Method, device and equipment for optimizing foundation pit construction through digital twinning

Abstract

The invention provides a method, device and equipment for optimizing foundation pit construction through digital twinning. The method comprises the steps that key feature data of foundation pit construction are obtained; based on the key feature data, constructing a dynamic digital twinborn model containing a time dimension; setting a working condition evolution time axis in the dynamic digital twin model, automatically advancing the time axis according to the construction progress, and synchronously updating the physical state of the foundation pit structure in the model; constructing a digital branch of construction equipment on the foundation pit construction site in the dynamic digital twinborn model, and mapping operation parameters of the construction equipment to the digital branch in real time; different construction equipment collaborative operation scenes are simulated in the dynamic digital twinborn model, and an optimal equipment collaborative operation strategy is generated; and according to the optimal equipment collaborative operation strategy, construction parameters of construction equipment are adjusted, and optimization of foundation pit construction is achieved. According to the method, dynamic changes in the foundation pit construction process are comprehensively reflected in real time, and optimization of foundation pit construction is achieved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, and equipment for optimizing foundation pit construction using digital twins. Background Technology

[0002] In the field of modern construction engineering, foundation pit construction is a fundamental link in underground structure construction, and its construction quality and safety directly affect the success or failure of the entire construction project. The foundation pit construction process is complex, involving multiple procedures such as earthwork excavation, support structure construction, and dewatering. It is also susceptible to various factors such as geological conditions, surrounding environment, and construction technology, and there are safety risks such as foundation pit collapse, excessive soil deformation, and settlement of surrounding buildings.

[0003] Traditional methods for monitoring and optimizing foundation pit construction mainly rely on manual inspections, sensor data collection, and experience-based construction plan development. Manual inspections suffer from high subjectivity, low efficiency, and difficulty in real-time monitoring; while sensor data collection can acquire some construction parameters, its data integration and analysis capabilities are limited, making it difficult to fully reflect the dynamic changes during foundation pit construction; experience-based construction plan development lacks the ability to accurately simulate and predict complex working conditions, cannot effectively cope with unexpected situations during construction, and cannot guarantee the safety and efficiency of foundation pit construction, thus hindering the optimization of foundation pit construction. Summary of the Invention

[0004] The main objective of this invention is to provide a method, apparatus, and equipment for optimizing foundation pit construction using digital twins, aiming to overcome the shortcomings of not being able to fully reflect the dynamic changes in the foundation pit construction process and not being able to achieve foundation pit construction optimization.

[0005] To achieve the above objectives, the present invention provides a method for optimizing foundation pit construction using digital twins, comprising the following steps:

[0006] Obtain key feature data of foundation pit construction; based on the key feature data, construct a dynamic digital twin model including the time dimension;

[0007] A working condition evolution timeline is set within the dynamic digital twin model, which can automatically advance the timeline according to the construction progress and synchronously update the physical state of the foundation pit structure in the model.

[0008] In the dynamic digital twin model, digital replicas of the construction equipment at the foundation pit construction site are constructed, and the operating parameters of the construction equipment are mapped to the digital replicas in real time.

[0009] The dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, the optimal equipment collaborative operation strategy is generated. Based on the optimal equipment collaborative operation strategy, the construction parameters of the construction equipment are adjusted to optimize the foundation pit construction.

[0010] Furthermore, the acquisition of key feature data for foundation pit construction includes:

[0011] Key feature data is collected by distributed edge computing nodes set up in the foundation pit construction area; each distributed edge computing node is connected to a sensor in the corresponding area, and the data collected by the sensor is processed by the distributed edge computing nodes to extract features, thereby obtaining the key feature data.

[0012] Furthermore, the construction parameters include the operating parameters of the construction equipment and the sequence of operations.

[0013] Furthermore, the construction of a dynamic digital twin model including a time dimension based on the key feature data includes:

[0014] Cross-validation was performed on the displacement sensor data, stress sensor data, and groundwater level sensor data in the key feature data to eliminate abnormal data.

[0015] Based on cross-validated data, a reconfigurable modular digital twin model is constructed. This reconfigurable modular digital twin model consists of a foundation structure module, an environmental impact module, and a construction technology module, which can be flexibly combined and have their parameters adjusted according to actual construction scenarios. The foundation structure module performs a detailed simulation of the interaction between the foundation pit support structure and the soil. The environmental impact module integrates meteorological and geological data to predict the impact of external factors on the stability of the foundation pit. The construction technology module digitizes each process in the construction flow, and the modules interact with each other in real time through message queues.

[0016] An event-driven time dimension management system is integrated into the reconfigurable modular digital twin model. Based on the construction schedule and real-time monitoring data, the event-driven time dimension management system uses key construction events as time nodes and automatically switches to a high-precision simulation mode before the event occurs. It also uses model reduction technology to reduce the amount of computation. The actual construction progress is precisely aligned with the model's time dimension. When there is a deviation between the actual progress and the model prediction, the model parameters are automatically backtracked and corrected to achieve dynamic simulation of the foundation pit construction process, thus obtaining the dynamic digital twin model.

[0017] Furthermore, the operating parameters of the construction equipment are mapped to the digital clone in real time, including:

[0018] Lightweight edge computing terminals are deployed on each construction device to collect equipment operating parameters in real time; each edge computing terminal and the central server work together to train the equipment operating status prediction model, enabling the digital clone to predict the operating trend of the equipment.

[0019] Align the device operating parameters with the time dimension in the digital twin model at different sampling frequencies; when the digital twin detects that the device operating parameters deviate from the normal fluctuation range, it quickly locates the cause of the anomaly and generates an early warning message containing maintenance suggestions by combining historical maintenance data;

[0020] By linking the entire lifecycle data of construction equipment with digital clones, the digital clones can dynamically adjust the weights and thresholds of real-time mapping of operating parameters based on the service life and maintenance records of the construction equipment.

[0021] Furthermore, the dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, an optimal collaborative operation strategy is generated, including:

[0022] Treating all types of equipment at the foundation pit construction site as individuals in a symbiotic relationship, each piece of equipment is assigned symbiotic requirement parameters based on its functional characteristics; the impact radiation generated by the construction equipment during operation is positively correlated with the intensity of the construction equipment's operation and the degree of its impact on the stability of the foundation pit;

[0023] In simulating collaborative operation scenarios of different construction equipment, when conflicts arise in the influence radiation between construction equipment, or when the stress and deformation of the foundation pit exceed the safe range, the construction equipment automatically adjusts its own state based on symbiotic adjustment rules; among them, the construction equipment prioritizes meeting the symbiotic needs that are beneficial to the stability maintenance of the foundation pit.

[0024] The system memorizes and evolves the symbiotic states of equipment that successfully reduced foundation pit risks and improved construction efficiency in each simulation. When a new collaborative operation scenario emerges, the system directly calls the historical best symbiotic state as the initial solution. During the actual simulation, the system dynamically fine-tunes the symbiotic relationship of the construction equipment based on real-time foundation pit stress and deformation data. Finally, it selects the equipment symbiotic mode that minimizes foundation pit stress and deformation and maximizes construction efficiency, and transforms it into the optimal equipment collaborative operation strategy.

[0025] Furthermore, the dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, an optimal collaborative operation strategy is generated, including:

[0026] A mechanical balance simulation system for collaborative construction equipment is constructed, which transforms the operational behavior of construction equipment into adjustable virtual force bundles. Each virtual force bundle has three basic attributes: direction, intensity, and duration of action, which correspond one-to-one with the digging angle, operating power, and duration of the construction equipment.

[0027] An intelligent damping network is introduced to cover the entire dynamic digital twin model; the intelligent damping network dynamically adjusts the force transmission resistance of each area based on the real-time stress monitoring data of the foundation pit support structure.

[0028] In simulating collaborative operation scenarios of different construction equipment, when there is a conflict between construction equipment or the local stress of the foundation pit exceeds the limit, the force beam adaptive adjustment protocol is activated; the virtual force beams corresponding to each construction equipment autonomously change their direction and intensity according to the resistance distribution of the intelligent damping network.

[0029] The force field distribution in each simulation is scored using a force field stability evaluation algorithm. After multiple rounds of simulation iterations, the combination of virtual force bundle parameters with the best score is selected and transformed into the optimal equipment collaborative operation strategy.

