BIM-based railway engineering labor inspection pricing management system construction method

Abstract

The invention discloses a BIM-based railway engineering inspection pricing management system construction method, and relates to the technical field of engineering management, and the method comprises the steps: carrying out the lightweight decomposition of a BIM model of railway engineering, generating an engineering structure tree, binding a structural unit of the engineering structure tree with a contract list, and generating a relation mapping table; geological radar data is acquired and analyzed into a karst development coefficient, acceptance status data is converted into a quality risk value, a handwritten annotation image is identified through an adversarial generative network, and an empirical risk coefficient is output; and according to the high-risk mark, a high-risk section corresponding to the geological radar data is marked, the measurement risk coefficient and the space coordinate of the high-risk section are stored in a block chain in an associated manner, and an inspection pricing sheet is generated. According to the method, deep fusion and real-time decision response of multi-source heterogeneous data are realized by constructing the geological risk mapping knowledge domain, and the limitation of traditional linear weighted fusion is broken through.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of engineering management, and particularly relates to a method for constructing a railway engineering inspection valuation management system based on BIM. BACKGROUND

[0002] With the deep application of intelligent construction in railway engineering, the BIM-based inspection valuation system gradually evolves from geometric modeling to data collaborative decision-making. In recent years, blockchain has been used in the field of engineering payment to ensure the traceability of inspection data. In the aspect of geological risk assessment, geological radar data is associated with the BIM model, but the technical bottleneck of dynamic measurement correction has not been broken through.

[0003] The prior art has the problem that geological risk factors cannot be organically integrated into the measurement payment decision chain, and traditional inspection valuation relies on static contract lists and manual progress verification, which cannot quantitatively integrate multi-source heterogeneous risk parameters such as karst development coefficients, quality risk values and experience risk coefficients, and lacks a blockchain credible evidence mechanism to support audit traceability. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a method for constructing a railway engineering inspection valuation management system based on BIM to solve the problem that railway engineering inspection valuation under complex geological conditions cannot integrate multi-source geological risks to achieve dynamic quantification and blockchain credible evidence.

[0006] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides a method for constructing a railway engineering inspection valuation management system based on BIM, which includes: decomposing the BIM model of railway engineering into a lightweight, generating an engineering structure tree, binding the structure units of the engineering structure tree to the contract list, and generating a relationship mapping table; Obtaining geological radar data to analyze karst development coefficients, converting acceptance state data into quality risk values, identifying handwritten annotation images through a generative adversarial network, and outputting experience risk coefficients; Constructing a geological risk knowledge graph, taking the karst development coefficients, quality risk values and experience risk coefficients as nodes, and integrating to generate a measurement risk coefficient; Calling the relationship mapping table, applying the measurement risk coefficient to the corresponding contract list quantities, and calculating the current inspection measurement value; Comparing the cumulative value of the current inspection measurement value with the total quantity of the contract list, if the cumulative value does not exceed the total quantity of the contract list, inputting the approval process, otherwise freezing the approval process and generating a high-risk marker; According to the high-risk mark, the high-risk section corresponding to the ground penetrating radar data is marked, the risk coefficient is stored in association with the spatial coordinates of the high-risk section to the block chain, and the construction pricing list is generated.

[0007] As a preferred scheme of the method for constructing the BIM-based railway engineering construction pricing management system, the engineering structure tree is generated, and the specific steps are as follows, Read the component geometric properties and dynamic load characteristics in the railway engineering BIM model, perform component semantic restructuring, and output the component topology structure with classification marked engineering priority; The component topology structure with engineering priority marked is subjected to frequency domain compression processing, and is spatiotemporally bound with the construction schedule, and a lightweight component set with dynamic weight marked is output; Based on the lightweight component set with dynamic weight marked, the spatial position attribute and the construction time sequence attribute are taken as nodes to generate the engineering structure tree.

