CN115492082A - Composite foundation treatment method, equipment and application for deep soft foundation - Google Patents
Composite foundation treatment method, equipment and application for deep soft foundation Download PDFInfo
- Publication number
- CN115492082A CN115492082A CN202211186947.7A CN202211186947A CN115492082A CN 115492082 A CN115492082 A CN 115492082A CN 202211186947 A CN202211186947 A CN 202211186947A CN 115492082 A CN115492082 A CN 115492082A
- Authority
- CN
- China
- Prior art keywords
- soil
- pile
- foundation
- cement
- block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E02—HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
- E02D—FOUNDATIONS; EXCAVATIONS; EMBANKMENTS; UNDERGROUND OR UNDERWATER STRUCTURES
- E02D3/00—Improving or preserving soil or rock, e.g. preserving permafrost soil
- E02D3/02—Improving by compacting
- E02D3/08—Improving by compacting by inserting stones or lost bodies, e.g. compaction piles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W30/00—Technologies for solid waste management
- Y02W30/50—Reuse, recycling or recovery technologies
- Y02W30/91—Use of waste materials as fillers for mortars or concrete
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Structural Engineering (AREA)
- Computing Systems (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Consolidation Of Soil By Introduction Of Solidifying Substances Into Soil (AREA)
- Agronomy & Crop Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Health & Medical Sciences (AREA)
- Environmental & Geological Engineering (AREA)
- Soil Sciences (AREA)
- Biophysics (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Mining & Mineral Resources (AREA)
- Paleontology (AREA)
- Civil Engineering (AREA)
- Biomedical Technology (AREA)
Abstract
The invention belongs to the technical field of foundation engineering treatment, and discloses a composite foundation treatment method, equipment and application for a deep soft foundation. The composite foundation treatment method for the deep soft foundation comprises the following steps: the light cement soil pile is pumped to long spiral drilling equipment through a pipeline to form a hole; pouring light cement soil in the drill lifting process to complete the construction of the light cement soil pile; then setting a mattress layer on the pile top to finally form the light cement-soil pile composite foundation. The light cement soil suspension pile processing technology combines a pile type composite foundation method (process) and a light embankment method (idea), and the light cement soil suspension pile processing technology has certain strength and the density of about 1g/cm 3 The light pile for soft soil foundationAnd (3) carrying out rapid reinforcement, wherein the pile body density is less than the sludge density, the pile body is in a suspension state in the sludge, part of embankment load is offset, and meanwhile, the replacement of the soft foundation by the light pile greatly reduces the self-weight stress of the foundation, so that the lower-load soft soil is effectively unloaded.
Description
Technical Field
The invention belongs to the technical field of foundation engineering treatment, and particularly relates to a composite foundation treatment method, equipment and application for a deep soft foundation.
Background
In the soft foundation settlement deformation, the effective stress principle is as follows: under the action of additional stress, the pressure of the excess pore water is continuously dissipated, and the effective stress is continuously increased; with the dissipation of the excess pore pressure, the free water in the soft soil is continuously discharged, and thus the soft soil is compressed in volume, which is expressed as the occurrence of sedimentation. Wherein the factors influencing the sedimentation deformation include: the additional stress (determining the final settlement of the soft soil layer), the consolidation degree (representing the proportion of the finished settlement to the total settlement at the current moment) and the post-construction settlement (determining the final effect of the roadbed treatment). Furthermore, two methods for controlling the post-construction settlement deformation in the prior art are as follows:
the first method comprises the following steps: reducing the additional stress of the soft soil layer; the common implementation method comprises the following steps: pile type composite foundation method (setting up stirring pile, CFG pile or prestressed pipe pile with certain density, transferring most of the overlying load to pile body, reducing the load and additional stress born by soft soil), light embankment method (adopting light embankment such as fly ash embankment, foam light soil embankment, etc., reducing embankment load and additional stress of soft soil layer).
And the second method comprises the following steps: the consolidation degree is improved when the completion is finished, and the post-construction settlement is controlled; the common implementation method is a drainage preloading method (vertical drainage and vacuum preloading (surcharge preloading) to accelerate the consolidation rate of the soft soil layer, improve the consolidation degree during completion and reduce the settlement after construction.
Through the above analysis, the problems and defects of the prior art are as follows:
(1) In the prior art, a pile type composite foundation method (process) and a light embankment method (idea) are not combined, so that the self-weight stress of the foundation cannot be effectively reduced, and the lower-negative soft soil cannot be effectively unloaded.
(2) The construction process in the prior art is complicated, the cost is high, and the foundation obtained in the prior art can not achieve the effects of drainage, buffering and deformation in coordination with a roadbed.
Disclosure of Invention
In order to overcome the problems in the related art, the disclosed embodiments of the present invention provide a method, device and application for processing a composite foundation for a deep and thick soft foundation.
