Assembly type horizontal electroosmosis vacuum preloading dehydration system and process

Abstract

The invention provides an assembly type horizontal electroosmosis vacuum preloading dehydration system and process, and relates to the technical field of road construction. The assembly type horizontal electroosmosis vacuum preloading dehydration system comprises the following modules: a horizontal pipe distribution module, an electroosmosis driving module, a vacuum strengthening module, an intelligent regulation and control module and a structure supporting module, and the horizontal pipe distribution module is used for optimizing the distribution density of the water collecting pipes through a genetic algorithm by adopting a self-adaptive path planning algorithm on the basis of the soil permeability coefficient, simulating the corrosion resistance of different materials in combination with a Monte Carlo method, and finally determining an optimal pipe distribution scheme and generating an optimized pipe distribution scheme. The arrangement density of the water collecting pipes is optimized through a self-adaptive path planning algorithm, multi-target parameter collaborative optimization is achieved in combination with a genetic algorithm and Monte Carlo simulation, the drainage pipe network layout efficiency is improved, the material corrosion resistance prediction accuracy is improved, a dynamic electric field regulation and control method integrates a PID control algorithm and electrochemical impedance spectroscopy analysis, and the comprehensive performance of the drainage pipe network is improved. And electrode voltage precision adjustment is realized.

Description

Technical Field

[0001] This invention relates to the field of highway construction technology, specifically to a prefabricated horizontal electroosmosis vacuum preloading dewatering system and process. Background Technology The technology field of highway construction mainly involves innovation in soft soil foundation treatment, especially in special geological conditions such as silt and fill soil with a water content exceeding 40%. Traditional drainage consolidation methods suffer from technical bottlenecks such as long construction period (usually 6-12 months) and low drainage efficiency (daily drainage <5L / m²). Current technological development focuses on the research of the combined action mechanism of electroosmosis and vacuum, the development of modular rapid construction equipment, and the integration of intelligent monitoring and control systems. It focuses on solving core problems in subgrade treatment such as pore water pressure dissipation rate (target value >0.8kPa / h) and post-construction settlement control accuracy (requirement <150mm). Among them, this technology refers to an innovative method to accelerate the dehydration of soft soil through the synergistic effect of prefabricated horizontal electroosmosis and vacuum preloading. It is mainly applied to the treatment of soft soil foundations with high water content. Its core value lies in shortening the traditional 3-6 month consolidation cycle to 15-30 days, while increasing the foundation bearing capacity to over 150kPa. It is particularly suitable for engineering fields that require the rapid formation of stable subgrades, such as coastal new city road construction and port yard expansion.

[0002] Traditional drainage consolidation methods employ vertical drainage board arrangements, resulting in a single drainage path and a small effective radius of influence. This necessitates repeated construction to achieve the designed degree of consolidation, significantly extending the construction period. Existing electroosmosis technologies mostly utilize fixed-voltage power supplies, lacking a response mechanism to dynamic changes in soil resistivity, which easily leads to electrode polarization and energy waste. Vacuum preloading construction relies on empirical formulas to estimate drainage board spacing, failing to consider the spatial variability of soil permeability, resulting in significant fluctuations in actual drainage efficiency. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a prefabricated horizontal electroosmosis vacuum pre-compression dewatering system and process, which solves the problems of traditional drainage consolidation methods that use vertical drainage board arrangements, resulting in a single drainage path, a small effective radius of influence, and the need for multiple repeated constructions to achieve the required degree of consolidation, thus significantly extending the construction period.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solution: a prefabricated horizontal electroosmosis vacuum pre-compression dehydration system, comprising the following modules: a horizontal pipe laying module, an electroosmosis drive module, a vacuum enhancement module, an intelligent control module, and a structural support module; The horizontal pipe laying module, based on the soil permeability coefficient, adopts an adaptive path planning algorithm, optimizes the layout density of water collection pipes through a genetic algorithm, and combines the Monte Carlo method to simulate the corrosion resistance of different materials, finally determining the optimal pipe laying scheme and generating an optimized pipe laying scheme. The horizontal pipe laying module includes a path planning submodule, a material matching submodule, and a water permeability verification submodule. The electroosmosis drive module, based on an optimized pipe layout scheme, adopts a dynamic electric field control method and adjusts the electrode voltage in real time through a PID control algorithm. Combined with electrochemical impedance spectroscopy analysis, a double-layer model is established to calculate the ion migration rate and generate an ion migration map. The electroosmosis drive module includes an electric field generation submodule, an impedance monitoring submodule, a corrosion prevention control submodule, and a power management submodule; The vacuum enhancement module, based on ion migration maps, employs a negative pressure gradient control method, adjusts the vacuum pump speed using a fuzzy logic algorithm, establishes a Darcy's law correction model to calculate the effective drainage radius, and generates a composite drainage field. The vacuum enhancement module includes a negative pressure generation submodule, a flow monitoring submodule, and a gas-water ionization module; The intelligent control module, based on the composite drainage field, adopts a multi-parameter coupling analysis method, predicts the dewatering progress through a BP neural network, and constructs a three-dimensional visualization model by combining digital twin technology to realize closed-loop control of current, vacuum degree, and flow rate, and generate a real-time operating condition model. The intelligent control module includes a data acquisition submodule, an algorithm decision-making submodule, a remote communication submodule, and a fault diagnosis submodule; The structural support module, based on a real-time working condition model, adopts the finite element dynamic verification method, simulates the stress distribution of the frame through ANSYS, and optimizes the connection nodes by combining modular design theory to achieve the structural stability of seismic standards and generate a dynamic support system. The structural support module includes a stress analysis submodule, a rapid assembly submodule, and an anti-corrosion treatment submodule.

[0005] Preferably, the path planning submodule uses a genetic optimization algorithm to calculate the optimal spacing of the water collection pipes based on soil permeability coefficient data, verifies the layout scheme through Monte Carlo simulation, and generates optimized pipe layout parameters. The material matching submodule, based on optimized pipe laying parameters, uses fuzzy comprehensive evaluation method to compare the corrosion resistance index of stainless steel and polyethylene to establish a material selection decision tree and generate an optimal material matrix. Based on the optimized material matrix, the permeability coefficient of the permeable fabric is calculated using a modified Darcy's law model. The water flow velocity is verified by particle image velocimetry technology, and an optimized pipe laying scheme is generated.

[0006] Preferably, the electric field generation submodule, based on the optimized pipe layout scheme, uses the finite element analysis method to construct a three-dimensional electric field model, optimizes the electrode spacing through the gradient descent algorithm, and generates an electric field distribution map; The impedance monitoring submodule measures soil resistivity using electrochemical impedance spectroscopy based on the electric field distribution map, establishes a dynamic double-layer model, and generates impedance characteristic spectra. The corrosion control submodule, based on impedance characteristic spectrum, uses an anodic oxidation process to generate an oxide film on the surface of titanium alloy, evaluates the corrosion resistance performance through Tafel curve, and generates a long-lasting electrode assembly. The power management submodule, based on the long-life electrode assembly, uses a PID control algorithm to adjust the constant current source output, eliminates current fluctuations through Kalman filtering, and generates ion migration maps.

[0007] Preferably, the negative pressure generation submodule uses fuzzy PID control to adjust the vacuum pump speed based on ion migration maps, establishes Bernoulli equation to calculate the negative pressure gradient, and generates a dynamic negative pressure field. The flow monitoring submodule, based on a dynamic negative pressure field, uses ultrasonic flow detection technology to measure the drainage speed and generates purified water flow data by denoising the signal through wavelet transform. The air-water ionization module, based on purified water flow data, uses cyclone separation technology to design a two-stage filtration device, optimizes separation efficiency through computational fluid dynamics, and generates a composite drainage field.

[0008] Preferably, the data acquisition submodule, based on the composite drainage field, uses multi-sensor fusion technology to collect current, vacuum, and flow data to generate a set of operating condition features; The algorithm decision-making submodule, based on the operating condition feature set, uses a deep reinforcement learning algorithm to train the control model, and uses Q-learning to achieve adaptive parameter adjustment and generate an optimized control strategy. The remote communication submodule, based on an optimized control strategy, uses the LoRaWAN protocol to build a wireless transmission network, ensures data security through AES-256 encryption, and generates a remote command stream; The fault diagnosis submodule, based on remote command streams, uses Bayesian network diagnostics to construct a fault tree model, identifies abnormal operating conditions through vibration spectrum analysis, and generates a real-time operating condition model.

[0009] Preferably, the stress analysis submodule, based on a real-time working condition model, uses the finite element dynamic analysis method to calculate the stress distribution of the frame, avoids resonance through modal analysis, and generates a safe load spectrum; The rapid assembly submodule is developed using a modular design method based on the safety load spectrum. The quick-connect snap-fit ​​structure is optimized through orthogonal experiments to generate prefabricated connection components. The anti-corrosion treatment submodule, based on prefabricated connection components, uses hot-dip galvanizing to form a coating, verifies corrosion resistance through salt spray testing, and generates a dynamic support system.

[0010] A prefabricated horizontal electroosmosis vacuum pre-compression dehydration process includes the following steps: S1: Based on the soil permeability characteristics, a horizontal directional pipe laying process is adopted. The axis of the water collection pipe is determined by a laser positioning instrument, and the stainless steel water collection pipe is accurately laid by a hydraulic jacking device. The spacing is controlled by an automatic pipe laying machine with PID adjustment, and the mechanical wrapping of the permeable cloth is completed simultaneously to generate a directional drainage network. S2: Based on the directional drainage network, a bipolar gradient electroosmosis process is adopted. A dynamic electric field is established through a constant current source, and the electrode spacing is adjusted in real time by applying electrochemical impedance monitoring technology. The surface of the titanium alloy anode plate is subjected to micro-arc oxidation treatment, and the cathode plate is equipped with an automatic descaling device to generate ion migration channels. S3: Based on ion migration channels, it adopts intelligent vacuum coupling technology, establishes a negative pressure gradient model by controlling the vacuum pump group through fuzzy PID, monitors the drainage volume in real time by combining wireless sensor network, and uses digital twin technology to construct a three-dimensional dewatering progress model to generate a stable dewatering field.

[0011] Preferably, the generation of a directional drainage network based on S1 includes the following steps: S101: Based on the on-site geological conditions, the dielectric constant of the deep soil layer is measured using ground-penetrating radar scanning technology. A stratigraphic profile is generated through three-dimensional point cloud reconstruction, and a three-dimensional geological model is produced. S102: Based on a three-dimensional geological model, a genetic algorithm is used to optimize the process and calculate the optimal pipe layout path. The drainage efficiency is verified through computational fluid dynamics simulation, and a set of pipe network layout parameters is generated. S103: Based on the pipeline layout parameter set, the laying of DN100 stainless steel pipes is controlled by hydraulic synchronous jacking process, and the connection between pipes is completed by automatic welding robot to generate directional drainage network.

[0012] Preferably, the generation of ion migration channels based on S2 includes the following steps: S201: Based on the coordinate data of the directional drainage network, titanium alloy electrodes are installed using laser positioning and guidance technology. Positioning accuracy is achieved through a six-axis robotic arm, and an electrode space matrix is ​​generated. S202: Based on the electrode space matrix, a dynamic electric field is established using a constant current source gradient control process. The field strength distribution is monitored in real time using a four-probe method to generate dynamic electric field parameters. S203: Based on dynamic electric field parameters, a ceramic film is generated on the anode surface using a micro-arc oxidation surface treatment process. The breakdown voltage is verified by an electrochemical workstation to generate a long-lasting electrode assembly. S204: Based on a long-lasting electrode assembly, the power supply frequency is adjusted in real time using impedance matching control technology, and the ion migration rate is optimized through phase-sensitive detection technology to generate ion migration channels.

[0013] Preferably, the generation of a stable dehydration field based on S3 includes the following steps: S301: Based on ion migration channel data, a fuzzy proportional-integral-derivative control process is used to adjust the vacuum pump group, establish a negative pressure gradient, and generate a dynamic negative pressure field. S302: Based on a dynamic negative pressure field, a dehydration prediction model is constructed using digital twin coupling technology. Parameter closed-loop control is achieved through edge computing devices to generate a stable dehydration field.

[0014] This invention provides a prefabricated horizontal electroosmosis vacuum pre-compression dehydration system and process. It has the following beneficial effects: 1. This invention optimizes the density of water collection pipes through an adaptive path planning algorithm, and achieves multi-objective parameter collaborative optimization by combining genetic algorithms and Monte Carlo simulation, thereby improving the efficiency of drainage network layout and the accuracy of material corrosion resistance prediction. The dynamic electric field control method integrates PID control algorithm and electrochemical impedance spectroscopy analysis to achieve precise adjustment of electrode voltage, reduce ion migration rate control error, and reduce energy consumption compared to traditional methods. The negative pressure gradient control method integrates fuzzy logic algorithm and Darcy's law correction model to shorten the response time of vacuum pump speed adjustment and control the effective drainage radius calculation error within a low range. The multi-parameter coupling analysis method combines BP neural network and digital twin technology to construct a three-variable closed-loop control system of current-vacuum degree-flow rate, which improves the sampling frequency of working parameters and the accuracy of dehydration progress prediction. The finite element dynamic verification method improves the seismic resistance of the structural support system, increases assembly efficiency, and increases the number of reuses through modular node optimization design. The three-dimensional visualization model realizes dynamic monitoring of all elements in the construction process, stabilizes the pore water pressure dissipation rate, controls the post-construction settlement within a small range, and improves the foundation bearing capacity test value.

[0015] 2. This invention integrates high-precision positioning and automated construction technology through horizontal directional pipe laying, achieving precise control of the water collection pipe layout, significantly improving construction efficiency and material compatibility. Dynamic electric field control technology effectively reduces energy consumption and stabilizes ion migration rate by optimizing electrode parameters in real time, extending the service life of electrode components. Intelligent vacuum coupling technology combined with advanced algorithm models greatly shortens the negative pressure gradient establishment time, improving the real-time performance and coverage density of drainage monitoring. Three-dimensional visualization technology comprehensively simulates the dehydration process, significantly improving the pore water pressure dissipation rate and foundation bearing capacity, ensuring the accuracy of post-construction settlement control. The coordinated design of the directional drainage network and electric field channels expands the negative pressure range, resulting in a breakthrough improvement in overall drainage efficiency. Attached Figure Description

[0016] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a schematic diagram of the main steps of the present invention; Figure 3 This is a detailed schematic diagram of S1 of the present invention; Figure 4 This is a detailed schematic diagram of S2 of the present invention; Figure 5 This is a detailed schematic diagram of S3 of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example: like Figure 1-5 As shown, this embodiment of the invention provides a prefabricated horizontal electroosmosis vacuum pre-compression dehydration system, comprising the following modules: a horizontal pipe laying module, an electroosmosis drive module, a vacuum enhancement module, an intelligent control module, and a structural support module; The horizontal pipe laying module, based on the soil permeability coefficient, adopts an adaptive path planning algorithm, optimizes the layout density of water collection pipes through a genetic algorithm, and combines the Monte Carlo method to simulate the corrosion resistance of different materials, finally determining the optimal pipe laying scheme and generating an optimized pipe laying scheme. The horizontal pipe laying module includes a path planning submodule, a material matching submodule, and a water permeability verification submodule. The electroosmosis drive module, based on an optimized pipe layout scheme, adopts a dynamic electric field control method and adjusts the electrode voltage in real time through a PID control algorithm. Combined with electrochemical impedance spectroscopy analysis, a double-layer model is established to calculate the ion migration rate and generate an ion migration map. The electroosmosis drive module includes an electric field generation submodule, an impedance monitoring submodule, a corrosion prevention control submodule, and a power management submodule; The vacuum enhancement module, based on ion migration maps, employs a negative pressure gradient control method, adjusts the vacuum pump speed using a fuzzy logic algorithm, establishes a Darcy's law correction model to calculate the effective drainage radius, and generates a composite drainage field. The vacuum enhancement module includes a negative pressure generation submodule, a flow monitoring submodule, and a gas-water ionization module; The intelligent control module, based on the composite drainage field, adopts a multi-parameter coupling analysis method, predicts the dewatering progress through a BP neural network, and constructs a three-dimensional visualization model by combining digital twin technology to realize closed-loop control of current, vacuum degree, and flow rate, and generate a real-time operating condition model. The intelligent control module includes a data acquisition submodule, an algorithm decision-making submodule, a remote communication submodule, and a fault diagnosis submodule; The structural support module, based on a real-time working condition model, adopts the finite element dynamic verification method, simulates the stress distribution of the frame through ANSYS, and optimizes the connection nodes by combining modular design theory to achieve the structural stability of seismic standards and generate a dynamic support system. The structural support module includes a stress analysis submodule, a rapid assembly submodule, and an anti-corrosion treatment submodule.

[0019] The path planning submodule uses soil permeability coefficient data and a genetic optimization algorithm to calculate the optimal spacing of the water collection pipes. The layout scheme is verified through Monte Carlo simulation, and optimized pipe layout parameters are generated. Based on soil permeability coefficient data, permeability coefficient test values ​​were collected at different locations on site. When the permeability coefficient was 1×10⁻⁻⁻⁻⁴ 5 Up to 5×10⁻ 5 When the permeability is in the range of cm / s, the optimal spacing interval is defined as 0.8–1.5 m. A genetic algorithm is used to initialize the population size to 50 groups. The fitness function is set as a weighted sum of drainage efficiency and material cost, with weight coefficients of 0.7 and 0.3 respectively. The top 20% of dominant individuals are retained using roulette wheel selection. A single-point crossover operation is performed to generate offspring, with a mutation probability of 5%. After 50 iterations, the optimal solution is output. For example, in a case of fill soil with a permeability coefficient of 3 × 10⁻⁻⁻⁶. 5 The algorithm converged at a speed of cm / s, and the spacing was determined to be 1.2m. Then, 1000 layout schemes were randomly generated through Monte Carlo simulation. The drainage efficiency threshold was set to >85%, and the proportion of schemes that met the requirements was selected. In one simulation, 978 schemes met the standard. Finally, the scheme with drainage efficiency of 88.5% and the lowest cost was selected to generate optimized pipe laying parameters.

[0020] The material matching submodule, based on optimized pipe laying parameters, uses fuzzy comprehensive evaluation method to compare the corrosion resistance index of stainless steel and polyethylene to establish a material selection decision tree and generate an optimal material matrix. Based on the pipe diameter and burial depth data in the optimized pipe layout parameters, the corrosion rate experimental data of stainless steel and polyethylene were extracted. When the chloride ion concentration > 5000 mg / L, the annual corrosion amount of stainless steel was defined as 0.02-0.05 mm, and that of polyethylene as 0.12-0.15 mm. An evaluation index weight matrix was constructed, with corrosion resistance weighted at 0.6, cost at 0.25, and compressive strength at 0.15. The score for the stainless steel sample was calculated as (0.6×90+0.25×70+0.15×85) = 84.5, and that of polyethylene as (0.6×75+0.25×90+0.15×70) = 78.5. A decision threshold of 80 points was set. In a coastal project, the chloride ion concentration reached 8000 mg / L, and stainless steel with a score of 83 points was selected, generating a preferred material matrix.

[0021] Based on the optimized material matrix, the permeability coefficient of the permeable fabric is calculated using a modified Darcy's law model. The water flow velocity is verified by particle image velocimetry technology, and an optimized pipe laying scheme is generated.

