength with the change of moisture, and the higher the stability coefficient; the larger the standard deviation, the greater the fluctuation of soil shear strength with the change of moisture, and the lower the stability coefficient; D4. Analyze the groundwater level curve. If the groundwater level data does not exceed the groundwater level threshold, the groundwater level value is normal and does not affect the prediction of tailings soil shear strength. If it exceeds the groundwater level threshold, the groundwater level value is too high, affecting the prediction of tailings soil shear strength. D5. Based on the above analysis results, the shear strength prediction process is intervened.

[0026] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A method for dynamically predicting the shear strength of ionic rare earth tailings soil, characterized in that: The method consists of the following steps: S1. Use a soil bulk density measuring instrument to obtain soil bulk density; S2. Use a direct shear instrument to obtain the shear force and displacement data of ionic rare earth tailings soil; use a tipping bucket rain gauge to obtain a pulse electrical signal carrying real-time rainfall data in the tailings soil shear strength prediction area; use a micro evaporator to obtain real-time soil water evaporation data; use an ultrasonic sensor to obtain groundwater level data in the tailings soil shear strength prediction area; S3. Use ArcGIS to create a three-dimensional map of soil bulk density; S4. Using the tailings soil shear force and displacement data, obtain the shear strength data, convert the pulse electrical signal carrying the real-time rainfall data into a digital signal and input it into the computer to obtain the real-time rainfall data; pre-process the shear strength data, real-time rainfall data and real-time soil moisture evaporation data, and use the multivariate regression algorithm to obtain a real-time dynamic model of soil moisture and rainfall data, and real-time soil moisture evaporation data; obtain a dynamic prediction model of tailings soil shear strength and stability coefficient based on soil moisture through the random forest algorithm; S5. Using a statistical polynomial regression model, determine the groundwater level threshold, draw a two-dimensional coordinate graph, and obtain a groundwater level curve graph; S6. Analyze the soil bulk density three-dimensional graph, shear strength data, real-time rainfall data, real-time soil water evaporation data and groundwater level curve, and intervene in the shear strength prediction process based on the analysis results.

2. The method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 1, characterized in that: The process of obtaining soil bulk density in S1 includes: The predicted area is selected, and the soil bulk density meter is based on the knife ring method to collect the original soil sample, dry it to constant weight, measure the volume of the knife ring and the drying weight of the soil sample, and calculate the soil bulk density.

3. The method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 2, characterized in that: The process of obtaining the shear force and displacement data of ionic rare earth tailings soil in S2 includes: The direct shear instrument simulates the forces received by the soil in actual scenarios by applying horizontal and vertical forces on the soil samples. The ionic rare earth tailings soil samples collected from the test area are placed in the shear box of the direct shear instrument, and a predetermined vertical load is applied. The vertical load simulates the gravity or other vertical stress above the soil layer. While the vertical load remains unchanged, the horizontal shear force is applied at a constant rate. During the shearing process, the direct shear instrument monitors the changes in shear force and displacement in real time to obtain the shear force and displacement data of the tailings soil.

4. The method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 3, characterized in that: The process of obtaining the pulse electric signal carrying real-time rainfall data in the tailings soil shear strength prediction area in S2 includes: The tipping bucket rain gauge has built-in rain receiver, water diversion funnel, tipping bucket, sensor and signal recorder. Rainwater enters the tipping bucket rain gauge through the rain receiver, flows through the water diversion funnel and enters the tipping bucket. The preset value of rainfall for the tipping bucket to flip is set to 0.1mm. When the rainfall inside the tipping bucket reaches the preset value, it loses balance due to gravity and flips. All the rainwater in the tipping bucket is discharged and reset to the initial state. The flipping action of the tipping bucket is connected to the sensor through a mechanical transmission device. When the tipping bucket flips, the sensor is triggered to generate a pulse electrical signal. The larger the number of pulse electrical signals, the greater the number of times the tipping bucket flips and the greater the rainfall. The signal recorder then records the number of times the tipping bucket flips to obtain a pulse electrical signal carrying real-time rainfall data.

5. The method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 4, characterized in that: The process of obtaining real-time soil water evaporation data in S2 includes: The micro-evaporator consists of a soil column, an outer barrel and a weighing device. Soil samples are collected in real time and filled into the soil column. The soil column is then placed in the outer barrel and the weighing device is installed to obtain the initial mass of the soil column. The micro-evaporator is placed in the test area to ensure that the sample soil is consistent with the evaporation environment of the test area. The mass of the soil column after evaporation is measured by the weighing device, and the mass of the soil column after evaporation is subtracted from the initial mass of the soil column to obtain the real-time water evaporation data of the tailings soil.

6. A method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 5, characterized in that: The process of obtaining groundwater level data in S2 includes: The ultrasonic sensor has a built-in ultrasonic transmitter and an ultrasonic receiver. The ultrasonic transmitter consists of a transmitter and a transducer. The transmitter generates an electrical signal, and the transducer consists of a piezoelectric chip. The piezoelectric chip generates mechanical vibration when it receives voltage excitation, and converts electrical energy into ultrasonic pulses. The ultrasonic transmitter transmits the pulse to the top of the groundwater body to be tested. When the ultrasonic pulse contacts the groundwater surface, based on the principle that sound waves will be reflected between different media, the ultrasonic receiver receives the reflected ultrasonic pulse and measures the round-trip time of the ultrasonic wave. The distance between the sensor and the groundwater surface is obtained by calculation, and then the distance from the sensor to the ground is subtracted to obtain the groundwater level data.

7. A method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 6, characterized in that: The process of establishing a three-dimensional soil bulk density map using ArcGIS in S3 includes: The soil bulk density data were cleaned and imported into ArcGIS 3D modeling software. A 3D network was created in the software, and the soil bulk density value was assigned to each node of the grid. The software visualization function was used to generate a 3D graph of the soil bulk density distribution.