The technical scheme is as follows: the composite foundation treatment method for the deep soft foundation comprises the following steps:
s1, pumping a light cement soil pile to a long spiral drilling device through a pipeline to form a hole;
s2, pouring lightweight cement soil in the drill lifting process to complete construction of the lightweight cement soil pile;
and S3, arranging a mattress layer on the pile top, and finally forming the light cement-soil pile composite foundation.
In one embodiment, in step S1, the lightweight cement pile is made of foamed lightweight cement, and a lightweight material is formed by introducing a large number of air bubbles into concrete by adding air bubbles, a foaming agent or an air-entraining agent.
In one embodiment, the foamed lightweight cementitious soil has a slurry volume weight of 1g/cm 3 The mass-volume ratio of the foaming agent or the added air entraining agent to the foamed lightweight cement soil is 0.01-0.05 v/m, and the gas of the bubbles is nitrogen or inert gas.
In one embodiment, the foamed lightweight cemented soil preparation method comprises:
i) The method comprises the steps that foam lightweight cement soil mixing equipment is utilized, a first flow sensor is utilized to measure a first flow L1 of foam cement slurry at an outlet of the foam lightweight cement soil mixing equipment, and a second flow L2 of the cement slurry entering the foam lightweight cement soil mixing equipment is measured by the first flow sensor;
ii) a target value of the density ρ 1 of the target foamed cement slurry is set in advance, and the density control device controls the density ρ 1 of the foamed cement slurry by controlling the second flow rate L2 through the following formula, wherein ρ 1= ρ 2 × L2/L1.
In one embodiment, the method for adjusting the density ρ 1, ρ 1= ρ 2 × L2/L1 of the foamed cement slurry by adjusting the second flow rate L2 by the density adjusting device according to the following formula includes:
constructing a foam cement slurry flow prediction model based on time and flow velocity;
determining a density regulation and prediction model framework structure based on a multilayer perceptron;
and step three, performing flow speed per hour modeling of the foam cement slurry based on the integrated learning of the multilayer perceptron, and obtaining a prediction result through a density regulation and control prediction model based on the multilayer perceptron.
In one embodiment, the construction of the time and flow rate based foamed cement slurry flow prediction model comprises:
providing a CNNS modeling of the convolutional neural network based on the flow rate dependence modeling of the convolutional neural network; extracting flow velocity information among the foam cement slurries through a convolutional neural network, and adopting a residual neural network;
the residual neural network adopts a ResNet50 network, and the level parameters of the ResNet50 network are explained as follows:
ZEROPAD block: filling the matrix (3, 3) for zero filling of 3 rows and 3 columns, namely the originally input foam cement slurry matrix is (2, 2) in size, the filled matrix is (5, 5) in size, and the excess part is all zero;
the CONV block: two-dimensional convolution with 64 convolution kernels of size (7, 7) and convolution step size (2, 2);
BatchNorm block: carrying out batch standardization;
a ReLU block: relu activation function, the formula is defined as follows:
MAXFOOL block, AVGPOOL block: maximum pooling layer, size (3, 3), step size (2, 2); average pooling, size (2, 2), step size (1, 1);
CONVBLOCK block: performing matrix addition operation on the results of the previous layer x by a short path, wherein the two-dimensional convolution operation and batch normalization operation are performed to scale the size of the foam cement slurry to the range of [ -1,1], and finally, a Relu activation function is output;
IDBLOCK x n Block: x directly carrying out matrix addition operation with the output result of the previous layer through a 'short path'; n represents a plurality of identical IDBLOCK blocks linked together;
flatten block: flattening the input into one-dimensional foam cement slurry; the size is (M, -1), wherein M represents the number of samples, and-1 represents the input sample matrix foam cement slurry synthesis;
FC block: the size of the full-connection layer is (H, N), wherein H represents the input dimension of the previous layer, and N represents the output dimension of the foam cement slurry required to be predicted;
BigRNN is a deep neural network, and Layer1 and Layer2 are additionally added residual block networks, and the output of Layer2 without an activation function is assumed to be z l+2 Then output a l+2 Is defined as follows:
a l+2 ＝g(z l+2 +a l )；
wherein g represents the Relu activation function; the expanded formula becomes:
a l+2 ＝g(w l+2 x+b l+2 +a l )；
wherein w l+2 、b l+2 The weight and bias of Layer2 Layer; if w is l+2 =0, while b l+2 =0, then a l+2 Will be equal to a l The performance of the network is not changed after the residual block is added;
suppose the output at a neural network that has not undergone batch normalization is z i Wherein i =1,2,3.. N denotes n number of samples, z & i The output result after batch normalization is represented, and the calculation formula is defined as follows:
wherein epsilon represents a minimal number not less than zero, and eta and beta are parameters obtained by neural network learning; suppose the input to the neural network is x t The selected batch size is gamma, wherein gamma is more than 0 and less than or equal to m, and m is the total number of samples, the batch normalization number is
In one embodiment, the calculation of η and β with the minimum batch gradient descent algorithm is as follows:
For t＝1，2，3……n；
at all x t Forward propagation on the basis of the batch normalization technique to obtain z E l And l denotes the ith layer of the neural network, and the respective gradients are calculated using a back propagation technique: dw l ,dη l ,dβ l ；
And (3) updating parameters: w is a l ＝w l -αdw l ，η l ＝η l -αdη l ，β l ＝β l -αdβ l Wherein α represents a learning rate; and obtaining the final needed prediction result.
