A foundation pit dewatering dynamic optimization method and system based on intelligent regulation

Abstract

This invention relates to the field of foundation pit dewatering technology, specifically to a dynamic optimization method and system for foundation pit dewatering based on intelligent control. The method includes the following steps: dynamically filtering real-time water level data to obtain water level prediction deviation; dynamically correcting the permeability coefficient through pore water pressure to obtain dynamic geological parameters; combining the water level prediction deviation, the dynamic geological parameters, and rainfall infiltration recharge to obtain precipitation demand characteristics; calculating a safety-efficiency trade-off coefficient based on precipitation efficiency and risk coefficient; and solving the dynamic dewatering scheme using an improved particle swarm optimization algorithm based on the precipitation demand characteristics and the safety-efficiency trade-off coefficient. This invention achieves high-precision water level control, reduces safety risks and energy consumption, and is adaptable to various scenarios such as soft soil and sandy soil, providing an effective solution for intelligent control of foundation pit dewatering through multi-source data dynamic feature extraction, scene-adaptive weight optimization, and multi-dimensional risk coupling assessment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of foundation pit dewatering, in particular to a dynamic optimization method and system for foundation pit dewatering based on intelligent regulation. BACKGROUND

[0002] The core goal of foundation pit dewatering, a key process in foundation pit excavation construction, is to control the underground water level in and around the pit within a safe range to avoid risks such as pit instability and surrounding building subsidence. The existing optimization methods for foundation pit dewatering have the following core defects:

[0003] Fixed thresholds or static algorithms are used to process core parameters such as water level and geology, which cannot adapt to changes in water level trends, periodic fluctuations (such as rainfall / tides), and dynamic evolution of parameters caused by soil consolidation, resulting in large water level prediction errors; risk assessment is one-sided, ignoring key pit parameters such as retaining structure horizontal displacement, pit bottom heave, and soil moisture content, which can easily miss risks such as pit bottom instability and retaining collapse; and the scene adaptation is insufficient, using a unified weight model for different geologies and excavation depths, resulting in large errors in the permeability coefficient of transitional soil and lagging in deep pit subsidence risk warning.

[0004] In summary, the existing methods cannot achieve the multi-objective balance of safety, efficiency, and energy consumption, and there is an urgent need for a dynamic optimization method for foundation pit dewatering that integrates dynamic feature coupling, scene adaptive weight, and intelligent optimization algorithm. SUMMARY

[0005] To address the deficiencies of existing methods and the needs of practical applications, the present application provides a dynamic optimization method for foundation pit dewatering based on intelligent regulation, which includes the following steps:

[0006] Dynamic filtering of real-time water level data to obtain water level prediction deviation; dynamic correction of permeability coefficient by pore water pressure to obtain dynamic geological parameters; combination of water level prediction deviation, dynamic geological parameters, and rainfall infiltration recharge to obtain dewatering demand characteristics; calculation of safety-efficiency trade-off coefficient based on dewatering efficiency and risk coefficient; and solving of dynamic dewatering scheme using an improved particle swarm optimization algorithm based on dewatering demand characteristics and safety-efficiency trade-off coefficient.

[0007] The present application dynamically corrects the permeability coefficient by pore water pressure, breaking the static limitations of traditional geological parameters and fitting the actual soil consolidation; integrates water level deviation, dynamic geological parameters, and rainfall recharge to quantify dewatering demand, determines the safety-efficiency trade-off relationship based on dewatering efficiency and risk coefficient, and finally outputs a dynamic scheme using an improved particle swarm optimization algorithm, effectively solving the problems of large water level error, rigid geological parameters, one-sided risk assessment, and high energy consumption, achieving high-precision water level control, ensuring foundation pit construction safety, saving engineering costs, and providing a practical technical solution for intelligent regulation of foundation pit dewatering.

[0008] Optionally, the dynamic filtering process is used to process real-time water level data to obtain a water level prediction deviation, including the following steps:

[0009] A water level trend slope, fluctuation amplitude and periodicity feature are introduced to calculate a dynamic water level attenuation coefficient; a state transition matrix is optimized according to the dynamic water level attenuation coefficient, and a water level prediction deviation is obtained based on a Kalman filtering algorithm. By introducing the water level trend slope, fluctuation amplitude and periodicity feature to calculate the dynamic water level attenuation coefficient, the limitations of traditional static attenuation coefficients are broken through, and the water level nonlinear change law is fitted. Then, the Kalman filtering state transition matrix is optimized accordingly, and the water level dynamic change is accurately modeled.

[0010] Optionally, the dynamic correction of the permeability coefficient by the pore water pressure is used to obtain dynamic geological parameters, including the following steps:

[0011] A first geological correction coefficient is obtained by using a permeability dominance index; and dynamic geological parameters are obtained by combining the first geological correction coefficient and the pore water pressure. The first geological correction coefficient is determined by using the permeability dominance index (combining initial permeability coefficient, pore ratio and other parameters), which breaks through the limitations of traditional fixed weights and adapts to different soil characteristics. Then, the permeability coefficient is dynamically corrected by combining the pore water pressure, which is related to the mechanical parameters of the soil and avoids errors of static geological parameters. This is conducive to accurately obtaining dynamic geological parameters, providing a reliable basis for calculating the dewatering rate, and ensuring the rationality of the dewatering scheme for the foundation pit.