[0008] As a preferred scheme of the method for constructing the BIM-based railway engineering construction pricing management system, the relationship mapping table is generated, and the specific steps are as follows, Traverse all nodes of the engineering structure tree, extract the spatial position attribute and the construction time sequence attribute of each node, and read the engineering quantity and the position range identifier of the contract list; Match and associate the spatial position attribute of the engineering structure tree with the position range identifier of the contract list, and generate a matching weight value by fusing the construction time sequence attribute; According to the matching weight value, correct the engineering quantity of the corresponding contract list item, generate an engineering quantity correction record with a construction time sequence influence factor, and integrate it into the relationship mapping table.

[0009] As a preferred scheme of the method for constructing the BIM-based railway engineering construction pricing management system, the specific steps for outputting the experience risk coefficient are as follows, According to the spatial position attribute in the relationship mapping table, locate the target section of the tunnel, collect the ground penetrating radar data and analyze it into a karst development coefficient; Based on the karst development coefficient, associate the construction time sequence attribute to match the corresponding inspection lot record, and calculate the quality risk value by using the fuzzy comprehensive evaluation algorithm; Taking the quality risk value as the weight reference, locate the construction drawing area associated with the nodes in the engineering structure tree, intercept the handwritten annotation image and analyze the semantic content by using the generative adversarial network, and identify the annotation semantics; Input the identified annotation semantics into the preset engineering experience risk knowledge base for quantitative mapping to generate the experience risk coefficient.

[0010] As a preferred scheme of the method for constructing the BIM-based railway engineering inspection valuation management system, the specific steps for generating the measurement risk coefficient in the fusion are as follows, The geological risk nodes, quality risk nodes and experience risk nodes are defined, the relationship attributes of spatial correlation, time sequence evolution and causal influence are established, and a geological risk knowledge graph basic framework is generated. According to the spatial position attribute, the karst development coefficient is bound to the geological risk node coordinate, the quality risk value is associated as a node chain according to the construction time sequence attribute, and the experience risk coefficient is injected into the experience risk node to form a geological risk knowledge graph. The risk interaction matrix in the geological risk knowledge graph is extracted, and three-dimensional orthogonal convolution coupling operation is implemented to output the updated geological risk knowledge graph. The updated geological risk knowledge graph is subjected to multi-scale wavelet packet decomposition, the frequency band energy features are extracted, and the measurement risk coefficient is generated by an entropy weight fusion algorithm.

[0011] As a preferred scheme of the method for constructing the BIM-based railway engineering inspection valuation management system, the specific steps for calculating the current inspection measurement value are as follows, The engineering quantity correction record with the construction time sequence influence factor in the relationship mapping table is called, and a four-dimensional tensor coupling matrix is constructed in combination with the measurement risk coefficient. The four-dimensional tensor coupling matrix is subjected to non-Abelian gauge field transformation by a Dirac matrix operator to generate a gauge field action functional under the constraint of the current space-time boundary. The gauge field action functional is input into a conformal deformation mapping engine, and a Riemann surface parameter integral is performed along the engineering partition boundary to output the current inspection measurement value.

[0012] As a preferred scheme of the method for constructing the BIM-based railway engineering inspection valuation management system, the specific steps for generating the high-risk label are as follows, The historical cumulative value is called and corrected by the time-dependent coefficient of the karst area geology to generate a correction result, and the correction result is superimposed with the current inspection measurement value to generate a cumulative value of the current inspection measurement value. The cumulative value of the current inspection measurement value is compared with the total amount of the contract list, if the cumulative value of the current inspection measurement value does not exceed the total amount of the contract list, a multi-level approval and signature process is automatically started, otherwise the approval process is frozen, and a high-risk label is generated at the associated spatial coordinate of the BIM model.