In one embodiment, in step S3, a mattress layer is placed on top of the pile for drainage, cushioning and deformation in cooperation with the subgrade.
The invention also aims to provide an application of the composite foundation treatment method for the deep and soft foundation in cross-sea bridge construction and soft soil layer traffic engineering pile pier construction.
Another object of the present invention is to provide a composite foundation treatment apparatus for a deep soft foundation, which implements the composite foundation treatment method for a deep soft foundation.
By combining all the technical schemes, the invention has the advantages and positive effects that:
firstly, aiming at the technical problems existing in the prior art and the difficulty in solving the problems, the technical problems to be solved by the technical scheme of the invention are closely combined with the technical scheme to be protected, the results in the research and development process, the foamed cement slurry and the like, the technical problems solved by the technical scheme of the invention are analyzed in detail and deeply, and some creative technical effects brought after the problems are solved are achieved: the foam light soil (foam concrete) related by the invention is a light material formed by introducing a large amount of bubbles into concrete by adding bubbles, a foaming agent or an air entraining agent and the like. The material has a reduced volume weight while maintaining the advantages of concrete, and therefore foamed lightweight soil is often used as an embankment filler. The foamed light soil used in the road construction at present is prepared by a cast-in-place method and a mode of combining physical foaming and chemical foaming. There are certain technical drawbacks. The invention also solves the problems of the complete set of construction process and quality control technology of the light cement-soil pile in the prior art. The introduction of new materials has great influence on the construction process and control measures, and a complete set of technology for site construction and quality control needs to be formed, but the prior art is not perfect.
Secondly, regarding the technical solution as a whole or from the perspective of products, the technical effects and advantages of the technical solution to be protected by the present invention are specifically described as follows: the invention provides a technical theory for processing the lightweight cement soil suspension pile by analyzing the combination of a pile type composite foundation method and a lightweight embankment method. The light cement soil suspension pile processing technology combines a pile type composite foundation method (process) and a light embankment method (idea), and the pile type composite foundation method (process) and the light embankment method (idea) have certain strength and the density of about 1g/cm 3 The light pile can quickly reinforce the soft soil foundation, because the density of the pile body is less than that of the sludge, the pile body is in a suspension state in the sludge to offset part of the embankment load, and meanwhile, the light pile can greatly replace the soft foundationThe self-weight stress of the foundation is reduced, and the lower negative soft soil is effectively unloaded. In the invention, the composite foundation treatment method consists of a light cement soil long spiral cast-in-place pile, soil between piles and a broken stone mattress layer, wherein the light cement soil pile is formed by mixing light cement soil slurry (volume weight 1 g/cm) 3 ) And pumping the pile to long spiral drilling equipment through a pipeline to form a hole, and pouring light cement soil in the drilling lifting process to finish the construction of the light cement soil pile. Then a mattress layer is arranged on the pile top to play the effects of drainage, buffering and deformation in cooperation with the roadbed, and finally the lightweight cement-soil pile composite foundation is formed. The volume weight of the slurry of the foamed lightweight cement soil is 1g/cm 3 (ii) a The mass-volume ratio of the foaming agent or the added air entraining agent to the foamed lightweight cement soil is 0.01-0.05 v/m; the gas of the bubbles is nitrogen or inert gas. By adopting the technical scheme, the gas of the bubbles can be effectively controlled to be the addition of nitrogen or inert gas, and the volume weight of the slurry of the foam lightweight cement soil can be effectively controlled to be 1g/cm 3 Left and right, intelligent degree is high.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure.
Fig. 1 is a flowchart of a composite foundation treatment method for a deep soft foundation according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention more comprehensible, embodiments accompanying figures are described in detail below. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein, but rather should be construed as broadly as the present invention is capable of modification in various respects, all without departing from the spirit and scope of the present invention.
1. Illustrative examples are illustrated:
examples
The embodiment of the invention provides a light cement soil suspension pile processing technical theory by analyzing the combination of a pile type composite foundation method and a light embankment method.