[0012] Optionally, the water level prediction deviation, the dynamic geological parameters and the rainfall infiltration recharge are combined to obtain dewatering demand characteristics, including the following steps:

[0013] A water level recharge amount model is constructed, and the rainfall infiltration recharge is obtained by using the water level recharge amount model; and based on a fuzzy neural network fusion algorithm, the water level prediction deviation, the dynamic geological parameters and the rainfall infiltration recharge are combined to obtain dewatering demand characteristics. By constructing the water level recharge amount model, the rainfall infiltration recharge is accurately calculated, and the estimation deviation of the recharge amount caused by traditional neglect of meteorological interference is avoided. Then, the water level prediction deviation, the dynamic geological parameters and the rainfall recharge are fused by using the fuzzy neural network, which solves the problems of multi-feature nonlinear coupling and dimension difference. This greatly improves the quantization accuracy of the dewatering demand, avoids insufficient or excessive dewatering caused by inaccurate demand estimation, provides a reliable target basis for subsequent dynamic dewatering scheme optimization, and ensures the pertinence and rationality of the dewatering regulation and control for the foundation pit.

[0014] Optionally, the safety efficiency trade-off coefficient is calculated according to the dewatering efficiency and the risk coefficient, including the following steps:

[0015] This invention assesses the dewatering efficiency of a cluster of dewatering wells; calculates the risk warning coefficient for the surrounding environment; and, based on a multi-objective ranking algorithm, obtains a safety-efficiency trade-off coefficient by combining the dewatering efficiency and the risk coefficient. First, it evaluates the efficiency of the dewatering well cluster to accurately quantify the dewatering execution capability; then, it calculates the risk warning coefficient for the surrounding environment to clarify the safety boundary, breaking through the limitations of traditional single-objective optimization. By integrating efficiency and risk through a multi-objective ranking algorithm, a safety-efficiency trade-off coefficient is obtained, dynamically balancing "maximizing efficiency" and "minimizing risk." This avoids the problem of traditional approaches that prioritize efficiency over safety or prioritize safety over efficiency, providing constraints and priorities for subsequent optimization, ensuring that the dewatering scheme balances safety and economy, and improving the scientific nature of foundation pit dewatering control.

[0016] Optionally, the assessment of the precipitation efficiency of the precipitation well group includes the following steps:

[0017] The method involves introducing a friction factor to calculate pipe resistance loss; using this pipe resistance loss to measure the effective flow rate of a single well; and then evaluating the dewatering efficiency of the dewatering well group based on the effective flow rate of the single well. By accurately calculating pipe resistance loss using the friction factor, and subsequently measuring the effective flow rate of a single well, the interference of factors such as pipe scaling on the flow rate is eliminated. This accurately quantifies the dewatering execution capacity, providing reliable efficiency data for subsequent safety-efficiency trade-offs. It avoids overly aggressive (high energy consumption) or conservative (insufficient dewatering) schemes due to efficiency misjudgments, ensuring the accuracy of the assessment of the foundation pit dewatering execution capacity and supporting scientific optimization decisions.

[0018] Optionally, the calculation of the risk warning coefficient of the surrounding environment includes the following steps:

[0019] This invention quantifies the risks of settlement of the original building, horizontal displacement of the retaining structure, heave at the bottom of the pit, and soil moisture content. Based on soil cohesion, compression modulus, and excavation depth, these risks are weighted to obtain a risk warning coefficient for the surrounding environment. By quantifying the three core risks—horizontal displacement of the retaining structure, heave at the bottom of the pit, and soil moisture content—and combining them with dynamic weighting based on soil cohesion, compression modulus, and excavation depth, this invention avoids the shortcomings of fixed weights that are unsuitable for different geological and excavation scenarios. It comprehensively covers all safety risk points in the foundation pit, making risk assessment more aligned with engineering realities, effectively avoiding missed risks such as foundation instability and retaining structure collapse, providing accurate risk basis for subsequent safety-efficiency trade-offs, and ensuring the safety of the surrounding environment of the foundation pit.

[0020] Optionally, the step of solving the dynamic precipitation scheme using an improved particle swarm optimization algorithm based on the precipitation demand characteristics and the safety-efficiency trade-off coefficient includes the following steps:

[0021] Based on the aforementioned precipitation demand characteristics and the aforementioned safety-efficiency trade-off coefficients, an objective function and constraints are constructed. A dynamic precipitation scheme is then solved using an improved particle swarm optimization algorithm based on these objective functions and constraints. This invention clarifies precipitation targets by combining precipitation demand characteristics, balances core requirements by relying on safety-efficiency trade-off coefficients, constructs an objective function and constraints that align with engineering realities, and then solves the problem using an improved particle swarm optimization algorithm. This effectively reduces energy consumption, decreases the rate of safety constraint violations, and outputs a precise dynamic precipitation scheme, ensuring both the safety and economy of foundation pit construction.

