Static pressure pile time-varying bearing capacity sensing and predicting method based on physical information network

Abstract

The invention provides a static pressure pile time-varying bearing capacity sensing and predicting method based on a physical information network. The method comprises the steps that pore pressure dissipation data are collected through a pile body sensor; inputting the pore pressure dissipation data and the calculation points into a PINNs forward and reverse integrated inversion model, and estimating excess pore water pressure in back propagation to obtain a final consolidation coefficient; inputting the final consolidation coefficient and the calculation point into a PINNs forward and reverse integrated forward model, and estimating excess pore water pressure in back propagation; and inputting the test set into the trained PINNs forward and reverse integrated forward model, outputting the excess pore water pressure dissipation degree of the pile side soil body, and predicting the time-varying bearing capacity of the static pressure pile according to the excess pore water pressure dissipation degree. According to the method, the soil layer consolidation coefficient can be accurately inversed and sensed through short-term pore pressure dissipation test data of the pile body pore pressure sensor, and then the pile side soil consolidation degree is predicted, so that accurate prediction of the time-varying bearing capacity of the static pressure pile is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of rock and soil mechanics and engineering technology, and specifically relates to a method for perceiving and predicting the time-varying bearing capacity of static pressure piles based on a physical information network. Background Art

[0002] During the process of static pile driving in natural saturated clay strata, the pile body shears and squeezes the soil to penetrate into the foundation, resulting in the remodeling of the surrounding soil and the generation of a high excess pore water pressure. After the pile driving is completed, the effective stress of the soil around the pile gradually recovers and increases with the dissipation of the excess pore water pressure. Therefore, the bearing capacity of static piles in saturated clay strata shows significant timeliness. However, due to the complexity of the stress and strain history of the soil around the static pile, the current research on the understanding of the pile driving effect and bearing capacity timeliness in saturated soft soil foundations is still insufficient. The static pile design and calculation process does not consider the dissipation of the excess pore water pressure generated by the pile driving, which causes the increase of the bearing capacity of the static pile after the pile driving is completed. As a result, the design of static piles in actual projects is often conservative, resulting in a huge waste of resources. Therefore, how to predict and make full use of the timeliness of the bearing capacity of static piles has become a key issue to be solved in the current research on soft soil pile foundations.

[0003] Domestic and foreign scholars have conducted extensive research on the time-effect of the bearing capacity of static pressure piles. For example, Randolph et al. used the undrained expansion of column holes to simulate the static pressure pile sinking process and conducted a theoretical study on the sinking effect of static pressure piles in saturated clay and the reconsolidation of soil around the piles. Skov et al. proposed an empirical calculation formula for the time-varying bearing capacity of static pressure piles by analyzing the measured data of the bearing capacity of static pressure piles in clay over time. Cao Quan et al. proposed a theoretical solution for calculating the time-effect of the bearing capacity of static pressure piles in soft soil based on the parameters of static penetration test according to the characteristics of static penetration probes and static pressure piles in soft clay. The above studies have great theoretical and engineering significance, but most theoretical studies ignore the in-situ mechanical properties of soil such as stress history and stress anisotropy, as well as the influence of soil stress relaxation effect on the time-varying bearing capacity of static pressure piles during the reconsolidation of soil around piles. In addition, most studies focus on the pile sinking process or the reconsolidation of soil around piles, which makes it difficult to systematically and comprehensively reflect the essence of the time-varying bearing capacity of static pressure piles.

[0004] Therefore, how to accurately predict the time-varying bearing capacity of static piles has become a key technical challenge that needs to be solved urgently. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a method for perceiving and predicting the time-varying bearing capacity of static piles based on a physical information network. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0006] In a first aspect, a method for perceiving and predicting time-varying bearing capacity of static piles based on a physical information network comprises:

[0007] S100, collecting pore pressure dissipation data using pile body sensors; the pore pressure dissipation data includes the deployment depth of the sensor, the excess pore water pressure collected by the sensor, and the collection time;

[0008] S200, the pore pressure dissipation data and calculation points are input into the PINNs forward and inversion integrated inversion model, and the excess pore water pressure is estimated in the back propagation to obtain the final consolidation coefficient;

[0009] S300, inputting the final consolidation coefficient and calculation points into the PINNs forward-inversion integrated forward model, and estimating the excess pore water pressure in the back propagation to obtain the trained PINNs forward-inversion integrated forward model;

[0010] S400, input the test set into the trained PINNs forward and inverse integrated forward model, output the excess pore water pressure dissipation of the pile side soil, and predict the time-varying bearing capacity of the static pile based on the excess pore water pressure dissipation of the pile side soil. The test set consists of several calculation points that need to be predicted, and the calculation points include time and radial distance.