The light cement soil suspension pile processing technology combines a pile type composite foundation method (process) and a light embankment method (idea), and the pile type composite foundation method (process) and the light embankment method (idea) have certain strength and the density of about 1g/cm 3 The light pile can quickly reinforce the soft soil foundation, because the density of the pile body is less than that of the sludge, the pile body is in a suspension state in the sludge to offset part of embankment load, and meanwhile, the replacement of the light pile on the soft foundation greatly reduces the self-weight stress of the foundation, and the soft soil under the load is effectively unloaded.
As shown in fig. 1, the method for processing a composite foundation for a deep soft foundation according to the embodiment of the present invention includes constructing a light cement-soil long spiral cast-in-place pile in combination with inter-pile soil and a gravel cushion layer, and specifically includes:
s101, the lightweight cement-soil pile is formed by lightweight cement-soil slurry (volume weight 1 g/cm) 3 ) Pumping the mixture to a long spiral drilling device through a pipeline to form a hole.
And S102, pouring lightweight cement soil in the drill lifting process to complete construction of the lightweight cement soil pile.
And S103, arranging a mattress layer on the pile top to achieve the effects of drainage, buffering and deformation in cooperation with the roadbed, and finally forming the light cement-soil pile composite foundation.
In a preferred embodiment, in step S101, the lightweight cement pile material is made of foamed lightweight cement, and a lightweight material is formed by introducing a large amount of air bubbles into concrete through the addition of air bubbles, the addition of a foaming agent, or the addition of an air-entraining agent.
The volume weight of the slurry of the foamed lightweight cement soil is 1g/cm 3 (ii) a The mass volume ratio of the foaming agent or the added air entraining agent to the foamed lightweight cement soil is 0.01v/m-0.05v/m; the gas of the bubbles is nitrogen or inert gas.
In a preferred embodiment, the method for manufacturing the foamed lightweight cement-soil comprises the following steps:
i) The method comprises the following steps of (1) carrying out by utilizing a foam lightweight cement soil mixing device:
measuring a first flow L1 of the foamed cement slurry at the outlet of the foamed lightweight cement-soil mixing device by using a first flow sensor, and measuring a second flow L2 of the cement slurry entering the foamed lightweight cement-soil mixing device by using the first flow sensor;
ii) presetting a target value of the density rho 1 of the target foamed cement slurry, and adjusting the density rho 1 of the foamed cement slurry by a density adjusting and controlling device through adjusting a second flow L2 according to the following formula;
ρ1＝ρ2*L2/L1。
in a preferred embodiment, the method for adjusting the density ρ 1 of the foamed cement slurry by adjusting the second flow L2 by the density adjusting device according to the following formula comprises:
constructing a foam cement slurry flow prediction model based on time and flow speed;
determining a density regulation and control prediction model framework structure based on a multilayer perceptron;
thirdly, performing flow speed per hour modeling of the foam cement slurry based on the integrated learning of the multilayer perceptron, and obtaining a prediction result through a density regulation and control prediction model based on the multilayer perceptron;
the construction of the foam cement slurry flow prediction model based on time and flow rate comprises the following steps:
convolutional neural network-based flow velocity dependence modeling
Providing a Convolutional Neural Network (CNNS) for modeling; extracting flow velocity information among the foam cement slurries through a convolutional neural network, and adopting a residual neural network;
a classical ResNet50 network is assumed, the class parameters of which are specified as follows:
ZEROPAD block: padding the matrix (3, 3), zero padding of 3 rows and 3 columns; namely, the originally input foam cement slurry matrix is (2, 2) in size, and the filled foam cement slurry matrix is (5, 5); the excess part is all zero;
the CONV block: two-dimensional convolution with 64 convolution kernels of size (7, 7) and convolution step size (2, 2);
BatchNorm block: carrying out batch standardization;
a ReLU block: relu activation function, the formula is defined as follows:
MAXFOOL block, AVGPOOL block: maximum pooling layer, size (3, 3), step size (2, 2); average pooling, size (2, 2), step size (1, 1);
CONVBLOCK block: performing matrix addition operation on the results of the previous layer x by a short path, wherein the two-dimensional convolution operation and batch normalization operation are performed to scale the size of the foam cement slurry to the range of [ -1,1], and finally, a Relu activation function is output;
IDBLOCK x n block: x directly carrying out matrix addition operation with the output result of the previous layer through a 'short path'; n represents a plurality of identical IDBLOCK blocks linked together;
the Flatten block: flattening the input into one-dimensional foam cement slurry; the size is (M, -1), wherein M represents the number of samples, and-1 represents the input sample matrix foam cement slurry synthesis;
FC block: the size of the full-connection layer is (H, N), wherein H represents the input dimension of the previous layer, and N represents the output dimension of the foam cement slurry required to be predicted;
the residual error neural network ResNet is used for solving the 'jump link' technology appeared by the deep neural network;