[0022] Optionally, the improved particle swarm optimization algorithm includes the following steps:

[0023] The inertial weights are corrected using the aforementioned safety-efficiency trade-off coefficient; the particle velocity update formula is improved based on the precipitation demand characteristics. Correcting the inertial weights with the safety-efficiency trade-off coefficient allows for dynamic adjustment of the search strategy based on either "safety priority" or "efficiency priority," avoiding the poor adaptability of fixed weights. Improving the particle velocity update formula based on precipitation demand characteristics guides particles to search in directions that meet precipitation demands, preventing the search from deviating from actual needs. These two optimization steps make the algorithm more suitable for foundation pit dewatering scenarios, improve solution accuracy, and contribute to outputting safe, efficient, and low-energy dynamic dewatering solutions.

[0024] Secondly, to efficiently execute the intelligent-controlled dynamic optimization method for foundation pit dewatering provided by this invention, this invention also provides an intelligent-controlled dynamic optimization system for foundation pit dewatering, including a processor, an input device, an output device, and a memory. The processor, input device, output device, and memory are interconnected. The memory stores a computer program containing program instructions. The processor is configured to call the program instructions to execute the intelligent-controlled dynamic optimization method for foundation pit dewatering as described in the first aspect of this invention. This intelligent-controlled dynamic optimization system for foundation pit dewatering is compact, stable in performance, and can stably execute the intelligent-controlled dynamic optimization method for foundation pit dewatering provided by this invention, further enhancing the overall applicability and practical application capability of this invention. Attached Figure Description

[0025] Figure 1 A flowchart of a dynamic optimization method for foundation pit dewatering based on intelligent control is provided for an embodiment of the present invention;

[0026] Figure 2 This is a framework diagram of a dynamic optimization system for foundation pit dewatering based on intelligent control, provided for an embodiment of the present invention. Detailed Implementation

[0027] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.

[0028] Throughout this specification, references to "an embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "in an embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0029] Please see Figure 1 To address the aforementioned problems, this invention provides a dynamic optimization method for foundation pit dewatering based on intelligent control, such as... Figure 1 As shown, in one embodiment, the method includes the following steps:

[0030] S1. Dynamically filter and process real-time water level data to obtain water level prediction deviation.

[0031] Real-time water level data from monitoring wells inside and around the foundation pit are collected by high-precision water level sensors deployed at different depths. Due to the complex electromagnetic environment and temperature fluctuations at the construction site, sensor noise is unavoidable in the data, such as abnormal signal jumps caused by electromagnetic interference and measurement deviations caused by temperature drift. To ensure data reliability, all sensors are calibrated in the laboratory and redundantly deployed during on-site installation to cross-verify the accuracy of the data.

[0032] Furthermore, the dynamic filtering process for real-time water level data to obtain the water level prediction deviation includes the following steps:

[0033] S11. Calculate the dynamic water level attenuation coefficient by introducing the water level trend slope, fluctuation amplitude, and periodic characteristics.

[0034] Specifically, the water level trend slope is extracted by linear regression of water level data over the past 24 hours, and satisfies the following:

[0035]

[0036] in, Indicates the slope of the water level trend. This indicates that the number of sampling points is 24. This represents the original water level data. Indicates the time sequence number. This indicates a long-term rise in water levels, necessitating efforts to slow natural decline and avoid underestimating the water level.

[0037] The fluctuation range reflects short-term stability through the standard deviation of historical water level fluctuations (data from the past 72 hours), satisfying the following:

[0038]

[0039] in, Indicates the fluctuation range. This represents the average water level over 72 hours. The larger the fluctuation range, the more drastic the short-term fluctuations in water level (such as frequent adjustments to pump frequency). It is necessary to avoid excessive attenuation coefficients that could lead to over-smoothing filtering and loss of fluctuation details.

[0040] The periodicity feature was extracted by using Fast Fourier Transform (FFT) to obtain the main period (unit: h) of the water level data for the past 7 days. The specific steps are as follows:

[0041] The frequency spectrum was obtained by performing FFT on the 7-day water level sequence (sampling interval 1 hour, total 168 points);

[0042] Find the maximum frequency corresponding to the spectral peak, and then obtain the main period;

[0043] Periodicity determination: If the peak power accounts for more than 30% of the total power, then a significant periodicity is considered to exist (such as a 24-hour periodicity caused by rainfall or a 12-hour periodicity caused by tides); otherwise, there is no significant periodicity.

[0044] Furthermore, the dynamic water level decay coefficient is calculated by introducing the water level trend slope, fluctuation amplitude, and periodic characteristics, and is divided into two scenarios: no significant period and significant period.

[0045] The significant periodic dynamic water level decay coefficient satisfies the following formula:

[0046]

[0047] The coefficient for water level decay without significant periodicity satisfies the following formula:

[0048]

[0049] in, This represents the dynamic water level attenuation coefficient. It represents the benchmark attenuation coefficient of soil water-holding capacity (determined through geotechnical tests or survey data). This indicates the largest historical fluctuation range over the past 30 days. Indicates the main period.

[0050] S12. Optimize the state transition matrix based on the dynamic water level attenuation coefficient, and obtain the water level prediction deviation based on the Kalman filter algorithm.

