A construction engineering survey data analysis method and system based on BIM

Abstract

The present invention relates to the field of data analysis technology, and in particular to a method and system for analyzing construction engineering survey data based on BIM. A construction engineering survey data analysis system based on BIM comprises: a BIM modeling module, a geological offset state analysis module, an alternative pipeline layout path determination module, and a collaborative learning module. The present invention determines a set of alternative pipeline layout paths taking into account the geological offset state through an ant colony algorithm, and sets up collaborative learning between a pipeline layout intelligent agent and an interference intelligent agent, so as to realize the consideration of pipeline layout under the future geological change environment and the consideration of pipeline layout under future human intervention, so that the optimal pipeline layout path finally obtained can have a high resistance to future geological changes and human intervention, thereby improving the service life of the pipeline.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a method and system for analyzing construction engineering survey data based on BIM. Background Art

[0002] In existing construction projects, underground pipeline layout typically relies on static survey data and conventional BIM modeling processes, with routing primarily based on current geological profiles, underground obstacle distribution, and design drawings. However, this approach suffers from two key technical flaws: First, it fails to fully account for the natural evolution of geological conditions over future timescales. Factors such as settlement evolution, fault activity, and groundwater level fluctuations may gradually develop during the pipeline's service life, ultimately leading to structural stress concentrations, joint dislocation, and even pipeline fractures. Second, it ignores the potential cumulative impact of future human interventions (such as urban development and construction, foundation disturbances, and infrastructure excavation). These disturbances are often nonlinear and sudden, easily disrupting the existing geological equilibrium and exacerbating pipeline stress and deformation instability. Because these two aspects are not systematically incorporated into routing and risk assessment processes, current underground pipeline systems exhibit common long-term operational issues: short service life, high maintenance frequency, and a high probability of sudden failures, seriously impacting the safety and economic efficiency of urban underground infrastructure. Summary of the Invention

[0003] The present invention uses an ant colony algorithm to determine a set of alternative pipeline layout paths taking into account geological offset conditions, and sets up collaborative learning between pipeline layout agents and interference agents, so as to realize the consideration of pipeline layout under future geological change environments and future human intervention. The optimal pipeline layout path finally obtained can have high resistance to future geological changes and human intervention, thereby improving the service life of the pipeline.

[0004] The present invention provides a construction engineering survey data analysis method based on BIM, comprising:

[0005] Based on BIM, a 3D model of the layout area corresponding to the area where the pipeline is to be laid is set in the 3D modeling software. The 3D model of the layout area includes a number of volume elements. The acquired construction engineering survey data is mapped to the volume elements in the 3D model of the layout area.

[0006] The construction engineering survey data corresponding to each voxel in the three-dimensional model of the layout area is sent to the geological offset state analysis model for processing, and the geological offset state corresponding to each voxel is output. The geological offset states include S0, S1, S2 and S3;

[0007] Based on the geological offset status of all voxels in the three-dimensional model of the layout area, an alternative pipeline layout path set is determined by an ant colony search algorithm, the alternative pipeline layout path set includes a plurality of alternative pipeline layout paths, and the alternative pipeline layout paths include a plurality of arranged voxel numbers;

[0008] A pipeline layout agent and an interference agent are set up. The pipeline layout agent is responsible for selecting alternative pipeline layout paths from the alternative pipeline layout path set and evaluating the resistance of the alternative pipeline layout path under the influence of the interference agent. The interference agent is used to predict the geological offset state corresponding to each voxel in the three-dimensional model of the layout area and set interference conditions to adjust the geological offset state corresponding to each voxel in the three-dimensional model of the layout area. Based on the pipeline layout agent and the interference agent, all alternative pipeline layout paths in the alternative pipeline layout path set are collaboratively learned, and the optimal pipeline layout path is output. The pipeline layout work in the area to be laid is executed through the optimal pipeline layout path; the interference condition is an interference condition that simulates human intervention.

[0009] Preferably, based on the geological offset status of all voxels in the three-dimensional model of the layout area, an ant colony search algorithm is used to determine a set of candidate pipeline layout paths, which specifically includes the following steps:

[0010] Step H1: Determine the voxel corresponding to the starting point and the voxel corresponding to the target point of the pipeline layout, record them as the starting voxel and the ending voxel, initialize the pheromone value between any two voxels, set the maximum number of iterations, and set a number of ants;

[0011] Step H2: For each ant, perform the following operations: each ant starts from the starting voxel and performs path simulation to obtain a simulated pipeline layout path. The simulated pipeline layout path includes several voxels, and the first and last voxels in the simulated pipeline layout path are the starting voxel and the end voxel respectively;

[0012] Step H3: Calculate the fitness corresponding to each simulated pipeline layout path, and update all pheromone values based on all simulated pipeline layout paths;

[0013] Step H4: Repeat steps H2-H3 until the number of iterations reaches the maximum number of iterations, and output the top N simulated pipeline layout paths with the best fitness values to form a set of candidate pipeline layout paths.