[0022] Based on the polypropylene permeable fabric parameters in the optimized material matrix, its original permeability coefficient was measured to be 2×10⁻³cm / s. Darcy's law gradient term was adjusted according to the 1.2m spacing between water collection pipes, and a pipe diameter influence factor of 0.92 was introduced into the correction formula. The effective permeability coefficient was calculated to be 1.84×10⁻³cm / s. In the verification stage, tracer particles with a diameter of 50μm were used. Water flow was captured at 1000 frames / s using a high-speed camera. The average particle displacement of adjacent frames was extracted to be 0.12mm, and the actual flow velocity was calculated to be 1.2cm / s. The deviation from the theoretical value was <5%, thus generating an optimized pipe layout scheme.

[0023] The electric field generation submodule, based on the optimized pipe layout scheme, uses the finite element analysis method to construct a three-dimensional electric field model, optimizes the electrode spacing through the gradient descent algorithm, and generates an electric field distribution map. Based on the coordinates and diameter parameters of the water collection pipes in the optimized pipe layout scheme, the coordinate data of the electrode arrangement area is extracted. When the spacing between the water collection pipes is 1.2m, the initial spacing between the electrodes is defined as 1.5m. The area is discretized into 500,000 elements using a tetrahedral mesh method. The initial soil resistivity is set to 20Ω·m. The applied voltage boundary conditions are 60V for the anode and 0V for the cathode. The electric field intensity distribution is calculated iteratively. Convergence is determined when the difference between two adjacent iterations is less than 0.5V / m. For example, in a coastal soft soil foundation case, the electric field intensity stabilized at 5.2V / m after 15 iterations. Subsequently, the gradient descent algorithm is used to adjust the electrode spacing with a step size of 0.1m. The objective function is the standard deviation of the electric field uniformity. The optimization stops when the standard deviation decreases from 1.8V / m to 0.9V / m, and the electric field distribution map is generated.

[0024] The impedance monitoring submodule measures soil resistivity using electrochemical impedance spectroscopy based on the electric field distribution map, establishes a dynamic double-layer model, and generates impedance characteristic spectra. Based on the equipotential line data in the electric field distribution map, 10 characteristic points were selected for impedance spectrum measurement. The frequency scanning range was set to 0.1Hz-10kHz, the excitation signal amplitude was 5mV, and phase angle and impedance modulus data were collected. When the phase angle showed a characteristic value of -45° at 100Hz, the double layer effect was determined to exist. When establishing the equivalent circuit model, the solution resistance Rs was set to 15Ω, the charge transfer resistance Rct to 80Ω, and the double layer capacitance Cdl to 0.02F. The parameters were adjusted by the constraint that the fitting error was less than 5%. In one measurement, the goodness of fit reached 96%, and the impedance characteristic spectrum was generated.

[0025] The corrosion control submodule, based on impedance characteristic spectrum, uses an anodic oxidation process to generate an oxide film on the surface of titanium alloy, evaluates the corrosion resistance performance through Tafel curve, and generates a long-lasting electrode assembly. Based on the charge transfer resistance value in the impedance characteristic spectrum, when Rct is less than 50Ω, it is determined that enhanced corrosion protection is required. A pulsed anodizing process is adopted, with the electrolyte set to 15% sulfuric acid solution, the current density set to 15A / dm², and the treatment time set to 30 minutes. The oxide film thickness is calculated to be 50μm using Coulomb's law. Subsequently, Tafel curve testing is performed, with a scan rate set to 1mV / s, and the self-corrosion current density is obtained as 0.12μA / cm². When the value is less than 0.2μA / cm², the corrosion protection is deemed qualified. After treatment, the self-corrosion potential of a certain titanium alloy electrode shifts positively by 0.25V, generating a long-lasting electrode assembly.

[0026] The power management submodule, based on the long-life electrode assembly, uses a PID control algorithm to adjust the constant current source output, eliminates current fluctuations through Kalman filtering, and generates ion migration maps.

[0027] Based on the impedance characteristics data of the long-life electrode assembly, the PID parameters were set as follows: proportional band 4%, integral time 0.5 min, derivative time 0.1 min, initial current 5 A. When the detected current fluctuation exceeded ±0.3 A, Kalman filtering was activated. The process noise covariance was set to 0.01, and the observation noise covariance was set to 0.001. After 10 iterations, the current standard deviation decreased from 0.4 A to 0.08 A. In one regulation, the output current stabilized in the range of 4.95 ± 0.05 A, and an ion mobility spectrum was generated.

[0028] The negative pressure generation submodule, based on ion migration maps, uses fuzzy PID control to adjust the vacuum pump speed, establishes the Bernoulli equation to calculate the negative pressure gradient, and generates a dynamic negative pressure field. Based on the current density distribution data in the ion mobility spectrum, the coordinates of the region with the maximum current density are extracted. When the detected value exceeds 5A / m², the universe of discourse of the fuzzy PID is set to [-3,3], the quantization factor is 0.4, the initial value of the proportional coefficient is set to 0.8, the integral time constant is 120s, and the derivative time constant is 30s. The vacuum pump speed deviation is divided into 7 levels through the membership function. In a certain control, a negative pressure deviation of -2kPa is detected. The speed increment is output as 200rpm through fuzzy inference. The flow velocity in the pipe section is calculated by combining the Bernoulli equation. When the pipe diameter is 100mm, the pressure difference ΔP and the flow velocity v satisfy ΔP=0.5ρv². Taking the water density as 1000kg / m³, the measured v=1.2m / s corresponds to ΔP=720Pa, which deviates from the actual sensor reading of 750Pa by 4%, and a dynamic negative pressure field is generated.

[0029] The flow monitoring submodule, based on a dynamic negative pressure field, uses ultrasonic flow detection technology to measure the drainage speed and generates purified water flow data by denoising the signal through wavelet transform. Based on the pressure gradient data in the dynamic negative pressure field, a pair of 2MHz ultrasonic transducers were installed in a DN100 pipeline. The downstream propagation time t1 and the upstream propagation time t2 were set. When the flow velocity v=(L / 2cosθ)(1 / t1-1 / t2), the channel length L=0.5m and the angle θ=45° were taken. The measured t1=332μs and t2=336μs were obtained, and the instantaneous flow velocity was calculated to be 0.98m / s. After collecting 100 sets of data, the db4 wavelet was used for 3-level decomposition. The threshold coefficient was set to 0.1, and the fluctuations less than 0.05m / s in the detail coefficients D1-D3 were filtered out. After reconstructing the signal, the flow velocity was stabilized at 0.95±0.03m / s, and the purified water flow data was generated.

[0030] The air-water ionization module, based on purified water flow data, uses cyclone separation technology to design a two-stage filtration device, optimizes separation efficiency through computational fluid dynamics, and generates a composite drainage field.

[0031] Based on the suspended solids concentration values ​​in the purified water flow data, when the detected value is >500mg / L, the first-stage hydrocyclone is set with a tangential inlet velocity of 6m / s, a diameter of 200mm, and a cone angle of 15°. The particle size d50 is calculated as √(9μD) / (πρsω²r). The water viscosity is 0.001Pa·s, the particle density is 2650kg / m³, the angular velocity is ω=300rad / s, and the radius is r=0.08m, resulting in d50=45μm. The cone angle of the second-stage hydrocyclone is adjusted to 10°, and the inlet velocity is increased to 8m / s, reducing d50 to 28μm. CFD simulation verification shows that the removal rate of particles larger than 150μm is 98%, generating a composite drainage field.

[0032] The data acquisition submodule, based on the composite drainage field, uses multi-sensor fusion technology to collect current, vacuum, and flow data, and generates a set of operating condition features. Based on the coordinates of vulnerable areas in the pipe network of the composite drainage field, Hall current sensors (range 0-10A, accuracy ±0.5%), piezoresistive vacuum sensors (range 0-100kPa, accuracy ±0.2%), and electromagnetic flowmeters (range 0-5m / s, accuracy ±1%) were deployed. The acquisition frequency was set to 10Hz. When the current data fluctuation exceeded ±0.3A, Kalman filtering was activated. The process noise covariance was set to 0.01, and the observation noise covariance was set to 0.001. The three sets of sensor data were weighted and fused with weighting coefficients of 0.4, 0.3, and 0.3. In one acquisition, the current was 4.2A, the vacuum degree was 85kPa, and the flow velocity was 1.2m / s. After normalization, a three-dimensional feature vector [0.42, 0.85, 0.24] was formed, generating a working condition feature set.

[0033] The algorithm decision-making submodule, based on the operating condition feature set, uses a deep reinforcement learning algorithm to train the control model, and uses Q-learning to achieve adaptive parameter adjustment and generate an optimized control strategy. Based on historical data sequences from the operating condition feature set, the state space is defined as a three-dimensional discrete space with current deviation ±0.5A, vacuum deviation ±5kPa, and flow deviation ±0.2m / s. The action space is set as a two-dimensional operation of voltage adjustment ±1V and pump speed adjustment ±5%. The reward function is R=100-10×|current deviation|-8×|vacuum deviation|-5×|flow deviation|. The Q-table is initialized as a 50×50×50×20×20 matrix, with a learning rate α=0.2, a discount factor γ=0.9, and an exploration rate ε=0.1. When the state [+0.3A, -4kPa, +0.1m / s] is detected, the action corresponding to the maximum Q value (+0.8V, -3% pump speed) is selected, and the Q value is updated Q(s,a)=Q(s,a)+α[R+γmaxQ(s',a')-Q(s,a)]. After 1000 iterations, the strategy converges, generating an optimized control strategy.

[0034] The remote communication submodule, based on an optimized control strategy, uses the LoRaWAN protocol to build a wireless transmission network, ensures data security through AES-256 encryption, and generates a remote command stream; Based on the device address encoding table in the optimized control strategy, 32 LoRa terminal nodes were configured, with a center frequency of 868MHz, a spreading factor of SF=10, and a bandwidth of 125kHz. Each data packet was given a 4-byte frame header (containing device ID and serial number). A 32-byte key was generated when AES-256 encryption was used. The "voltage 4.5V, pump speed 82%" command was converted into hexadecimal code 0x120xAF 0x34, encrypted in CBC mode, and then a 48-byte ciphertext was generated. A 2-byte checksum was added through CRC-16 verification. The actual packet loss rate was <0.5% in a transmission distance of 3km, and a remote command stream was generated.

[0035] The fault diagnosis submodule, based on remote command streams, uses Bayesian network diagnostics to construct a fault tree model, identifies abnormal operating conditions through vibration spectrum analysis, and generates a real-time operating condition model.

[0036] Based on the device response time-series data in the remote command stream, a Bayesian network of 12 nodes (including power failure, sensor failure, mechanical jamming, etc.) is constructed. The prior probabilities are set to power failure 0.05 and sensor failure 0.08. When the vibration spectrum exhibits an amplitude > 0.5g feature in the 100-150Hz range, the conditional probability P(mechanical jamming|high-frequency vibration) is updated to 0.85. Accelerometer data is collected (sampling rate 5kHz), and the 125Hz component amplitude of 0.6g is extracted through FFT transformation. The posterior probability P(mechanical jamming) is calculated to be 0.78. An alarm is triggered when the threshold of 0.7 is exceeded, and a real-time operating condition model is generated.

[0037] The stress analysis submodule, based on a real-time working condition model, uses the finite element dynamic analysis method to calculate the stress distribution of the frame, avoids resonance through modal analysis, and generates a safe load spectrum. Based on the load time history data in the real-time working condition model, the combined working condition of maximum axial force 120kN and bending moment 85kN·m was extracted. The steel frame was discretized into 5000 hexahedral elements, and the material elastic modulus was set to 206GPa and Poisson's ratio to 0.3. Fixed constraints were applied to the foundation connection points. When the calculated maximum equivalent stress was 285MPa, compared with the yield strength of Q355 steel (355MPa), and with a safety factor of 1.25, the stress distribution was deemed acceptable. During modal analysis, the frequency scanning range was set to 0-50Hz. The overlap between the third modal frequency (48Hz) and the equipment operating frequency (47Hz) was found to be >90%. The beam section height was adjusted from 400mm to 450mm, and after recalculation, the modal frequency shifted to 51Hz, generating a safe load spectrum.

[0038] The rapid assembly submodule is developed using a modular design method based on the safety load spectrum. The quick-connect snap-fit ​​structure is optimized through orthogonal experiments to generate prefabricated connection components. Based on the bolt preload requirements in the safety load spectrum, the quick-connect buckle was designed with a contact surface inclination angle of 55°, a tooth pitch of 8mm, and a tooth depth of 2mm. Two materials, ABS and Nylon 66, were selected for orthogonal experiments. A factor level table was set up with three factors and three levels, including material type (2 levels), assembly angle (30° / 45° / 60°), and pressing force (3kN / 5kN / 7kN). Nine sets of experiments were conducted using an L9(3^4) orthogonal table. The fifth set (Nylon 66, 45°, 5kN) was found to have the shortest assembly time (12 seconds) and the largest disassembly force (8.5kN). Range analysis showed that the material type contributed 58%. After optimization, the assembly process was determined to be: insertion of locating pin → 45° rotation → 5kN hydraulic pressing, generating a prefabricated connection assembly.