[0030] Furthermore, a working condition evolution timeline is set within the dynamic digital twin model, which can automatically advance the timeline according to the construction progress and synchronously update the physical state of the foundation pit structure in the model, including:

[0031] Build an event-driven timeline engine to define key processes in foundation pit construction as key time nodes on the timeline;

[0032] During construction, real-time construction progress is collected, and the actual construction progress is dynamically matched with key time nodes on the timeline. When the actual occurrence time of one event deviates from the key time node by more than a preset threshold, the adaptive adjustment program of the timeline is automatically triggered: if the event is completed ahead of schedule, the time nodes of subsequent events are accelerated; if the event is delayed, the time arrangement of the affected events is recalculated according to the logical dependencies of the events.

[0033] To ensure the synchronous updating of the physical state of the foundation pit structure in the model, the time axis is divided into millisecond-level equipment operation time slices, minute-level process execution time slices, and hour-level construction progress time slices. Within the millisecond-level time slices, the local stress state of the foundation pit is updated in real time based on the real-time operating parameters of the equipment. Within the minute-level time slices, the stress and strain distribution of the foundation pit structure is calculated in conjunction with the process execution status. Within the hour-level time slices, the overall stability of the foundation pit is assessed and predicted by comprehensively considering the construction progress and environmental changes.

[0034] During the timeline progression, a causal traceability chain for changes in physical state is established. Each time the physical state of the foundation pit structure is updated, the key factors that trigger changes in physical state are recorded. When an abnormal state is detected, the source event and time node that caused the abnormality are quickly located along the causal traceability chain, so as to realize the synchronous update of the working condition evolution timeline and the physical state of the foundation pit structure.

[0035] Furthermore, the dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, an optimal collaborative operation strategy is generated, including:

[0036] The operation of construction equipment is abstracted into interactive virtual forces, and the magnitude, direction, and duration of the force are correlated with the equipment operating parameters in real time. In the digital twin model, the virtual forces interact to form a dynamic force field, which directly maps the stress state of the foundation pit.

[0037] In the process of simulating the collaborative operation of different construction equipment, when the local stress of the foundation pit is abnormal or the operation of construction equipment conflicts, based on the principle of force field adaptive diffusion, the abnormal area automatically generates repulsive force, which forces the surrounding virtual force to change the direction and intensity of action, and guides the construction equipment to adjust the operation parameters.

[0038] By setting a force field balance index, the uniformity of force distribution and construction efficiency of the foundation pit under different collaborative operation scenarios are quantitatively evaluated. After multiple rounds of simulation, the virtual force combination with the optimal force field balance is selected, which is then converted into the actual operating parameters of the construction equipment to generate the optimal equipment collaborative operation strategy.

[0039] The present invention also provides an apparatus for optimizing foundation pit construction using digital twins, comprising:

[0040] An acquisition unit is used to acquire key feature data of foundation pit construction; based on the key feature data, a dynamic digital twin model containing the time dimension is constructed.

[0041] The setting unit is used to set the working condition evolution time axis in the dynamic digital twin model, which can automatically advance the time axis according to the construction progress and synchronously update the physical state of the foundation pit structure in the model.

[0042] The construction unit is used to build digital clones of construction equipment at the foundation pit construction site in the dynamic digital twin model, and to map the operating parameters of the construction equipment to the digital clones in real time.

[0043] The simulation unit is used to simulate different construction equipment collaborative operation scenarios in the dynamic digital twin model. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, the optimal equipment collaborative operation strategy is generated.

[0044] The adjustment unit is used to adjust the construction parameters of the construction equipment according to the optimal equipment collaborative operation strategy, so as to optimize the foundation pit construction.

[0045] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods described above.

[0046] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods described above.

[0047] The present invention provides a method, apparatus, and equipment for optimizing foundation pit construction using digital twins, comprising: acquiring key feature data of foundation pit construction; constructing a dynamic digital twin model including a time dimension based on the key feature data; setting a working condition evolution timeline within the dynamic digital twin model, which can automatically advance the timeline according to the construction progress and synchronously update the physical state of the foundation pit structure in the model; constructing digital replicas of construction equipment at the foundation pit construction site within the dynamic digital twin model, and mapping the operating parameters of the construction equipment to the digital replicas in real time; simulating collaborative operation scenarios of different construction equipment within the dynamic digital twin model, and generating an optimal equipment collaborative operation strategy by comparing the stress and deformation of the foundation pit under different collaborative operation scenarios; and adjusting the construction parameters of the construction equipment according to the optimal equipment collaborative operation strategy to optimize foundation pit construction. In this invention, by constructing a dynamic digital twin model including a time dimension, the dynamic changes during the foundation pit construction process are reflected in real time and comprehensively; by simulating collaborative operation scenarios of different construction equipment within the dynamic digital twin model, an optimal equipment collaborative operation strategy is generated, thereby optimizing foundation pit construction. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the method steps for optimizing foundation pit construction using digital twins in one embodiment of the present invention;

[0049] Figure 2 This is a structural block diagram of a device for optimizing foundation pit construction using digital twins in one embodiment of the present invention;

[0050] Figure 3 This is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0051] The implementation, functional features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0053] Reference Figure 1 One embodiment of the present invention provides a method for optimizing foundation pit construction using digital twins, comprising the following steps:

[0054] Step S1: Obtain key feature data of foundation pit construction; based on the key feature data, construct a dynamic digital twin model including the time dimension;

[0055] Step S2: Set a working condition evolution timeline in the dynamic digital twin model. The timeline can be automatically advanced according to the construction progress, and the physical state of the foundation pit structure in the model can be updated synchronously.

[0056] Step S3: Construct a digital clone of the construction equipment at the foundation pit construction site in the dynamic digital twin model, and map the operating parameters of the construction equipment to the digital clone in real time.

[0057] Step S4: Simulate collaborative operation scenarios of different construction equipment in the dynamic digital twin model, and generate the optimal collaborative operation strategy of equipment by comparing the stress and deformation of the foundation pit under different collaborative operation scenarios.

[0058] Step S5: Adjust the construction parameters of the construction equipment according to the optimal equipment collaborative operation strategy to optimize the foundation pit construction.

[0059] In this embodiment, as described in step S1 above, firstly, key characteristic data of the foundation pit construction are obtained through multiple channels, including geological exploration, on-site sensor networks, construction drawings, and relevant historical engineering data. This data covers information that significantly impacts the foundation pit construction process and results, such as geological soil parameters, hydrological conditions, support structure design parameters, and construction schedule. Based on the obtained key characteristic data, advanced modeling techniques are used to integrate the time dimension into the model building process, creating a digital twin model that reflects the dynamic changes throughout the entire lifecycle of foundation pit construction. This model not only visually presents the three-dimensional spatial structure of the foundation pit but also simulates the changes in the construction process over time, providing a basic framework for subsequent construction monitoring and optimization.

[0060] As described in step S2 above, a timeline for the evolution of working conditions is set within the constructed dynamic digital twin model. This timeline serves as a bridge connecting the actual construction progress with the dynamic updates of the model. The timeline sets initial time nodes based on the construction schedule and automatically advances its progress by associating with data collected in real-time from the construction site. As the timeline progresses, the model synchronously updates the physical state of the foundation pit structure in the model, including parameters such as stress distribution, deformation, and stability indices, based on actual changes in working conditions at each stage of construction, such as increased excavation depth, gradual installation of support structures, and rises and falls in groundwater levels. This achieves accurate simulation and real-time reflection of the dynamic changes in the foundation pit construction process.

[0061] As described in step S3 above, a digital replica of each piece of construction equipment at the foundation pit construction site is constructed based on the dynamic digital twin model. By deploying IoT sensors on the construction equipment, real-time operating parameters are collected, such as the digging force, travel speed, and engine status of excavators, and the lifting weight, rotation angle, and boom length of cranes. Using data transmission and processing technologies, these real-time operating parameters are accurately mapped to the corresponding digital replicas of the equipment, enabling the digital replicas to realistically reproduce the actual operating status of the construction equipment within the model. This provides reliable data support for subsequent analysis of equipment operation status and optimization of equipment scheduling and management.