the reasons why the ResNet network performs well are summarized below: suppose there is a deeper neural network with x as input and a as output l (ii) a Adding a residual block structure, wherein all the activation functions in the network are Relu activation functions, namely all the outputs are more than or equal to zero;
BigRNN is a deep neural network, and Layer1 and Layer2 are additionally added residual block networks, and the output of Layer2 without an activation function is assumed to be z l+2 Then output a l+2 The formula of (c) is defined as follows:
a l+2 ＝g(z l+2 +a l )；
wherein g represents a Relu activation function; the expanded formula becomes:
a l+2 ＝g(w l+2 x+b l+2 +a l )；
wherein, w l+2 、b l+2 The weights and offsets of Layer 2; if w is l+2 =0, while b l+2 =0, then a l+2 Will be equal to a l The performance of the network is not changed after the residual block is added;
suppose the output at the neural network without batch normalization is z i Wherein i =1,2,3.. N denotes the number of n samples,
the output result after batch normalization is represented, and the calculation formula is defined as follows:
wherein epsilon represents a minimal number not less than zero, and eta and beta are parameters obtained by neural network learning; suppose the input to the neural network is x t The selected batch size is gamma, wherein gamma is more than 0 and less than or equal to m, and m is the total number of samples, the batch normalization quantity is
In one embodiment, the calculation process for η and β using the minimum batch gradient descent algorithm as an example is as follows:
For t＝1，2，3……n；
at all x t Forward propagation is performed;
obtained by using batch normalization technique
l represents the l layer of the neural network;
the individual gradients were calculated using a back propagation technique: dw l ,dη l ,dβ l ；
And (3) updating parameters: w is a l ＝w l -αdw l ，η l ＝η l -αdη l ，β l ＝β l -αdβ l Wherein α represents a learning rate;
and obtaining the final needed prediction result.
In the above embodiments, the description of each embodiment has its own emphasis, and reference may be made to the related description of other embodiments for parts that are not described or recited in any embodiment.
2. The application example is as follows:
application example
The Xianghai bridge program is located in the Changtian district of the Zhuhai, the Jinwan district, the Mantoumen district and Tanzhou town of Zhongshan city, and has a total length of 17.344 km. After the Xianghai bridge is built into a traffic vehicle, the Xianghai bridge will be the second main road connecting the western regions of the urban area, which becomes an important vein for getting through the rapid development of the Chinese and western economy, brings more convenient life for citizens in the surrounding areas, and can relieve traffic pressure in the urban area and treat blockage and slow blockage. The route starts from the neighborhood of a shellfish road made in a winter area of the Zhuhai city, is butted with a branch line of a bridge of the Xianghai, and successively intersects with a Tanzhou town of the West Jingzhongshan city, a sluice area and a Jinwan area of the Zhuhai city, namely a Guanao high speed, a Zhongshan ancient magic highway (planning) and a Jiangzhu high speed, and finally ends at a heart road of the S272 lake, the total length of the route is 20.235km, and the origin-destination mileage is as follows: k1+ 194.855-K21 +429.822, wherein the length of the independent bridge section is 17.548km (including the main bridge, the east approach and the west approach), and the length of the connecting line is 2.890km.
The poor soft land base widely distributed in the Zhuhai area has great influence on the speed and quality of road and bridge engineering construction. Three sections of soft soil foundations are arranged in the foundation engineering of the Shanghai bridge engineering, the total length is 1561.49m, wherein the sections K17+ 668.33-K18 +260 are 591.67m long, and the maximum filling height is 5.82m; k19+ 520-K20 +100 sections, the length is 580m, and the maximum filling height is 5.15m; k21+ 040-K21 +429.82 sections with the length of 389.82m and the maximum filling height of 1.86m.
The soft soil mainly comprises 2-2 layers of silt and 2-3 layers of silt clay: (1) 2-2 layers of sludge (Q) 4 mc ): distributed in K3+ 150-K21 +429.82 alluvial plain region; dark gray, gray black, flowing plastic, a small amount of coarse sand and shell fragments in local parts, and high compressibility; the layer thickness is 5.80-24.80 m, and the average thickness is 15.57m; basic allowable value of bearing capacity of foundation soil ao ]=40kPa; standard value q of frictional resistance of side soil of drilled pile ik =5kPa. (2) 2-3 layers of silty clay (Q4) mc ): distributed in K1+ 200-K1 +550, K2+ 080-K2 +150, K5+ 900-K21 +429.82; grey, grey black, flowing plastic state, local containing rotten wood and a small amount of fine sand, high compressibility; the layer thickness is 0.40-39.80 m, and the average thickness is 14.67m; basic allowable value of bearing capacity of foundation soil ao ]=60kPa; pile side soil frictional resistance standard value q of drilled pile ik ＝8kPa。
On the whole, the soft pearl-sea soil has the characteristics of high water content, large porosity ratio, high compressibility, low strength and low permeability. Due to the special properties of the soft soil, the foundation containing the soft soil has the characteristics of low bearing capacity, large settling amount, long settling and solidifying stabilization time and the like, increases the construction difficulty, and is a key part in the quality control of roadbed engineering.