[0014] Preferably, the fitness corresponding to the simulated pipeline layout path is calculated, specifically comprising the following steps: Statistically assuming that the geological offset state in the simulated pipeline layout path is S k The number of voxels Q k , k=0,1,2,3, calculate the geological migration score R=[(∑ k W (S k )Q k) / q], where W(S k ) is the geological offset state S k The corresponding geological state weight, q is the total number of voxels in the simulated pipeline layout path, and the fitness δ corresponding to the simulated pipeline layout path is calculated by the following formula: δ=ηD -1 +εR -1 , where η and ε are the distance correlation coefficient and geological scoring coefficient, respectively, and D is the total distance of the simulated pipeline layout path.

[0015] Preferably, the pipeline layout agent and the interference agent collaboratively learn all the alternative pipeline layout paths in the alternative pipeline layout path set and output the optimal pipeline layout path, which specifically includes the following steps:

[0016] In each iteration of collaborative learning, the interference agent predicts a three-dimensional model of the predicted layout area at a random time point based on the construction engineering survey data prediction network, and the predicted layout area three-dimensional model includes the predicted geological offset states corresponding to all voxels; and the interference agent randomly selects interference conditions from the interference condition library, and adjusts the predicted geological offset states corresponding to all voxels in the predicted layout area three-dimensional model based on the interference conditions to obtain the interference layout area three-dimensional model, and the interference layout area three-dimensional model includes the interference geological offset states corresponding to all voxels; the pipeline layout agent selects an alternative pipeline layout path from the alternative pipeline layout path set, and calculates the geological offset score corresponding to the selected alternative pipeline layout path based on the interference layout area three-dimensional model;

[0017] The interference agent predicts the three-dimensional model of the predicted layout area at a random time point based on the construction engineering survey data prediction network. Specifically, the interference agent includes the following contents: for any volume element, the construction engineering survey data corresponding to the volume element is sent to the construction engineering survey data prediction network for processing to obtain predicted construction engineering survey data, and then the predicted construction engineering survey data is sent to the geological offset state analysis model to obtain the predicted geological offset state corresponding to the volume element;

[0018] The interference intelligent agent randomly selects interference conditions from the interference condition library, and adjusts the predicted geological offset states corresponding to all elements in the three-dimensional model of the predicted layout area based on the interference conditions to obtain the three-dimensional model of the interference layout area, which specifically includes the following contents: setting interference conditions and predicted construction engineering survey data corresponding to all elements in the finite element analysis software, adjusting the predicted construction engineering survey data under the influence of the interference conditions through finite element analysis to obtain interference construction engineering survey data, and then sending the interference construction engineering survey data into the geological offset state analysis model to obtain the interference geological offset state corresponding to the element.

[0019] Preferably, the resistance corresponding to the alternative pipeline layout path is calculated based on the geological offset score of the alternative pipeline layout path in all iteration rounds, specifically including the following steps: counting the number of times G that the geological offset score is higher than the geological offset score threshold in all iteration rounds, and the resistance μ corresponding to the alternative pipeline layout path is: μ=G / U, where U is the total number of iteration rounds.

[0020] Preferably, the geological migration state analysis model is established based on a random forest model; and training the geological migration state analysis model specifically includes the following steps:

[0021] A number of geological offset state analysis training samples are obtained, wherein the geological offset state analysis training samples are construction engineering survey data, and the geological offset state analysis training samples are labeled by geological offset state. All labeled geological offset state analysis training samples are combined into a geological offset state analysis training set, and a geological offset state analysis model is trained using the geological offset state analysis training set, with the labeled geological offset state as the target during training.

