A foundation pit excavation scheme multi-dimensional collaborative optimization method and system

Abstract

The present application relates to the technical field of foundation pit construction, in particular to a method and system for multi-dimensional collaborative optimization of foundation pit excavation scheme. The method comprises the following steps: obtaining foundation pit monitoring data, extracting dynamic deformation parameters of the retaining structure according to the foundation pit monitoring data; constructing a soil pressure calculation model to obtain dynamic soil pressure parameters through the soil pressure calculation model; extracting construction period optimization features, cost optimization features and environmental resource coordination features, combining the dynamic deformation parameters of the retaining structure, the dynamic soil pressure parameters and the environmental resource coordination features to obtain safety environmental resource coordination parameters; completing multi-dimensional collaborative optimization of the foundation pit excavation scheme through the construction period optimization features, the cost optimization features and the safety environmental resource coordination parameters. The present application balances multiple targets of safety, economy, efficiency and environment, and solves the problems of low optimization precision, poor adaptability and difficulty in meeting the construction requirements of large foundation pits under complex geological conditions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of foundation pit construction, in particular to a foundation pit excavation scheme multi-dimensional collaborative optimization method and system. BACKGROUND

[0002] Foundation pit excavation is a core process in the early stage of construction engineering, and its scheme optimization needs to balance multiple objectives such as safety (deformation, soil pressure), economy (cost), efficiency (construction period), and environment (noise, groundwater). However, the existing technology has the following key defects, which makes it difficult to meet the needs of complex engineering, including:

[0003] The extraction of safety parameters is limited, and the deformation parameters of the enclosure structure are mainly dependent on static monitoring and fixed filtering algorithms, without considering dynamic parameters such as soil type and monitoring reliability for correction, resulting in low accuracy of deformation prediction. The calculation of soil pressure ignores the coupling effects of soil aging characteristics (such as internal friction angle decay) and excavation speed, and cannot reflect the dynamic changes of soil pressure during construction.

[0004] The multi-objective characteristics are fragmented, and the construction period optimization only considers single factors such as weather and equipment, without considering the constraints of geological conditions (such as soil cohesion) on process efficiency. The cost optimization lacks the correlation between risk cost (such as deformation repair) and time-varying parameters (such as material price fluctuations), resulting in a lag in cost warning. The extraction of environmental and resource characteristics does not deeply integrate geological, equipment, and construction stage data, and the utilization rate of multi-source information is low.

[0005] The collaborative optimization mechanism is missing, and the safety, environmental, and resource collaborative parameters are mainly based on empirical threshold settings, without quantitative derivation through objective weights (such as dynamic entropy weight) and optimal solution evaluation (such as TOPSIS). The final scheme optimization does not construct a reinforcement learning framework for the coupling of safety, construction period, and cost, which is prone to the contradiction between "high safety redundancy" and "excessive construction period and cost".

[0006] The above defects result in low precision and poor adaptability of existing scheme optimization, and there is an urgent need for a technical method covering the whole process of "multi-dimensional feature extraction, collaborative parameter derivation, and multi-objective optimization". SUMMARY

[0007] In view of the deficiencies of existing methods and the needs of practical applications, in order to solve the above problems, on the one hand, the present application provides a foundation pit excavation scheme multi-dimensional collaborative optimization method, including the following steps:

[0008] Obtain foundation pit monitoring data, extract the dynamic deformation parameters of the enclosure structure according to the foundation pit monitoring data; construct a soil pressure calculation model, and obtain dynamic soil pressure parameters through the soil pressure calculation model; extract the construction period optimization characteristics, cost optimization characteristics and environmental resource coordination characteristics, combine the dynamic deformation parameters of the enclosure structure, the dynamic soil pressure parameters and the environmental resource coordination characteristics to obtain the safety environmental resource coordination parameters; and complete the multi-dimensional collaborative optimization of the foundation pit excavation scheme through the construction period optimization characteristics, the cost optimization characteristics and the safety environmental resource coordination parameters.

[0009] The present application extracts the dynamic deformation parameters of the enclosure structure accurately through monitoring data, combines the soil characteristics and construction dynamics to construct a soil pressure model to obtain dynamic soil pressure parameters, solves the problem of low precision of traditional safety parameters and difficulty in reflecting construction dynamics, and lays a solid foundation for safety evaluation; extracts the construction period, cost and environmental resource coordination characteristics, breaks the limitations of multi-target characteristics, and makes each dimension of the characteristics fit the actual project; fuses safety and environmental resource parameters to obtain quantized safety environmental resource coordination parameters, avoids experience threshold deviation; couples multi-dimensional characteristics to complete optimization, realizes safety (deformation and soil pressure controllable), economy (less cost overrun), efficiency (small construction period deviation), environment (noise and groundwater up to standard) coordination, adapts to complex scenarios, improves the scientificity and practicality of the foundation pit scheme, and ensures the safe and efficient implementation of the project.

[0010] Optionally, the extracting the dynamic deformation parameters of the enclosure structure according to the foundation pit monitoring data comprises the following steps:

[0011] Adjusting coefficients are constructed according to soil types and monitoring reliability, and the state equation is corrected by using the adjusting coefficients; a dynamic weight factor is constructed by using a deformation rate and cohesion to optimize Kalman gain, and a geological adaptive filtering mechanism is obtained; and the dynamic deformation parameters of the enclosure structure are extracted in combination with the corrected state equation and the geological adaptive filtering mechanism. The present application constructs adjusting coefficients according to soil types, monitoring reliability and the like, corrects the state equation to make it fit the geological characteristics and data quality, and avoids pure static hypothesis deviation; introduces a dynamic weight factor to optimize Kalman gain, constructs a geological adaptive filtering mechanism, and balances model prediction and measured data; and extracts parameters in combination with the two, greatly improves the precision of parameters such as deformation peak value and rate, provides accurate data support for foundation pit safety evaluation, avoids safety risks caused by inaccurate parameters, and is a safety data basis for subsequent multi-dimensional collaborative optimization.

[0012] Optionally, the constructing a soil pressure calculation model and obtaining dynamic soil pressure parameters through the soil pressure calculation model comprises the following steps:

[0013] According to the time-dependent characteristics of the soil body and the excavation speed, a dynamic active earth pressure coefficient is obtained; an earth pressure calculation model is constructed in combination with the effective unit weight, the pore water pressure and the dynamic active earth pressure coefficient, and a dynamic earth pressure parameter is obtained through the earth pressure calculation model. The dynamic active earth pressure coefficient is obtained in combination with the time-dependent characteristics of the soil body and the excavation speed, so that the coefficient is fitted to the time-dependent change of the soil mechanics and the operation rhythm of the construction; the model is constructed in combination with the effective unit weight and the pore water pressure, so that the stress mechanism of the earth pressure under the actual working condition is comprehensively reflected; finally, the dynamic earth pressure parameter is accurately obtained, reliable data support is provided for the safety evaluation of the foundation pit support design, deformation prevention and control, and the safety risk caused by the deviation of the earth pressure calculation is avoided, which is an important safety data basis for subsequent multi-dimensional collaborative optimization.

[0014] Optionally, the construction period optimization feature extraction includes the following steps:

[0015] According to the construction period influencing factors, the key process time is calculated; and the first fitness function is constructed to extract the construction period optimization feature according to the first fitness function. The key process time is calculated in combination with the geological factors (soil cohesion, moisture content), personnel factors (attendance rate) and material factors (supply delay), so as to avoid the deviation of the static assumption, and make the process time fit the actual construction dynamics; the first fitness function is constructed to take into account the construction period, equipment resources, personnel cost, material waste and quality risk, so as to avoid the invalid scheme of "meeting the construction period but exceeding the cost and quality rework", and not simply pursue the shortest construction period; and finally, the optimized key path and construction period redundancy are extracted, which provides accurate construction period data support for subsequent multi-dimensional collaborative optimization, and is an important link to realize the balance of "safety-economy-efficiency".

[0016] Optionally, the cost optimization feature extraction includes the following steps:

[0017] The time-varying parameter is introduced, the dynamic earned value is calculated based on the time-varying parameter; and the cost optimization feature is extracted based on the dynamic earned value and by using the particle swarm optimization algorithm. The dynamic earned value is calculated by introducing the material fluctuation coefficient (fitting the market dynamics) and the risk cost coefficient (related to the deformation overrun risk), so as to reflect the real change of the cost in real time and avoid the problem of large deviation between the static budget and the actual situation; the particle swarm optimization is used based on the dynamic earned value, the resource allocation is optimized with the target of minimum cost + progress guarantee, and the contradiction between the simple cost reduction and the construction period delay is avoided; and finally, the cost deviation warning and the optimal resource ratio are extracted, which provides accurate cost data support for subsequent multi-dimensional collaborative optimization, and is the core economic dimension guarantee to realize the balance of safety-construction period-cost.

[0018] Optionally, the environmental resource collaborative feature extraction includes the following steps:

[0019] According to the geological-equipment-environment-construction stage multi-dimensional parameters, the forgetting gate weight of the long short-term memory network is improved; based on the multi-dimensional time sequence data, the improved long short-term memory network is used to extract the environment resource coordination features. The application improves the weight by combining the geological risk, equipment load, environmental interference and construction stage multi-dimensional engineering parameters, so that the model can dynamically judge the historical information retention value (such as equipment failure, pit bottom construction, more key data retention); based on multi-dimensional time sequence data (groundwater level, geological parameters, equipment state, etc.), the improved LSTM is used to extract features, so that the underground water control threshold, equipment scheduling priority and other results are more suitable for the actual working condition of the foundation pit, and the disconnection between features and engineering caused by “pure data driving” is avoided; finally, accurate environment resource dimension support is provided for subsequent safety environment resource coordination parameter derivation, which is an important link to realize the safety-environment-resource coordination.

[0020] Optionally, the safety environment resource coordination parameters are obtained by combining the enclosure dynamic deformation parameters, the dynamic soil pressure parameters and the environment resource coordination features, including the following steps:

[0021] The weighted evaluation index matrix is constructed based on dynamic entropy weight by combining the enclosure dynamic deformation parameters, the dynamic soil pressure parameters and the environment resource coordination features; the scheme coordination is evaluated by using the distance method, and the safety environment resource coordination parameters are obtained by combining the scheme coordination and the weighted evaluation index matrix. The application first fuses the enclosure deformation, soil pressure (safety core) and environment resource features (environment-resource dimension), constructs a weighted matrix based on dynamic entropy weight, and dynamically adjusts the weight with the construction stage (such as increasing the deformation parameter weight at the pit bottom stage and increasing the noise parameter weight in the residential area), so as to avoid the fixed weight from deviating from the engineering practice; then the scheme coordination is quantified by TOPSIS, and the optimal matching parameters in multiple dimensions are objectively selected to avoid subjective deviation of the experience threshold. Finally, the coordination safety threshold, resource-safety balance coefficient and other parameters are obtained, which lay a solid safety-environment data foundation for subsequent “safety-duration-cost” multi-dimensional optimization, ensure that the optimization process does not ignore the safety bottom line and environmental constraints, and avoid the imbalance problems of “emphasizing efficiency and ignoring safety” or “emphasizing safety and ignoring environment”.

[0022] Optionally, the safety environment resource coordination parameters are obtained by combining the enclosure dynamic deformation parameters, the dynamic soil pressure parameters and the environment resource coordination features, including the following steps:

[0023] The optimal trade-off point of duration and cost and the resource progress matching degree are obtained by the duration optimization feature and the cost optimization feature; the multi-dimensional coordinated optimization of the foundation pit excavation scheme is completed by combining the optimal trade-off point of duration and cost, the resource progress matching degree and the safety environment resource coordination parameters.

[0024] Optionally, the construction period cost optimal trade-off point, the resource progress matching degree and the safety environment resource coordination parameter are combined to complete multi-dimensional collaborative optimization of the foundation pit excavation scheme, including the following steps:

[0025] The construction period cost optimal trade-off point, the resource progress matching degree and the safety environment resource coordination parameter are combined to construct a key indicator state vector and define a construction decision action vector; a multi-objective collaborative optimization function is designed, and based on the key indicator state vector, the construction decision action vector and the multi-objective collaborative optimization function, multi-dimensional collaborative optimization of the foundation pit excavation scheme is completed by using a deep reinforcement learning framework. The optimal trade-off point and the resource progress matching degree are obtained by the construction period and cost characteristics, avoiding imbalance of short construction period and high cost or low cost and over construction period; the safety environment parameter is fused to construct the key indicator state vector (including safety threshold and construction period cost) and the construction decision action vector (construction parameter), so that the optimization is anchored to the engineering actual scene; finally, safety, economy, efficiency and environment are coordinated by using the multi-objective function and the deep reinforcement learning, and a multi-dimensional collaborative closed loop is formed to provide a scientific and feasible optimal scheme for a complex foundation pit.

[0026] In order to efficiently execute the multi-dimensional collaborative optimization method of the foundation pit excavation scheme provided by the present application, the present application further provides a multi-dimensional collaborative optimization system of the foundation pit excavation scheme, which comprises a processor, an input device, an output device and a memory, and the processor, the input device, the output device and the memory are connected with each other, wherein the memory is used for storing a computer program, the computer program contains program instructions, and the processor is configured to call the program instructions to execute the multi-dimensional collaborative optimization method of the foundation pit excavation scheme according to the first aspect of the present application. The multi-dimensional collaborative optimization system of the foundation pit excavation scheme has compact structure and stable performance, can stably execute the multi-dimensional collaborative optimization method of the foundation pit excavation scheme provided by the present application, and further improves the overall applicability and practical application ability of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 A multi-dimensional collaborative optimization method flow chart of the foundation pit excavation scheme provided by the embodiment of the present application is provided.

[0028] Figure 2 A multi-dimensional collaborative optimization system block diagram of the foundation pit excavation scheme provided by the embodiment of the present application is provided. DETAILED DESCRIPTION

[0029] Specific embodiments of the present invention will be described in detail below. 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.

[0030] 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.

[0031] Please see Figure 1 To address the aforementioned problems, this invention provides a multi-dimensional collaborative optimization method for foundation pit excavation schemes, such as... Figure 1 As shown, in one embodiment, the method includes the following steps:

[0032] S1. Obtain foundation pit monitoring data, and extract dynamic deformation parameters of the retaining structure based on the foundation pit monitoring data.

[0033] The foundation pit monitoring data includes two core types of data: horizontal displacement and vertical displacement, which are collected in real time through high-frequency sampling.

[0034] Specifically, the step of extracting dynamic deformation parameters of the retaining structure based on the foundation pit monitoring data includes the following steps:

[0035] S11. Construct adjustment coefficients based on soil type and monitoring reliability, and use the adjustment coefficients to correct the state equation.

[0036] Soil type data refers to the proportion and distribution of sandy soil and cohesive soil within the influence area of ​​the foundation pit. Adjustment coefficients are constructed based on soil type and monitoring reliability, satisfying the following formula:

[0037]

[0038] in, This indicates that the adjustment coefficient is used to control the decay rate of the process noise covariance. Indicates the proportion of sandy soil. Standard deviation of horizontal displacement monitoring value in the last 5 times, Mean value of horizontal displacement monitoring value in the last 5 times. As a monitoring data reliability coefficient, it reflects the reliability of displacement monitoring data. The higher the reliability is, the lower the adjustment coefficient can be appropriately reduced to avoid excessive attenuation of process noise covariance.

[0039] Further, the state equation is corrected by using the adjustment coefficient, and the state equation satisfies:

[0040]

[0041] Wherein, Indicates the state vector at time k, Indicates the vector-valued function, Indicates the control input, Indicates the process noise.

[0042] The state vector integrates the key state variables of the foundation pit, which satisfies: , Indicate the horizontal displacement of the foundation pit wall and the vertical displacement of the foundation bottom respectively. By arranging high-precision total station, static level and other monitoring equipment, millimeter-level deformation can be captured in real time, providing data basis for construction safety warning; Indicate the soil cohesion and internal friction angle respectively.

[0043] The vector-valued function corresponds to the transfer sub-function of each component (horizontal displacement, vertical displacement, cohesion, internal friction angle) in the state vector.

[0044] The horizontal displacement transfer sub-function satisfies:

[0045]

[0046] Wherein, Indicates the groundwater depth, Indicates the safety cohesion threshold.

[0047] The vertical displacement transfer sub-function satisfies:

[0048]

[0049] Wherein, Indicates the vertical displacement coefficient (mm / m), which is the vertical displacement of the foundation caused by unit excavation depth (such as 4-6 for sandy soil and 2-4 for cohesive soil, which is calibrated by indoor triaxial test), Indicates the maximum internal friction angle.

[0050] The cohesion transfer sub-function satisfies:

[0051]

[0052] The internal friction angle transfer sub-function satisfies:

[0053]

[0054] wherein, represents the cohesion attenuation coefficient, represents the soil type basis attenuation coefficient is 0.013, represents the internal friction angle attenuation coefficient, represents the soil type basis attenuation coefficient is 0.008, represents the actual construction mechanical power, represents the reference mechanical power, and the power of a medium-sized machine commonly used in foundation pit engineering is taken as the correction reference.

[0055] Control input Focused excavation depth This key control variable directly drives the evolution of the deformation of the foundation pit; considering the layered and segmented excavation process, it can be further decomposed into single-layer excavation thickness, layered excavation sequence and other sub-parameters. For example, when constructing by the top-down method, the excavation depth needs to be time-matched with the erection time of the support structure to avoid excessive stress release of the soil due to long exposure time caused by excavation.

[0056] As an unmeasurable disturbance term in a dynamic system, its essence is a mathematical abstraction of random factors such as geological condition mutation, construction equipment vibration, and groundwater seepage during foundation pit excavation, and it obeys zero-mean Gaussian distribution , denotes the covariance matrix, satisfying: , represents the initial covariance matrix.

[0057] S12, a dynamic weight factor is constructed by the deformation rate and the cohesion to optimize the Kalman gain, and a geological adaptive filtering mechanism is obtained.

[0058] In the embodiment, a dynamic weight factor is constructed by the deformation rate and the cohesion, satisfying the following formula:

[0059]

[0060] wherein, represents the dynamic weight factor, represents the deformation rate, represents the deformation rate safety threshold, which is determined based on specifications and engineering experience, is the geological risk degree, reflecting the gap between the current cohesion and the safety threshold.

[0061] Further, a dynamic weight factor is introduced on the basis of a traditional Kalman gain formula to construct a geological adaptive filtering mechanism to meet the following conditions:

[0062]

[0063] wherein, K represents a Kalman gain matrix, P represents a state covariance matrix, H represents an observation matrix, R represents an observation noise covariance matrix.

[0064] S13, combined with the corrected state equation and the geological adaptive filtering mechanism, the dynamic deformation parameters of the enclosure structure are extracted.

[0065] Combined with the corrected state equation and the geological adaptive filtering mechanism, the dynamic deformation parameters of the enclosure structure are extracted, the dynamic deformation parameters of the enclosure structure include a dynamic deformation peak value, a deformation rate and a deformation stability coefficient, and the deformation stability coefficient refers to a ratio of a horizontal displacement of a foundation pit wall to the dynamic deformation peak value.

[0066] The present application integrates geological parameters into the state equation and gain calculation of the extended Kalman filter in real time, dynamically balances the model prediction and monitoring data through a dynamic weight factor, and is more in line with the physical mechanism of geological control deformation (the greater the cohesion, the more stable the deformation, and the filter weight should tend to the model).

[0067] S2, a soil pressure calculation model is constructed, and dynamic soil pressure parameters are obtained through the soil pressure calculation model.

[0068] In the embodiment, the soil pressure calculation model is constructed, and the dynamic soil pressure parameters are obtained through the soil pressure calculation model, including the following steps:

[0069] S21, a dynamic active soil pressure coefficient is obtained according to the time-dependent characteristics of the soil body and the excavation speed.

[0070] The dynamic active soil pressure coefficient comprehensively considers the double influences of the time-dependent characteristics of the soil body and the excavation speed, and its expression is:

[0071]

[0072] Kd represents a dynamic active soil pressure coefficient, represents a time-dependent internal friction angle, which reflects the attenuation law of the shear strength parameter of the soil body with time, is an initial internal friction angle of the soil body, which is provided by a geological survey report, is a disturbance attenuation coefficient, and the disturbance attenuation coefficient of sandy soil is 0.02 and the disturbance attenuation coefficient of clayey soil is 0.005, represents a real-time excavation speed, represents the maximum excavation speed determined based on the engineering safety threshold, when the excavation speed approaches the value, the soil pressure release obviously lags behind, through is corrected.

[0073] S22, combining the effective gravity, the pore water pressure and the dynamic active soil pressure coefficient, a soil pressure calculation model is constructed, and a dynamic soil pressure parameter is obtained through the soil pressure calculation model.

[0074] The effective gravity is the actual gravity of the soil considering the buoyancy of groundwater, and the calculation formula is:

[0075]

[0076] represents the effective gravity, represents the saturated gravity, represents the depth of the pore water pressure, the pore water pressure increases with the increase of the depth of the soil body, and the value is obtained by linear interpolation through the data of the multi-layer water level monitoring points arranged around the foundation pit.

[0077] Further, the soil pressure calculation model is constructed by combining the effective gravity, the pore water pressure and the dynamic active soil pressure coefficient, and the following formula is satisfied:

[0078]

[0079] wherein, represents the dynamic soil pressure.

[0080] The dynamic soil pressure parameter is obtained through the soil pressure calculation model, including the dynamic soil pressure peak value, the soil pressure distribution uniformity and the soil pressure-excavation speed sensitivity .

[0081] The traditional Rankine formula is a classic soil pressure calculation method and has been used in foundation pit engineering design for a long time. However, the formula essentially belongs to a static calculation model and has significant limitations in actual engineering scenarios. Firstly, the influence of speed change on soil pressure release during foundation pit excavation is not considered, while different excavation speeds in actual construction will lead to dynamic change characteristics of soil pressure; secondly, the time-dependent characteristics of geological materials under construction disturbance are ignored, and the decay law of soil mechanical parameters with time cannot be reflected. When the soil particles cannot be rearranged quickly due to rapid excavation, the soil pressure is too large, and long-term construction will lead to the decay of the internal friction angle, and the invention simultaneously integrates the excavation speed (construction dynamics) and the time-dependent internal friction angle (geological dynamics), through excavation speed correction and time-dependent internal friction angle correction, which is more consistent with the actual construction physical process.

[0082] S3, extract the construction period optimization feature, the cost optimization feature and the environment resource coordination feature, combine the enclosure dynamic deformation parameter, the dynamic soil pressure parameter and the environment resource coordination feature, and obtain a safe environment resource coordination parameter.

[0083] The extraction of the construction period optimization feature includes the following steps:

[0084] S31, calculate the critical process time according to the construction period influencing factors.

[0085] The time of the foundation pit excavation critical process (such as layered excavation, support construction, and dewatering operation) is affected by the multi-dimensional coupling of “geology-equipment-personnel-material-weather”, and the calculation formula is:

[0086]

[0087] Among them, represents the i-th critical process time, represents the reference time of the i-th critical process, which is determined by the process standard, such as the reference time of 8 hours per layer for sandy soil layered excavation;

[0088] represents the geological condition influence coefficient, reflecting the restriction of soil mechanical properties on process efficiency, represents the water content, represents the saturated water content of the soil;

[0089] represents the equipment failure influence coefficient, represents the equipment repair time, represents the equipment failure level coefficient (1 for light failure, 1.5 for moderate failure, and 2 for severe failure);

[0090] represents the personnel configuration influence coefficient, reflecting the influence of actual on-site personnel on process efficiency, represents the planned on-site rate, represents the real-time personnel on-site rate;

[0091] represents the material supply influence coefficient, reflecting the influence of support material delay on the process, and the longer the delay, the larger the coefficient, represents the actual arrival time of materials, represents the planned arrival time of materials;

[0092] represents the weather influence coefficient, represents the hourly rainfall influence coefficient, no rain , light rain , moderate rain , and above heavy rain , denotes the average wind speed influence coefficient, no wind , light wind , strong wind .

[0093] S32, construct a first fitness function, and extract the construction period optimization features according to the first fitness function.

[0094] Specifically, based on the five types of targets of shortest construction period, equipment saving, personnel cost control, material waste rate reduction and process quality risk avoidance, a first fitness function of multi-dimensional coordination is constructed, which meets:

[0095]

[0096] wherein, denotes the first fitness function value, denotes the total construction period of the critical path, denotes the critical path, denotes the actual number of input equipment, denotes the total number of available equipment, denotes the actual personnel cost, denotes the planned personnel cost, denotes the material waste rate, denotes the actual amount of supporting material, denotes the theoretical amount, denotes the process quality risk, denotes the actual concrete strength, denotes the design strength, denotes the actual installation deviation of steel support, denotes the allowable deviation.

[0097] Further, the constraint conditions include personnel configuration constraints, material waste constraints, quality risk constraints and resource limitations, and the first fitness function is taken as the objective function to maximize, and the construction period optimization features including the optimized critical path, process float time, construction period redundancy, personnel cost redundancy, material waste control rate and quality risk control rate are solved by genetic algorithm.

[0098] In the embodiment, the cost optimization features include the following steps:

[0099] S33, introduce time-varying parameters, and calculate dynamic earned value based on the time-varying parameters.

[0100] The time-varying parameter refers to a parameter that dynamically changes with the construction time t and needs to be adjusted based on real-time / periodic update data. The core role is to break through the limitations of static cost calculation and make the cost optimization fit the dynamic reality of "material price fluctuations, safety risk changes and resource unit price adjustments" in foundation pit excavation.

[0101] Specifically, a time-varying parameter is introduced, and a dynamic earned value is calculated based on the time-varying parameter, satisfying the following formula:

[0102]

[0103] wherein, represents the dynamic earned value, represents the i-th process budget value determined according to the initial planning of the foundation pit excavation project, represents a material price fluctuation coefficient, which is used to quantify the influence of material market price fluctuation on cost, represents the current price of the material at time t, represents the material reference price, which is taken from market research data at the time of project establishment, represents a risk cost coefficient, represents the maximum deformation of the foundation pit at time t, represents the allowable deformation threshold, represents the deformation repair cost, which is estimated according to historical similar engineering repair cases and current market labor and material prices.

[0104] S34, based on the dynamic earned value, a particle swarm optimization algorithm is used to extract cost optimization features.

[0105] The particle swarm optimization algorithm is used to solve the optimal solution of resource input of each process of foundation pit excavation, and the labor, equipment, material and other resource inputs are abstracted as optimization variables.

[0106] Specifically, the objective function is constructed with the goal of minimizing the total cost C of the project:

[0107]

[0108] wherein, represents the optimization variable, represents the unit price of the i-th resource at time t.

[0109] Further, the inertia weight of the particle swarm optimization algorithm adopts an adaptive adjustment strategy, satisfying:

[0110]

[0111] represents the inertia weight, represents the upper limit of the budget, represents the theoretical minimum cost. By automatically reducing the inertia weight when the cost is close to the upper limit of the budget, the algorithm search granularity is refined, which helps to find a better resource allocation scheme in the case of resource shortage; when the cost is far from the upper limit, the inertia weight is increased to speed up the global search efficiency.

[0112] Further, the extracted cost optimization features include cost deviation early warning values, optimal resource allocation ratios, and risk cost reserves.

[0113] In yet another embodiment, the environment resource coordination features are extracted, including the following steps:

[0114] S35, according to the geological-equipment-environment-construction stage multi-dimensional parameters, the long short-term memory network forget gate weight is improved.

[0115] The geological-equipment-environment-construction stage multi-dimensional parameters include groundwater level, soil cohesion, internal friction angle, construction noise, equipment operating load and excavation speed. According to the geological-equipment-environment-construction stage multi-dimensional parameters, the long short-term memory network forget gate weight is improved to meet:

[0116]

[0117] Wherein, The long short-term memory network forget gate weight is represented by The Sigmoid activation function is represented by The weight coefficients of geological risk, equipment risk, construction stage coefficient and environmental interference degree are determined by the analytic hierarchy process combined with the priority of foundation pit engineering, The geological risk degree is represented by, which reflects the degree of soil cohesion deviating from the safety threshold. The higher the risk, the more historical geological data needs to be retained.