At present, in the design of TJ4 standard sections of the Shanghai bridge, in order to ensure the stability of filling soil and meet the requirements of bearing capacity and post-construction settlement, a main line adopts vacuum preloading and a vertical drainage body and is combined with a light embankment for treatment, and the traditional cement mixing pile and foam light soil combination mode is adopted for foundation treatment in the splicing stage.
The foamed light soil replaces the traditional compacted road foundation soil, has the effects of reducing overlying load and improving foundation rigidity, ensures that the settlement of roads and bridge head sections meets the standard requirement, but has three problems which are not solved yet: 1) During the construction process, the water loss can cause the drying and cracking of the foamed light soil and influence the performance of the light roadbed; 2) The long-term performance degradation problem under the action of dry-wet cycle of traffic dynamic load and atmospheric environment; 3) And finally, the settlement is difficult to predict, so that the design method has defects. As a mature foundation treatment method, the cement soil stirring method still has great limitations in the use process: 1) The depth of the bottom surface of the soft soil is not more than 15m, and the pile end of the vertical bearing mixing pile in the deep soft soil is difficult to enter a relatively hard layer; 2) The plasticity index of the treated soft soil is generally not more than 22.
The application of the lightening technology in the foundation can avoid the influence of traffic dynamic load and atmospheric action, the method is combined with the traditional foundation treatment method, the self-weight stress of the foundation is greatly reduced, the lower negative deep soft soil is effectively unloaded, the cement soil pile can be designed in a suspension mode, and the method is a new idea for foundation treatment. The light cement soil is a porous light solidified soil formed by mixing cement (solidified material), water, soil and additive in a certain proportion to form slurry, then mixing the slurry with a certain amount of stable bubbles and stirring the mixture and finally hardening the mixture. Compared with the traditional cement soil, the lightweight cement soil has smaller volume weight, can effectively reduce the self-weight load of the foundation, increases the bearing capacity of the composite foundation in a phase-changing manner, reduces the settlement and deformation of the foundation, particularly soft soil which cannot be effectively treated at the lower part of a reinforcing area, and has huge potential advantages when being applied to the treatment of deep soft soil foundations (larger than 20 m). However, at present, no engineering practice for reference exists for the application of the lightweight cement-soil pile, and the construction process of the lightweight cement-soil pile needs to be analyzed.
The foamed light soil is used as roadbed filling, a plurality of engineering practices exist in China, but the application effect of the foamed light soil to the engineering construction of the Shanghai bridge is not ideal.
The processing method provided by the embodiment of the invention is mainly characterized by complete construction process, and good effect when being applied to construction of the Shanghai bridge.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, and any modification, equivalent replacement, and improvement made by those skilled in the art within the technical scope of the present invention disclosed herein, which is within the spirit and principle of the present invention, should be covered by the present invention.
Claims (10)
1. A composite foundation treatment method for deep and thick soft foundation is characterized by comprising the following steps:
s1, pumping a light cement soil pile to a long spiral drilling device through a pipeline to form a hole;
s2, pouring lightweight cement soil in the drill lifting process to complete construction of the lightweight cement soil pile;
and S3, arranging a mattress layer on the pile top, and finally forming the light cement-soil pile composite foundation.
2. The method for treating a composite foundation for a deep soft foundation according to claim 1, wherein the lightweight cement-soil piles are made of foamed lightweight cement-soil, and lightweight materials formed after a large number of air bubbles are introduced into the concrete by adding air bubbles, a foaming agent or an air-entraining agent in step S1.
3. The method of claim 2, wherein the foamed lightweight cement soil has a slurry volume weight of 1g/cm 3 The mass-volume ratio of the foaming agent or the added air entraining agent to the foamed lightweight cement soil is 0.01-0.05 v/m, and the gas of the bubbles is nitrogen or inert gas.
4. The method for treating a composite foundation for a deep soft foundation according to claim 3, wherein the method for preparing foamed lightweight cement soil comprises:
i) The method comprises the steps of measuring a first flow L1 of foam cement slurry at an outlet of the foam lightweight cement-soil mixing device by using a first flow sensor, and measuring a second flow L2 of the cement slurry entering the foam lightweight cement-soil mixing device by using the first flow sensor;
ii) a target value of the density ρ 1 of the target foamed cement slurry is set in advance, and the density control device controls the density ρ 1 of the foamed cement slurry by controlling the second flow rate L2 through the following formula, wherein ρ 1= ρ 2 × L2/L1.