[0011] In a second aspect, the present invention provides a method for perceiving and predicting the time-varying bearing capacity of static piles based on a physical information network, comprising:

[0012] A collection module is used to collect pore pressure dissipation data using pile body sensors; the pore pressure dissipation data includes the sensor deployment depth, excess pore water pressure collected by the sensor and collection time;

[0013] The inversion module is used to input the pore pressure dissipation data and calculation points into the PINNs forward and inversion integrated inversion model, and estimate the excess pore water pressure in the back propagation to obtain the final consolidation coefficient;

[0014] The forward modeling module is used to input the final consolidation coefficient and calculation points into the PINNs forward-inverse integrated forward model, and estimate the excess pore water pressure in the back propagation to obtain the trained PINNs forward-inverse integrated forward model;

[0015] The prediction module is used to input the test set into the trained PINNs forward and inversion integrated forward model, output the excess pore water pressure dissipation of the pile side soil, and predict the time-varying bearing capacity of the static pile based on the excess pore water pressure dissipation of the pile side soil. The test set consists of several calculation points that need to be predicted, and the calculation points include time and radial distance.

[0016] Beneficial effects:

[0017] The present invention provides a method for perceiving and predicting the time-varying bearing capacity of a static pressure pile based on a physical information network, including: collecting pore pressure dissipation data using a pile body sensor; inputting the pore pressure dissipation data and calculation points into a PINNs forward and inverse integrated inversion model, and estimating the excess pore water pressure in the back propagation to obtain the final consolidation coefficient; inputting the final consolidation coefficient and calculation points into a PINNs forward and inverse integrated forward model, and estimating the excess pore water pressure in the back propagation to obtain the final excess pore water pressure; inputting a test set into the trained PINNs forward and inverse integrated forward model, outputting the dissipation degree of the excess pore water pressure of the pile side soil, and predicting the time-varying bearing capacity of the static pressure pile according to the excess pore water pressure of the pile side soil. The present invention can accurately invert the perceived soil consolidation coefficient through the short-term pore pressure dissipation test data of the pile body pore pressure sensor, and then predict the consolidation degree of the pile side soil, thereby realizing the accurate prediction of the time-varying bearing capacity of the static pressure pile.

[0018] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of a method for perceiving and predicting the time-varying bearing capacity of static pressure piles based on a physical information network provided by an embodiment of the present invention;

[0020] Figure 2 It is a schematic diagram of a static pressure pile and its sensor arrangement provided by an embodiment of the present invention;

[0021] Figure 3 is a schematic diagram of a PINNs forward and inversion integrated model provided by an embodiment of the present invention;

[0022] Figure 4 It is a schematic diagram of the result of the inversion consolidation coefficient of a PINNs forward and inversion integrated model provided by an embodiment of the present invention;

[0023] Figure 5 It is a schematic diagram of the result of predicting the dissipation of excess pore water pressure by using a PINNs forward and inversion integrated model provided by an embodiment of the present invention;

[0024] Figure 6 It is a comparison curve diagram of the time-varying bearing capacity prediction result and the measured bearing capacity of a PINNs static pressure pile provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0026] Embodiment 1

[0027] In view of the challenges faced by the current prediction methods of the time-varying bearing capacity of jacked piles, such as the failure to consider the influence of the in-situ mechanical properties of soil such as soil stress history and stress anisotropy and the stress relaxation effect of soil during the reconsolidation process of the soil around the pile on the time-varying bearing capacity of jacked piles, as well as the low efficiency of the prediction of the time-varying bearing capacity of jacked piles, it is difficult to achieve a fast and accurate prediction of the time-varying bearing capacity of jacked piles in practical applications.

[0028] Therefore, see Figure 1 , Figure 1 1 is a flow chart of a method for perceiving and predicting the time-varying bearing capacity of a static pressure pile based on a physical information network provided by an embodiment of the present invention. The method for perceiving and predicting the time-varying bearing capacity of a static pressure pile based on a physical information network provided by the present invention comprises:

[0029] S100, collecting pore pressure dissipation data using pile body sensors;

[0030] Among them, the pore pressure dissipation data include the deployment depth of the sensor, the excess pore water pressure collected by the sensor, and the collection time;

[0031] Figure 2 The figure shows a static pressure pile and a schematic diagram of the arrangement of sensors provided by an embodiment of the present invention. Figure 2 , pore pressure sensors can be arranged on the body of the static pressure pile, and then a pile driving test can be carried out to collect the measured values ​​of pore water pressure at different depths through the pore pressure sensors.

[0032] In order to better illustrate the beneficial effects of the present invention, a centrifugal model experiment was carried out:

[0033] Specifically, the model pile used in the experiment is a pore pressure static penetration instrument with an effective length L of 300 mm and an axis diameter D of 10 mm. The experimental steps are as follows: First, the model pile is pushed into the soil sample at a constant speed of 90 mm / min until it reaches the specified depth of 240 mm; then, the penetration process is stopped and the pore pressure dissipation test is performed until the pore pressure reading decays to the static pore water pressure level. During the experiment, the pore pressure dissipation data was recorded in detail.