[0118] The equipment risk degree is represented by, which reflects the degree of equipment load exceeding the limit or failure. The higher the risk, the more historical equipment data needs to be retained, The real-time equipment load is represented by The rated maximum load is represented by

[0119] The construction stage coefficient is represented by, which reflects the difference in the demand for historical information in different construction stages. More geological data needs to be retained in the later period.

[0120] The environmental interference degree is represented by, which reflects the interference degree of environmental factors such as heavy rain and high noise. The higher the interference, the more historical environmental data needs to be retained, The real-time hourly rainfall is represented by The heavy rain threshold is represented by The real-time noise decibel is represented by The construction noise limit value is represented by

[0121] S36, based on multi-dimensional time series data, environment resource coordination features are extracted using an improved long short-term memory network.

[0122] According to the improved long short-term memory network forgetting gate weight adjustment hidden layer state update formula, the environment resource coordination characteristics based on multi-dimensional time sequence data are output, including the groundwater control threshold, the geological risk partition, the equipment scheduling priority and the noise early warning value.

[0123] Further, the safety environment resource coordination parameters are obtained by combining the envelope structure dynamic deformation parameters, the dynamic soil pressure parameters and the environment resource coordination characteristics, including the following steps:

[0124] S37, the weighted evaluation index matrix is constructed based on dynamic entropy weight by combining the envelope structure dynamic deformation parameters, the dynamic soil pressure parameters and the environment resource coordination characteristics.

[0125] In the embodiment, the evaluation index matrix containing 7 indexes is constructed based on the core control elements of foundation pit engineering, and the following conditions are met: wherein, represents the maximum horizontal displacement of the foundation pit, represents the displacement rate, represents the maximum active soil pressure, represents the groundwater level change, represents the total amount of resource input, covering the cost of manpower, materials and equipment, represents the duration, represents the construction noise decibel value.

[0126] Further, the basic entropy weight is adjusted by introducing a stage adjustment coefficient, and then the weighted evaluation index matrix is constructed, and the following conditions are met: wherein, represents the dynamic weight of the jth index, represents the stage adjustment coefficient of the jth index, represents the basic entropy weight of the jth index.

[0127] The core role of the stage adjustment coefficient is to correct the weight of each evaluation index (such as safety risk, cost deviation, and progress compliance rate) according to the differences in core objectives (safety in the early stage, efficiency in the middle stage, and cost in the later stage) of different construction stages (such as the early, middle, and late stages of excavation), avoiding one-size-fits-all evaluation. Using industry standards (such as the "Technical Standard for Monitoring of Building Foundation Pit Engineering") or historical project data of the enterprise, "stage key threshold values" (such as deformation limit values and cost overrun warning lines) are set in advance, and the "threshold value compliance situation" is directly mapped to the adjustment coefficient. For example, by checking the specifications / historical data, the "key threshold values" of each stage (such as a deformation limit value of ≤30 mm in stage 2 and a cost overrun of ≤5%) are determined, the "threshold value compliance degree" (such as compliance degree = actual value / threshold value, for example, actual deformation of 25 mm, compliance degree = 25 / 30 ≈ 0.83) is defined, and the mapping rule of "compliance degree → adjustment coefficient" (such as coefficient = 0.6 + 0.8 × compliance degree) is established. The construction stage characteristics are quantified as weight adjustment coefficients in the present application, realizing "construction stage-multi-objective" dynamic coordination.

[0128] S38, the scheme coordination is evaluated by using the TOPSIS method, and the safety environment resource coordination parameters are obtained by combining the scheme coordination and the weighted evaluation index matrix.

[0129] Specifically, the scheme coordination is evaluated by using the TOPSIS method, and the positive ideal solution is defined as: and the negative ideal solution is: , which respectively represent the optimal and worst values of each index in all schemes.

[0130] Further, the closeness degree of each scheme to the ideal solution is calculated by the closeness degree calculation formula, which satisfies: , , wherein represents the closeness degree of the i-th scheme to the ideal solution, i.e., the scheme coordination, and the closer the closeness degree value is to 1, the better the scheme performs in multi-index coordination, , wherein represents the i-th row of the evaluation index matrix.

[0131] Further, the safety environment resource coordination parameters are obtained by combining the scheme coordination and the weighted evaluation index matrix, including:

[0132] The coordination safety threshold value: when the closeness degree is ≥0.8, the critical value of the corresponding index constitutes the coordination safety threshold value, which provides a quantitative standard for construction safety control;

[0133] The resource-safety balance coefficient: the ratio of the closeness degree to the duration redundancy, which reflects the balance between safety guarantee and resource utilization by relating the closeness degree to the duration redundancy;

[0134] The environment-construction coordination degree: taking the closeness degree as the numerator and the ratio of the noise actual value to the environmental standard threshold value as the denominator, it quantifies the coordination degree of construction activities and environmental constraints.

[0135] S4, complete multi-dimensional collaborative optimization of the foundation pit excavation scheme through the construction period optimization feature, the cost optimization feature, and the safety environment resource coordination parameter.

[0136] The multi-dimensional collaborative optimization of the foundation pit excavation scheme through the construction period optimization feature, the cost optimization feature, and the safety environment resource coordination parameter includes the following steps:

[0137] S41, obtain a construction period cost optimal trade-off point and a resource progress matching degree through the construction period optimization feature and the cost optimization feature.

[0138] Based on the non-dominated sorting genetic algorithm II (NSGA-II) framework, a double-objective optimization function is set for the construction period optimization feature and the cost optimization feature, i.e., minimizing the construction period and the cost. After obtaining a Pareto frontier solution set, a non-dominated solution with uniform distribution and wide coverage is selected by calculating the individual crowding degree index.

[0139] Further, by quantifying the deviation degree of the construction period and the cost relative to the initial plan, the optimal balance point of the construction period-cost double objective is accurately located with the objective of minimizing the Euclidean distance, thereby providing a scientific and reasonable foundation pit excavation scheme for decision makers.

[0140] Resource progress matching degree As a core index for quantifying the coordination of resource input and construction progress in the foundation pit excavation process, the matching efficiency of resource consumption and time progress is reflected through weighted normalization, which satisfies: wherein, represents the standardized input intensity of the ith resource (such as manpower, machinery, materials, etc.), which is determined by the ratio of the actual resource input amount to the unit time reference input amount, represents the standardized duration of the ith construction phase, which is represented by the ratio of the actual construction period to the planned reference construction period. The numerator item reflects the comprehensive contribution of the single resource in the corresponding construction phase, and the denominator item summarizes all resource-time products, achieving dimensionless processing of the index. Through normalized weight distribution, potential contradictions between resource input and progress arrangement can be accurately identified, providing a quantitative basis for dynamically optimizing resource allocation schemes and balancing construction efficiency and cost.

[0141] S42, complete multi-dimensional collaborative optimization of the foundation pit excavation scheme in combination with the construction period cost optimal trade-off point, the resource progress matching degree, and the safety environment resource coordination parameter.

[0142] The multi-dimensional collaborative optimization of the foundation pit excavation scheme in combination with the construction period cost optimal trade-off point, the resource progress matching degree, and the safety environment resource coordination parameter includes the following steps:

[0143] First, by combining the optimal trade-off point between construction period and cost, the resource schedule matching degree, and the safety, environment, and resource coordination parameters, a key indicator state vector is constructed and a construction decision action vector is defined.

[0144] Key Indicator State Vector , Indicates the depth of the foundation pit. This indicates environmental impact indicators, such as the settlement of surrounding buildings. These represent the actual project duration and cost, respectively. This vector comprehensively depicts the overall status of safety, environment, resources, schedule, and cost.

[0145] Construction decision-making action vector It includes adjustable construction decision variables. Indicates the excavation speed of the foundation pit. Indicates the timing of support structure construction. This indicates the intensity of resource input, such as the quantity of machinery and equipment deployed. This indicates the rate of dewatering and pumping out of the foundation pit.

[0146] Then, a multi-objective collaborative optimization function is designed. Based on the key indicator state vector, the construction decision action vector, and the multi-objective collaborative optimization function, a deep reinforcement learning framework is used to complete the multi-dimensional collaborative optimization of the foundation pit excavation scheme.

[0147] Specifically, the multi-objective collaborative optimization function satisfies:

[0148]

[0149] in, This represents the reward function value. Indicates the deformation safety threshold. Indicates the planned construction period. Indicates budget cost, This indicates the permissible threshold for environmental impact. This indicates a penalty for exceeding the limit. Or when the maximum active earth pressure exceeds the earth pressure safety threshold, Strong negative feedback will be given for any violations to ensure that the plan meets the bottom line of safety.

[0150] Furthermore, a deep reinforcement learning framework is employed. The policy network outputs action suggestions based on the current key indicator state vector, while the value network evaluates the long-term value of the actions. An experience replay mechanism stores historical state-action-reward sequences, reducing data correlation and improving learning stability. The target network periodically updates parameters to reduce fluctuations during training. After multiple rounds of iterative optimization, when the optimal key indicator state vector is reached, the policy network outputs the final optimal mining scheme, achieving scheme optimization under multi-objective collaboration.

[0151] Referring to Figure 2 In the embodiment, in order to efficiently implement the foundation pit excavation scheme multi-dimensional collaborative optimization method provided by the present application, the present application further provides a foundation pit excavation scheme multi-dimensional collaborative optimization system, which comprises an input device, an output device, a processor and a memory, the input device, the output device, the processor and the memory are connected to each other, the memory contains program instructions, and the program instructions are used for the steps of the foundation pit excavation scheme multi-dimensional collaborative optimization method. The foundation pit excavation scheme multi-dimensional collaborative optimization system of the present application has compact structure and stable performance, can stably implement the foundation pit excavation scheme multi-dimensional collaborative optimization method of the present application, and further improves the overall applicability and practical application ability of the present application.

[0152] In the embodiment, the processor can be a central processing unit, and can also be other general-purpose processors, digital signal processors, application-specific integrated circuits, ready programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The input device can be used to obtain data information. The output device can be used to output the results obtained by the program instructions contained in the computer program stored in the memory provided by the present application. The memory can include read-only memory and random access memory, and provide instructions and data to the processor. Part of the memory can also include non-volatile random access memory.

[0153] In a possible implementation, the memory can include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required by a function, etc.; the data storage area can store data created during use. In addition, the memory can include read-only memory and random access memory, and provide instructions and data to the processor. The memory stores an operating system and operation instructions, executable modules or data structures, or subsets thereof, or an expanded set thereof, wherein the operation instructions can include various operation instructions for implementing various operations. The operating system can include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0154] The embodiment also provides a storage medium having a computer program stored thereon, and the computer program is executed by the processor to implement the steps of the foundation pit excavation scheme multi-dimensional collaborative optimization method.

[0155] The storage medium can include a U disk, a mobile hard disk, a read-only memory, a random access memory, a magnetic disk or an optical disk, and various media that can store program codes.

[0156] To sum up, the application extracts the dynamic deformation parameters of the enclosure structure accurately by monitoring data, constructs the dynamic soil pressure model by combining the soil characteristics and the construction dynamics to obtain the dynamic soil pressure parameters, solves the problem of low precision of traditional safety parameters and difficulty in reflecting the construction dynamics, and lays a solid foundation for safety evaluation; extracts the construction period, cost, and environmental resource coordination characteristics, breaks the limitations of multi-target characteristics, and makes each dimension characteristic fit the engineering practice; fuses the safety and environmental resource parameters to obtain the quantitative safety and environmental resource coordination parameters, avoids the deviation of experience threshold, and couples the multi-dimensional characteristics to complete optimization and realize the safety, economy, efficiency, and environmental coordination, adapt to complex scenes, improve the scientificity and practicality of the foundation pit scheme, and ensure the safe and efficient implementation of the project.

[0157] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope recorded in the present application.

Claims

1. A method for multi-dimensional collaborative optimization of a foundation pit excavation scheme, characterized in that, The method comprises the following steps: obtaining foundation pit monitoring data, and extracting enclosure dynamic deformation parameters according to the foundation pit monitoring data; constructing a soil pressure calculation model, and obtaining dynamic soil pressure parameters through the soil pressure calculation model; extracting construction period optimization features, cost optimization features and environmental resource coordination features, combining the enclosure dynamic deformation parameters, the dynamic soil pressure parameters and the environmental resource coordination features to obtain safety environmental resource coordination parameters; completing multi-dimensional collaborative optimization of the foundation pit excavation scheme through the construction period optimization features, the cost optimization features and the safety environmental resource coordination parameters; The method of extracting enclosure dynamic deformation parameters according to the foundation pit monitoring data comprises the following steps: constructing an adjustment coefficient according to soil types and monitoring reliability, and correcting a state equation by using the adjustment coefficient; constructing a dynamic weight factor optimization Kalman gain by using deformation rate and cohesion to obtain a geological adaptive filtering mechanism; extracting enclosure dynamic deformation parameters by combining the corrected state equation and the geological adaptive filtering mechanism; constructing a dynamic weight factor by using deformation rate and cohesion to meet the following formula: wherein, denotes a dynamic weight factor, denotes a deformation rate, denotes a deformation rate safety threshold, determined based on specifications and engineering experience, is a geological risk degree, reflecting the gap between the current cohesion and the safety threshold; introducing a dynamic weight factor on the basis of a traditional Kalman gain formula to construct a geological adaptive filtering mechanism to meet: wherein, denotes a Kalman gain matrix, denotes a state covariance matrix, denotes an observation matrix, denotes an observation noise covariance matrix.

2. The method of claim 1, wherein, The method of constructing a soil pressure calculation model and obtaining dynamic soil pressure parameters through the soil pressure calculation model comprises the following steps: obtaining a dynamic active soil pressure coefficient according to soil aging characteristics and excavation speed; constructing a soil pressure calculation model by combining effective specific gravity, pore water pressure and the dynamic active soil pressure coefficient, and obtaining dynamic soil pressure parameters through the soil pressure calculation model.

3. The method of claim 1, wherein, The method of extracting construction period optimization features comprises the following steps: calculating key process time according to construction period influencing factors; constructing a first fitness function to extract the construction period optimization features according to the first fitness function.

4. The method of claim 1, wherein, The method of extracting cost optimization features comprises the following steps: introducing time-varying parameters, calculating dynamic earned value based on the time-varying parameters; extracting cost optimization features by using a particle swarm optimization algorithm based on the dynamic earned value.

5. The method of claim 1, wherein, The method of extracting environmental resource coordination features comprises the following steps: improving a long short-term memory network forgetting gate weight according to geological-equipment-environment-construction stage multi-dimensional parameters; extracting environmental resource coordination features by using an improved long short-term memory network based on multi-dimensional time series data.

6. The method of claim 1, wherein, The method of combining the enclosure dynamic deformation parameters, the dynamic soil pressure parameters and the environmental resource coordination features to obtain safety environmental resource coordination parameters comprises the following steps: constructing a weighted evaluation index matrix based on dynamic entropy weight by combining the enclosure dynamic deformation parameters, the dynamic soil pressure parameters and the environmental resource coordination features; evaluating scheme coordination by using a superior-inferior solution distance method, combining the scheme coordination and the weighted evaluation index matrix to obtain safety environmental resource coordination parameters.

7. The method of claim 1, wherein, The method of completing multi-dimensional collaborative optimization of the foundation pit excavation scheme through the construction period optimization features, the cost optimization features and the safety environmental resource coordination parameters comprises the following steps: obtaining a construction period cost optimal trade-off point and a resource progress matching degree through the construction period optimization features and the cost optimization features; The construction period cost optimal trade-off point, the resource progress matching degree and the safety environment resource coordination parameter are combined to complete multi-dimensional collaborative optimization of the foundation pit excavation scheme.

8. The method of claim 7, wherein, The construction period cost optimal trade-off point, the resource progress matching degree and the safety environment resource coordination parameter are combined to complete multi-dimensional collaborative optimization of the foundation pit excavation scheme. The construction period cost optimal trade-off point, the resource progress matching degree and the safety environment resource coordination parameter are combined to construct a key index state vector and define a construction decision action vector. A multi-target collaborative optimization function is designed, and based on the key index state vector, the construction decision action vector and the multi-target collaborative optimization function, a deep reinforcement learning framework is used to complete multi-dimensional collaborative optimization of the foundation pit excavation scheme.

9. A foundation pit excavation scheme multi-dimensional collaborative optimization system, characterized in that, The multi-dimensional collaborative optimization system of the foundation pit excavation scheme comprises an input device, an output device, a processor and a memory, which are connected with each other, and the memory comprises program instructions for executing the multi-dimensional collaborative optimization method of the foundation pit excavation scheme according to any one of claims 1-8.

Images