[0051] Traditional Kalman filtering algorithms suffer from poor filtering performance when processing foundation pit water level data due to neglecting the constraint of the nonlinear rate of change of water level. This invention introduces a water level gradient penalty term and constructs a penalty function to constrain the abnormal rate of change of water level. The specific design is as follows:

[0052] To accurately describe the dynamic changes in water level, a state vector is defined. ,in, This represents the actual water level in the foundation pit at time t, reflecting the current water level status. The rate of change of water level reflects the dynamic evolution trend of water level, and the combination of the two can fully depict the spatiotemporal variation characteristics of water level.

[0053] Taking into full account the effects of natural water level decay and artificial precipitation, a state equation is constructed. The natural water level decay process follows the principles of groundwater dynamics and is influenced by geological parameters such as aquifer permeability and porosity. Artificial precipitation is regulated by controllable factors such as the pumping volume and pumping time of dewatering wells. Through parameter identification and model calibration, the equation accurately simulates the dynamic changes in the foundation pit water level, providing a reliable basis for intelligent control.

[0054] The state equations satisfy:

[0055]

[0056] in, Represents the state transition matrix. , Indicates the sampling interval. Represents the control matrix. This indicates the influence coefficient of the water pump. express Pump frequency at all times Represent process noise and obey .

[0057] The observation equation satisfies: , This represents the observed water level at time t. Represents the observation matrix. Represent the observed noise and obey , Indicates the sensor's accuracy level.

[0058] S2. Dynamic geological parameters are obtained by dynamically correcting the permeability coefficient through pore water pressure.

[0059] Specifically, the process of dynamically correcting the permeability coefficient through pore water pressure to obtain dynamic geological parameters includes the following steps:

[0060] S21. Using the permeability-dominant index, obtain the first geological correction coefficient.

[0061] The permeability dominance index is used to quantify the comprehensive influence of inherent soil properties (permeability, compressibility, stiffness) on the "dynamic correction of permeability coefficient." It intuitively reflects the dominant priority of "permeability parameters" and "pore water pressure-sensitive parameters and compression modulus" in the correction process, satisfying the following:

[0062]

[0063] in, Indicates the penetration dominance index. This represents the standardized parameter of the initial permeability coefficient. This represents the standardized parameter of the initial void ratio. This represents the standardized parameters of the compression modulus. The standardized parameters are obtained through the maximum-minimum normalization algorithm. The closer the standardized initial permeability coefficient is to 1, the stronger the soil permeability. The closer the standardized initial void ratio is to 1, the more developed the soil pores and the higher the compressibility. The closer the standardized compression modulus is to 1, the greater the soil stiffness and the stronger the resistance to deformation.

[0064] Furthermore, using the permeability-dominant index, the first geological correction coefficient is obtained, satisfying:

[0065]

[0066] in, This represents the first geological correction coefficient, which transforms discrete weights into a continuous function, breaking through the binary limitation of "either soft or sandy".

[0067] S22. Combine the first geological correction coefficient and pore water pressure to obtain dynamic geological parameters.

[0068] During precipitation, as pore water pressure dissipates, the effective stress between soil particles increases, causing changes in the soil pore structure and ultimately affecting its permeability. The static initial permeability coefficient cannot accurately reflect changes in soil permeability during precipitation, leading to calculation errors. This invention combines the first geological correction coefficient and pore water pressure to obtain dynamic geological parameters that satisfy:

[0069]

[0070] in, Represents the dynamic permeability coefficient. Represents the initial permeability coefficient. Indicates the soil compression modulus. Indicates the initial void ratio. Indicates the initial pore water pressure. This indicates the real-time pore water pressure.

[0071] Furthermore, the porosity correction formula satisfies:

[0072]

[0073] in, Indicates dynamic porosity. This represents the initial porosity. This represents the increment of effective stress. As precipitation progresses, pore water pressure decreases, and effective stress increases, leading to consolidation and compression of the soil, resulting in a decrease in porosity. This formula, using the increment of effective stress and the soil compression modulus, accurately calculates the change in soil porosity at different times, providing key parameter support for the accurate analysis of soil seepage characteristics during precipitation.

[0074] S3. By combining the water level prediction deviation, the dynamic geological parameters, and rainfall infiltration recharge, the precipitation demand characteristics are obtained.

[0075] In this embodiment, obtaining precipitation demand characteristics by combining the water level prediction deviation, the dynamic geological parameters, and rainfall infiltration recharge includes the following steps:

[0076] S31. Construct a water level recharge model and use the water level recharge model to obtain the rainfall infiltration recharge.

[0077] Specifically, the water level replenishment model satisfies the following formula:

[0078]

[0079] in, This indicates rainfall infiltration replenishment. This indicates the amount of rainfall collected. This indicates vegetation coverage (e.g., 70% for lawns and 10% for paved surfaces). This represents the infiltration coefficient: 0.6~0.8 for sandy soil and 0.3~0.5 for clayey soil. This indicates the amount of evaporation.

[0080] S32. Based on the fuzzy neural network fusion algorithm, combined with the water level prediction deviation, the dynamic geological parameters and rainfall infiltration recharge, the precipitation demand characteristics are obtained.

[0081] Specifically, fuzzification (mapping features to the [0,1] interval): transforms continuous numerical values ​​into linguistic fuzzy variables through membership functions, enhancing the model's ability to express complex uncertainties.

[0082] The fuzzy subset of water level deviation is defined as three linguistic variables: {small, medium, large}. A Gaussian membership function is used to dynamically adjust the function width to adapt to water level fluctuation characteristics under different engineering geological conditions, and it satisfies the following: ,in, Indicates the degree of membership of water level deviation. This represents the water level deviation, which is the difference between the water level at time t and the predicted water level at time t+1. express The historical average.

[0083] Fuzzy subset of permeability coefficient: Divided into {low, medium, high}, and permeability performance is described using a Sigmoid membership function, satisfying: , The standard permeability coefficient reference value obtained from engineering survey is used in the function. An inflection point is generated at this point, enabling a smooth transition between different penetration levels.

[0084] The supply quantity fuzzy subset is set as {negative, zero, positive}, based on the hyperbolic tangent function. Establish a hierarchical relationship. As the largest historical replenishment volume, it ensures that the function can still effectively map the replenishment characteristics under extreme hydrological conditions.

[0085] Furthermore, a three-layer feedforward neural network is constructed, consisting of an input layer, a hidden layer, and an output layer. The input layer receives the three types of fuzzy feature vectors mentioned above, and the output layer directly maps the precipitation demand value. The loss function is set as follows: ,in, Indicates the target water level. This represents the real-time water level prediction value. This represents the volume affected by dewatering in the foundation pit. The loss function integrates the target water level deviation, soil water supply characteristics, and the volume affected by the foundation pit, ensuring that the model output meets actual engineering requirements. The training process employs the Adam algorithm with an adaptive learning rate, iteratively optimizing the model to converge the loss on the training set to the global optimum.

[0086] Dynamic precipitation demand As a core control variable, it reflects the amount of water pumped to reach the target water level at the current moment; demand rate. This characterizes the time-varying properties of precipitation demand, which is used to predict future trends in precipitation intensity and provide a basis for advanced decision-making in intelligent control systems. By calculating the differential approximate derivative of dynamic precipitation demand at adjacent times, the demand rate is updated dynamically in real time.

[0087] S4. Calculate the safety-efficiency trade-off coefficient based on precipitation efficiency and risk coefficient.

[0088] In another embodiment, calculating the safety-efficiency trade-off coefficient based on precipitation efficiency and risk coefficient includes the following steps:

[0089] S41. Evaluate the precipitation efficiency of the precipitation well group.

[0090] Due to the influence of factors such as pipe scaling and corrosion on flow rate, there is a discrepancy between theoretical calculations and actual results. By introducing a pipe resistance loss calculation module and combining it with real-time monitoring data, the flow rate parameters are dynamically corrected, significantly improving the calculation accuracy.

[0091] First, the friction factor is introduced to calculate the pipe resistance loss, which satisfies the following:

[0092]

[0093] in, Indicates pipe resistance loss. This represents the friction coefficient obtained through Newton's iteration of the Colebrook formula. Indicates the length of the pipe. Indicates the pipe diameter. This indicates the real-time flow rate of water within the pipe. Represents gravitational acceleration. This indicates the cross-sectional area of ​​the pipe.

[0094] Then, the effective flow rate of a single well is calculated using the pipe resistance loss, and the precipitation efficiency of the precipitation well group is evaluated based on the effective flow rate of the single well.

[0095] Specifically, the effective flow rate of a single well is calculated using the aforementioned pipe resistance loss, satisfying the following: , This represents the measured flow rate of a single well;

[0096] The precipitation efficiency of the precipitation well group is evaluated based on the effective flow rate of the single well, satisfying the following conditions: , Indicates the number of wells. This represents the rated flow rate of the i-th well.

[0097] S42. Calculate the risk warning coefficient of the surrounding environment.

[0098] Specifically, the risks of settlement of the original building, horizontal displacement of the retaining structure, heave of the pit bottom, and soil moisture content are quantified.

[0099] The settlement risk of the original building is quantified according to the following formula:

[0100]

[0101] in, Indicates the risk of settlement of the original building. Indicates the settling rate. This indicates the dynamic allowable settlement value. This represents the original static allowable settlement value. Indicates the soil moisture content. Indicates the saturated water content of the soil. Indicates the amount of building settlement. Indicates the first 30 days The maximum value.

[0102] The original allowable static settlement value is determined based on the building structure type and relevant standards. Taking brick-concrete structures as an example, due to the relatively weak deformation resistance of their wall materials, the maximum allowable static settlement value is typically set at 20mm according to the "Code for Design of Building Foundations" (GB50007). This value is a critical threshold for ensuring the structural safety of the building and preventing wall cracking and functional failure. In the process of multi-foundation pit parameter coupling optimization, it needs to be used as an important benchmark parameter for dynamic analysis and adjustment.

[0103] The risk of horizontal displacement of the building envelope is quantified according to the following formula:

[0104]

[0105] in, This indicates the risk of horizontal displacement of the building envelope. Indicates the horizontal displacement rate. Indicates the dynamic allowable horizontal displacement. This represents the initial maximum allowable value of horizontal displacement of the enclosure structure. Indicates cohesion. Indicates initial cohesion. Indicates the first 30 days The maximum value.

[0106] The risk of a trough bottoming out can be quantified by the following formula:

[0107]

[0108] in, This indicates a risk of the pit bottom heaving. Indicates the rate of bulging. Indicates the dynamic allowable bulge value. This represents the real-time effective stress of the soil. This represents the effective stress of the soil in the initial state of foundation pit construction. This indicates the amount of bulge at the bottom of the pit. Indicates the first 30 days The maximum value.

[0109] The risk of soil moisture content is quantified by the following formula:

[0110]

[0111] in, Indicates the risk of soil moisture content. Indicates the rate of change of soil moisture content. This indicates the moisture content limit to prevent soil saturation and subsequent rapid strength drop. Indicates the first 30 days The maximum value.

[0112] Next, the risk of settlement of the original building, the risk of horizontal displacement of the retaining structure, the risk of heave at the bottom of the pit, and the risk of soil moisture content are weighted according to the soil cohesion, compression modulus, and excavation depth of the foundation pit to obtain the risk warning coefficient of the surrounding environment.

[0113] Specifically, the risk of settlement decreases with increasing soil stiffness (greater stiffness results in less settlement) and increases with increasing excavation depth (greater depth results in a wider range of settlement impact). Therefore: ;

[0114] The risk of retaining wall displacement increases with increasing soil cohesion and stiffness (sandy soil has high stiffness, making the retaining wall prone to excessive displacement due to concentrated lateral pressure). ;

[0115] The risk of pit bottom heave decreases with increasing soil cohesion (sandy soils have strong resistance to heave) and increases with increasing excavation depth (greater depth results in greater stress difference at the base). Therefore: ;

[0116] Moisture content is an indirect risk factor, and its weight should remain relatively stable, only slightly adjusted according to soil properties (soft soil is more sensitive to moisture content). ,in, Standardized parameters representing excavation depth, Standardized parameters representing cohesion A standardized parameter representing the compressibility modulus.

[0117] Furthermore, the risk warning coefficient of the surrounding environment is obtained, satisfying the following formula:

[0118]

[0119] in, This indicates the risk warning coefficient of the surrounding environment.

[0120] S43. Based on the multi-objective ranking algorithm, and combining the precipitation efficiency and the risk coefficient, a safety-efficiency trade-off coefficient is obtained.

[0121] In this embodiment, the Pareto multi-objective sorting algorithm is used to balance the two mutually restrictive core objectives of "well group efficiency" and "settlement safety".

[0122] Construct objective function 1, By multiplying the pumping efficiency by the sum of the effective pumping volumes of each well, the system comprehensively considers both equipment operating efficiency and rainfall scale, thereby maximizing the assessment of rainfall efficiency.

[0123] Construct objective function 2, ( By using the reciprocal of the radius of influence of precipitation as a risk metric, the radius of influence is inversely correlated with the degree of risk, ensuring that the scope of precipitation’s impact on the surrounding environment is effectively controlled during the algorithm optimization process.

[0124] Using Pareto optimal solution selection, strict threshold conditions are set during solution space selection, retaining only those that meet the criteria. and The solution, where, As a minimum efficiency threshold, it ensures the basic efficiency requirements of precipitation operations; As an acceptable risk threshold, it ensures the safety and controllability of the surrounding environment.

[0125] Furthermore, the safety-efficiency trade-off coefficient serves as the core basis for dynamic control, enabling flexible switching of objective priorities through the ratio of objective function 2 to objective function 1. When the safety-efficiency trade-off coefficient is greater than 1, the weight of the risk minimization objective is increased, and the system prioritizes ensuring the settlement safety of the surrounding environment of the foundation pit; when the safety-efficiency trade-off coefficient is less than 1, the efficiency maximization objective takes precedence, and the system focuses on improving the efficiency of dewatering operations to meet actual needs such as the construction period.

[0126] S5. Based on the precipitation demand characteristics and the safety-efficiency trade-off coefficient, the dynamic precipitation scheme is solved using an improved particle swarm optimization algorithm.

[0127] The process of solving the dynamic precipitation scheme using an improved particle swarm optimization algorithm based on the precipitation demand characteristics and the safety-efficiency trade-off coefficient includes the following steps:

[0128] S51. Based on the precipitation demand characteristics and the safety-efficiency trade-off coefficient, construct the objective function and constraints.

[0129] In the embodiment, the objective function satisfies:

[0130]

[0131] in, Indicates the pump frequency at well point i. By fitting the pump characteristic curves, the nonlinear relationship between pump power and frequency was obtained. This was achieved by analyzing the data at each well point. The accumulation of internal power consumption enables accurate calculation of the energy consumption of the entire precipitation system.

[0132] Constraints include precipitation demand constraints, security constraints, efficiency constraints, and trade-off constraints.

[0133] The precipitation demand constraint means that at any given time, the effective drainage volume of all well points is not less than the real-time demand for foundation pit dewatering.

[0134] Safety constraints aim to keep the impact of precipitation on the surrounding environment within a safe range, and to avoid geological disasters and building damage caused by precipitation.

[0135] Efficiency constraints are designed to ensure that precipitation systems operate at high efficiency, improve energy utilization efficiency, and reduce operating costs.

[0136] Trade-off constraints are designed to flexibly balance the relationship between precipitation demand and other constraints, such as: , This represents the safety-efficiency trade-off coefficient, when... When the pressure is increased, the system will focus more on meeting precipitation requirements; conversely, it will tend to optimize other constraints.

[0137] S52. Based on the objective function and the constraints, solve the dynamic precipitation scheme using the improved particle swarm optimization algorithm.

[0138] In this embodiment, the improved particle swarm optimization algorithm includes the following steps:

[0139] First, the inertia weight is corrected using the aforementioned safety-efficiency trade-off coefficient.

[0140] Inertia weight, as a core parameter of the particle swarm optimization algorithm, directly determines the balance between a particle's ability to "explore the global environment" and "utilize local resources" in the search space. A safety-efficiency trade-off coefficient is introduced to construct a dynamic control model, enabling adaptive optimization of the search strategy under different operating conditions, satisfying the following:

[0141]

[0142] in, This indicates a correction of the inertia weight. This represents the maximum inertia weight. This represents the median of the inertia weights. Indicates the maximum number of iterations. This represents the midpoint of the iteration count. This represents the minimum inertia weight.

[0143] Then, the particle velocity update formula is improved based on the precipitation demand characteristics.

[0144] Specifically, the particle velocity update formula is improved based on the precipitation demand characteristics to satisfy the following formula:

[0145]

[0146] in, Indicates the first The particle in the first Speed ​​in the next iteration These are individual learning factors and social learning factors, respectively. A random number within the interval [0,1]. For the first The best position in the history of each particle This is the best historical position for the group. This represents the pump frequency of the i-th precipitation well in the t-th iteration. Indicates demand deviation, when (Calculation of insufficient drainage capacity): When positive, the newly added term drives particles to update in the direction of "increasing frequency," rapidly increasing the total displacement; when (Calculated drainage volume exceeds demand) If the value is negative, the new term will push particles to update in the direction of "reduced frequency" to avoid excessive precipitation leading to energy waste.

[0147] Specifically, based on the objective function and the constraints, the dynamic precipitation scheme is solved using an improved particle swarm optimization algorithm, including:

[0148] Determine the optimization variables: The pump frequency at each well point is the core optimization variable;

[0149] Initialize the particle swarm: Set the initial values ​​of particle positions (corresponding frequencies) and velocities, and determine the number of iterations and the convergence threshold;

[0150] Calculate particle fitness: Based on the objective function, combined with precipitation demand, safety, efficiency, and trade-off constraints, an adaptive penalty function is used to handle constraint violations and obtain the fitness value;

[0151] Iterative particle update: Adjust particle velocity and position according to dynamic inertia weight and precipitation demand guide term; if local stagnation occurs, activate escape mechanism to disturb particles.

[0152] Output scheme: After iterative convergence, the optimal pump frequency combination is output, i.e., the dynamic precipitation scheme.

[0153] Please see Figure 2In an embodiment, to efficiently execute the intelligent-controlled dynamic optimization method for foundation pit dewatering provided by this invention, the present invention also provides an intelligent-controlled dynamic optimization system for foundation pit dewatering, comprising: an input device, an output device, a processor, and a memory, wherein the input device, output device, processor, and memory are interconnected, and the memory contains program instructions for executing the steps of the intelligent-controlled dynamic optimization method for foundation pit dewatering. The intelligent-controlled dynamic optimization system for foundation pit dewatering of this invention has a compact structure and stable performance, and can stably execute the intelligent-controlled dynamic optimization method for foundation pit dewatering of this invention, further enhancing the overall applicability and practical application capability of this invention.

[0154] In this embodiment, the processor may be a central processing unit, but it can also be other general-purpose processors, digital signal processors, application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (OPGs), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor. Input devices can be used to acquire data. Output devices can be used to output the results obtained by storing program instructions contained in a computer program in the memory provided by this invention. The memory may include read-only memory and random access memory (RAM), and provides instructions and data to the processor. A portion of the memory may also include non-volatile random access memory (RAM).

[0155] In one possible implementation, the memory may include a stored program area and a stored data area. The stored program area may store the operating system and applications required for at least one function; the stored data area may store data created during use. Furthermore, the memory may include read-only memory and random access memory, and provides instructions and data to the processor. The memory stores the operating system and operating instructions, executable modules, or data structures, or subsets thereof, or extended sets thereof. The operating instructions may include various operation instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and handling hardware-based tasks.

[0156] The embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described dynamic optimization method for foundation pit dewatering based on intelligent control.

[0157] The storage medium can include various media that can store program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0158] In summary, this invention dynamically corrects the permeability coefficient by pore water pressure, breaking through the limitations of traditional static geological parameters and aligning with the actual soil consolidation. It integrates water level deviation, dynamic geological parameters, and rainfall replenishment to quantify precipitation demand, and combines precipitation efficiency with risk coefficients to determine the safety-efficiency trade-off. Finally, it uses an improved particle swarm optimization algorithm to output a dynamic solution, effectively addressing problems such as large water level errors, rigid geological parameters, one-sided risk assessment, and high energy consumption. This achieves high-precision water level control, ensures the safety of foundation pit construction, saves engineering costs, and provides a practical technical solution for intelligent regulation of foundation pit dewatering.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the present invention.