[0022] The present invention also provides a BIM-based construction engineering survey data analysis system, comprising:

[0023] The BIM modeling module is used to set up a three-dimensional model of the layout area corresponding to the area where the pipeline is to be laid in the three-dimensional modeling software based on BIM. The three-dimensional model of the layout area includes a number of volume elements, and the acquired construction engineering survey data is mapped to the volume elements in the three-dimensional model of the layout area;

[0024] The geological offset state analysis module is used to input the construction engineering survey data corresponding to each voxel in the three-dimensional model of the layout area into the geological offset state analysis model for processing, and output the geological offset state corresponding to each voxel. The geological offset states include S0, S1, S2 and S3;

[0025] An alternative pipeline layout path determination module is used to determine an alternative pipeline layout path set based on the geological offset status of all voxels in the three-dimensional model of the layout area through an ant colony search algorithm, wherein the alternative pipeline layout path set includes a plurality of alternative pipeline layout paths, and the alternative pipeline layout paths include a plurality of arranged voxel numbers;

[0026] The collaborative learning module is used to set up the pipeline layout agent and the interference agent. The pipeline layout agent is responsible for selecting alternative pipeline layout paths from the alternative pipeline layout path set and evaluating the resistance of the alternative pipeline layout path under the influence of the interference agent. The interference agent is used to predict the geological offset state corresponding to each element in the three-dimensional model of the layout area and set interference conditions to adjust the geological offset state corresponding to each element in the three-dimensional model of the layout area. Based on the pipeline layout agent and the interference agent, all alternative pipeline layout paths in the alternative pipeline layout path set are collaboratively learned, and the optimal pipeline layout path is output. The pipeline layout work in the area to be laid is performed through the optimal pipeline layout path.

[0027] The present invention has the following advantages:

[0028] The present invention uses an ant colony algorithm to determine a set of alternative pipeline layout paths taking into account geological offset conditions, and sets up collaborative learning between pipeline layout agents and interference agents, so as to realize the consideration of pipeline layout under future geological change environments and future human intervention. The optimal pipeline layout path finally obtained can have high resistance to future geological changes and human intervention, thereby improving the service life of the pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a schematic diagram of the structure of a BIM-based construction engineering survey data analysis system adopted in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0031] Example 1, a BIM-based construction engineering survey data analysis method, comprising:

[0032] Based on BIM, a three-dimensional model of the layout area corresponding to the area where the pipeline is to be laid is set in the three-dimensional modeling software. The three-dimensional model of the layout area includes a number of individual elements. A body number is set for each body element. In this embodiment, the length of the three-dimensional model of the layout area is generally divided into I parts, the width is divided into J parts, and the height is divided into K parts. Correspondingly, the three-dimensional model of the layout area includes I×J×K individual elements. Each body element represents a local position in the area where the pipeline is to be laid. The acquired construction engineering survey data is mapped to the body elements in the three-dimensional model of the layout area. The construction engineering survey data here includes each body element in the area where the pipeline is to be laid. The information of the layer position, thickness, lithology and interlayer (derived from the borehole histogram and geological profile), friction resistance, tip resistance and N value (derived from in-situ test data) and compression coefficient, creep modulus, internal friction angle and cohesion (derived from indoor geotechnical test data) are obtained. It should be noted that since the acquisition of the construction engineering survey data in the area where the pipeline is to be laid is sampled, it is impossible to obtain the construction engineering survey data corresponding to each position in the area where the pipeline is to be laid. Therefore, for the volume elements corresponding to the actually acquired positions in the area where the pipeline is to be laid, the mapping of the construction engineering survey data is generally achieved through spatial interpolation.

[0033] The construction engineering survey data corresponding to each voxel in the three-dimensional model of the layout area is sent to the geological offset state analysis model for processing, and the geological offset state corresponding to each voxel is output. The geological offset states include S0, S1, S2 and S3, among which S0 is a stable state, which indicates that the geology at the location corresponding to the voxel will not offset or will offset slightly in the short term, and the pipeline can be laid normally. S1 is a micro-offset state, which indicates that the geology at the location corresponding to the voxel will offset slightly in the long term, and a flexible interface is required when laying the pipeline. S2 is a medium-offset state, which indicates that the geology at the location corresponding to the voxel will offset moderately, which will cause the pipeline to bend and become empty, etc., and reinforcement or direct avoidance is required when laying the pipeline. S3 is a high-offset state, which indicates that the geology at the location corresponding to the voxel will offset severely, which will cause the pipeline to break, and pipeline laying is strictly prohibited.

[0034] The geological migration state analysis model here is established based on the random forest model. Training the geological migration state analysis model specifically includes the following steps:

[0035] Obtain several geological offset state analysis training samples, which are construction engineering survey data. The geological offset state analysis training samples here are obtained by operators from experiments performed after actual collection. The geological offset state analysis training samples are labeled by geological offset state, and all labeled geological offset state analysis training samples are composed of a geological offset state analysis training set. The geological offset state analysis model is trained using the geological offset state analysis training set. During the training, the labeled geological offset state is used as the target, and feature selection and decision tree construction are performed to determine whether the accuracy of the geological offset state analysis model meets expectations. If the accuracy of the geological offset state analysis model meets expectations, the trained geological offset state analysis model is output; otherwise, the geological offset state analysis model is continued to be trained using the geological offset state analysis training set.