[0034] S200, the pore pressure dissipation data and calculation points are input into the PINNs forward and inversion integrated inversion model, and the excess pore water pressure is estimated in the back propagation to obtain the final consolidation coefficient;

[0035] Among them, the PINNs forward and inversion integrated inversion model includes a first neural network; the consolidation coefficient is obtained by training the first neural network, and the training process includes: repeatedly inputting pore pressure dissipation data and calculation points into the first neural network, and outputting a first estimated value of excess pore water pressure, using the first estimated value of excess pore water pressure to calculate the physical information-data loss function, and reversely updating the weight and bias of the first neural network according to the physical information-data loss function until the training is terminated, and taking the consolidation coefficient at the end of the training as the final consolidation coefficient.

[0036] The first neural network of the present invention includes two first input layers and a first output layer, the two first input layers input depth and time respectively, and the first output layer outputs the estimated excess pore water pressure. For example, the first neural network includes 5 hidden layers, the number of neural network nodes in each hidden layer is 30, the selected activation function is Tanh, the optimizer is set to Adam, and the initial value of the consolidation coefficient to be inverted is set to 1.

[0037] Therefore, the calculation points are first input into the first neural network, and the excess pore water pressure is estimated in the back propagation process, so that the first estimated value of the excess pore water pressure can be obtained. Secondly, the physical-data loss function is calculated based on the first estimated value of the excess pore water pressure using the chain rule, and the residual of the physical-data loss function can be obtained. Then, based on the residual of the physical-data loss function, the chain rule is used to continuously reversely update the weights and bias of the first neural network and generate an estimated value of the consolidation coefficient, so that the residual of the physical-data loss function is minimized, and the consolidation coefficient under this condition is used as the final consolidation coefficient.

[0038] In a specific embodiment of the present invention, reference Figure 3 , the training process of the first neural network includes:

[0039] a1, inputting the collected pore pressure dissipation data and calculation points into the PINNs forward and inversion integrated inversion model to obtain a first estimated value of excess pore water pressure; the first neural network includes: a first input layer, a first output layer and a first hidden layer;

[0040] b1, calculation of the physical information-data loss function based on the first estimate of the excess pore water pressure;

[0041] The physical information-data loss function includes the loss function of the first radial consolidation equation, the loss function of the first initial condition, the loss function of the first boundary condition, and the excess pore pressure dissipation data loss function;

[0042] The physical information-data loss function is expressed as:

[0043]

[0044] in, represents the physical information-data loss function, represents the loss function for the excess pore pressure dissipation data, represents the loss function of the first radial consolidation equation, represents the loss function of the first initial condition, represents the loss function of the first boundary condition;

[0045] During the jacked pile sinking process, a large excess pore pressure is mainly generated in the radial direction. Therefore, the radial consolidation equation is used to simulate the dissipation process of excess pore pressure in the soil on the side of the jacked pile sinking. The radial consolidation equation can be expressed as follows:

[0046]

[0047] Where: G is the shear modulus, v' is the effective Poisson's ratio, c h is the horizontal consolidation coefficient, z is the depth, t is the time, and u(z,t) is the excess pore water pressure at depth z and time t.

[0048] The loss function of the first radial consolidation equation is constructed through the radial consolidation equation and is expressed as:

[0049]

[0050] Among them, λ P represents the weight of the loss function of the radial consolidation equation, N P represents the number of calculation points in the calculation domain of the radial consolidation equation, represents the i1th calculation point in the calculation domain of the radial consolidation equation, represents the radial distance of the i1th calculation point in the calculation domain of the radial consolidation equation, represents the time of the i1th calculation point in the calculation domain of the radial consolidation equation, represents the first estimate of the excess pore water pressure at the i1th calculation point in the calculation domain of the radial consolidation equation, c h represents the consolidation coefficient;

[0051] During the dissipation of excess pore pressure on the pile side, the pore pressure change at the radius of the elastic-plastic zone is extremely small, which can be taken as the drainage boundary. Since the pile body is impermeable, the pile diameter can be set as a non-drainage boundary condition, and the following is obtained:

[0052]

[0053] The loss function of the first boundary condition is expressed as:

[0054]

[0055] In the formula, represents the weight of the loss function of the boundary condition at the pile diameter, The weight of the loss function for the boundary condition at the elastic-plastic radius, N B represents the number of calculation points from the boundary condition calculation domain, represents the i2th calculation point in the boundary condition calculation domain, Represents the radial distance of the i2th calculation point in the boundary condition calculation domain, represents the time of the i2th calculation point in the boundary condition calculation domain, represents the first estimated value of the excess pore water pressure at the i2th calculation point within the boundary condition calculation domain;

[0056] Loss function for the first initial condition It is related to the initial pore water pressure, so the loss function of the first initial condition is expressed as:

[0057]

[0058] Among them, λ I represents the weight of the loss function for calculating the initial conditions, N I represents the number of calculation points in the initial condition calculation domain, represents the i3th calculation point in the initial condition calculation domain, represents the radial distance of the i3th calculation point in the initial condition calculation domain, represents the estimated time of the i3th calculation point in the initial condition calculation domain, represents the first estimate of the excess pore water pressure at the i3th calculation point in the initial condition calculation domain, represents the initial pore water pressure at the i3th calculation point in the initial condition calculation domain;

[0059] Loss function for excess pore pressure dissipation data It is constructed from the super-pore pressure dissipation data collected by the sensor, so the loss function of the super-pore pressure dissipation data is expressed as:

[0060]

[0061] Among them, λ D represents the weight of the loss function for calculating excess pore pressure dissipation data, N D represents the number of calculation points in the calculation domain from the excess pore pressure dissipation data, represents the i4th calculation point in the calculation domain of excess pore pressure dissipation data, represents the radial distance of the i4th calculation point in the calculation domain of excess pore pressure dissipation data, represents the time of the i4th calculation point in the calculation domain of excess pore pressure dissipation data, represents the excess pore pressure dissipation data of the i4th calculation point in the excess pore pressure dissipation data calculation domain, Represents the first estimate of the excess pore water pressure at the i4th calculation point in the excess pore pressure dissipation data calculation domain.

[0062] c1, based on the residual of the physical information-data loss function, uses the chain rule to reversely update the weights and biases of the first neural network and serves as the first neural network for the next cycle;

[0063] d1, repeat a1-c1 until the residual of the physical information-data loss function is less than the predetermined first criterion, then the training is terminated, and the consolidation coefficient in the physical information-data loss function at the training cutoff is used as the final consolidation coefficient.

[0064] refer to Figure 3 The present invention can set the PINNs forward-inversion integrated inversion model and the PINNs forward-inversion integrated forward model in the inversion module and the forward module respectively, and then call these two modules to execute the training process of the corresponding models.

[0065] This embodiment updates the weight and bias of the first neural network according to the first estimated value of the excess pore water pressure, and calculates the consolidation coefficient in the physical information-data loss function at this time as the estimated value of the consolidation coefficient, and then continues to execute according to the above steps until the residual of the physical information-data loss function is less than the predetermined first standard, and calculates the consolidation coefficient in the physical information-data loss function at this time as the final consolidation coefficient. It should be noted that the first standard can be set according to actual conditions, and this embodiment does not specifically limit this. The first standard is, for example, e -2 .

[0066] In addition, during the training process of the PINNs forward and inversion integrated inversion model, the consolidation coefficient c is saved once every 1000 trainings. V The estimated value of the last training is taken as the consolidation coefficient c v The final result of the inversion. In order to qualitatively measure the error between the true value and the inverted / predicted value, L w The relative error is used as a metric, and the error is calculated as follows:

[0067]

[0068] In the formula, and u Respectively represent the inversion / prediction value and the true value, which can represent the inversion / prediction value and the true value of excess pore water pressure, or the inversion / prediction value and the true value of consolidation coefficient; N represents the number of inversion / prediction values. Figure 4 As shown in the figure, after 40,000 trainings, the consolidation coefficient c of the PINNs forward and inversion integrated inversion model isv The inversion error is only 0.04, which shows that the PINNs forward and inversion integrated inversion model constructed in the present invention can quickly and accurately invert the consolidation coefficient c based on the short-term pore water pressure excess pore pressure dissipation data. h .

[0069] S300, inputting the final consolidation coefficient and calculation points into the PINNs forward-inversion integrated forward model, and estimating the excess pore water pressure in the back propagation to obtain the trained PINNs forward-inversion integrated forward model;

[0070] Among them, the PINNs forward-inverse integrated forward model includes a second neural network; the final excess pore water pressure is obtained by training the second neural network, and the training process includes: repeatedly inputting pore pressure dissipation data and calculation points into the second neural network, and outputting a second estimated value of the excess pore water pressure, using the second estimated value of the excess pore water pressure to calculate the physical information loss function, and reversely updating the weight and bias of the second neural network according to the physical information loss function until the training is terminated, and taking the second estimated value at the training end as the final excess pore water pressure.

[0071] It is worth noting that the calculation points in steps S200 and S300 are randomly selected.

[0072] In a specific embodiment of the present invention, reference Figure 3 , the training process of the second neural network includes:

[0073] a2, taking the final consolidation coefficient as the consolidation coefficient of the physical information loss function of the second neural network, and inputting the calculation point into the second neural network to obtain a second estimated value of the excess pore water pressure; the second neural network includes: a second input layer, a second output layer and a second hidden layer;

[0074] The second neural network in this step includes two second input layers and one second output layer, the two second input layers input depth and time respectively, and the second output layer outputs the estimated excess pore water pressure. For example, the second neural network includes 5 hidden layers, the number of neural network nodes in each hidden layer is 30, the selected activation function is Tanh, and the optimizer is set to Adam.

[0075] b2, calculating the physical information loss function according to the second estimated value;

[0076] The physical information loss function includes a second partial differential equation loss function, a second initial condition loss function, and a second boundary condition loss function;

[0077] The physical information loss function is expressed as:

[0078]

[0079] in, represents the physical information loss function, represents the loss function of the second radial consolidation equation, represents the loss function of the second initial condition, represents the loss function of the second boundary condition;

[0080] The loss function of the second radial consolidation equation is expressed as:

[0081]

[0082] Among them, λ P represents the weight of the loss function of the radial consolidation equation, N P represents the number of calculation points in the calculation domain of the radial consolidation equation, represents the i5th calculation point in the calculation domain of the radial consolidation equation. For example, it is set to 10000. represents the radial distance of the i5th calculation point in the calculation domain of the radial consolidation equation, represents the time of the i5th calculation point in the calculation domain of the radial consolidation equation, represents the second estimated value of the excess pore water pressure at the i5th calculation point in the calculation domain of the radial consolidation equation;

[0083] The loss function of the second boundary condition is expressed as:

[0084]

[0085] In the formula, represents the weight of the loss function of the boundary condition at the pile diameter, The weight of the loss function for the boundary condition at the elastic-plastic radius, N B represents the number of calculation points from the boundary condition calculation domain, Indicates the i6th calculation point in the boundary condition calculation domain. For example, it is set to 1000. represents the radial distance of the i6th calculation point in the boundary condition calculation domain, represents the time of the i6th calculation point in the boundary condition calculation domain, represents the second estimated value of the excess pore water pressure at the i6th calculation point in the boundary condition calculation domain;

[0086] The loss function of the second initial condition is expressed as:

[0087]

[0088] Among them, λ I Represents the weight of the loss function for calculating the initial conditions, for example, set to 1, N IIndicates the number of calculation points in the initial condition calculation domain. For example, if it is set to 1000, represents the i7th calculation point in the initial condition calculation domain, represents the radial distance of the i7th calculation point in the initial condition calculation domain, represents the estimated time of the i7th calculation point in the initial condition calculation domain, represents the second estimated value of the excess pore water pressure at the i7th calculation point in the calculation domain of the initial condition, Represents the initial pore water pressure at the i7th calculation point in the initial condition calculation domain.

[0089] c2, based on the residual of the physical information loss function, and using the chain rule to reversely update the weights and biases of the second neural network;

[0090] d2, repeat a2-c2 until the residual of the physical information loss function is less than the predetermined second standard, then the training is terminated.

[0091] This embodiment updates the weight and bias of the second neural network according to the second estimated value, and then continues to execute according to the above steps until the residual of the physical information loss function is less than the predetermined second standard, completing the training of the PINNs forward and inverse integrated forward model. It should be noted that the second standard can be set according to actual conditions, and this embodiment does not specifically limit this. The second standard is, for example, e -2 .

[0092] S400, input the test set into the trained PINNs forward and inverse integrated forward model, output the excess pore water pressure dissipation of the pile side soil, and predict the time-varying bearing capacity of the static pile based on the excess pore water pressure dissipation of the pile side soil. The test set consists of several calculation points that need to be predicted, and the calculation points include time and radial distance.

[0093] The test set consists of several calculation points that need to be predicted, including time and radial distance. The test set is input into the trained PINNs forward and inverse integrated forward model, and the excess pore water pressure dissipation of the pile side soil is output. The excess pore water pressure dissipation of the pile side soil is input into the calculation formula of the time-varying bearing capacity of the static pile, and the time-varying bearing capacity of the static pile is calculated.

[0094] The calculation formula of the time-varying bearing capacity of static pressure pile is expressed as:

[0095]

[0096] In the formula, Q t (t), Q s (t) and Q b (t) represent the time-varying bearing capacity, pile side resistance and pile end resistance of static pile respectively; h irepresents the depth of the pile driven into the i-th layer of soil; f si (t) and q bi (t) represent the unit pile side resistance and unit pile end resistance in the i-th soil layer; C s and A b are the pile cross-sectional perimeter and pile cross-sectional area respectively;

[0097] Unit pile tip resistance q b (t) is:

[0098]

[0099] Where N c is the end bearing capacity coefficient, s u0tc is the shear strength of the in-situ soil under triaxial compression, Λ is the plastic volume strain ratio, OCR is the overconsolidation ratio, p′(t) is the average effective stress changing with time, and p′0 is the initial average effective stress.

[0100] The average effective stress p′(t) is:

[0101]

[0102] In the formula, u0-u t is the dissipation degree of excess pore water pressure of the soil on the pile side, u0 is the initial excess pore water pressure of the soil on the pile side, u t is the excess pore water pressure of the soil at the pile side;

[0103] Unit pile side resistance f s (t) is:

[0104]

[0105] In the formula, and ψ f is the stress transformation parameter, φ′ is the effective internal friction angle, and M is the slope of the critical state line.

[0106] For example, the test set consists of 100 sets of calculation points in the range [t∈(0,100), z=0.25]. Figure 5 The results of the prediction of excess pore water pressure dissipation by the PINNs forward and inverse integrated model are shown. It can be seen that the prediction results of PINN are in good agreement with the measured solution, which shows that PINN can accurately predict the excess pore water pressure dissipation of the soil on the pile side. The excess pore water pressure dissipation of the soil on the pile side is input into the calculation formula of the time-varying bearing capacity of the static pressure pile to predict the time-varying bearing capacity of the static pressure pile. Figure 6The comparison results of the predicted value of the time-varying bearing capacity of the PINNs static pressure pile and the measured bearing capacity are presented, and the error value is only 0.01. It can be found that the prediction results of the PINNs prediction model established by the present invention are in good agreement with the measured values, proving that the method of the present invention can accurately invert the perceived soil consolidation coefficient through the short-term pore pressure dissipation test data of the pile body pore pressure sensor, and then predict the consolidation degree of the soil on the pile side, thereby realizing the accurate prediction of the time-varying bearing capacity of the static pressure pile.

[0107] The present invention provides a method for perceiving and predicting the time-varying bearing capacity of a static pressure pile based on a physical information network, including: using a pile body sensor to collect pore pressure dissipation data; inputting the pore pressure dissipation data and calculation points into a PINNs forward and inverse integrated inversion model, and estimating the excess pore water pressure in the back propagation to obtain the final consolidation coefficient; inputting the final consolidation coefficient and calculation points into a PINNs forward and inverse integrated forward model, and estimating the excess pore water pressure in the back propagation; inputting a test set into the trained PINNs forward and inverse integrated forward model, outputting the dissipation degree of the excess pore water pressure of the soil on the pile side, and predicting the time-varying bearing capacity of the static pressure pile based on the value. The present invention can accurately invert the perceived soil consolidation coefficient through the short-term pore pressure dissipation test data of the pile body pore pressure sensor, and then predict the consolidation degree of the soil on the pile side, thereby realizing the accurate prediction of the time-varying bearing capacity of the static pressure pile.

[0108] The present invention also provides a method for perceiving and predicting the time-varying bearing capacity of static piles based on a physical information network, comprising:

[0109] A collection module is used to collect pore pressure dissipation data using pile body sensors; the pore pressure dissipation data includes the sensor deployment depth, excess pore water pressure collected by the sensor and collection time;

[0110] The inversion module is used to input the pore pressure dissipation data and calculation points into the PINNs forward and inversion integrated inversion model, and estimate the excess pore water pressure in the back propagation to obtain the final consolidation coefficient;

[0111] The forward modeling module is used to input the final consolidation coefficient and calculation points into the PINNs forward-inverse integrated forward model, and estimate the excess pore water pressure in the back propagation to obtain the trained PINNs forward-inverse integrated forward model;

[0112] The prediction module is used to input the test set into the trained PINNs forward and inversion integrated forward model, output the excess pore water pressure dissipation of the pile side soil, and predict the time-varying bearing capacity of the static pile based on the excess pore water pressure dissipation of the pile side soil. The test set consists of several calculation points that need to be predicted, and the calculation points include time and radial distance.

[0113] The present invention provides a device for sensing and predicting the time-varying bearing capacity of a static pressure pile based on a physical information network, which uses a pile body sensor to collect pore pressure dissipation data; inputs the pore pressure dissipation data and calculation points into a PINNs forward and inverse integrated inversion model, and estimates the excess pore water pressure in the back propagation to obtain the final consolidation coefficient; inputs the final consolidation coefficient and calculation points into a PINNs forward and inverse integrated forward model, and estimates the excess pore water pressure in the back propagation; inputs a test set into the trained PINNs forward and inverse integrated forward model, outputs the excess pore water pressure dissipation degree of the pile side soil, and predicts the time-varying bearing capacity of the static pressure pile based on the value. The present invention can accurately invert the sensing soil consolidation coefficient through the short-term pore pressure dissipation test data of the pile body pore pressure sensor, and then predict the consolidation degree of the pile side soil, thereby realizing the accurate prediction of the time-varying bearing capacity of the static pressure pile.

[0114] It is worth noting that the terms "first" and "second" in the present invention are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0115] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art may understand and implement other variations of the disclosed embodiments by reviewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality of components or steps.

[0116] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A method for perceiving and predicting the time-varying bearing capacity of static piles based on a physical information network, characterized in that: include: S100, collecting pore pressure dissipation data using pile body sensors; the pore pressure dissipation data includes the deployment depth of the sensor, the excess pore water pressure collected by the sensor, and the collection time; S200, inputting the pore pressure dissipation data and the calculation points into the PINNs forward and inversion integrated inversion model, and estimating the excess pore water pressure in the back propagation to obtain the final consolidation coefficient; S300, inputting the final consolidation coefficient and the calculation point into the PINNs forward-inversion integrated forward model, and estimating the excess pore water pressure in the back propagation to obtain the trained PINNs forward-inversion integrated forward model; S400, input the test set into the trained PINNs forward-inverse integrated forward model, output the excess pore water pressure dissipation of the pile side soil, and predict the time-varying bearing capacity of the static pile based on the excess pore water pressure dissipation of the pile side soil, wherein the test set is composed of a number of calculation points that need to be predicted, and the calculation points include time and radial distance.

2. The method for perceiving and predicting the time-varying bearing capacity of static piles based on physical information network according to claim 1 is characterized in that: The PINNs forward and inversion integrated inversion model includes a first neural network; The consolidation coefficient is obtained by training the first neural network, and the training process includes: repeatedly inputting pore pressure dissipation data and calculation points into the first neural network, and outputting a first estimated value of excess pore water pressure, calculating a physical information-data loss function using the first estimated value of excess pore water pressure, and reversely updating the weights and biases of the first neural network according to the physical information-data loss function until the training is terminated, and taking the consolidation coefficient at the end of the training as the final consolidation coefficient.