8. The method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 7, characterized in that: The process of obtaining the dynamic model of soil moisture and real-time rainfall data, soil real-time moisture evaporation data and the dynamic prediction model of tailings soil shear strength and stability coefficient based on soil moisture in S4 includes: B1. Establish a two-dimensional coordinate system, with the horizontal axis representing the position and the vertical axis representing the shear force. Map the shear force and its displacement data to the two-dimensional coordinate system and draw a shear force and displacement relationship curve. If the curve has a peak, the shear force corresponding to the peak is the shear strength. If the curve has no peak, when the shear force corresponding to the shear displacement reaches stability, the shear force is the shear strength. B2. Amplify and filter the collected pulse electrical signals carrying real-time rainfall data, convert the analog electrical signals into digital signals through an analog-to-digital converter and input them into a computer for data decoding, data calibration and data storage, and obtain real-time rainfall data; B3. Clean and standardize the soil shear strength data, real-time rainfall data, and real-time soil water evaporation data; B4. Using the multiple regression algorithm, set the dependent variable as soil moisture content, the independent variables as real-time rainfall and real-time soil moisture evaporation, use the statistical software SPSS to input data, select the least squares method for model fitting, obtain the SPSS output model, and check the consistency of the SPSS output model with the assumptions of linear regression; B5. When checking the inconsistency between the assumptions of the SPSS model and the linear regression, compare the difference between the model prediction value and the actual observation value, take measures to transform variables, delete abnormal data and use weighted least squares method to optimize the SPSS output model, and obtain the dynamic model of soil moisture and real-time rainfall data and soil real-time moisture evaporation data; B6. Using the dynamic model of soil moisture, real-time rainfall data and real-time soil moisture evaporation data, the real-time soil moisture data is calculated based on the fact that the real-time soil moisture data is equal to the real-time rainfall data minus the real-time soil moisture evaporation data; B7. Use the random forest algorithm to set random forest parameters. The data set consists of real-time soil moisture data and soil shear strength data. The data set is divided into training set, validation set and test set. Samples are extracted from the training set to train the random forest model. The validation set is input into the random forest model. The model fit is adjusted through parameters. The test set is then input into the adjusted random forest model to evaluate the model performance and obtain a dynamic prediction model for tailings soil shear strength and its stability coefficient based on soil moisture.

9. The method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 8, characterized in that: The process of obtaining the groundwater level curve map in S5 includes: C1. Clean and standardize the groundwater level data and tailings soil shear strength data; C2. Importing the preprocessed data into SPSS, using the statistical software SPSS to fit the polynomial regression model, starting from a quadratic polynomial to fit the nonlinear relationship between the groundwater level data and the shear strength, increasing the order of the polynomial until a polynomial regression model that fits the nonlinear relationship between the groundwater level data and the shear strength is obtained; C3. The polynomial regression model of the nonlinear relationship between groundwater level data and shear strength includes the statistical results of regression coefficient, R-square value, F statistic and P value. The fitting effect of the model is evaluated according to the statistical results. When the fitting is good and the P value is less than 0.05, the safety threshold of tailings soil shear strength is determined by using the regression results. The corresponding groundwater level data is obtained by regression equation calculation, and the obtained groundwater level data is set as the threshold that does not affect the shear strength prediction result. C4. Establish a two-dimensional coordinate system, where the horizontal axis is time and the vertical axis is the groundwater level value. Map the groundwater level data and the groundwater level threshold into the two-dimensional coordinate system to obtain a groundwater level curve.

10. The method for dynamically predicting the shear strength of ionic rare earth tailings soil according to claim 9, characterized in that: The process of analyzing the soil bulk density three-dimensional graph, shear strength data, real-time rainfall data, real-time soil water evaporation data and groundwater level curve graph in S6, and intervening in the shear strength prediction process based on the analysis results includes: D1. Analyze the three-dimensional graph of soil bulk density. The larger the soil bulk density in the horizontal direction, the predicted regional climate is dry, the terrain is steep, and the vegetation coverage rate is low. The smaller the soil bulk density in the horizontal direction, the predicted regional climate is humid, the terrain is flat, and the vegetation coverage rate is high. The larger the soil bulk density in the vertical direction, the greater the soil depth. The smaller the soil bulk density in the vertical direction, the soil is affected by human disturbance and natural disasters, and the soil depth shows a decreasing trend. D2. Analyze the real-time rainfall data and the real-time soil moisture evaporation data, and combine the dynamic model of soil moisture with the real-time rainfall data and the real-time soil moisture evaporation data to obtain the relationship and dynamic change law between the real-time soil moisture and the real-time rainfall and the real-time soil moisture evaporation; D3. Through the dynamic prediction model of shear strength and its stability coefficient based on soil moisture, real-time soil moisture data is input, and the model outputs the corresponding soil shear strength data. Among them, the stability coefficient is the degree to which the soil shear strength remains stable with the change of soil moisture. By analyzing the fluctuation degree of soil shear strength with the change of soil moisture, the standard deviation of soil shear strength under different soil moisture conditions is calculated as the stability coefficient. The smaller the standard deviation, the smaller the fluctuation of soil shear strength with the change of moisture, and the higher the stability coefficient; the larger the standard deviation, the greater the fluctuation of soil shear strength with the change of moisture, and the lower the stability coefficient; D4. Analyze the groundwater level curve. If the groundwater level data does not exceed the groundwater level threshold, the groundwater level value is normal and does not affect the prediction of tailings soil shear strength. If it exceeds the groundwater level threshold, the groundwater level value is too high, affecting the prediction of tailings soil shear strength. D5. Based on the above analysis results, the shear strength prediction process is intervened.

Images