[0036] Based on the geological offset status of all voxels in the three-dimensional model of the layout area, an ant colony search algorithm is used to determine a set of alternative pipeline layout paths. The set of alternative pipeline layout paths includes several alternative pipeline layout paths, and the alternative pipeline layout paths include several arranged voxel numbers. It should be noted that during the execution of underground pipeline layout, all voxels in the three-dimensional model of the layout area will be searched based on the pre-set starting and target points. The ant colony search algorithm is used to determine several expected alternative pipeline layout paths. These alternative pipeline layout paths represent feasible pipeline layout solutions.

[0037] Set up pipeline layout intelligent agent and interference intelligent agent, pipeline layout intelligent agent is responsible for selecting alternative pipeline layout path from the alternative pipeline layout path set and evaluating the resistance of the alternative pipeline layout path under the influence of interference intelligent agent, interference intelligent agent is used to predict the geological offset state corresponding to each voxel in the three-dimensional model of the layout area and set interference conditions to adjust the geological offset state corresponding to each voxel in the three-dimensional model of the layout area, based on the pipeline layout intelligent agent and interference intelligent agent, all alternative pipeline layout paths in the alternative pipeline layout path set are collaboratively learned, and the optimal pipeline layout path is output. The pipeline layout work in the area where the pipeline is to be laid is performed along the path. It should be noted that the interference agent can perform time-series prediction of the geological offset state, and then determine the adaptability of the alternative pipeline layout path under the future geological environment. It also sets interference conditions that simulate human intervention, such as tunneling operations, excavation operations, piling operations, and grouting operations corresponding to human construction. These operations will affect the use of the pipeline. Therefore, through collaborative learning, the resistance of the alternative pipeline layout path to human intervention is evaluated, so that the optimal pipeline layout path can take into account future geological changes and human intervention, thereby improving the service life of the pipeline.

[0038] This application uses an ant colony algorithm to determine a set of alternative pipeline layout paths taking into account geological offset conditions, and sets up collaborative learning between pipeline layout agents and interference agents to achieve consideration of pipeline layout under future geological change environments and future human intervention, so that the optimal pipeline layout path finally obtained can have a high resistance to future geological changes and human intervention, thereby improving the service life of the pipeline.

[0039] Based on the geological offset status of all voxels in the 3D model of the layout area, an ant colony search algorithm is used to determine a set of alternative pipeline layout paths. The specific steps include the following:

[0040] Step H1: Determine the voxel corresponding to the starting point and the voxel corresponding to the target point of the pipeline layout, record them as the starting voxel and the ending voxel, initialize the pheromone value between any two voxels, set the maximum number of iterations, and set a number of ants;

[0041] Step H2: For each ant, perform the following operations: each ant starts from the starting voxel and performs path simulation to obtain a simulated pipeline layout path. The simulated pipeline layout path includes several voxels, and the first and last voxels in the simulated pipeline layout path are the starting voxel and the end voxel respectively;

[0042] It should be noted that during the path simulation, the ant will calculate the selection probability of all the voxels adjacent to the current voxel based on the pheromone value and the heuristic value. For example, if there are M voxels adjacent to the current voxel, then the selection probability of the mth adjacent voxel is Fm=[(Am) α × (Bm) β ) / ∑ i (Am) α × (Bm) β ), i∈(1, 2, 3, …, M)], m=1, 2, 3, …, M, where Am is the pheromone between the current voxel and the mth adjacent voxel, Bm is the heuristic value between the current voxel and the mth adjacent voxel, that is, the inverse of the distance between the current voxel and the mth adjacent voxel, α and β are both weight coefficients, and the next voxel is selected by a roulette wheel selection algorithm based on the corresponding selection probabilities of all voxels adjacent to the current voxel, and the voxel is not repeatedly selected until the next voxel selected is the end voxel or there is no voxel to be selected, completing the path simulation, and forming all voxels traversed during the path simulation into a candidate layout path, and the candidate layout path whose first and last positions are the starting voxel and the end voxel in all the candidate layout paths are recorded as the simulated pipeline layout path; in the process of path simulation in this application, the voxels with a geological offset state of S3 and the inaccessible voxels obtained through construction engineering survey and analysis are also avoided in a priori manner;

[0043] Step H3: Calculate the fitness corresponding to each simulated pipeline layout path, and update all pheromone values based on all simulated pipeline layout paths;