3. The method for perceiving and predicting the time-varying bearing capacity of static piles based on physical information network according to claim 2 is characterized in that: The training process of the first neural network includes: a1, inputting the collected pore pressure dissipation data and calculation points into the PINNs forward and inversion integrated inversion model to obtain a first estimated value of excess pore water pressure; the first neural network includes: a first input layer, a first output layer and a first hidden layer; b1, calculating the residual of the physical information-data loss function according to the first estimated value of the excess pore water pressure; c1, according to the residual of the physical information-data loss function, and using the chain rule to reversely update the weights and biases of the first neural network, and serve as the first neural network for the next cycle; d1, repeat a1-c1 until the residual of the physical information-data loss function is less than a predetermined first criterion, then the training is terminated, and the consolidation coefficient in the physical information-data loss function at the training termination is used as the final consolidation coefficient.

4. The method for perceiving and predicting the time-varying bearing capacity of static piles based on physical information network according to claim 3 is characterized in that: The physical information-data loss function includes a loss function of the first radial consolidation equation, a loss function of the first initial condition, a loss function of the first boundary condition, and an excess pore pressure dissipation data loss function; The physical information-data loss function is expressed as: in, represents the physical information-data loss function, represents the loss function for the excess pore pressure dissipation data, represents the loss function of the first radial consolidation equation, represents the loss function of the first initial condition, represents the loss function of the first boundary condition; The loss function of the first radial consolidation equation is expressed as: Among them, λ P represents the weight of the loss function of the radial consolidation equation, N P represents the number of calculation points in the calculation domain of the radial consolidation equation, represents the i1th calculation point in the calculation domain of the radial consolidation equation, represents the radial distance of the i1th calculation point in the calculation domain of the radial consolidation equation, represents the time of the i1th calculation point in the calculation domain of the radial consolidation equation, represents the first estimate of the excess pore water pressure at the i1th calculation point in the calculation domain of the radial consolidation equation, c h represents the consolidation coefficient; The loss function of the first boundary condition is expressed as: In the formula, represents the weight of the loss function of the boundary condition at the pile diameter, The weight of the loss function for the boundary condition at the elastic-plastic radius, N B represents the number of calculation points from the boundary condition calculation domain, represents the i2th calculation point in the boundary condition calculation domain, Represents the radial distance of the i2th calculation point in the boundary condition calculation domain, represents the time of the i2th calculation point in the boundary condition calculation domain, represents the first estimated value of the excess pore water pressure at the i2th calculation point within the boundary condition calculation domain; The loss function of the first initial condition is expressed as: Among them, λ I represents the weight of the loss function for calculating the initial conditions, N I represents the number of calculation points in the initial condition calculation domain, represents the i3th calculation point in the initial condition calculation domain, represents the radial distance of the i3th calculation point in the initial condition calculation domain, represents the estimated time of the i3th calculation point in the initial condition calculation domain, represents the first estimate of the excess pore water pressure at the i3th calculation point in the initial condition calculation domain, represents the initial pore water pressure at the i3th calculation point in the initial condition calculation domain; The loss function of the excess pore pressure dissipation data is expressed as: Among them, λ D represents the weight of the loss function for calculating excess pore pressure dissipation data, N D represents the number of calculation points in the calculation domain from the excess pore pressure dissipation data, represents the i4th calculation point in the calculation domain of excess pore pressure dissipation data, represents the radial distance of the i4th calculation point in the calculation domain of excess pore pressure dissipation data, represents the time of the i4th calculation point in the calculation domain of excess pore pressure dissipation data, represents the excess pore pressure dissipation data of the i4th calculation point in the excess pore pressure dissipation data calculation domain, Represents the first estimate of the excess pore water pressure at the i4th calculation point in the excess pore pressure dissipation data calculation domain.

5. The method for perceiving and predicting the time-varying bearing capacity of static piles based on physical information network according to claim 1 is characterized in that: The PINNs forward-inversion integrated forward model includes a second neural network; the training process of the PINNs forward-inversion integrated forward model includes: repeatedly inputting calculation points into the second neural network and outputting a second estimated value of excess pore water pressure, calculating a physical information loss function using the second estimated value of excess pore water pressure, and reversely updating the weight and bias of the second neural network according to the physical information loss function until the training is terminated.

6. The method for perceiving and predicting the time-varying bearing capacity of static piles based on physical information network according to claim 5 is characterized in that: The training process of the second neural network includes: a2, using the final consolidation coefficient as the consolidation coefficient of the physical information loss function of the second neural network, and inputting the calculation point into the second neural network to obtain a second estimated value of the excess pore water pressure; the second neural network includes: a second input layer, a second output layer and a second hidden layer; b2, calculating the physical information loss function according to the second estimated value; c2, according to the residual of the physical information loss function, and using the chain rule to reversely update the weights and biases of the second neural network; d2, repeat a2-c2 until the residual of the physical information loss function is less than a predetermined second standard, then the training is terminated to obtain a trained PINNs forward and inversion integrated forward model.