5. The composite foundation treatment method for the deep soft foundation according to claim 4, wherein the method for adjusting the density ρ 1 of the foamed cement slurry by adjusting the second flow rate L2 by the density control device through the following formula, wherein ρ 1= ρ 2 × L2/L1 comprises:
constructing a foam cement slurry flow prediction model based on time and flow velocity;
determining a density regulation and control prediction model framework structure based on a multilayer perceptron;
and step three, performing flow speed per hour modeling of the foam cement slurry based on the integrated learning of the multilayer perceptron, and obtaining a prediction result through a density regulation and control prediction model based on the multilayer perceptron.
6. The composite foundation treatment method for the deep soft foundation according to claim 5, wherein the construction of the time and flow rate based foam cement slurry flow prediction model comprises:
providing a CNNS modeling of the convolutional neural network based on the flow rate dependence modeling of the convolutional neural network; extracting flow velocity information among the foam cement slurries through a convolutional neural network, and adopting a residual neural network;
the residual neural network adopts a ResNet50 network, and the level parameters of the ResNet50 network are described as follows:
ZEROPAD block: filling the matrix (3, 3) for zero filling of 3 rows and 3 columns, namely the originally input foam cement slurry matrix is (2, 2) in size, the filled matrix is (5, 5) in size, and the excess part is all zero;
the CONV block: two-dimensional convolution with 64 convolution kernels of size (7, 7) and convolution step size (2, 2);
BatchNorm block: carrying out batch standardization;
a ReLU block: relu activation function, the formula is defined as follows:
MAXFOOL block, AVGPOOL block: maximum pooling layer, size (3, 3), step size (2, 2); average pooling, size (2, 2), step size (1, 1);
CONVBLOCK block: performing matrix addition operation on the results of the previous layer x by a short path, wherein the two-dimensional convolution operation and batch normalization operation are performed to scale the size of the foam cement slurry to the range of [ -1,1], and finally, a Relu activation function is output;
IDBLOCK x n Block: x directly carrying out matrix addition operation with the output result of the previous layer through a 'short path'; n represents a plurality of identical IDBLOCK blocks linked together;
the Flatten block: flattening the input into one-dimensional foam cement slurry; the size is (M, -1), wherein M represents the number of samples, and-1 represents the synthesis of foam cement slurry of an input sample matrix;
FC block: the size of the full-connection layer is (H, N), wherein H represents the input dimension of the previous layer, and N represents the output dimension of the foam cement slurry required to be predicted;
BigRNN is a deep neural network, and Layer1 and Layer2 are additionally added residual block networks, and the output of Layer2 without an activation function is assumed to be z l+2 Then output a l+2 Is defined as follows:
a l+2 ＝g(z l+2 +a l )；
wherein g represents a Relu activation function; the expanded formula becomes:
a l+2 ＝g(w l+2 x+b l+2 +a l )；
wherein, w l+2 、b l+2 The weights and offsets of Layer 2; if w is l+2 =0, while b l+2 =0, then a l+2 Will be equal to a l The performance of the network is not changed after the residual block is added;
suppose the output at the neural network without batch normalization is z i Wherein i =1,2,3.. N represents n sample numbers, z · en i The output result after batch normalization is shown, and the calculation formula is defined as follows:
wherein epsilon represents a minimum number not less than zero, eta and beta are parameters obtained by neural network learning; suppose the input to the neural network is x t The selected batch size is gamma, wherein gamma is more than 0 and less than or equal to m, and m is the total number of samples, the batch normalization quantity is
7. The composite foundation treatment method for the deep soft foundation according to claim 5, wherein the calculation process of eta and beta by the minimum batch gradient descent algorithm is as follows:
For t＝1，2，3……n；
at all x t Forward propagation is carried out, and the forward propagation is obtained by using a batch normalization technology
l denotes the l-th layer of the neural network, the respective gradients are calculated using a back propagation technique: dw l ,dη l ,dβ l ；
Updating parameters: w is a l ＝w l -αdw l ，η l ＝η l -αdη l ，β l ＝β l -αdβ l Wherein α represents a learning rate; and obtaining the final needed prediction result.
8. The method for treating a composite foundation for a deep soft foundation according to claim 1, wherein a mattress layer is provided on a pile top for drainage, buffering and deformation in cooperation with a roadbed in the step S3.
9. Use of a method of composite foundation treatment for deep soft foundations according to any one of claims 1 to 7 in cross-sea bridge construction and soft soil layer traffic engineering pier construction.