[0044] Calculate the fitness corresponding to the simulated pipeline layout path, specifically including the following steps: Statistically simulate the geological offset state in the pipeline layout path as S k The number of voxels Q k , k=0,1,2,3, calculate the geological migration score R=[(∑ k W (S k )Q k ) / q], where W(S k ) is the geological offset state S k The corresponding geological state weight, generally S0=0, S1=0.3, S2=0.7, S3=1.0, is used to reflect the risk severity corresponding to the geological state. q is the total number of voxels in the simulated pipeline layout path. The fitness δ corresponding to the simulated pipeline layout path is calculated by the following formula: δ=ηD -1 +εR -1 , where η and ε are the distance correlation coefficient and geological score coefficient, respectively, reflecting the degree of attention corresponding to the distance and geological offset score. The fitness δ corresponding to the simulated pipeline layout path is greater when the distance is shorter and the geological offset score is lower. D is the total distance of the simulated pipeline layout path.

[0045] It should be noted that the basis for updating all pheromone values based on all simulated pipeline layout paths is that ants release pheromones on the edges they pass through in this round according to the length of the paths they construct. The shorter the paths constructed by the ants, the more pheromones they release; and the more times an edge is crossed by an ant, the more pheromones it obtains.

[0046] Step H4: Repeat steps H2-H3 until the number of iterations reaches the maximum number of iterations, and output the top N simulated pipeline layout paths with the best fitness values to form a set of candidate pipeline layout paths.

[0047] Based on the pipeline layout agent and the interference agent, all the alternative pipeline layout paths in the alternative pipeline layout path set are collaboratively learned to output the optimal pipeline layout path. The specific steps include the following:

[0048] In each iteration of collaborative learning, the interference agent predicts a three-dimensional model of the predicted layout area at a random time point based on the construction engineering survey data prediction network. The predicted layout area three-dimensional model includes the predicted geological offset states corresponding to all voxels; and the interference agent randomly selects interference conditions from the interference condition library, and adjusts the predicted geological offset states corresponding to all voxels in the predicted layout area three-dimensional model based on the interference conditions to obtain the interference layout area three-dimensional model. The interference layout area three-dimensional model includes the interference geological offset states corresponding to all voxels. The interference geological offset state here has the same form as the geological offset state and is also represented by S0, S1, S2 and S3. The pipeline layout agent selects an alternative pipeline layout path from the alternative pipeline layout path set, and calculates the geological offset score corresponding to the selected alternative pipeline layout path based on the interference layout area three-dimensional model;

[0049] It should be noted that the interference agent predicts the three-dimensional model of the predicted layout area at a random time point based on the construction engineering survey data prediction network, which specifically includes the following contents: for any voxel, the construction engineering survey data corresponding to the voxel is sent to the construction engineering survey data prediction network for processing to obtain predicted construction engineering survey data, and then the predicted construction engineering survey data is sent to the geological offset state analysis model to obtain the predicted geological offset state corresponding to the voxel. It should be noted here that the construction engineering survey data prediction network is established based on a recurrent neural network, and the time interval is set to 1 year. The construction engineering survey data prediction network is trained through the actual collected historical construction engineering survey data;

[0050] The interference agent randomly selects interference conditions from the interference condition library, and adjusts the predicted geological offset states corresponding to all voxels in the three-dimensional model of the predicted layout area based on the interference conditions to obtain the three-dimensional model of the interference layout area, which specifically includes the following contents: the interference condition library includes several interference conditions, and the interference conditions include interference methods (such as piling, excavation, grouting and drainage, etc.), corresponding interference intensities (such as piling energy corresponding to piling, grouting pressure corresponding to grouting) and interference center points (corresponding to the center of the voxel), each interference condition corresponds to an influence, and the predicted geological offset states corresponding to all voxels in the three-dimensional model of the predicted layout area are adjusted based on the interference conditions. Specifically, the interference conditions and the predicted construction engineering survey data corresponding to all voxels are set in the finite element analysis software (such as PLAXIS), and the predicted construction engineering survey data under the influence of the interference conditions are adjusted through finite element analysis to obtain interference construction engineering survey data, and then the interference construction engineering survey data are sent to the geological offset state analysis model to obtain the interference geological offset state corresponding to the voxel;

[0051] The geological offset score is calculated as follows: the interference geological offset state in the alternative pipeline layout path is Sk The number of voxels Q k , k=0,1,2,3, calculate the geological migration score R=[(∑ k W (S k )Q k ) / q], where W(S k ) is the interference geological migration state S k The corresponding geological state weight, generally S0=0, S1=0.3, S2=0.7, S3=1.0, is used to reflect the risk severity corresponding to the geological state. q is the total number of voxels in the alternative pipeline layout path;

[0052] Until the number of collaborative learning iterations reaches the maximum number of collaborative learning iterations, the resistance corresponding to the alternative pipeline layout path is calculated based on the geological deviation score of the alternative pipeline layout path in all iterative rounds. The resistance can reflect the stability of the alternative pipeline layout path under geological changes and random human intervention; the alternative pipeline layout path with the largest resistance is selected as the optimal pipeline layout path output.

[0053] The resistance of the alternative pipeline layout path is calculated based on the geological offset score of the alternative pipeline layout path in all iteration rounds, which specifically includes the following steps:

[0054] Count the number of times G that the geological offset score exceeds the geological offset score threshold in all iterations. The geological offset score threshold here is set by the operator. The resistance μ corresponding to the alternative pipeline layout path is: μ = G / U, where U is the total number of iterations, that is, the maximum number of iterations of collaborative learning.

[0055] Example 2, a BIM-based construction engineering survey data analysis system, see Figure 1 ,include:

[0056] The BIM modeling module is used to set a three-dimensional model of the layout area corresponding to the area where the pipeline is to be laid in the three-dimensional modeling software based on BIM. The three-dimensional model of the layout area includes a number of individual elements. A number of individual elements is set for each element. In this embodiment, the length of the three-dimensional model of the layout area is generally divided into I parts, the width is divided into J parts, and the height is divided into K parts. Correspondingly, the three-dimensional model of the layout area includes I×J×K individual elements. Each element represents a local position in the area where the pipeline is to be laid. The acquired construction engineering survey data is mapped to the elements in the three-dimensional model of the layout area. The construction engineering survey data here includes the pipeline to be laid. Information on the stratigraphic position, thickness, lithology, and interlayers at each location in the region (derived from borehole histograms and geological profiles), friction resistance, tip resistance, and N value (derived from in-situ test data), and compression coefficient, creep modulus, internal friction angle, and cohesion (derived from indoor geotechnical test data). It should be noted that since the acquisition of construction engineering survey data in the area where the pipeline is to be laid is sampled, it is impossible to obtain the construction engineering survey data corresponding to each location in the area where the pipeline is to be laid. Therefore, for the voxels corresponding to the actually acquired locations in the area where the pipeline is to be laid, the mapping of the construction engineering survey data is generally achieved through spatial interpolation.

[0057] The geological offset state analysis module is used to send the construction engineering survey data corresponding to each voxel in the three-dimensional model of the layout area into the geological offset state analysis model for processing, and output the geological offset state corresponding to each voxel. The geological offset states include S0, S1, S2 and S3, among which S0 is a stable state, indicating that the geology at the location corresponding to the voxel will not offset or will offset slightly in the short term, and the pipeline can be laid normally. S1 is a micro-offset state, indicating that the geology at the location corresponding to the voxel will offset slightly in the long term, and a flexible interface is required when laying the pipeline. S2 is a medium offset state, indicating that the geology at the location corresponding to the voxel will offset moderately, which will cause the pipeline to bend and become hollow, etc., and reinforcement or direct avoidance is required when laying the pipeline. S3 is a high offset state, indicating that the geology at the location corresponding to the voxel will offset severely, which will cause the pipeline to break, and pipeline laying is strictly prohibited.

[0058] The module for determining alternative pipeline layout paths is used to determine a set of alternative pipeline layout paths using an ant colony search algorithm based on the geological offset status of all voxels in the three-dimensional model of the layout area. The set of alternative pipeline layout paths includes several alternative pipeline layout paths, each of which includes several arranged voxel numbers. It should be noted that during the execution of underground pipeline layout, all voxels in the three-dimensional model of the layout area will be searched based on the pre-set starting and target points, and several alternative pipeline layout paths that meet the expectations will be determined using the ant colony search algorithm. These alternative pipeline layout paths represent feasible pipeline layout solutions;

[0059] The collaborative learning module is used to set up pipeline layout agents and interference agents. The pipeline layout agent is responsible for selecting alternative pipeline layout paths from the alternative pipeline layout path set and evaluating the resistance of the alternative pipeline layout path under the influence of the interference agent. The interference agent is used to predict the geological offset state corresponding to each element in the three-dimensional model of the layout area and set interference conditions to adjust the geological offset state corresponding to each element in the three-dimensional model of the layout area. Based on the pipeline layout agent and the interference agent, all alternative pipeline layout paths in the alternative pipeline layout path set are collaboratively learned and the optimal pipeline layout path is output. The pipeline layout path executes the pipeline layout work in the area where the pipeline is to be laid. It should be noted that the interference intelligent agent can perform time-series prediction of the geological offset state, and then judge the adaptability of the alternative pipeline layout path under the future geological environment. It also sets interference conditions that simulate human intervention, such as tunneling operations, excavation operations, piling operations and grouting operations corresponding to human construction. These operations will affect the use of the pipeline. Therefore, the resistance of the alternative pipeline layout path to human intervention is evaluated through collaborative learning, so that the optimal pipeline layout path can take into account future geological changes and human intervention, thereby improving the service life of the pipeline.

[0060] It should be understood that those skilled in the art may make improvements or modifications based on the above description, and all such improvements and modifications shall fall within the scope of protection of the appended claims. Any portion of this specification not described in detail herein is prior art known to those skilled in the art.

Claims

1. A construction engineering survey data analysis method based on BIM, characterized in that: include: Based on BIM, a 3D model of the layout area corresponding to the area where the pipeline is to be laid is set in the 3D modeling software. The 3D model of the layout area includes a number of volume elements. The acquired construction engineering survey data is mapped to the volume elements in the 3D model of the layout area. The construction engineering survey data corresponding to each voxel in the three-dimensional model of the layout area is sent to the geological offset state analysis model for processing, and the geological offset state corresponding to each voxel is output. The geological offset states include S0, S1, S2 and S3; Based on the geological offset status of all voxels in the three-dimensional model of the layout area, an alternative pipeline layout path set is determined by an ant colony search algorithm, the alternative pipeline layout path set includes a plurality of alternative pipeline layout paths, and the alternative pipeline layout paths include a plurality of arranged voxel numbers; A pipeline layout agent and an interference agent are set up. The pipeline layout agent is responsible for selecting an alternative pipeline layout path from the alternative pipeline layout path set and evaluating the resistance of the alternative pipeline layout path under the influence of the interference agent. The interference agent is used to predict the geological offset state corresponding to each voxel in the three-dimensional model of the layout area and set interference conditions to adjust the geological offset state corresponding to each voxel in the three-dimensional model of the layout area. Based on the pipeline layout agent and the interference agent, all alternative pipeline layout paths in the alternative pipeline layout path set are collaboratively learned, and the optimal pipeline layout path is output. The pipeline layout work in the pipeline area to be laid is executed through the optimal pipeline layout path; the interference condition is an interference condition simulating human intervention; Based on the pipeline layout agent and the interference agent, all the alternative pipeline layout paths in the alternative pipeline layout path set are collaboratively learned to output the optimal pipeline layout path. The specific steps include the following: In each iteration of collaborative learning, the interference agent predicts a three-dimensional model of the predicted layout area at a random time point based on the construction engineering survey data prediction network, and the predicted layout area three-dimensional model includes the predicted geological offset states corresponding to all voxels; and the interference agent randomly selects interference conditions from the interference condition library, and adjusts the predicted geological offset states corresponding to all voxels in the predicted layout area three-dimensional model based on the interference conditions to obtain the interference layout area three-dimensional model, and the interference layout area three-dimensional model includes the interference geological offset states corresponding to all voxels; the pipeline layout agent selects an alternative pipeline layout path from the alternative pipeline layout path set, and calculates the geological offset score corresponding to the selected alternative pipeline layout path based on the interference layout area three-dimensional model; The interference agent predicts a three-dimensional model of the predicted layout area at a random time point based on the construction engineering survey data prediction network. Specifically, the interference agent includes the following contents: for any volume element, the construction engineering survey data corresponding to the volume element is sent to the construction engineering survey data prediction network for processing to obtain predicted construction engineering survey data, and then the predicted construction engineering survey data is sent to the geological offset state analysis model to obtain the predicted geological offset state corresponding to the volume element; The interference intelligent agent randomly selects interference conditions from the interference condition library, and adjusts the predicted geological offset states corresponding to all elements in the three-dimensional model of the predicted layout area based on the interference conditions to obtain the three-dimensional model of the interference layout area, which specifically includes the following contents: setting interference conditions and predicted construction engineering survey data corresponding to all elements in the finite element analysis software, adjusting the predicted construction engineering survey data under the influence of the interference conditions through finite element analysis to obtain interference construction engineering survey data, and then sending the interference construction engineering survey data into the geological offset state analysis model to obtain the interference geological offset state corresponding to the element.

2. The method for analyzing construction engineering survey data based on BIM according to claim 1, characterized in that: Based on the geological offset status of all voxels in the 3D model of the layout area, an ant colony search algorithm is used to determine a set of alternative pipeline layout paths. The specific steps include the following: Step H1: Determine the voxel corresponding to the starting point and the voxel corresponding to the target point of the pipeline layout, record them as the starting voxel and the ending voxel, initialize the pheromone value between any two voxels, set the maximum number of iterations, and set a number of ants; Step H2: For each ant, perform the following operations: each ant starts from the starting voxel and performs path simulation to obtain a simulated pipeline layout path. The simulated pipeline layout path includes several voxels, and the first and last voxels in the simulated pipeline layout path are the starting voxel and the end voxel respectively; Step H3: Calculate the fitness corresponding to each simulated pipeline layout path, and update all pheromone values based on all simulated pipeline layout paths; Step H4: Repeat steps H2-H3 until the number of iterations reaches the maximum number of iterations, and output the top N simulated pipeline layout paths with the best fitness values to form a set of candidate pipeline layout paths.

3. The method for analyzing construction engineering survey data based on BIM according to claim 2, characterized in that: Calculate the fitness corresponding to the simulated pipeline layout path, specifically including the following steps: Statistically simulate the geological offset state in the pipeline layout path as S k The number of voxels Q k , k=0,1,2,3, calculate the geological migration score R=[(∑ k W (S k )Q k ) / q], where W(S k ) is the geological offset state S k The corresponding geological state weight, q is the total number of voxels in the simulated pipeline layout path, and the fitness δ corresponding to the simulated pipeline layout path is calculated by the following formula: δ=ηD -1 +εR -1 , where η and ε are the distance correlation coefficient and geological scoring coefficient, respectively, and D is the total distance of the simulated pipeline layout path.

4. The method for analyzing construction engineering survey data based on BIM according to claim 3, characterized in that: The resistance corresponding to the alternative pipeline layout path is calculated based on the geological offset score of the alternative pipeline layout path in all iteration rounds. The specific steps include: counting the number of times G that the geological offset score is higher than the geological offset score threshold in all iteration rounds. The resistance μ corresponding to the alternative pipeline layout path is: μ=G / U, where U is the total number of iteration rounds.

5. The method for analyzing construction engineering survey data based on BIM according to claim 4, characterized in that: The geological migration state analysis model is established based on the random forest model. Training the geological migration state analysis model includes the following steps: A number of geological offset state analysis training samples are obtained, wherein the geological offset state analysis training samples are construction engineering survey data, and the geological offset state analysis training samples are labeled by geological offset state. All labeled geological offset state analysis training samples are combined into a geological offset state analysis training set, and a geological offset state analysis model is trained using the geological offset state analysis training set, with the labeled geological offset state as the target during training.

6. A BIM-based construction engineering survey data analysis system, characterized in that: The system applies the BIM-based construction engineering survey data analysis method according to any one of claims 1 to 5, comprising: The BIM modeling module is used to set up a three-dimensional model of the layout area corresponding to the area where the pipeline is to be laid in the three-dimensional modeling software based on BIM. The three-dimensional model of the layout area includes a number of volume elements, and the acquired construction engineering survey data is mapped to the volume elements in the three-dimensional model of the layout area; The geological offset state analysis module is used to input the construction engineering survey data corresponding to each voxel in the three-dimensional model of the layout area into the geological offset state analysis model for processing, and output the geological offset state corresponding to each voxel. The geological offset states include S0, S1, S2 and S3; An alternative pipeline layout path determination module is used to determine an alternative pipeline layout path set based on the geological offset status of all voxels in the three-dimensional model of the layout area through an ant colony search algorithm, wherein the alternative pipeline layout path set includes a plurality of alternative pipeline layout paths, and the alternative pipeline layout paths include a plurality of arranged voxel numbers; The collaborative learning module is used to set up the pipeline layout agent and the interference agent. The pipeline layout agent is responsible for selecting alternative pipeline layout paths from the alternative pipeline layout path set and evaluating the resistance of the alternative pipeline layout path under the influence of the interference agent. The interference agent is used to predict the geological offset state corresponding to each element in the three-dimensional model of the layout area and set interference conditions to adjust the geological offset state corresponding to each element in the three-dimensional model of the layout area. Based on the pipeline layout agent and the interference agent, all alternative pipeline layout paths in the alternative pipeline layout path set are collaboratively learned, and the optimal pipeline layout path is output. The pipeline layout work in the area to be laid is performed through the optimal pipeline layout path.

Images