[0062] As described in step S4 above, the constructed dynamic digital twin model and equipment digital clones are used to simulate various collaborative operation scenarios of construction equipment. During the simulation, key data such as the stress distribution and deformation of the foundation pit under different collaborative operation scenarios are obtained through methods such as mechanical analysis and finite element calculation. These data are then compared and analyzed in depth to evaluate the impact of different collaborative operation schemes on the safety and construction efficiency of the foundation pit. Based on the analysis results, optimization algorithms are used to select the equipment collaborative operation scheme that maximizes the stress on the foundation pit, minimizes deformation, and achieves the highest construction efficiency from numerous simulation scenarios, thus generating the optimal equipment collaborative operation strategy.

[0063] As described in step S5 above, the generated optimal equipment collaborative operation strategy is transformed into specific construction equipment control instructions and parameter adjustment schemes. These adjustment instructions are transmitted to each construction device on the construction site through the construction equipment management system or IoT communication technology. Based on the received instructions, the construction equipment automatically adjusts construction parameters, such as operating speed, operating sequence, and operating force, thereby implementing the optimization strategy in the actual construction process. This achieves intelligent and precise optimization of the entire foundation pit construction process, effectively ensuring the safety and efficiency of foundation pit construction.

[0064] In one embodiment, obtaining key feature data of foundation pit construction includes:

[0065] Key feature data is collected by distributed edge computing nodes set up in the foundation pit construction area; each distributed edge computing node is connected to a sensor in the corresponding area, and the data collected by the sensor is processed by the distributed edge computing nodes to extract features, thereby obtaining the key feature data.

[0066] In this embodiment, the construction area of ​​the foundation pit is divided into multiple monitoring sub-regions according to spatial distribution characteristics and monitoring needs, with at least one distributed edge computing node deployed in each sub-region. These nodes have independent data processing capabilities and are connected to a central server via wired or wireless communication networks. Each node is equipped with a dedicated interface that can connect to various types of sensors, forming a regional data acquisition unit.

[0067] Within each sub-region, different types of sensors are deployed according to the monitoring target:

[0068] Geomechanical sensors: monitor parameters such as soil stress and strain, groundwater level changes, and pore water pressure.

[0069] Structural health sensors: monitor parameters such as internal forces, deformation, and bolt preload of the support structure.

[0070] Construction process sensors: monitor equipment operating parameters (such as excavation depth, crane load), construction progress, etc.

[0071] Environmental sensors: monitor the impact of environmental factors such as temperature, precipitation, and wind speed on the foundation pit.

[0072] After the raw data collected by the sensor (such as analog signals, time-domain waveforms, etc.) is transmitted to the edge computing node, the node processes it according to a preset algorithm as follows:

[0073] Data cleaning: removing noise, outliers, and redundant data;

[0074] Feature extraction: Key features can be extracted from multiple dimensions such as time domain, frequency domain, and time-frequency domain. For example, features such as the rate of change and peak value can be extracted from the groundwater level fluctuation curve; frequency features can be extracted from the vibration signal of the support structure for structural damage identification; power fluctuation features can be extracted from equipment operation data for construction efficiency evaluation.

[0075] Data compression: Wavelet transform, principal component analysis (PCA), and other methods can be used to reduce the dimensionality of feature data and compress it.

[0076] Edge decision-making: Based on preset thresholds, feature data is judged in real time to generate early warning information.

[0077] The processed key feature data can be transmitted to the central server via low-power wide-area networks (such as LoRa and NB-IoT), while the edge nodes retain local copies of the data. The central server performs fusion analysis on the global data and continuously optimizes the feature extraction model of the edge nodes through machine learning algorithms, forming an intelligent data acquisition system that combines edge computing with cloud collaboration.

[0078] The extracted key feature data serves as input to the dynamic digital twin model, driving changes in the time dimension of the model:

[0079] Geological parameters are used to update the geotechnical constitutive relationship of the model; construction progress features are used to advance the time axis of the working condition evolution; equipment operation features are used to drive the real-time mapping of the digital clones of construction equipment; environmental features are used to simulate the impact of external factors on the stability of the foundation pit.

[0080] Through the aforementioned distributed edge computing architecture, real-time acquisition, edge intelligent processing, and precise feature extraction of foundation pit construction data were achieved, providing a high-quality data foundation for the subsequent construction and application of dynamic digital twin models. This effectively solved the problems of high transmission latency, large consumption of computing resources, and weak real-time response capability that exist in traditional centralized data processing modes.

[0081] In one embodiment, the construction parameters include the operating parameters of the construction equipment and the work sequence.

[0082] In one embodiment, constructing a dynamic digital twin model including a time dimension based on the key feature data includes:

[0083] Cross-validation was performed on the displacement sensor data, stress sensor data, and groundwater level sensor data in the key feature data to eliminate abnormal data.

[0084] Based on cross-validated data, a reconfigurable modular digital twin model is constructed. This reconfigurable modular digital twin model consists of a foundation structure module, an environmental impact module, and a construction technology module, which can be flexibly combined and have their parameters adjusted according to actual construction scenarios. The foundation structure module performs a detailed simulation of the interaction between the foundation pit support structure and the soil. The environmental impact module integrates meteorological and geological data to predict the impact of external factors on the stability of the foundation pit. The construction technology module digitizes each process in the construction flow, and the modules interact with each other in real time through message queues.

[0085] An event-driven time dimension management system is integrated into the reconfigurable modular digital twin model. Based on the construction schedule and real-time monitoring data, the event-driven time dimension management system uses key construction events as time nodes and automatically switches to a high-precision simulation mode before the event occurs. It also uses model reduction technology to reduce the amount of computation. The actual construction progress is precisely aligned with the model's time dimension. When there is a deviation between the actual progress and the model prediction, the model parameters are automatically backtracked and corrected to achieve dynamic simulation of the foundation pit construction process, thus obtaining the dynamic digital twin model.

[0086] In this embodiment, the displacement sensor data, stress sensor data, and groundwater level sensor data from the key feature data are first cross-validated. Since data collected by different types of sensors reflect the state of the foundation pit from different perspectives and have inherent physical correlations, mathematical constraint equations based on physical laws such as mechanical equilibrium and seepage theory are established, and the sensor data are substituted into these equations for calculation. Simultaneously, a Kalman filter algorithm is used to fuse the multi-source data, estimate the true state of the data in real time, and calculate the residual between the actual measured value and the estimated value. Based on this, the Grubbs criterion is used to set a threshold, and statistical analysis is performed on the residuals. Data exceeding the threshold range is identified as outlier and removed, thereby ensuring the reliability of the data used for model construction and laying the foundation for the accuracy of subsequent models.

[0087] Based on cross-validated data, a reconfigurable modular digital twin model is constructed. This model consists of three core modules: basic structure module, environmental impact module, and construction technology module. Each module can be flexibly combined and its parameters adjusted according to the actual construction scenario.

[0088] The basic structure module employs finite element analysis to perform a detailed simulation of the interaction between the foundation pit support structure and the soil. By establishing a suitable geotechnical constitutive model (such as the Mohr-Coulomb model), it accurately describes the mechanical properties of the soil under different stress states. Simultaneously, it considers the material properties and boundary conditions of the support structure, achieving precise simulation of the stress and deformation of the foundation pit structure. The environmental impact module integrates meteorological data (such as temperature, humidity, precipitation, and wind speed) and geological data (such as soil layer distribution and groundwater characteristics). By constructing a multiphysics coupling model, it predicts the impact of external environmental factors on the stability of the foundation pit, such as analyzing the weakening effect of groundwater level rise caused by precipitation on the shear strength of the soil. The construction process module digitally models each procedure in the construction process, breaking it down into standardized atomic operations. It clarifies the sequence, duration, and required construction equipment and resources for each procedure, while establishing a mapping relationship between construction equipment operating parameters and soil disturbance, achieving dynamic simulation of the construction process. All modules achieve real-time asynchronous data interaction through message queues, using the Protobuf format to define the data exchange protocol, ensuring efficient and accurate data transmission.

[0089] An event-driven time-dimensional management system is integrated into a reconfigurable modular digital twin model. This system first analyzes the construction schedule, transforming it into a series of ordered key construction events (such as excavation to a specific depth, completion of support structure installation), and sets trigger conditions, duration, and state transition functions for each event. Based on the construction schedule and real-time monitoring data, the system manages the timeline progression using key construction events as time nodes. Before a key event occurs, the system automatically switches to a high-precision simulation mode, reducing computational load through model reduction techniques (such as intrinsic orthogonal decomposition), thus improving computational efficiency while maintaining simulation accuracy. By establishing a quantitative index system for the deviation between the actual construction progress and the model's time dimension, the system compares the actual construction progress with the model prediction in real time. When the deviation exceeds a preset threshold, the system activates a Bayesian parameter update algorithm, automatically backtracking and correcting the model parameters based on real-time monitoring data. This ensures the model accurately reflects the dynamic changes in the foundation pit construction process, ultimately achieving dynamic simulation of the foundation pit construction process and obtaining a complete dynamic digital twin model.

[0090] In one embodiment, mapping the operating parameters of the construction equipment to a digital clone in real time includes:

[0091] Lightweight edge computing terminals are deployed on each construction device to collect equipment operating parameters in real time; each edge computing terminal and the central server work together to train the equipment operating status prediction model, enabling the digital clone to predict the operating trend of the equipment.

[0092] Align the device operating parameters with the time dimension in the digital twin model at different sampling frequencies; when the digital twin detects that the device operating parameters deviate from the normal fluctuation range, it quickly locates the cause of the anomaly and generates an early warning message containing maintenance suggestions by combining historical maintenance data;

[0093] By linking the entire lifecycle data of construction equipment with digital clones, the digital clones can dynamically adjust the weights and thresholds of real-time mapping of operating parameters based on the service life and maintenance records of the construction equipment.

[0094] In this embodiment, firstly, a lightweight edge computing terminal with low power consumption and high integration is deployed on each construction device. This edge computing terminal is equipped with multiple adapter interfaces, allowing direct connection to the device's built-in sensors (such as engine speed sensors, hydraulic pressure sensors, and displacement sensors) to achieve real-time acquisition of equipment operating parameters (including but not limited to engine torque, boom tilt angle, travel speed, oil temperature, and oil pressure). The edge computing terminal employs multi-threaded data acquisition technology, enabling simultaneous processing of data streams from multiple sensors. It also utilizes a data caching mechanism to locally store acquired data during temporary network interruptions, automatically retransmitting it upon network recovery. This ensures the continuity and integrity of equipment operating parameter acquisition, providing a reliable data source for subsequent data processing and model training.

[0095] Equipment operating parameter data collected by each edge computing terminal is transmitted to the central server via 5G or a dedicated wireless network. The central server and edge computing terminals build a collaborative training system based on a federated learning framework, enabling distributed training of the equipment operating status prediction model without transmitting raw data. Edge computing terminals use locally collected data to perform initial model training and upload the trained model parameters to the central server. The central server collects parameter updates from all edge terminals, performs global model aggregation, and then distributes the updated model back to each edge terminal. Through multiple rounds of iterative training, the model learns the historical patterns and potential characteristics of equipment operating parameters. The trained equipment operating status prediction model is deployed in a digital avatar. The digital avatar analyzes and processes the real-time collected equipment operating parameters based on this model, combining time series prediction algorithms (such as Long Short-Term Memory networks, LSTM) to predict future operating trends of the equipment, such as predicting engine failure probabilities and hydraulic system performance degradation trends, providing forward-looking decision-making basis for equipment maintenance and construction scheduling.

[0096] Due to differences in sampling frequencies among different types of sensors (e.g., high-frequency vibration sensors can sample at frequencies up to kilohertz, while some environmental sensors only sample at the minute level), a multi-scale spatiotemporal alignment algorithm is employed to achieve accurate matching between equipment operating parameters and the time dimension of the digital twin model. This algorithm first divides the time dimension of the digital twin model into millisecond-level micro-operation time slices, second-level equipment action time slices, minute-level process execution time slices, and hour-level construction progress time slices. For data at different time scales, corresponding data processing techniques are employed: for millisecond-level high-frequency data, Kalman filtering is used for noise reduction and data smoothing; for minute-level and higher low-frequency data, interpolation algorithms (such as cubic spline interpolation) are used to expand or compress the time dimension. By calculating the time offset between each sampling moment and the time node of the digital twin model, the equipment operating parameters are resampled and timestamped to ensure that equipment operating parameters with different sampling frequencies can be accurately mapped to the corresponding time dimension of the digital twin model, thereby realistically reproducing the dynamic time-series process of equipment operation in the model.

[0097] The digital avatar continuously monitors equipment operating parameters and makes real-time judgments based on preset normal fluctuation range thresholds (determined through equipment design parameters and statistical analysis of historical operating data). When equipment operating parameters deviate from the normal fluctuation range, the digital avatar activates a two-layer anomaly localization mechanism: the first layer uses rule-based reasoning based on knowledge graphs to match similar fault cases from the equipment operation knowledge network (including equipment operation manuals, component structural relationships, common fault modes, etc.) to initially locate the possible location and cause of the anomaly; the second layer uses a pre-trained deep neural network (such as a convolutional neural network CNN combined with an attention mechanism) to extract and analyze the features of the abnormal parameters, further accurately identifying the fault type. Simultaneously, the digital avatar connects to a historical maintenance database and, based on the located anomaly cause, retrieves maintenance records for similar faults, including information such as the equipment status at the time of the fault, maintenance measures, and maintenance effects. This comprehensive analysis generates a warning message containing the cause of the anomaly, its impact, and maintenance recommendations, which is promptly sent to equipment managers and construction technicians via SMS, APP push notifications, etc., so that they can quickly take countermeasures and reduce the impact of equipment failure on the construction progress.

[0098] The entire lifecycle data of construction equipment (including factory parameters, installation and commissioning records, service life, cumulative working hours, maintenance records, and parts replacement records) is deeply correlated with a digital avatar. The digital avatar establishes a mapping relationship between the equipment's lifecycle data and operating parameters to analyze the performance changes of the equipment at different stages of use. Based on this, the digital avatar employs a reinforcement learning-driven dynamic weight adjustment mechanism. According to factors such as the equipment's service life, maintenance records, and real-time operating conditions, the weight and threshold of the real-time mapping of operating parameters are automatically adjusted through reinforcement learning algorithms. For example, for equipment with a longer service life and more maintenance frequency, the monitoring weight of key operating parameters (such as engine temperature and bearing vibration values) is increased, and the threshold is appropriately reduced to enhance sensitivity to potential faults; while for new equipment, a more lenient monitoring strategy is adopted. By dynamically adjusting weights and thresholds, the digital avatar can better match the actual state of the equipment, achieving adaptive and accurate mapping and intelligent monitoring of the construction equipment's operating parameters.

[0099] In one embodiment, the dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, an optimal collaborative operation strategy is generated, including:

[0100] Treating all types of equipment at the foundation pit construction site as individuals in a symbiotic relationship, each piece of equipment is assigned symbiotic requirement parameters based on its functional characteristics; the impact radiation generated by the construction equipment during operation is positively correlated with the intensity of the construction equipment operation and the degree of impact on the stability of the foundation pit.

[0101] In simulating collaborative operation scenarios of different construction equipment, when conflicts arise in the influence radiation between construction equipment, or when the stress and deformation of the foundation pit exceed the safe range, the construction equipment automatically adjusts its own state based on symbiotic adjustment rules; among them, the construction equipment prioritizes meeting the symbiotic needs that are beneficial to the stability maintenance of the foundation pit.

[0102] The system memorizes and evolves the symbiotic states of equipment that successfully reduced foundation pit risks and improved construction efficiency in each simulation. When a new collaborative operation scenario emerges, the system directly calls the historical best symbiotic state as the initial solution. During the actual simulation, the system dynamically fine-tunes the symbiotic relationship of the construction equipment based on real-time foundation pit stress and deformation data. Finally, it selects the equipment symbiotic mode that minimizes foundation pit stress and deformation and maximizes construction efficiency, and transforms it into the optimal equipment collaborative operation strategy.

[0103] In this embodiment, various equipment at the foundation pit construction site, such as excavators, cranes, and anchor drilling rigs, are first abstracted as individual units in a symbiotic relationship, and their unique symbiotic requirement parameters are assigned based on their functional characteristics. For example, the symbiotic requirement parameters for excavators include operating radius, digging depth, and required operating space dimensions; for cranes, they include lifting height, turning radius, and lifting capacity limits. These parameters constitute the basic behavioral constraints of the equipment in the symbiotic system. Simultaneously, to quantify the impact of equipment operations on the foundation pit and its surrounding environment, the concept of "impact radiation" is defined. Its intensity is positively correlated with the operating intensity of the construction equipment (such as the digging force of excavators and the lifting frequency of cranes) and the degree of impact on the stability of the foundation pit (such as the range of soil disturbance caused by equipment operation and the magnitude of additional loads). Impact radiation spreads outward from the equipment as the center, forming a dynamically changing impact field in the digital twin model, used to characterize the comprehensive effect of equipment operation behavior on the foundation pit and other equipment.

[0104] During the simulated collaborative operation of construction equipment, the interaction of influence radiation between equipment and the stress and deformation data of the foundation pit are continuously monitored. When the influence radiation between equipment overlaps or conflicts (such as overlapping work spaces, resulting in superimposed loads causing local stress exceeding limits in the foundation pit), or when the stress and deformation of the foundation pit exceed a pre-set safety threshold, the symbiotic adjustment mechanism of the equipment is triggered. Each piece of equipment automatically adjusts its own operating status according to the symbiotic adjustment rules, prioritizing the maintenance of foundation pit stability. For example, when the influence radiation generated by the excavator operating near the foundation pit support structure may threaten the support safety, the equipment prioritizes reducing the digging force and adjusting the working angle to reduce the adverse impact on the foundation pit structure. When multiple pieces of equipment are operating simultaneously, through a dynamic negotiation mechanism, based on their respective symbiotic demand parameters and influence radiation intensity, the operating sequence and time interval are automatically coordinated to avoid mutual interference. This adjustment process simulates the mutual adaptation and co-evolution mechanism in biological symbiosis, enabling the equipment group to achieve autonomous collaboration under complex working conditions.

[0105] A symbiotic state memory database is established to comprehensively record the symbiotic combinations of equipment that successfully reduced foundation pit risks and improved construction efficiency in each simulation, including key information such as equipment operating parameters, collaborative operation sequence, and impact radiation distribution. When a new collaborative operation scenario emerges, a similarity matching algorithm is first used to retrieve the historically optimal symbiotic state closest to the current working condition from the memory database, using it as the initial simulation scheme, significantly shortening the strategy generation time. During actual simulation, the system collects the stress and deformation data of the foundation pit in real time, and combines it with the current equipment operating status, using a dynamic optimization algorithm to fine-tune the symbiotic relationship of the equipment. For example, based on the deformation rate of a certain area of ​​the foundation pit, the operating intensity and sequence of equipment near that area are dynamically adjusted; through reinforcement learning algorithms, the effects of different fine-tuning strategies are evaluated, strengthening the adjustment behaviors that effectively improve foundation pit stability and construction efficiency, and weakening operations with ineffective or negative impacts. After multiple rounds of iterative simulation, the equipment symbiotic mode that minimizes foundation pit stress and deformation and maximizes construction efficiency is finally selected. The equipment operating parameters, collaborative logic, and other information corresponding to this mode are transformed into an executable optimal equipment collaborative operation strategy to guide equipment scheduling and operation arrangement on the actual construction site.

[0106] In one embodiment, the dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, an optimal collaborative operation strategy is generated, including:

[0107] A mechanical balance simulation system for collaborative construction equipment is constructed, which transforms the operational behavior of construction equipment into adjustable virtual force bundles. Each virtual force bundle has three basic attributes: direction, intensity, and duration of action, which correspond one-to-one with the digging angle, operating power, and duration of the construction equipment.

[0108] An intelligent damping network is introduced to cover the entire dynamic digital twin model; the intelligent damping network dynamically adjusts the force transmission resistance of each area based on the real-time stress monitoring data of the foundation pit support structure.

[0109] In simulating collaborative operation scenarios of different construction equipment, when there is a conflict between construction equipment or the local stress of the foundation pit exceeds the limit, the force beam adaptive adjustment protocol is activated; the virtual force beams corresponding to each construction equipment autonomously change their direction and intensity according to the resistance distribution of the intelligent damping network.

[0110] The force field distribution in each simulation is scored using a force field stability evaluation algorithm. After multiple rounds of simulation iterations, the combination of virtual force bundle parameters with the best score is selected and transformed into the optimal equipment collaborative operation strategy.

[0111] In this embodiment, a mechanical equilibrium simulation system for the coordinated operation of construction equipment is first established. The operational behaviors of various construction equipment such as excavators, cranes, and anchor drilling rigs at the foundation pit construction site are transformed into adjustable virtual force bundles with physical characteristics. Each virtual force bundle is assigned three basic attributes: direction, intensity, and duration. These attributes are precisely mapped to the actual operating parameters of the construction equipment: the direction of the virtual force bundle corresponds to the equipment's digging angle, lifting direction, or drilling orientation; the intensity is directly related to the equipment's operating power, digging force, and lifting weight; and the duration is consistent with the duration of a single operation by the equipment. In this way, the complex operational behaviors of construction equipment are transformed into a quantifiable and analyzable mechanical model, providing a basic framework for subsequent simulation and optimization. This allows the dynamic digital twin model to intuitively reflect the operational status of the construction equipment and its impact on the foundation pit from a mechanical perspective.

[0112] A smart damping network is introduced into the dynamic digital twin model, covering the entire simulated foundation pit area, acting as a smart "protective net" for the pit structure. The smart damping network dynamically adjusts based on real-time stress monitoring data of the foundation pit support structure. Specifically, high-precision stress sensors are deployed at key parts of the foundation pit support structure (such as support beams and retaining piles) to collect structural stress data in real time and transmit it to the smart damping network system. According to the stress distribution, the force transmission resistance coefficient of each area is dynamically adjusted: in areas with low stress and structural stability, the force transmission resistance is reduced, allowing virtual force beams to pass smoothly; while in areas where stress is close to or exceeds the design threshold, the force transmission resistance is significantly increased, forming a "resistance barrier," forcing the virtual force beams to change their transmission direction and guiding the operation of construction equipment in a direction more favorable to the stability of the foundation pit, thereby avoiding the threat to foundation pit safety posed by localized stress concentration.

[0113] During the simulation of collaborative operations of different construction equipment, the system continuously monitors operational conflicts between equipment and the stress state of various parts of the foundation pit. Once an operational conflict is detected between construction equipment (such as overlapping workspaces or intersecting equipment paths), or if the local stress in the foundation pit exceeds safety limits, the force beam adaptive adjustment protocol is immediately activated. At this time, the virtual force beams corresponding to each construction equipment will autonomously adjust their direction and intensity based on the real-time resistance distribution of the intelligent damping network. For example, when the working areas of two excavators overlap, causing force field disturbances, their corresponding virtual force beams will automatically change direction due to the high resistance set by the intelligent damping network in the overlapping area, guiding the equipment to adjust its working position. If the stress in a corner of the foundation pit exceeds the limit, the virtual force beams of nearby equipment will reduce their intensity or change their direction, directing the force to a more stable area. This adjustment process requires no manual intervention; the virtual force beams possess autonomous "perception-decision-action" capabilities, enabling the construction equipment group to automatically achieve mechanical equilibrium under complex working conditions.

[0114] In this embodiment, a built-in force field stability assessment algorithm is implemented, which quantifies and scores the force field distribution generated in each simulation from multiple dimensions. The scoring indicators include core elements such as the uniformity of stress on the foundation pit (measured by calculating the standard deviation of stress at each location), deformation (displacement data obtained from finite element analysis), and construction efficiency (calculated based on equipment operating time and completed work volume). During multiple simulation iterations, the system continuously records the force field distribution and its scoring results for each simulation. Through comparative analysis, the system gradually selects the optimal combination of virtual force bundle parameters. The corresponding construction equipment operating parameters (such as excavation angle, operating power, and operating sequence) and collaborative logic of this combination constitute the optimal equipment collaborative operation strategy. Finally, this strategy is transformed into executable instructions and transmitted to each construction device on the construction site, achieving precise optimization of the construction process, ensuring the foundation pit maintains mechanical stability throughout construction, and improving overall construction efficiency.

[0115] In one embodiment, a working condition evolution timeline is set within the dynamic digital twin model, which can automatically advance the timeline according to the construction progress and synchronously update the physical state of the foundation pit structure in the model, including:

[0116] Build an event-driven timeline engine to define key processes in foundation pit construction as key time nodes on the timeline;

[0117] During construction, real-time construction progress is collected, and the actual construction progress is dynamically matched with key time nodes on the timeline. When the actual occurrence time of one event deviates from the key time node by more than a preset threshold, the adaptive adjustment program of the timeline is automatically triggered: if the event is completed ahead of schedule, the time nodes of subsequent events are accelerated; if the event is delayed, the time arrangement of the affected events is recalculated according to the logical dependencies of the events.

[0118] To ensure the synchronous updating of the physical state of the foundation pit structure in the model, the time axis is divided into millisecond-level equipment operation time slices, minute-level process execution time slices, and hour-level construction progress time slices. Within the millisecond-level time slices, the local stress state of the foundation pit is updated in real time based on the real-time operating parameters of the equipment. Within the minute-level time slices, the stress and strain distribution of the foundation pit structure is calculated in conjunction with the process execution status. Within the hour-level time slices, the overall stability of the foundation pit is assessed and predicted by comprehensively considering the construction progress and environmental changes.

[0119] During the timeline progression, a causal traceability chain for changes in physical state is established. Each time the physical state of the foundation pit structure is updated, the key factors that trigger changes in physical state are recorded. When an abnormal state is detected, the source event and time node that caused the abnormality are quickly located along the causal traceability chain, so as to realize the synchronous update of the working condition evolution timeline and the physical state of the foundation pit structure.

[0120] In this embodiment, an event-driven timeline engine is first constructed as the core component connecting the construction plan with the dynamic evolution of the digital twin model. Based on the foundation pit construction organization design and engineering drawings, this engine defines key processes such as earthwork excavation to a specific depth, support structure installation, and foundation slab pouring as critical time nodes on the timeline. Each critical time node is associated with detailed construction parameters and logical rules: clearly defining the preconditions for the process (e.g., support installation must be carried out after the corresponding area's earthwork excavation is completed), duration estimation, and dependencies with other processes. In this way, the linear construction schedule is transformed into a structured sequence of events, laying the foundation for subsequent dynamic monitoring and adaptive adjustments, ensuring that the timeline accurately reflects the key milestones of the construction process.

[0121] During construction, real-time construction progress information (such as the amount of earthwork completed and the number of supporting structures installed) is obtained through multiple data acquisition channels, including IoT sensors and BIM progress management systems deployed on the construction site. The actual construction progress is dynamically matched with key time nodes on the timeline, and a time deviation calculation algorithm is used to quantify the difference between the actual occurrence time of an event and the preset time node. When the deviation between the actual occurrence time of a key event and the preset time node exceeds a preset threshold (e.g., more than 10% earlier or later than the planned duration), an adaptive adjustment procedure for the timeline is automatically triggered: if the event is completed ahead of schedule, the time nodes of subsequent events are accelerated synchronously based on the logical dependencies between work processes, optimizing the scheduling of construction resources; if the event is delayed, the timing of the affected event is recalculated, and the impact on the overall construction progress is minimized by adjusting the start and end times of subsequent work processes and the priority of resource allocation, ensuring that the timeline always aligns with the actual construction rhythm on site.

[0122] To synchronously update the physical state of the foundation pit structure in the model, the time axis is divided into three time slices with different granularities: milliseconds, minutes, and hours, enabling layered and refined simulation of the construction process. Within the millisecond time slice, based on real-time operating parameters of construction equipment (such as excavator digging force and crane lifting speed), the local stress state of the foundation pit is updated in real time using rigid body dynamics equations and local mechanical models, capturing the dynamic load changes on the soil and support structure instantaneously caused by equipment operation. In the minute-level time slice, combined with the execution status of work processes (such as earthwork excavation progress and application of prestressing), the stress-strain distribution of the foundation pit structure is calculated using finite element analysis, evaluating the mechanical response of the structure during phased construction. In the hour-level time slice, considering both construction progress (such as cumulative excavation depth and completed work processes) and environmental changes (such as groundwater level fluctuations and temperature changes), a macroscopic mechanical model and stability assessment algorithm are used to assess and predict the overall stability of the foundation pit, identifying potential risks and providing early warnings. This multi-scale time slice update mechanism ensures both accurate simulation of equipment operation details and macroscopic control of the overall construction process.

[0123] As the timeline progresses, a causal traceability chain for changes in physical state is established to record the complete logical context of the evolution of the foundation pit structure's physical state. Each time the foundation pit structure's physical state is updated, key factors triggering the change are automatically captured, including construction equipment operating parameters, process execution status, environmental variables, and time nodes, and a causal relationship graph is constructed in chronological order. When an abnormal state is detected in the foundation pit (such as excessive stress in the support structure or excessive soil deformation), reverse reasoning is performed along the causal traceability chain, and correlation analysis is used to quickly locate the source event and time node causing the anomaly. For example, if abnormal stress is found in a certain section of the support beam, it can be traced back to initial causes such as excessively deep excavation in that area or concentrated equipment loads, providing construction managers with accurate diagnostic basis to help them develop targeted solutions. It also accumulates data experience for subsequent construction simulation and strategy optimization, achieving deep integration and collaborative updating of the working condition evolution timeline and the foundation pit structure's physical state.

[0124] In one embodiment, the dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, an optimal collaborative operation strategy is generated, including:

[0125] The operation of construction equipment is abstracted into interactive virtual forces, and the magnitude, direction, and duration of the force are correlated with the equipment operating parameters in real time. In the digital twin model, the virtual forces interact to form a dynamic force field, which directly maps the stress state of the foundation pit.

[0126] In the process of simulating the collaborative operation of different construction equipment, when the local stress of the foundation pit is abnormal or the operation of construction equipment conflicts, based on the principle of force field adaptive diffusion, the abnormal area automatically generates repulsive force, which forces the surrounding virtual force to change the direction and intensity of action, and guides the construction equipment to adjust the operation parameters.

[0127] By setting a force field balance index, the uniformity of force distribution and construction efficiency of the foundation pit under different collaborative operation scenarios are quantitatively evaluated. After multiple rounds of simulation, the virtual force combination with the optimal force field balance is selected, which is then converted into the actual operating parameters of the construction equipment to generate the optimal equipment collaborative operation strategy.

[0128] In this embodiment, the operational behavior of construction equipment is first abstracted into interactive virtual forces, establishing a real-time mapping relationship between virtual force attributes and equipment operating parameters. Each piece of construction equipment (such as excavators, cranes, and anchor drilling rigs) generates corresponding virtual forces during operation. The magnitude of these forces is directly related to the equipment's operating power and load intensity; for example, the digging force of an excavator is determined by the engine output power and bucket digging resistance. The direction depends on the equipment's operating angle or trajectory; for example, when a crane is hoisting, the virtual force direction is vertically upward along the sling. The duration of action is consistent with the duration of a single operation. In the dynamic digital twin model, the virtual forces generated by each piece of construction equipment superimpose and interact, forming a force field that dynamically changes over time. This force field is closely related to the actual stress state of the foundation pit. The distribution and intensity changes of forces within the force field can intuitively reflect the stress conditions of various parts of the foundation pit, providing a visualized and quantifiable mechanical model foundation for subsequent simulation and analysis.

[0129] During the simulation of collaborative operations of different construction equipment, the stress state of various parts of the foundation pit and the operational conflicts between equipment are continuously monitored. When local stress anomalies are detected in the foundation pit (such as stress exceeding the design threshold or stress concentration), or operational conflicts exist between construction equipment (such as overlapping workspaces or interference in equipment operating paths), an adjustment mechanism based on the principle of adaptive diffusion of force field is triggered. At this time, the abnormal area automatically generates a repulsive force in the force field. The magnitude of this repulsive force is proportional to the degree of anomaly, and its direction points towards the surrounding area. The surrounding virtual forces are acted upon by the repulsive force, and according to the principle of force composition and decomposition, their direction and intensity are automatically changed. For example, if the stress in a corner of the foundation pit exceeds the limit, the virtual force corresponding to the nearby excavator will reduce its intensity and change its digging direction under the action of the repulsive force, avoiding further load on that area; when multiple equipment have conflicting workspaces, their virtual forces repel each other, prompting the equipment to adjust its work position or sequence. Through this adaptive diffusion adjustment, the construction equipment is guided to adjust its operating parameters in real time, so that the force field returns to a state of equilibrium, ensuring the safety of the foundation pit structure and the smooth progress of construction.

[0130] To quantitatively evaluate the advantages and disadvantages of different collaborative operation scenarios of construction equipment, a force field balance index was established. This index comprehensively considers two core elements: uniformity of stress in the foundation pit and construction efficiency. Regarding the uniformity of stress in the foundation pit, the dispersion of stress distribution is assessed by calculating statistical quantities such as the standard deviation and coefficient of variation of stress in each region of the force field; the smaller the dispersion, the more uniform the stress distribution. Regarding construction efficiency, an efficiency evaluation model is constructed by combining data such as equipment operating time and completed work volume. Through multiple rounds of simulation with different virtual force combinations (i.e., different equipment collaborative operation schemes), the force field balance index is calculated for each simulation. By comparing the balance scores of each simulation, the virtual force combination with the best score is selected. The equipment operating parameters (such as digging angle, operating power, and operation sequence) corresponding to this combination are considered the optimal equipment collaborative operation strategy. Finally, this strategy is translated into executable instructions and transmitted to the construction equipment at the construction site, guiding the equipment to operate according to the optimized parameters, thereby achieving high efficiency and scientific management of equipment collaborative operation during foundation pit construction.

[0131] In one embodiment, after adjusting the construction parameters of the construction equipment to optimize the foundation pit construction, the process includes:

[0132] Obtain the identification characters of each construction equipment at the foundation pit construction site and construct an undirected identification graph; wherein, each character in the identification characters is added to the node of the undirected graph in sequence;

[0133] Obtain the model information of the distributed edge computing nodes set up in the foundation pit construction area, and construct a model information matrix;

[0134] Obtain the construction progress trend curve of the foundation pit; the construction progress trend curve is simulated by the increase in construction progress over the same time interval;

[0135] Based on the construction progress change trend curve and the model information matrix, a first coding table transformation factor is generated;

[0136] Based on the construction progress change trend curve and the undirected graph of the identifier, a second coding table transformation factor is generated;

[0137] The transformation factors of the first encoding table and the transformation factors of the second encoding table are then fused to obtain a fused transformation factor.

[0138] The preset encoding table is transformed based on the fusion transformation factor to obtain the transformed encoding table; the transformed encoding table is used to encode the construction data during the foundation pit construction process and then send it to the management terminal.

[0139] In this embodiment, firstly, the unique identifier character of each construction equipment at the foundation pit construction site is obtained through the equipment IoT module. This identifier character consists of information such as the equipment manufacturer code, model number, and serial number, and is globally unique. These identifier characters are then mapped sequentially to the nodes of an undirected graph to construct an undirected graph of construction equipment identifiers. In the graph structure, each node represents a character, and nodes corresponding to adjacent characters are connected by undirected edges. The weight of the edges can be assigned based on the semantic relationship between characters (such as the association field in the model code of the same equipment), or no edge weight can be assigned. This undirected identifier graph not only completely records the equipment identification information but also reflects the inherent logical relationship of the identifier characters through its graph structure, providing a structured representation basis for subsequent data processing.

[0140] Model information was collected from each distributed edge computing node deployed in the foundation pit construction area, including key attributes such as hardware configuration (processor model, memory capacity, storage specifications), supported communication protocols, and data processing capabilities. The model information for each node was organized into a row vector, then arranged according to node deployment location or numbering to construct a model information matrix. The rows of the matrix represent different edge computing nodes, and the columns correspond to the model attribute parameters. This matrix can intuitively present the hardware resource distribution and performance characteristics of edge computing nodes within the construction area, providing a quantitative basis for data processing resource scheduling and algorithm adaptation.

[0141] Real-time data on the construction progress of the foundation pit is collected at fixed time intervals (e.g., hourly, daily), including key indicators such as the completed earthwork excavation volume, the number of support structures installed, and the volume of concrete poured. Based on this data, time series analysis is used to fit the increasing construction progress data within the same time interval, generating a construction progress trend curve. This curve uses a smooth spline interpolation algorithm to eliminate data fluctuation noise, clearly showing the trend of construction progress over time. Simultaneously, the curve's slope reflects fluctuations in construction efficiency, providing a visual analysis tool for construction progress prediction and resource allocation.

[0142] Joint analysis of the construction progress trend curve and the model information matrix can be achieved by: firstly, extracting key feature parameters of the trend curve, such as slope, curvature, and inflection point positions, to quantify the rate of change and stage characteristics of the construction progress; simultaneously, performing principal component analysis (PCA) on the model information matrix to extract key feature vectors characterizing the performance of edge computing nodes. Based on these two types of feature data, a mapping relationship between construction progress and node performance is established using a machine learning regression model (such as random forest regression), generating a first encoding table transformation factor. This factor incorporates the dynamic changes in data processing requirements due to the construction progress, as well as adaptation parameters for the hardware performance of edge computing nodes, guiding the encoding algorithm to dynamically adjust data processing strategies according to the construction progress. The specific process for generating the first encoding table transformation factor is complex and will not be detailed here.

[0143] For the construction progress trend curve and the undirected graph of equipment identification, the time series data of the trend curve is first converted into node features and fused with the undirected graph to construct a spatiotemporal correlation graph model. A graph convolutional network (GCN) is used to extract features from the fused graph to uncover the potential relationship between construction progress changes and equipment identification, such as the correlation between the commissioning time of different equipment and construction progress nodes. Based on the extracted features, a reinforcement learning algorithm is used to optimize the encoding strategy and generate a second encoding table transformation factor. This factor integrates the synergistic relationship between equipment identification information and construction progress, enabling differentiated encoding processing of data generated by different equipment. The specific process of generating the second encoding table transformation factor involves many steps and will not be detailed here.

[0144] Deeply fusing the transformation factors of the first and second coding tables can be achieved using a combination of weighted summation and feature concatenation. Based on the importance of the two types of factors in construction data processing, weighting coefficients are determined through cross-validation experiments, highlighting the impact of the relationship between construction progress and node performance adaptation, and the relationship between equipment identification and progress coordination, on the coding strategy. The fused transformation factors integrate multi-dimensional information such as hardware resource adaptation, equipment identification association, and dynamic changes in construction progress, forming a unified basis for adjusting the coding strategy and providing core parameters for generating an efficient transformation coding table.

[0145] Based on the fusion transformation factor, the preset encoding table is dynamically transformed. First, the encoding rules of the encoding table are adjusted according to the parameters in the transformation factor, such as changing the data compression algorithm type, adjusting the encryption strength level, and optimizing the data fragmentation strategy. Then, through a parameter mapping mechanism, the numerical information of the transformation factor is converted into specific encoding table configuration items. The generated transformed encoding table is used to encode various types of data generated during the foundation pit construction process (such as equipment operating parameters, monitoring sensor data, and construction progress records), compressing data transmission volume, improving data security, and sending it through a communication protocol agreed upon between edge computing nodes and the management terminal, achieving efficient and accurate transmission and management of construction data.

[0146] Reference Figure 2 In another embodiment of the present invention, an apparatus for optimizing foundation pit construction using digital twins is also provided, comprising:

[0147] An acquisition unit is used to acquire key feature data of foundation pit construction; based on the key feature data, a dynamic digital twin model containing the time dimension is constructed.

[0148] The setting unit is used to set the working condition evolution time axis in the dynamic digital twin model, which can automatically advance the time axis according to the construction progress and synchronously update the physical state of the foundation pit structure in the model.

[0149] The construction unit is used to build digital clones of construction equipment at the foundation pit construction site in the dynamic digital twin model, and to map the operating parameters of the construction equipment to the digital clones in real time.

[0150] The simulation unit is used to simulate different construction equipment collaborative operation scenarios in the dynamic digital twin model. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, the optimal equipment collaborative operation strategy is generated.

[0151] The adjustment unit is used to adjust the construction parameters of the construction equipment according to the optimal equipment collaborative operation strategy, so as to optimize the foundation pit construction.

[0152] In this embodiment, the specific implementation of each unit in the above device embodiment is described in the above method embodiment, and will not be repeated here.

[0153] Reference Figure 3 This invention also provides a computer device, which may be a server, and its internal structure may be as follows: Figure 3As shown, the computer device includes a processor, memory, display screen, input device, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores the data corresponding to this embodiment. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the above-described method.

[0154] Those skilled in the art will understand that Figure 3 The structures shown are merely block diagrams of some structures related to the present invention and do not constitute a limitation on the computer devices on which the present invention is applied.

[0155] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. It is understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0156] In summary, the method, apparatus, and equipment for optimizing foundation pit construction using digital twins provided in this embodiment of the invention include: acquiring key feature data of foundation pit construction; constructing a dynamic digital twin model containing a time dimension based on the key feature data; setting a working condition evolution timeline within the dynamic digital twin model, which can automatically advance the timeline according to the construction progress and synchronously update the physical state of the foundation pit structure in the model; constructing digital replicas of the construction equipment at the foundation pit construction site within the dynamic digital twin model, and mapping the operating parameters of the construction equipment to the digital replicas in real time; simulating collaborative operation scenarios of different construction equipment within the dynamic digital twin model, and generating an optimal equipment collaborative operation strategy by comparing the stress and deformation of the foundation pit under different collaborative operation scenarios; and adjusting the construction parameters of the construction equipment according to the optimal equipment collaborative operation strategy to optimize foundation pit construction. In this invention, by constructing a dynamic digital twin model containing a time dimension, the dynamic changes during the foundation pit construction process are reflected in real time and comprehensively; by simulating collaborative operation scenarios of different construction equipment within the dynamic digital twin model, an optimal equipment collaborative operation strategy is generated, thereby optimizing foundation pit construction.

[0157] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the present invention and embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can 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 memory bus dynamic RAM, etc.

[0158] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0159] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for optimizing foundation pit construction using digital twins, characterized in that, Includes the following steps: Obtain key feature data of foundation pit construction; based on the key feature data, construct a dynamic digital twin model including the time dimension; A working condition evolution timeline is set within the dynamic digital twin model, which can automatically advance the timeline according to the construction progress and synchronously update the physical state of the foundation pit structure in the model. In the dynamic digital twin model, digital replicas of the construction equipment at the foundation pit construction site are constructed, and the operating parameters of the construction equipment are mapped to the digital replicas in real time. The dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, the optimal equipment collaborative operation strategy is generated. Based on the optimal equipment collaborative operation strategy, the construction parameters of the construction equipment are adjusted to optimize the foundation pit construction.

2. The method for optimizing foundation pit construction using digital twins according to claim 1, characterized in that, The key feature data obtained from the foundation pit construction includes: Key feature data is collected by distributed edge computing nodes set up in the foundation pit construction area; each distributed edge computing node is connected to a sensor in the corresponding area, and the data collected by the sensor is processed by the distributed edge computing nodes to extract features, thereby obtaining the key feature data.

3. The method for optimizing foundation pit construction using digital twins according to claim 1, characterized in that, The construction parameters include the operating parameters of the construction equipment and the sequence of operations.

4. The method for optimizing foundation pit construction using digital twins according to claim 1, characterized in that, The construction of a dynamic digital twin model including a time dimension based on the key feature data includes: Cross-validation was performed on the displacement sensor data, stress sensor data, and groundwater level sensor data in the key feature data to eliminate abnormal data. Based on cross-validated data, a reconfigurable modular digital twin model is constructed. This reconfigurable modular digital twin model consists of a foundation structure module, an environmental impact module, and a construction technology module, which can be flexibly combined and have their parameters adjusted according to actual construction scenarios. The foundation structure module performs a detailed simulation of the interaction between the foundation pit support structure and the soil. The environmental impact module integrates meteorological and geological data to predict the impact of external factors on the stability of the foundation pit. The construction technology module digitizes each process in the construction flow, and the modules interact with each other in real time through message queues. An event-driven time dimension management system is integrated into the reconfigurable modular digital twin model. Based on the construction schedule and real-time monitoring data, the event-driven time dimension management system uses key construction events as time nodes and automatically switches to a high-precision simulation mode before the event occurs. It also uses model reduction technology to reduce the amount of computation. The actual construction progress is precisely aligned with the model's time dimension. When there is a deviation between the actual progress and the model prediction, the model parameters are automatically backtracked and corrected to achieve dynamic simulation of the foundation pit construction process, thus obtaining the dynamic digital twin model.

5. The method for optimizing foundation pit construction using digital twins according to claim 1, characterized in that, The operating parameters of construction equipment are mapped to the digital clone in real time, including: Lightweight edge computing terminals are deployed on each construction device to collect equipment operating parameters in real time; each edge computing terminal and the central server work together to train the equipment operating status prediction model, enabling the digital clone to predict the operating trend of the equipment. Align the device operating parameters with the time dimension in the digital twin model at different sampling frequencies; when the digital twin detects that the device operating parameters deviate from the normal fluctuation range, it quickly locates the cause of the anomaly and generates an early warning message containing maintenance suggestions by combining historical maintenance data; By linking the entire lifecycle data of construction equipment with digital clones, the digital clones can dynamically adjust the weights and thresholds of real-time mapping of operating parameters based on the service life and maintenance records of the construction equipment.

6. The method for optimizing foundation pit construction using digital twins according to claim 1, characterized in that, The dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, the optimal equipment collaborative operation strategy is generated, including: Treating all types of equipment at the foundation pit construction site as individuals in a symbiotic relationship, each piece of equipment is assigned symbiotic requirement parameters based on its functional characteristics; the impact radiation generated by the construction equipment during operation is positively correlated with the intensity of the construction equipment's operation and the degree of its impact on the stability of the foundation pit; In simulating collaborative operation scenarios of different construction equipment, when conflicts arise in the influence radiation between construction equipment, or when the stress and deformation of the foundation pit exceed the safe range, the construction equipment automatically adjusts its own state based on symbiotic adjustment rules; among them, the construction equipment prioritizes meeting the symbiotic needs that are beneficial to the stability maintenance of the foundation pit. The system memorizes and evolves the symbiotic states of equipment that successfully reduced foundation pit risks and improved construction efficiency in each simulation. When a new collaborative operation scenario emerges, the system directly calls the historical best symbiotic state as the initial solution. During the actual simulation, the system dynamically fine-tunes the symbiotic relationship of the construction equipment based on real-time foundation pit stress and deformation data. Finally, it selects the equipment symbiotic mode that minimizes foundation pit stress and deformation and maximizes construction efficiency, and transforms it into the optimal equipment collaborative operation strategy.

7. The method for optimizing foundation pit construction using digital twins according to claim 1, characterized in that, The dynamic digital twin model simulates collaborative operation scenarios of different construction equipment. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, the optimal equipment collaborative operation strategy is generated, including: A mechanical balance simulation system for collaborative construction equipment is constructed, which transforms the operational behavior of construction equipment into adjustable virtual force bundles. Each virtual force bundle has three basic attributes: direction, intensity, and duration of action, which correspond one-to-one with the digging angle, operating power, and duration of the construction equipment. A smart damping network is introduced to cover the entire dynamic digital twin model; the smart damping network dynamically adjusts the force transmission resistance of each area based on the real-time stress monitoring data of the foundation pit support structure. In simulating collaborative operation scenarios of different construction equipment, when there is a conflict between construction equipment or the local stress of the foundation pit exceeds the limit, the force beam adaptive adjustment protocol is activated; the virtual force beams corresponding to each construction equipment autonomously change their direction and intensity according to the resistance distribution of the intelligent damping network. The force field distribution in each simulation is scored using a force field stability evaluation algorithm. After multiple rounds of simulation iterations, the combination of virtual force bundle parameters with the best score is selected and transformed into the optimal equipment collaborative operation strategy.

8. A device for optimizing foundation pit construction using digital twins, characterized in that, include: An acquisition unit is used to acquire key feature data of foundation pit construction; based on the key feature data, a dynamic digital twin model containing the time dimension is constructed. The setting unit is used to set the working condition evolution time axis in the dynamic digital twin model, which can automatically advance the time axis according to the construction progress and synchronously update the physical state of the foundation pit structure in the model. The construction unit is used to build digital clones of construction equipment at the foundation pit construction site in the dynamic digital twin model, and to map the operating parameters of the construction equipment to the digital clones in real time. The simulation unit is used to simulate different construction equipment collaborative operation scenarios in the dynamic digital twin model. By comparing the stress and deformation of the foundation pit under different collaborative operation scenarios, the optimal equipment collaborative operation strategy is generated. The adjustment unit is used to adjust the construction parameters of the construction equipment according to the optimal equipment collaborative operation strategy, so as to optimize the foundation pit construction.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images