10. A composite ground treatment apparatus for a deep soft foundation, which implements the composite ground treatment method for a deep soft foundation according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211186947.7A CN115492082B (en) | 2022-09-28 | 2022-09-28 | Composite foundation treatment method, equipment and application for deep soft foundation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211186947.7A CN115492082B (en) | 2022-09-28 | 2022-09-28 | Composite foundation treatment method, equipment and application for deep soft foundation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115492082A true CN115492082A (en) | 2022-12-20 |
| CN115492082B CN115492082B (en) | 2023-09-19 |
Family
ID=84472942
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211186947.7A Active CN115492082B (en) | 2022-09-28 | 2022-09-28 | Composite foundation treatment method, equipment and application for deep soft foundation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115492082B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105200879A (en) * | 2015-10-16 | 2015-12-30 | 北京科技大学 | Water-permeable foam concrete pile compound foundation |
| CN105780764A (en) * | 2016-03-30 | 2016-07-20 | 东南大学 | Light cemented soil composite mixing pile |
| CN111364453A (en) * | 2020-03-24 | 2020-07-03 | 中交一公局第七工程有限公司 | Method for reinforcing deep soft soil foundation by light cement soil long spiral cast-in-place pile |
| JP2021169757A (en) * | 2020-04-14 | 2021-10-28 | 太平洋セメント株式会社 | Quality prediction method of compacting material for strengthening ground |
| CN114036821A (en) * | 2021-10-21 | 2022-02-11 | 北京科技大学 | Thickener control method and device based on non-deterministic hidden space model |
-
2022
- 2022-09-28 CN CN202211186947.7A patent/CN115492082B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105200879A (en) * | 2015-10-16 | 2015-12-30 | 北京科技大学 | Water-permeable foam concrete pile compound foundation |
| CN105780764A (en) * | 2016-03-30 | 2016-07-20 | 东南大学 | Light cemented soil composite mixing pile |
| CN111364453A (en) * | 2020-03-24 | 2020-07-03 | 中交一公局第七工程有限公司 | Method for reinforcing deep soft soil foundation by light cement soil long spiral cast-in-place pile |
| JP2021169757A (en) * | 2020-04-14 | 2021-10-28 | 太平洋セメント株式会社 | Quality prediction method of compacting material for strengthening ground |
| CN114036821A (en) * | 2021-10-21 | 2022-02-11 | 北京科技大学 | Thickener control method and device based on non-deterministic hidden space model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115492082B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Akbulut et al. | Estimating the groutability of granular soils: a new approach | |
| CN109914173B (en) | Foam light soil pouring table back structure in expressway reconstruction and extension project and construction method thereof | |
| CN101220588A (en) | Soft base processing method | |
| CN206902488U (en) | A kind of soft soil foundation high-filled embankment of effectively control settlement after construction | |
| CN110983892A (en) | Foamed light soil and pouring construction process thereof | |
| Zhu et al. | Utilization of excavated loess and gravel soil in controlled low strength material: Laboratory and field tests | |
| Arai et al. | Interpretation of concrete top base foundation behaviour on soft ground by coupled stress flow finite element analysis | |
| CN111321725A (en) | Collapsible loess foundation reinforcing method and built variable-rigidity ramming pile | |
| CN102839683B (en) | Granular pile-permeable concrete stake dual compound foundation and processing method | |
| CN115492082A (en) | Composite foundation treatment method, equipment and application for deep soft foundation | |
| CN103195051B (en) | Limestone soil and pervious concrete pile composite foundation and treatment method thereof | |
| SE512058C2 (en) | Procedure for soil stabilization during road construction | |
| CN110172979B (en) | Roller compacted concrete damming method | |
| He et al. | Consolidation behavior and microstructure properties of cement-treated dredged soil during the stress curing | |
| Mulyono | Laboratory experiment: pervious concrete for permeable pavement, focus in compressive strength and permeability | |
| CN114508109A (en) | Construction method using foam lightweight concrete as foundation pit backfill material | |
| CN106906812A (en) | A kind of method that utilization river and lake silt builds artificial ecologic island | |
| Umanath et al. | Assessment of innovative dented sheet liner on the improvement of hydraulic properties of pervious concrete pile | |
| Li et al. | Application of artificial solidified crust–PVDs method in ground improvement: A case study | |
| Mahmood et al. | Effect of lime, cement, and lime-cement stabilisation on low to medium plasticity clayey soil | |
| CN115583821A (en) | Material for lightweight cement soil cast-in-place pile, preparation method and application | |
| Mulyono | Sustainable material for local road using pervious concrete with various aggregates | |
| CN109778913A (en) | A kind of steel-shelled concrete can intelligent positioning casting system | |
| Wang et al. | Physical model test study on strength characteristics of gravel pile composite foundation | |
| CN110485229B (en) | Construction process for bubble mixed light soil |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |