Construction engineering intelligent dewatering control method and system

Abstract

The application relates to the technical field of building engineering foundation pit dewatering, and provides a building engineering intelligent dewatering control method and system, which comprises the following steps: screening a plurality of feasible dewatering schemes in a dewatering scheme database; obtaining uncertainty conditions by carrying out probability distribution sampling on a permeability coefficient and a water level depth, and selecting a plurality of optimal dewatering schemes from the feasible dewatering schemes; selecting a plurality of optimal dewatering schemes by using an approximation of ideal solution sequencing method; encoding engineering requirements and the feasible dewatering schemes into feature vectors, predicting the application probability of each feasible dewatering scheme by using a classifier, and selecting a plurality of optimal dewatering schemes; constructing a Pareto front for all the optimal dewatering schemes, screening non-dominated solutions, calculating objective functions for the non-dominated solutions, and obtaining final optimal dewatering schemes; and carrying out dewatering based on the final optimal dewatering schemes, and controlling automatic starting and stopping of water pumps and water pump power. The scientificity and accuracy of dewatering scheme selection are improved.

Description

Technical Field

[0001] This invention belongs to the field of dewatering technology for foundation pits in building engineering, and particularly relates to an intelligent dewatering control method and system for building engineering. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] In recent years, with the continuous deepening and expansion of urban vertical construction, large-scale buildings are increasingly facing construction scenarios under complex geological conditions. For example, high groundwater levels can affect the progress of foundation pit engineering during construction. Foundation pit dewatering is a key process to ensure construction safety and foundation stability. Therefore, foundation pit dewatering has become an important research topic.

[0004] In the field of construction engineering, the urgency of developing intelligent dewatering technology for foundation pits stems from the efficiency bottlenecks and safety and environmental shortcomings of traditional processes. Traditional technologies rely on manual monitoring and experience-based operation, which suffers from problems such as lagging water level control, insufficient equipment adaptability (e.g., blockage by silt layers, uneven permeability in karst formations), low pumping efficiency, and high energy consumption. This can easily lead to water accumulation at the bottom of the pit, settlement of surrounding buildings, and even piping accidents. At the same time, the risks of a humid environment for electrically driven equipment, excessive sand content caused by the failure of gravel filter media, and the lack of design for the recycling of groundwater resources further exacerbate construction costs and environmental burdens.

[0005] As deep foundation pit engineering develops towards ultra-deep and complex geology, the industry urgently needs intelligent solutions that integrate the Internet of Things, automation, and data analysis: real-time closed-loop control of water level can be achieved through level sensors and logic control systems, and pumping strategies can be dynamically optimized by combining multi-source data (geological parameters, settlement monitoring, construction progress), and anti-clogging hardware (such as self-cleaning filters and crushing structures) can be integrated to improve equipment reliability; at the same time, intelligent prediction models can balance dewatering efficiency and environmental protection, and water resource recycling systems can be used to reduce energy consumption.

[0006] However, existing intelligent dewatering technologies for foundation pits have the following drawbacks: In the process of dewatering construction projects, facing complex and diverse geological conditions and engineering requirements, traditional methods are unable to quickly and accurately select the optimal solution from numerous dewatering schemes; geological parameters (such as permeability coefficient, water level depth, etc.) have a certain degree of uncertainty, which will affect the selection and implementation effect of dewatering schemes. Summary of the Invention

[0007] To address the technical problems mentioned above, this invention provides an intelligent precipitation control method and system for building engineering. After filtering feasible solutions using rules, multiple optimal precipitation schemes are selected through a comprehensive multi-method approach. Finally, a non-dominated solution is selected using the Pareto front, and the objective function is calculated to obtain the final precipitation scheme. This method comprehensively considers geological conditions, engineering requirements, and various uncertainties, selecting the most suitable optimal precipitation scheme for a specific project from numerous options. This improves the scientific rigor and accuracy of precipitation scheme selection, laying a foundation for the smooth progress of subsequent precipitation work.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] The first aspect of this invention provides an intelligent precipitation control method for building engineering, comprising:

[0010] Obtain soil layer types and engineering requirements, and select several feasible precipitation schemes from the precipitation scheme database through geological adaptability rules and engineering constraint rules;

[0011] The probability distribution of permeability coefficient and water level depth is sampled to obtain uncertainty conditions. Under these uncertainty conditions, several optimal precipitation schemes are selected from the feasible precipitation schemes.

[0012] Among the feasible precipitation schemes, positive ideal solutions and negative ideal solutions are selected, and several optimal precipitation schemes are selected by the approximation ideal solution ranking method.

[0013] The engineering requirements and feasible precipitation schemes are encoded into feature vectors. A classifier is used to predict the applicability probability of each feasible precipitation scheme and select several optimal precipitation schemes.

[0014] For all optimal precipitation schemes, Pareto fronts are constructed to screen out non-dominated solutions. For the non-dominated solutions, the objective function is calculated to obtain the final optimal precipitation scheme.

[0015] Rainfall is carried out based on the final optimal precipitation plan, and the automatic start and stop of water pumps and water pump power are controlled.

[0016] Furthermore, the project requirements include the depth of the foundation pit, the drawdown of the water level, and the budget.

[0017] Furthermore, each precipitation scheme includes precipitation methods, efficiency, cost, and environmental impact factors.

[0018] Furthermore, the objective function is a weighted sum of efficiency, cost, and environmental impact factors.

[0019] Furthermore, under conditions of uncertainty, several optimal precipitation schemes are selected from feasible precipitation schemes by minimizing the maximum regret model.

[0020] Furthermore, the power of the water pump is adjusted based on the monitored water level, soil density and moisture content, meteorological data, and ground subsidence.

[0021] Furthermore, the permeability coefficient is assumed to follow a log-normal distribution, and the water level depth H is assumed to follow a uniform or gamma distribution. The rationality of the parameter distribution assumptions is verified by the KS test.

[0022] A second aspect of the present invention provides an intelligent precipitation control system for building engineering, comprising:

[0023] The feasible solution screening module is configured to: obtain soil layer type and engineering requirements, and screen out several feasible precipitation schemes from the precipitation scheme database through geological adaptability rules and engineering constraint rules;

[0024] The first selection module is configured to: sample the probability distribution of the permeability coefficient and the water level depth to obtain uncertainty conditions, and select several optimal precipitation schemes from the feasible precipitation schemes under the uncertainty conditions;

[0025] The second selection module is configured to: select positive ideal solutions and negative ideal solutions from feasible precipitation schemes, and select several optimal precipitation schemes by using the approximation ideal solution ranking method;

[0026] The third selection module is configured to: encode engineering requirements and feasible precipitation schemes into feature vectors, predict the applicability probability of each feasible precipitation scheme through a classifier, and select several optimal precipitation schemes.

[0027] The final scheme selection module is configured to: construct the Pareto front for all optimal precipitation schemes, filter out non-dominated solutions, and calculate the objective function for the non-dominated solutions to obtain the final optimal precipitation scheme;

[0028] The control module is configured to: carry out precipitation based on the final optimal precipitation plan, and control the automatic start and stop of the water pump and the water pump power.

[0029] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the intelligent precipitation control method for building engineering as described above.

[0030] A fourth aspect of the present invention provides a computer device including a computer-readable storage medium, a processor, and a computer program stored on the computer-readable storage medium and executable on the processor, wherein the processor executes the program to implement the steps of the intelligent precipitation control method for building engineering as described above.

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] This invention, after screening feasible solutions through rules, comprehensively screens multiple optimal precipitation schemes through multiple methods, and finally selects non-dominated solutions through Pareto front screening and calculates the objective function to obtain the final precipitation scheme. It can comprehensively consider geological conditions, engineering requirements and various uncertainties, and select the optimal precipitation scheme most suitable for the specific project from many precipitation schemes, thereby improving the scientificity and accuracy of precipitation scheme selection and laying the foundation for the smooth progress of subsequent precipitation work.

[0033] This invention samples the probability distribution of permeability coefficient and water level depth, and selects the optimal solution under uncertainty conditions through a reasonable model (such as the minimization of maximum regret model). This makes the precipitation scheme more adaptable and robust when facing the uncertainty of formation parameters, reduces the risk of precipitation scheme failure due to changes in formation parameters, and improves the reliability and safety of the project.

[0034] This invention automatically adjusts the power of water pumps and controls their start and stop based on multi-source information such as real-time monitoring of water level, soil density and moisture content, meteorological data, and ground subsidence. This achieves intelligent and precise control of the precipitation process. This precise control not only ensures that the precipitation effect meets the engineering requirements, but also effectively saves energy, reduces the impact on the surrounding environment (such as ground subsidence), and improves the overall efficiency of precipitation in building engineering.

[0035] This invention, by selecting the optimal precipitation scheme and achieving precise control, can avoid unnecessary waste of resources and reasonably control precipitation costs. At the same time, by comprehensively considering the objective function of efficiency, cost and environmental impact factors, it achieves optimal resource allocation while meeting engineering requirements, thereby improving the economic and environmental benefits of the project. Attached Figure Description

[0036] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0037] Figure 1 This is a flowchart of an intelligent precipitation control method for building engineering according to Embodiment 1 of the present invention;

[0038] Figure 2 This is a schematic diagram of the structure of a computer device according to Embodiment 4 of the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0040] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0041] Example 1

[0042] This embodiment provides a method for intelligent precipitation control in building engineering.

[0043] In this embodiment, taking the construction of a high-rise commercial complex in the central area of ​​a city as an example, the excavation depth of the foundation pit is 15 meters, the site strata are mainly silt layer, silty clay layer and gravel layer, and the groundwater level is buried at a depth of about 3 meters; the project budget is 5 million yuan, and the water level is required to be lowered to a depth of more than 12 meters to meet the conditions for dry construction of the foundation pit.

[0044] This embodiment provides an intelligent precipitation control method for building engineering, such as... Figure 1 As shown, it includes the following steps:

[0045] Step 101: Data preparation and scheme selection.

[0046] (1) Determine the input parameter domain: Input surface data and engineering requirements, D={G,E}, where G=(k,S,H) represents the stratum data (permeability coefficient k, soil type S, water level depth H), and E=(d,Δh,B) represents the engineering requirements (foundation pit depth d, water level drawdown Δh, budget B).

[0047] For example, after geological survey, the permeability coefficient k of the silt layer is about 10m / d, the soil layer type S includes silt layer, silty clay layer and gravel layer, the water level depth H = 3m; the foundation pit depth d = 15m, the water level drawdown requirement Δh ≥ 12m, and the budget B = 5 million yuan.

[0048] (2) Establish the scheme space: S={si∣si=(Mi,Fi,Ci,Ei)}, where Mi is the dewatering method (e.g., open ditch plus collection well dewatering, light well point dewatering, pipe well point dewatering, etc.), Fi is the efficiency, Ci is the cost, and Ei is the environmental impact factor.

[0049] For example, the dewatering method Mi considers three common methods: open ditch plus catch well dewatering, lightweight wellpoint dewatering, and tubular wellpoint dewatering. For each scheme si, efficiency Fi, cost estimation Ci, and environmental impact factor assessment Ei are performed. For instance, open ditch plus catch well dewatering has a relatively low cost but a high environmental impact factor (potentially leading to significant ground subsidence in the surrounding area); tubular wellpoint dewatering has a higher cost but a relatively smaller environmental impact.

[0050] (3) Construct the objective function: Maximize F(si)=αEfficiency(si)+βCost(si)+γEnvironment(si), where α, β and γ are weight coefficients, satisfying α+β+γ=1. Efficiency(si), Cost(si) and Environment(si) are efficiency, cost and environmental impact related indicators, namely Fi, Ci and Ei, respectively. The comprehensive score is calculated based on previous projects and expert scores to determine the level.

[0051] The determination of γ typically employs the entropy weight method, which assigns objective weights by quantifying the degree of data variation in environmental indicators. The specific steps are as follows:

[0052] First, data standardization: positive indicators (such as the vegetation recovery rate after precipitation treatment): Negative indicators (such as noise index, i.e., the percentage of time during construction when noise levels exceed a threshold): Ensure all data are normalized to the [0,1] interval to eliminate dimensional differences. Where y ij x represents the proportion of the i-th sample in the j-th indicator. In a multi-indicator system, x j This can represent the specific value of the j-th indicator. For example, in the standardization formula, x... ij Let x represent the j-th index value of the i-th sample. j It can represent a decision variable (such as environmental impact value, noise index, etc.).

[0053] Secondly, calculate the information entropy and the difference coefficient:

[0054] Information entropy: in, Where n is the number of samples;

[0055] Coefficient of difference: g j =1-e j The larger the difference coefficient, the higher the information content of the indicator and the greater its weight.

[0056] Finally, determine the weight γ: Where m is the total number of indicators.

[0057] Among these, adjusting β requires combining a linear programming model with budget constraints to dynamically optimize cost priorities.

[0058] First, construct the budget constraints: Among them, C i Let B be the cost of the solution, and let x be the total budget. i Select variables for the scheme. The scheme selection variable can be understood as choosing which precipitation scheme to use, such as wellpoint precipitation with pipes or precipitation with open ditches and catch wells.

[0059] Secondly, objective function and sensitivity analysis:

[0060] Optimization objective: min∑C i ·x i +λ·Efficiency, where λ is the trade-off coefficient between efficiency and cost.

[0061] Weighting adjustment: If the budget is reduced, the beta value needs to be increased to strengthen cost constraints.

[0062] (4) Parameter matching module processing.

[0063] Rule base construction: Rules are defined based on domain knowledge: geological adaptability rules (applicable if S∈soil_list thensi) and engineering constraint rules (if d≥d min andΔh≥Δh req Then, if feasible, the precipitation schemes are screened in the precipitation scheme database to obtain a set of feasible precipitation schemes.

[0064] For example, due to the high permeability coefficient of the silty sand layer, wellpoint dewatering and lightweight wellpoint dewatering are applicable to this stratum; in terms of engineering constraints, all three schemes are initially feasible in terms of the requirements for foundation pit depth and water level drawdown.

[0065] (5) Uncertainty handling.

[0066] Uncertainty handling utilizes Monte Carlo simulation to sample the probability distribution of formation parameters k and H, calculating the expected success rate of the proposed scheme. Furthermore, robust optimization is employed to construct a model that minimizes the maximum regret. Where U is the set of parameter uncertainties; F(s) i ) indicates that in decision option s i The objective function value, where s is... i s represents the decision variables to be chosen in an optimization problem (such as engineering design schemes, resource allocation strategies, etc.), while F is an indicator for evaluating the performance of that scheme (such as cost, benefit, success rate, etc.). Example: In the water resource scheduling problem, s i It could be the operating scheme of the water pump, F(s) i This corresponds to the total energy consumption of the system or the error in water level control. This represents the optimal objective function value under a specific combination of uncertain parameters D∈U. Wherein, This represents the optimal decision when parameter D is known, while U is the set of parameter uncertainties (such as the possible ranges of formation parameters k and H). Example: If the true value of the formation permeability coefficient k is a fixed value D, then... It is the minimum energy consumption under the optimal scheduling scheme corresponding to D.

[0067] In other words, Monte Carlo simulations can generate a large number of random samples, helping decision-makers to better understand and quantify uncertainty; while the maximum regret minimization model can select several optimal decision options under these uncertain conditions.

[0068] Monte Carlo simulation uses probability distribution modeling and random sampling, combined with statistical inference, to calculate the expected value of the success rate of a plan. Its core steps are as follows:

[0069] First, model the probability distribution of the parameters.

[0070] Input parameter modeling: Fit probability distributions to formation parameters k (permeability coefficient) and H (water level depth). For example, the permeability coefficient k may follow a log-normal distribution (common in hydrogeological parameters); the water level depth H can be assumed to be uniformly distributed or gamma-ray distributed (this needs to be verified based on measured data).

[0071] Distribution verification: The reasonableness of the parameter distribution hypothesis is verified by the Kolmogorov-Smirnov test.

[0072] Then, random sampling and simulation experiments were conducted.

[0073] Random sampling: Generating a large number of random samples (e.g., 10,000 samplings) from the parameter distribution. For example: simulating the effect of a precipitation scheme using a numerical model (e.g., a groundwater seepage model) for each parameter combination (ki, Hi).

[0074] Success determination: Define the criteria for a successful scheme (e.g., drawdown Δh ≥ Δhreq), and count the number of samples that meet the criteria.

[0075] Finally, statistical calculations and expected value determination are performed.

[0076] Expected success rate: E (success rate) = total number of samples / number of successful samples × 100%. For example, if the Δh requirement is met in 8,500 out of 10,000 simulations, the expected success rate is 85%.

[0077] Monte Carlo simulations, through probabilistic modeling and statistical experiments, transform the uncertainty of complex geological parameters into a quantifiable expected success rate. This value is not only a core indicator for risk decision-making but also provides a quantitative basis for multi-objective optimization and dynamic adjustment. In a digital twin platform, parameter distributions are updated based on real-time monitoring data, and the expected success rate is recalculated to dynamically optimize precipitation plans, thereby improving the reliability of decision-making.

[0078] (6) Multi-attribute decision making.

[0079] The advantages and disadvantages of using TOPSIS (Topology-Solution Ranking) to quantify the effectiveness of multi-attribute decision-making schemes:

[0080] Positive ideal solution: s + =(max Efficiency,min Cost,min Environment);

[0081] Negative ideal solution: s - =(min Efficiency,max Cost,max Environment);

[0082] Proximity: Several optimal precipitation schemes are selected based on Wi ranking. This represents the distance from solution i to the ideal solution s+; Wi represents the distance from solution i to the negative ideal solution s-; Wi represents the proximity of solution i, used for final ranking, with a larger value (closer to 1) indicating a better solution; s + and s - These are the benchmark reference points, used to calculate the distance from each scheme to these two points. and

[0083] Suppose there are n options, each with m attributes (such as efficiency, cost, and environmental impact). Let Z represent the j-th attribute of the i-th option. ij Eliminate the influence of units, for example, by standardizing the value of the j-th attribute:

[0084]

[0085] (7) Machine learning enhancement module.

[0086] Feature engineering theory encodes feasible precipitation schemes and input parameters into feature vectors: X = [OneHot(S), d, Δh, B], and eliminates the influence of dimensions through standardization (Z-score).

[0087] Classification model selection: Train a classifier based on historical data (such as random forest, SVM), with the objective function as follows: Among them, y i The label represents the historical best solution, and L is the cross-entropy loss function.

[0088] Input: Feasible precipitation schemes and input parameters.

[0089] Processing: Feature standardization (Z-score) and encoding (OneHot) are performed to eliminate dimensional differences; a classifier is used to predict the applicability probability of each feasible precipitation scheme.

[0090] Output: the applicability probability of each feasible precipitation scheme. Select the schemes with the highest applicability probabilities as the optimal schemes.

[0091] (8) Multi-objective optimization.

[0092] Input: Several optimal solutions obtained from uncertainty handling, multi-attribute decision-making, and machine learning enhancement modules.

[0093] Construct the Pareto front and screen for non-dominated solutions (e.g., solutions that are not inferior to other solutions in terms of efficiency, cost, and environment). Pareto optimal front: Define the non-dominated relationship between solutions: if solution s1 is not inferior to s2 in terms of efficiency, cost, and environment, then s1 dominates s2.

[0094] Dynamic weight allocation: For the environmental priority type, γ is calculated using the entropy weight method, while for the cost-sensitive type, β is adjusted through linear programming constraints.

[0095] For non-dominated solutions, calculate the objective function to obtain the final sorted set of optimal solutions S″.

[0096] (9) Verification and extension of the theoretical model.

[0097] Verification method: Inversion verification is performed through examples. Typical engineering cases (such as a soft soil foundation pit in a project) are input to check the consistency between the model output results and human expert decisions, thereby verifying the accuracy of the model. If the model output results deviate too much from the expert decisions, the model output results are further optimized and iterated according to the expert decisions until the model achieves the expected use effect.

[0098] Sensitivity analysis:

[0099] By changing the weights of α, β, and γ, the stability of the scheme ranking is observed. The Spearman rank correlation coefficient is used to calculate the correlation coefficient of the scheme ranking before and after the weight adjustment. If the coefficient is >0.9, the ranking stability is considered high; if it is <0.5, the weight allocation logic needs to be optimized and further iterative optimization is required until the scheme ranking is stable.

[0100] The system's robustness is verified by the rate of change of the Pareto front. The proportion of the original Pareto optimal solution retained after weight adjustment reflects the robustness of the scheme. For example, if 80% of the original optimal solution is still located on the new Pareto front under a weight perturbation of α ± 10%, the system has strong robustness.

[0101] Further directions: Digital twin integration, embedding theoretical models into the BIM platform to achieve dynamic optimization of precipitation schemes; considering the multi-party interest game among construction units, owners, and environmental protection departments, constructing an optimal decision-making model under Nash equilibrium.

[0102] Step S102: Automatic start / stop control of the water pump.

[0103] Based on the optimal precipitation plan, the logic control system integrates sensing technology, algorithm optimization, and actuators to achieve multi-dimensional collaborative control of the automatic start and stop of water pumps. Its core lies in achieving precise regulation through dynamic parameter feedback and intelligent decision-making. When the water level and other data monitored by the real-time monitoring system reach the engineering requirements, the logic control system stops precipitation. The automatic water pump control of the logic control system achieves a leap from extensive management to refined regulation through a closed-loop mechanism of "perception-decision-execution-feedback."

[0104] Based on a closed-loop mechanism of "perception-decision-execution-feedback", the system achieves steady-state operation (such as maintaining the target water level) by dynamically adjusting the output through real-time data. The theoretical model can be expressed as: Where u(t) is the control variable (pump power), e(t) = r(t) - y(t) is the water level deviation (set value - measured value); the proportional term K p e(t) function: responds to the current error and quickly adjusts the output; parameter Kp (proportional gain): the larger the value, the faster the response, but too large a value will lead to oscillation or overshoot; too small a value will result in a slow response, and steady-state error may remain. Integral term Accumulate historical errors to eliminate steady-state errors (such as long-term small deviations). Parameter Ki (integral gain): a value that is too small will result in slow error elimination; a value that is too large will lead to integral saturation (severe output fluctuations). Example: Continuous small water level deviations are gradually corrected through the integral term. e(τ): represents the error at a certain point in the past, τ. Differential term. Function: To predict error trends, suppress oscillations, and improve stability. Parameter Kd (differential gain): A value that is too high will amplify noise interference; a value that is too low will fail to effectively suppress overshoot.

[0105] Step 103: Real-time data monitoring and processing.

[0106] Water level monitoring: The water level sensor uses a float-type or capacitive sensor to monitor the water level changes in the pit in real time, with an accuracy of ±0.1cm. The data acquisition frequency can be manually adjusted (e.g., once / minute).

[0107] Environmental parameter monitoring: The environmental parameter module integrates external parameters such as soil density and moisture content, meteorological data (such as rainfall forecast), and dynamically adjusts start-up and shutdown strategies.

[0108] Flow and pressure monitoring: Electromagnetic flow meters and differential pressure sensors are installed at the inlet and outlet of the pumping pipe to monitor the actual pumping volume and head, preventing pipe blockage or pump overload. When the pumping volume is too low, the monitoring alarm module will issue a warning.

[0109] Data feedback: The system processes and feeds back data such as monitored water level, soil density and moisture content, meteorological data, and ground subsidence to the logic control system, which automatically adjusts the water pump power, controls the precipitation efficiency, and simultaneously realizes real-time monitoring and alarm functions for pump blockage.

[0110] Step 104: Data collection, storage, and intelligent analysis.

[0111] Data upload and storage: The IoT module integrates 4G / 5G or LoRa communication modules, supporting real-time data upload to the cloud platform with a coverage range of up to 5km. The real-time monitoring system continues to collect and store relevant information.

[0112] Multi-terminal interaction: Display water level trends, energy consumption statistics and equipment status through WeChat mini-programs and web interfaces, and support historical data comparison, analysis, export and report generation.

[0113] Local decision caching: When communication is interrupted, the PLC's built-in algorithm can still maintain basic operation based on the last valid data.

[0114] Machine learning iteration: Train neural network models using historical data to predict the optimal pumping rate under different geological conditions, fully record the entire process of intelligent precipitation in building engineering, and import the data into the digital twin system as a source for data optimization.

[0115] Traditional water level monitoring mainly relies on manual rope measurement, which results in low data collection frequency, large errors, and inability to provide real-time warnings of water levels exceeding safety limits. Water pump start-up and shutdown require manual operation, leading to significant response delays and potential water level loss or energy waste. This invention provides real-time monitoring based on a closed-loop mechanism of "system perception-decision-execution-feedback," which can optimize pumping power to reduce energy waste and automatically alarm for abnormal situations, such as pump blockage during precipitation or uneven ground subsidence, thereby reducing accident losses.

[0116] Example 2

[0117] This embodiment provides an intelligent precipitation control system for building engineering, which specifically includes:

[0118] The feasible solution screening module is configured to: obtain soil layer type and engineering requirements, and screen out several feasible precipitation schemes from the precipitation scheme database through geological adaptability rules and engineering constraint rules;

[0119] The first selection module is configured to: sample the probability distribution of the permeability coefficient and the water level depth to obtain uncertainty conditions, and select several optimal precipitation schemes from the feasible precipitation schemes under the uncertainty conditions;

[0120] The second selection module is configured to: select positive ideal solutions and negative ideal solutions from feasible precipitation schemes, and select several optimal precipitation schemes by using the approximation ideal solution ranking method;

[0121] The third selection module is configured to: encode engineering requirements and feasible precipitation schemes into feature vectors, predict the applicability probability of each feasible precipitation scheme through a classifier, and select several optimal precipitation schemes.

[0122] The final scheme selection module is configured to: construct the Pareto front for all optimal precipitation schemes, filter out non-dominated solutions, and calculate the objective function for the non-dominated solutions to obtain the final optimal precipitation scheme;

[0123] The control module is configured to: carry out precipitation based on the final optimal precipitation plan, and control the automatic start and stop of the water pump and the water pump power.

[0124] It should be noted that each module in this embodiment corresponds one-to-one with each step in Embodiment 1, and their specific implementation processes are the same, so they will not be repeated here.

[0125] Example 3

[0126] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the intelligent precipitation control method for building engineering as described in Embodiment 1 above.

[0127] Example 4

[0128] This embodiment provides a computer device, such as... Figure 2 As shown, the system includes a display device, an input device, a computer-readable storage medium (volatile memory and non-volatile storage medium), a processor, a communication interface (i.e., a network interface), and a computer program stored on the computer-readable storage medium and executable on the processor. The processor, communication interface, and computer-readable storage medium can be connected via a bus or other means. The communication interface is used to receive and send data, and when the processor executes the program, it implements the steps of the intelligent precipitation control method for building engineering described in Embodiment 1 above.

[0129] Any references to memory, storage, database, or other media used in this application and embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0130] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0131] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0132] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0133] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent precipitation control in building engineering, characterized in that, include: Obtain soil layer types and engineering requirements, and select several feasible precipitation schemes from the precipitation scheme database through geological adaptability rules and engineering constraint rules; The probability distribution of permeability coefficient and water level depth is sampled to obtain uncertainty conditions. Under these uncertainty conditions, several optimal precipitation schemes are selected from the feasible precipitation schemes. Among the feasible precipitation schemes, positive ideal solutions and negative ideal solutions are selected, and several optimal precipitation schemes are selected by the approximation ideal solution ranking method. The engineering requirements and feasible precipitation schemes are encoded into feature vectors. A classifier is used to predict the applicability probability of each feasible precipitation scheme and select several optimal precipitation schemes. For all optimal precipitation schemes, Pareto fronts are constructed to screen out non-dominated solutions. For the non-dominated solutions, the objective function is calculated to obtain the final optimal precipitation scheme. Rainfall is carried out based on the final optimal precipitation plan, and the automatic start and stop of water pumps and water pump power are controlled.

2. The intelligent precipitation control method for building engineering as described in claim 1, characterized in that, The project requirements include the depth of the foundation pit, the drawdown of the water level, and the budget.

3. The intelligent precipitation control method for building engineering as described in claim 1, characterized in that, Each precipitation plan includes precipitation methods, efficiency, cost, and environmental impact factors.

4. The intelligent precipitation control method for building engineering as described in claim 1, characterized in that, The objective function is a weighted sum of efficiency, cost, and environmental impact factors.

5. The intelligent precipitation control method for building engineering as described in claim 1, characterized in that, Under uncertainty, several optimal precipitation schemes are selected from feasible precipitation schemes by minimizing the maximum regret model.

6. The intelligent precipitation control method for building engineering as described in claim 1, characterized in that, The power of the water pump is adjusted based on the monitored water level, soil density and moisture content, meteorological data, and ground subsidence.

7. The intelligent precipitation control method for building engineering as described in claim 1, characterized in that, The permeability coefficient is assumed to follow a log-normal distribution, and the water level depth H is assumed to follow a uniform or gamma distribution. The rationality of the parameter distribution assumptions is verified by the KS test.

8. A smart precipitation control system for building engineering, characterized in that, include: The feasible solution screening module is configured to: obtain soil layer type and engineering requirements, and screen out several feasible precipitation schemes from the precipitation scheme database through geological adaptability rules and engineering constraint rules; The first selection module is configured to: sample the probability distribution of the permeability coefficient and the water level depth to obtain uncertainty conditions, and select several optimal precipitation schemes from the feasible precipitation schemes under the uncertainty conditions; The second selection module is configured to: select positive ideal solutions and negative ideal solutions from feasible precipitation schemes, and select several optimal precipitation schemes by using the approximation ideal solution ranking method; The third selection module is configured to: encode engineering requirements and feasible precipitation schemes into feature vectors, predict the applicability probability of each feasible precipitation scheme through a classifier, and select several optimal precipitation schemes. The final scheme selection module is configured to: construct the Pareto front for all optimal precipitation schemes, filter out non-dominated solutions, and calculate the objective function for the non-dominated solutions to obtain the final optimal precipitation scheme; The control module is configured to: carry out precipitation based on the final optimal precipitation plan, and control the automatic start and stop of the water pump and the water pump power.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the intelligent precipitation control method for building engineering as described in any one of claims 1-7.

10. A computer device comprising a computer-readable storage medium, a processor, and a computer program stored on the computer-readable storage medium and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the intelligent precipitation control method for building engineering as described in any one of claims 1-7.

Images