[0039] The anti-corrosion treatment submodule, based on prefabricated connection components, uses hot-dip galvanizing to form a coating, verifies corrosion resistance through salt spray testing, and generates a dynamic support system.

[0040] Based on the surface roughness data of the prefabricated connecting components, when the Ra value is in the range of 6.3-12.5μm, the oxide layer is removed by pickling (15% hydrochloric acid concentration, 25℃ temperature, 8 minutes). The temperature of the hot-dip zinc bath is controlled at 450±5℃, the immersion time is 3 minutes, and the coating thickness is measured to be 85-100μm by a magnetic thickness gauge. The salt spray test is set with 5% NaCl solution, pH 6.5-7.2, chamber temperature 35℃, and continuous spraying for 48 hours. The white rust area on the sample surface is <5%, and the red rust area is <0.1%, which determines that the corrosion resistance level reaches ISO 9223 C4 level, and a dynamic support system is generated.

[0041] A prefabricated horizontal electroosmosis vacuum pre-compression dehydration process includes the following steps: S1: Based on the soil permeability characteristics, a horizontal directional pipe laying process is adopted. The axis of the water collection pipe is determined by a laser positioning instrument, and the stainless steel water collection pipe is accurately laid by a hydraulic jacking device. The spacing is controlled by an automatic pipe laying machine with PID adjustment, and the mechanical wrapping of the permeable cloth is completed simultaneously to generate a directional drainage network. S2: Based on the directional drainage network, a bipolar gradient electroosmosis process is adopted. A dynamic electric field is established through a constant current source, and the electrode spacing is adjusted in real time by applying electrochemical impedance monitoring technology. The surface of the titanium alloy anode plate is subjected to micro-arc oxidation treatment, and the cathode plate is equipped with an automatic descaling device to generate ion migration channels. S3: Based on ion migration channels, it adopts intelligent vacuum coupling technology, establishes a negative pressure gradient model by controlling the vacuum pump group through fuzzy PID, monitors the drainage volume in real time by combining wireless sensor network, and uses digital twin technology to construct a three-dimensional dewatering progress model to generate a stable dewatering field.

[0042] The generation of a directional drainage network based on S1 includes the following steps: S101: Based on the on-site geological conditions, the dielectric constant of the deep soil layer is measured using ground-penetrating radar scanning technology. A stratigraphic profile is generated through three-dimensional point cloud reconstruction, and a three-dimensional geological model is produced. Based on the water content and soil type data from the field geological conditions, a 1GHz ground-penetrating radar antenna was used to collect reflected wave signals every 0.5m along the survey line. When an abnormal area of ​​dielectric constant in the silt layer was detected, the scanning depth resolution was set to 0.1m. Interpolation was performed on the cross-section with a travel time difference of Δt=2.3ns for the reflected wave to generate a grid point cloud with a spacing of 0.2m×0.2m. A three-dimensional surface was constructed by Delaunay triangulation. In a coastal soft soil foundation project, the dielectric constant ε_r=25 was measured at a depth of 5m, and the converted water content w=(ε_r-5) / 0.3=66.7%, which deviated from the soil test result of 65.2% by 2.3%. A three-dimensional geological model was generated.

[0043] S102: Based on a three-dimensional geological model, a genetic algorithm is used to optimize the process and calculate the optimal pipe layout path. The drainage efficiency is verified through computational fluid dynamics simulation, and a set of pipe network layout parameters is generated. Based on the permeability coefficient distribution data in the 3D geological model, the fitness function F = 0.6 × drainage efficiency + 0.4 × construction cost is defined. The population size is initialized with 100 path schemes, the crossover probability is set to 0.8, the mutation probability is set to 0.05, and after 50 iterations, the top 10% elite individuals are selected. In one scheme, the pipe length is optimized from 120m to 98m, and the drainage efficiency is increased from 82% to 87%. Through CFD simulation, the pipe diameter is set to 100mm and the slope is 0.5%. The Reynolds number Re = 2300 is calculated to be in the laminar flow critical state. When the flow velocity v = 0.35m / s, the drainage volume Q = π × (0.05)^2 × 0.35 = 0.0027m³ / s. The daily drainage volume of 233m³ is verified to meet the standard, and the pipeline layout parameter set is generated.

[0044] S103: Based on the pipeline layout parameter set, the laying of DN100 stainless steel pipes is controlled by hydraulic synchronous jacking process, and the connection between pipes is completed by automatic welding robot to generate directional drainage network.

[0045] Based on the coordinate elevation data of the pipeline layout parameter set, four 200t hydraulic jacking devices were configured to work synchronously. The jacking speed was set to 2cm / min, the correction threshold was ±3cm for lateral deviation and ±0.5% for longitudinal slope deviation. When the 15th pipe section was detected to have a lateral offset of 2.8cm, the No. 2 hydraulic cylinder was activated to increase the pressure to 5MPa for correction. The automatic welding robot was set to a current of 120A, a voltage of 22V, and a wire feeding speed of 5m / min. After the circumferential weld was completed, the porosity was detected by X-ray and found to be <1.5%, thus generating a directional drainage pipeline network.

[0046] The generation of S2-based ion migration channels includes the following steps: S201: Based on the coordinate data of the directional drainage network, titanium alloy electrodes are installed using laser positioning and guidance technology. Positioning accuracy is achieved through a six-axis robotic arm, and an electrode space matrix is ​​generated. Based on the three-dimensional coordinates of the water collection pipe nodes in the directional drainage network coordinate data, a reference coordinate system was established using a 635nm wavelength laser positioning system. The repeatability of the six-axis robotic arm was set to ±0.1mm. When the coordinate deviation of the 15th electrode installation point reached 0.15mm, the PID correction algorithm was triggered to adjust the joint angle. After three iterations, the positioning error was reduced to 0.05mm. After the installation of 24 sets of titanium alloy electrodes was completed, the coordinates of the center points of each electrode were recorded to form an 8×3 matrix, generating the electrode space matrix.

[0047] S202: Based on the electrode space matrix, a dynamic electric field is established using a constant current source gradient control process. The field strength distribution is monitored in real time using a four-probe method to generate dynamic electric field parameters. Based on the adjacent electrode spacing data in the electrode space matrix, the initial output current of the constant current source was set to 5A, with a gradient step size of 0.5A / min. When the spacing between adjacent electrode pairs was 1.2m, the control voltage was gradually increased from 12V to 18V. When using the four-probe method for measurement, the probe spacing was set to 0.3m, and the measured potential difference ΔV=4.2V was obtained. The electric field strength E=ΔV / d=14V / m was calculated. The current was adjusted to 5.3A to stabilize the electric field strength in the range of 15±0.5V / m, thus generating dynamic electric field parameters.