7. The method for perceiving and predicting the time-varying bearing capacity of static piles based on physical information network according to claim 6 is characterized in that: The physical information loss function includes a second partial differential equation loss function, a second initial condition loss function and a second boundary condition loss function; The physical information loss function is expressed as: in, represents the physical information loss function, represents the loss function of the second radial consolidation equation, represents the loss function of the second initial condition, represents the loss function of the second boundary condition; The loss function of the second radial consolidation equation is expressed as: Among them, λ P represents the weight of the loss function of the radial consolidation equation, N P represents the number of calculation points in the calculation domain of the radial consolidation equation, represents the i5th calculation point in the calculation domain of the radial consolidation equation, represents the radial distance of the i5th calculation point in the calculation domain of the radial consolidation equation, represents the time of the i5th calculation point in the calculation domain of the radial consolidation equation, represents the second estimated value of the excess pore water pressure at the i5th calculation point in the calculation domain of the radial consolidation equation; The loss function of the second boundary condition is expressed as: In the formula, represents the weight of the loss function of the boundary condition at the pile diameter, The weight of the loss function for the boundary condition at the elastic-plastic radius, N B represents the number of calculation points from the boundary condition calculation domain, represents the i6th calculation point in the boundary condition calculation domain, represents the radial distance of the i6th calculation point in the boundary condition calculation domain, represents the time of the i6th calculation point in the boundary condition calculation domain, represents the second estimated value of the excess pore water pressure at the i6th calculation point in the boundary condition calculation domain; The loss function of the second initial condition is expressed as: Among them, λ I represents the weight of the loss function for calculating the initial conditions, N I represents the number of calculation points in the initial condition calculation domain, represents the i7th calculation point in the initial condition calculation domain, represents the radial distance of the i7th calculation point in the initial condition calculation domain, represents the estimated time of the i7th calculation point in the initial condition calculation domain, represents the second estimated value of the excess pore water pressure at the i7th calculation point in the calculation domain of the initial condition, Represents the initial pore water pressure at the i7th calculation point in the initial condition calculation domain.

8. The method for perceiving and predicting the time-varying bearing capacity of static pressure piles based on physical information network according to claim 1 is characterized in that: S400 includes: The test set is input into the trained PINNs forward and inverse integrated forward model, and the dissipation degree of excess pore water pressure of the soil on the pile side is output. The dissipation degree of excess pore water pressure of the soil on the pile side is input into the calculation formula of the time-varying bearing capacity of the jacked pile to obtain the predicted time-varying bearing capacity of the jacked pile.

9. The method for perceiving and predicting the time-varying bearing capacity of static piles based on physical information network according to claim 8 is characterized in that: The calculation formula of the time-varying bearing capacity of the static pressure pile is expressed as: In the formula, Q t (t), Q s (t) and Q b (t) represent the time-varying bearing capacity, pile side resistance and pile end resistance of static pile respectively; h i represents the depth of the pile driven into the i-th layer of soil; f si (t) and q bi (t) represent the unit pile side resistance and unit pile end resistance in the i-th soil layer; C s and a b are the pile cross-sectional perimeter and pile cross-sectional area respectively; The unit pile tip resistance q b (t) is: Where N c is the end bearing capacity coefficient, s u0tc is the shear strength of the in-situ soil under triaxial compression, Λ is the plastic volume strain ratio, OCR is the overconsolidation ratio, and p ′ (t) is the average effective stress changing with time, p ′ 0 is the initial average effective stress; The average effective stress p ′ (t) is: In the formula, u0-u t is the dissipation degree of excess pore water pressure of the soil on the pile side, u0 is the initial excess pore water pressure of the soil on the pile side, u t is the excess pore water pressure of the soil at the pile side; The unit pile side resistance f s (t) is: In the formula, and ψ f is the stress transformation parameter, φ ′ is the effective internal friction angle, and M is the slope of the critical state line.

10. A method for perceiving and predicting the time-varying bearing capacity of static piles based on a physical information network, characterized in that: include: A collection module for collecting pore pressure dissipation data using pile body sensors; The pore pressure dissipation data includes the deployment depth of the sensor, the excess pore water pressure collected by the sensor and the collection time; An inversion module, for inputting the pore pressure dissipation data and calculation points into the PINNs forward and inversion integrated inversion model, and estimating the excess pore water pressure in the back propagation to obtain the final consolidation coefficient; A forward modeling module, used for inputting the final consolidation coefficient and the calculation point into the PINNs forward-inverse integrated forward model, and estimating the excess pore water pressure in the back propagation to obtain the trained PINNs forward-inverse integrated forward model; The prediction module is used to input the test set into the trained PINNs forward and inversion integrated forward model, output the excess pore water pressure dissipation of the pile side soil, and predict the time-varying bearing capacity of the static pile based on the excess pore water pressure dissipation of the pile side soil. The test set consists of a number of calculation points that need to be predicted, and the calculation points include time and radial distance.

Images