Claims

1. A dynamic optimization method for foundation pit dewatering based on intelligent control, characterized in that, Includes the following steps: Dynamic filtering is used to process real-time water level data to obtain water level prediction deviation. Dynamic geological parameters are obtained by dynamically correcting the permeability coefficient using pore water pressure. By combining the water level prediction deviation, the dynamic geological parameters, and rainfall infiltration recharge, the precipitation demand characteristics are obtained; Calculate the safety-efficiency trade-off coefficient based on precipitation efficiency and risk coefficient; Based on the precipitation demand characteristics and the safety-efficiency trade-off coefficient, an improved particle swarm optimization algorithm is used to solve the dynamic precipitation scheme.

2. The method for dynamic optimization of foundation pit dewatering based on intelligent control according to claim 1, characterized in that, The dynamic filtering process for real-time water level data to obtain water level prediction deviation includes the following steps: The dynamic water level attenuation coefficient is calculated by incorporating water level trend slope, fluctuation amplitude, and periodic characteristics. The state transition matrix is ​​optimized based on the dynamic water level attenuation coefficient, and the water level prediction deviation is obtained based on the Kalman filter algorithm.

3. The method for dynamic optimization of foundation pit dewatering based on intelligent control according to claim 1, characterized in that, The process of dynamically correcting the permeability coefficient through pore water pressure to obtain dynamic geological parameters includes the following steps: The first geological correction coefficient was obtained using the permeability dominance index; Dynamic geological parameters are obtained by combining the first geological correction coefficient and pore water pressure.

4. The method for dynamic optimization of foundation pit dewatering based on intelligent control according to claim 1, characterized in that, The process of combining the water level prediction deviation, the dynamic geological parameters, and rainfall infiltration recharge to obtain precipitation demand characteristics includes the following steps: Construct a water level recharge model, and use the water level recharge model to obtain the rainfall infiltration recharge; Based on the fuzzy neural network fusion algorithm, the precipitation demand characteristics are obtained by combining the water level prediction deviation, the dynamic geological parameters, and the rainfall infiltration recharge.

5. The method for dynamic optimization of foundation pit dewatering based on intelligent control according to claim 1, characterized in that, The calculation of the safety-efficiency trade-off coefficient based on precipitation efficiency and risk coefficient includes the following steps: Evaluate the precipitation efficiency of the precipitation well group; Calculate the risk warning coefficient of the surrounding environment; Based on a multi-objective ranking algorithm, a safety-efficiency trade-off coefficient is obtained by combining the precipitation efficiency and the risk warning coefficient.

6. The method for dynamic optimization of foundation pit dewatering based on intelligent control according to claim 5, characterized in that, The assessment of the precipitation efficiency of the precipitation well group includes the following steps: Introduce the friction factor to calculate pipe resistance loss; The effective flow rate of a single well is calculated by measuring the pipe resistance loss, and the precipitation efficiency of the precipitation well group is evaluated based on the effective flow rate of the single well.

7. The method for dynamic optimization of foundation pit dewatering based on intelligent control according to claim 5, characterized in that, The calculation of the risk warning coefficient of the surrounding environment includes the following steps: Quantify the risks of settlement of the original building, horizontal displacement of the retaining structure, heave of the pit bottom, and soil moisture content; The risk warning coefficients for the surrounding environment are obtained by weighting the risks of settlement of the original building, horizontal displacement of the retaining structure, heave of the pit bottom, and moisture content of the soil based on soil cohesion, compression modulus, and excavation depth.

8. The method for dynamic optimization of foundation pit dewatering based on intelligent control according to claim 1, characterized in that, The process of solving the dynamic precipitation scheme using an improved particle swarm optimization algorithm based on the precipitation demand characteristics and the safety-efficiency trade-off coefficient includes the following steps: Based on the precipitation demand characteristics and the safety-efficiency trade-off coefficient, an objective function and constraints are constructed. Based on the objective function and the constraints, an improved particle swarm optimization algorithm is used to solve the dynamic precipitation scheme.

9. The method for dynamic optimization of foundation pit dewatering based on intelligent control according to claim 8, characterized in that, The improved particle swarm optimization algorithm includes the following steps: The inertia weight is corrected using the aforementioned safety-efficiency trade-off coefficient; The particle velocity update formula is improved based on the precipitation demand characteristics.

10. A dynamic optimization system for foundation pit dewatering based on intelligent control, characterized in that, The intelligent control-based dynamic optimization system for foundation pit dewatering includes: an input device, an output device, a processor, and a memory. The input device, output device, processor, and memory are interconnected. The memory is used to store a computer program. The computer program contains program instructions. The processor is configured to call the program instructions. The program instructions are used to execute the intelligent control-based dynamic optimization method for foundation pit dewatering according to any one of claims 1-9.

Images