[0048] S203: Based on dynamic electric field parameters, a ceramic film is generated on the anode surface using a micro-arc oxidation surface treatment process. The breakdown voltage is verified by an electrochemical workstation to generate a long-lasting electrode assembly. Based on the maximum field strength value in the dynamic electric field parameters, the micro-arc oxidation electrolyte was configured as 8 g / L sodium silicate and 2 g / L potassium hydroxide. The pulse frequency was set to 500 Hz, the duty cycle to 30%, and the treatment time to 20 minutes. When the anolyte current density reached 15 A / dm², the oxide film growth rate was monitored to be 2 μm / min. After the treatment, a three-electrode system was used for testing. The reference electrode was a saturated calomel electrode, and the scan rate was 1 mV / s. The breakdown voltage was measured to increase from the initial 350 V to 850 V, thus generating a long-lasting electrode assembly.

[0049] S204: Based on a long-lasting electrode assembly, the power supply frequency is adjusted in real time using impedance matching control technology, and the ion migration rate is optimized through phase-sensitive detection technology to generate ion migration channels.

[0050] Based on the impedance phase angle data of the long-life electrode assembly, the target impedance matching tolerance is set to ±5°. When a phase angle deviation of -8° is detected under 50Hz operating conditions, the power supply frequency is adjusted to 52Hz. The signal phase is detected at a resolution of 0.1Hz through a lock-in amplifier. After optimization, the ion migration rate is increased from 1.2mm / s to 1.5mm / s, and an ion migration channel is generated.

[0051] The generation of a stable dehydration field based on S3 includes the following steps: S301: Based on ion migration channel data, a fuzzy proportional-integral-derivative control process is used to adjust the vacuum pump group, establish a negative pressure gradient, and generate a dynamic negative pressure field. Based on the current density distribution in the ion migration channel data, the coordinates of the maximum current region are extracted. The deviation domain of the fuzzy PID is set to [-3kPa, +3kPa], with quantization factors K_e=0.5, K_de=0.3, proportional coefficient K_p=0.8, integral time T_i=120s, and derivative time T_d=30s. When a negative pressure deviation of -2.5kPa is detected, the vacuum pump speed increment is output by 300rpm through fuzzy inference. After adjustment, the negative pressure gradient increases from -0.8kPa / m to -1.2kPa / m. In the case of soft soil foundation, the drainage rate increases from 0.6m³ / h to 0.9m³ / h, generating a dynamic negative pressure field.

[0052] S302: Based on a dynamic negative pressure field, a dehydration prediction model is constructed using digital twin coupling technology. Parameter closed-loop control is achieved through edge computing devices to generate a stable dehydration field.

[0053] Based on pore water pressure monitoring data in a dynamic negative pressure field, a digital twin mesh with 5000 nodes was constructed, and the soil permeability coefficient was set to k = 1 × 10⁻ 6 m / s, compression modulus E_s=5MPa, the model parameters are updated every 5 minutes by the edge computing device. When the pore water pressure in a certain area drops from 80kPa to 65kPa, the vacuum setting value is automatically adjusted from -90kPa to -95kPa. After three iterations, the drainage consolidation degree reaches 92%, and a stable dewatering field is generated.

[0054] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A prefabricated horizontal electroosmosis vacuum pre-compression dehydration system, characterized in that... It includes the following modules: horizontal pipe laying module, electroosmosis drive module, vacuum strengthening module, intelligent control module, and structural support module; The horizontal pipe laying module, based on the soil permeability coefficient, adopts an adaptive path planning algorithm, optimizes the layout density of water collection pipes through a genetic algorithm, and combines the Monte Carlo method to simulate the corrosion resistance of different materials, finally determining the optimal pipe laying scheme and generating an optimized pipe laying scheme. The horizontal pipe laying module includes a path planning submodule, a material matching submodule, and a water permeability verification submodule. The electroosmosis drive module, based on an optimized pipe layout scheme, adopts a dynamic electric field control method and adjusts the electrode voltage in real time through a PID control algorithm. Combined with electrochemical impedance spectroscopy analysis, a double-layer model is established to calculate the ion migration rate and generate an ion migration map. The electroosmosis drive module includes an electric field generation submodule, an impedance monitoring submodule, a corrosion prevention control submodule, and a power management submodule; The vacuum enhancement module, based on ion migration maps, employs a negative pressure gradient control method, adjusts the vacuum pump speed using a fuzzy logic algorithm, establishes a Darcy's law correction model to calculate the effective drainage radius, and generates a composite drainage field. The vacuum enhancement module includes a negative pressure generation submodule, a flow monitoring submodule, and a gas-water ionization module; The intelligent control module, based on the composite drainage field, adopts a multi-parameter coupling analysis method, predicts the dewatering progress through a BP neural network, and constructs a three-dimensional visualization model by combining digital twin technology to realize closed-loop control of current, vacuum degree, and flow rate, and generate a real-time operating condition model. The intelligent control module includes a data acquisition submodule, an algorithm decision-making submodule, a remote communication submodule, and a fault diagnosis submodule; The structural support module, based on a real-time working condition model, adopts the finite element dynamic verification method, simulates the stress distribution of the frame through ANSYS, and optimizes the connection nodes by combining modular design theory to achieve the structural stability of seismic standards and generate a dynamic support system. The structural support module includes a stress analysis submodule, a rapid assembly submodule, and an anti-corrosion treatment submodule.

2. The prefabricated horizontal electroosmosis vacuum pre-compression dehydration system according to claim 1, characterized in that: The path planning submodule uses soil permeability coefficient data and a genetic optimization algorithm to calculate the optimal spacing of the water collection pipes. The layout scheme is verified through Monte Carlo simulation, and optimized pipe layout parameters are generated. The material matching submodule, based on optimized pipe laying parameters, uses fuzzy comprehensive evaluation method to compare the corrosion resistance index of stainless steel and polyethylene to establish a material selection decision tree and generate an optimal material matrix. Based on the optimized material matrix, the permeability coefficient of the permeable fabric is calculated using a modified Darcy's law model. The water flow velocity is verified by particle image velocimetry technology, and an optimized pipe laying scheme is generated.

3. The prefabricated horizontal electroosmosis vacuum pre-compression dehydration system according to claim 1, characterized in that: The electric field generation submodule, based on the optimized pipe layout scheme, uses the finite element analysis method to construct a three-dimensional electric field model, optimizes the electrode spacing through the gradient descent algorithm, and generates an electric field distribution map. The impedance monitoring submodule measures soil resistivity using electrochemical impedance spectroscopy based on the electric field distribution map, establishes a dynamic double-layer model, and generates impedance characteristic spectra. The corrosion control submodule, based on impedance characteristic spectrum, uses an anodic oxidation process to generate an oxide film on the surface of titanium alloy, evaluates the corrosion resistance performance through Tafel curve, and generates a long-lasting electrode assembly. The power management submodule, based on the long-life electrode assembly, uses a PID control algorithm to adjust the constant current source output, eliminates current fluctuations through Kalman filtering, and generates ion migration maps.

4. The prefabricated horizontal electroosmosis vacuum pre-compression dehydration system according to claim 1, characterized in that: The negative pressure generation submodule, based on ion migration maps, uses fuzzy PID control to adjust the vacuum pump speed, establishes the Bernoulli equation to calculate the negative pressure gradient, and generates a dynamic negative pressure field. The flow monitoring submodule, based on a dynamic negative pressure field, uses ultrasonic flow detection technology to measure the drainage speed and generates purified water flow data by denoising the signal through wavelet transform. The air-water ionization module, based on purified water flow data, uses cyclone separation technology to design a two-stage filtration device, optimizes separation efficiency through computational fluid dynamics, and generates a composite drainage field.

5. The prefabricated horizontal electroosmosis vacuum pre-compression dehydration system according to claim 1, characterized in that: The data acquisition submodule, based on the composite drainage field, uses multi-sensor fusion technology to collect current, vacuum, and flow data, and generates a set of operating condition features. The algorithm decision-making submodule, based on the operating condition feature set, uses a deep reinforcement learning algorithm to train the control model, and uses Q-learning to achieve adaptive parameter adjustment and generate an optimized control strategy. The remote communication submodule, based on an optimized control strategy, uses the LoRaWAN protocol to build a wireless transmission network, ensures data security through AES-256 encryption, and generates a remote command stream; The fault diagnosis submodule, based on remote command streams, uses Bayesian network diagnostics to construct a fault tree model, identifies abnormal operating conditions through vibration spectrum analysis, and generates a real-time operating condition model.

6. The prefabricated horizontal electroosmosis vacuum pre-compression dehydration system according to claim 1, characterized in that: The stress analysis submodule, based on a real-time working condition model, uses the finite element dynamic analysis method to calculate the stress distribution of the frame, avoids resonance through modal analysis, and generates a safe load spectrum. The rapid assembly submodule is developed using a modular design method based on the safety load spectrum. The quick-connect snap-fit ​​structure is optimized through orthogonal experiments to generate prefabricated connection components. The anti-corrosion treatment submodule, based on prefabricated connection components, uses hot-dip galvanizing to form a coating, verifies corrosion resistance through salt spray testing, and generates a dynamic support system.

7. A prefabricated horizontal electroosmosis vacuum pre-compression dehydration process, characterized in that... Includes the following steps: S1: Based on the soil permeability characteristics, a horizontal directional pipe laying process is adopted. The axis of the water collection pipe is determined by a laser positioning instrument, and the stainless steel water collection pipe is accurately laid by a hydraulic jacking device. The spacing is controlled by an automatic pipe laying machine with PID adjustment, and the mechanical wrapping of the permeable cloth is completed simultaneously to generate a directional drainage network. S2: Based on the directional drainage network, a bipolar gradient electroosmosis process is adopted. A dynamic electric field is established through a constant current source, and the electrode spacing is adjusted in real time by applying electrochemical impedance monitoring technology. The surface of the titanium alloy anode plate is subjected to micro-arc oxidation treatment, and the cathode plate is equipped with an automatic descaling device to generate ion migration channels. S3: Based on ion migration channels, it adopts intelligent vacuum coupling technology, establishes a negative pressure gradient model by controlling the vacuum pump group through fuzzy PID, monitors the drainage volume in real time by combining wireless sensor network, and uses digital twin technology to construct a three-dimensional dewatering progress model to generate a stable dewatering field.

8. The prefabricated horizontal electroosmosis vacuum pre-compression dehydration process according to claim 7, characterized in that: The generation of a directional drainage network based on S1 includes the following steps: S101: Based on the on-site geological conditions, the dielectric constant of the deep soil layer is measured using ground-penetrating radar scanning technology. A stratigraphic profile is generated through three-dimensional point cloud reconstruction, and a three-dimensional geological model is produced. S102: Based on a three-dimensional geological model, a genetic algorithm is used to optimize the process and calculate the optimal pipe layout path. The drainage efficiency is verified through computational fluid dynamics simulation, and a set of pipe network layout parameters is generated. S103: Based on the pipeline layout parameter set, the laying of DN100 stainless steel pipes is controlled by hydraulic synchronous jacking process, and the connection between pipes is completed by automatic welding robot to generate directional drainage network.

9. The prefabricated horizontal electroosmosis vacuum pre-compression dehydration process according to claim 7, characterized in that: The generation of S2-based ion migration channels includes the following steps: S201: Based on the coordinate data of the directional drainage network, titanium alloy electrodes are installed using laser positioning and guidance technology. Positioning accuracy is achieved through a six-axis robotic arm, and an electrode space matrix is ​​generated. S202: Based on the electrode space matrix, a dynamic electric field is established using a constant current source gradient control process. The field strength distribution is monitored in real time using a four-probe method to generate dynamic electric field parameters. S203: Based on dynamic electric field parameters, a ceramic film is generated on the anode surface using a micro-arc oxidation surface treatment process. The breakdown voltage is verified by an electrochemical workstation to generate a long-lasting electrode assembly. S204: Based on a long-lasting electrode assembly, the power supply frequency is adjusted in real time using impedance matching control technology, and the ion migration rate is optimized through phase-sensitive detection technology to generate ion migration channels.

10. The prefabricated horizontal electroosmosis vacuum pre-compression dehydration process according to claim 7, characterized in that: The generation of a stable dehydration field based on S3 includes the following steps: S301: Based on ion migration channel data, a fuzzy proportional-integral-derivative control process is used to adjust the vacuum pump group, establish a negative pressure gradient, and generate a dynamic negative pressure field. S302: Based on a dynamic negative pressure field, a dehydration prediction model is constructed using digital twin coupling technology. Parameter closed-loop control is achieved through edge computing devices to generate a stable dehydration field.

Images