[0013] As a preferred scheme of the method for constructing the BIM-based railway engineering inspection valuation management system, the specific steps for generating the inspection valuation single are as follows, According to the spatial coordinates of the high-risk mark, intelligent segmentation processing of the geological radar data is triggered to generate a spatial coordinate set of the high-risk section; Based on the spatial coordinate set of the high-risk section, a time-space risk vector is constructed by fusing the metrological risk coefficient and stored by the block chain to generate a time-space digital fingerprint; The dynamic risk weight is extracted by using the time-space digital fingerprint, the risk correction metrology is performed on the associated contract list engineering quantity, and the correction metrology result is generated; According to the correction metrology result and the time-space digital fingerprint, a check engineering pricing sheet with a block chain verification mark and a risk coordinate is generated.

[0014] In a second aspect, the present application provides a computer device comprising a memory and a processor, the memory storing a computer program, wherein: the computer program is executed by the processor to implement any step of the method for constructing a BIM-based railway engineering check engineering pricing management system according to the first aspect of the present application.

[0015] In a third aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein: the computer program is executed by the processor to implement any step of the method for constructing a BIM-based railway engineering check engineering pricing management system according to the first aspect of the present application.

[0016] The present application has the beneficial effects that: by constructing a geological risk knowledge graph, multi-source heterogeneous data deep fusion and real-time decision response are realized, and the limitations of traditional linear weighted fusion are broken through. The topological structure of the geological risk knowledge graph captures the nonlinear correlation of geological-geological inspection-experience data; the GCN layer iterative updating mechanism enables the metrological risk coefficient to have dynamic response capability, supporting accurate correction of metrological values and block chain storage and auditing. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Fig. 1 The flowchart of the method for constructing a BIM-based railway engineering check engineering pricing management system.

[0019] Fig. 2 The flowchart for generating the engineering structure tree and the relationship mapping table.

[0020] Fig. 3 The flowchart for constructing a geological risk knowledge graph and generating a metrological risk coefficient.

[0021] Fig. 4 Flow chart for generating the inspection and pricing sheet. DETAILED DESCRIPTION

[0022] In order to make the above objectives, features and advantages of the present application more apparent, specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0023] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, it will be apparent to one skilled in the art that the present application can be practiced without the specific details given herein, that the present application can be practiced with other different ways, and that the present application is not limited to the embodiments given herein. In other instances, well-known methods, procedures, components, and networks have not been described in detail so as not to unnecessarily obscure aspects of the present application.

[0024] Secondly, the term 'one embodiment' or 'an embodiment' as used herein means that a particular implementation can include a particular feature, structure, or characteristic, but every embodiment can not necessarily include the particular feature, structure, or characteristic. Furthermore, the following terms, as used in the description and associated claims, shall not be limited to the specifically recited uses.

[0025] Reference Signs List Figs. 1-4 For one embodiment of the present application, the embodiment provides a method for constructing a BIM-based railway engineering inspection and pricing management system, comprising the following steps: S1, decompose the BIM model of the railway engineering, generate an engineering structure tree, bind the structure units of the engineering structure tree with the contract list, and generate a relationship mapping table.

[0026] Further, read the component geometric properties and dynamic load characteristics in the railway engineering BIM model, perform component semantic reorganization, and output the component topology structure classified and marked with engineering priority; Specifically, access the railway engineering BIM model, extract component geometric properties such as shape, size, and spatial coordinates, and dynamic load characteristics such as force direction, value, and change period, process the component geometric properties and dynamic load characteristics, group the component geometric properties according to role function and material properties through component semantic reorganization, classify the component reorganization results, assign engineering priority marks, and generate the component topology structure classified and marked with engineering priority.

[0027] Implement frequency domain compression processing on the component topology structure with engineering priority marks, and perform space-time binding with the construction schedule, and output the lightweight component set with dynamic weight marks; Specifically, the geometric attribute data of the component topology with the engineering priority marker is subjected to a discrete Fourier transform, frequency domain coefficients are extracted, the frequency domain coefficients are divided according to the time nodes of the construction schedule plan, the progress time sequence of the construction schedule plan is mapped to the frequency domain space, the energy distribution of the frequency domain coefficients is redistributed according to the mapping result, wavelet threshold denoising and information entropy weighting operations are performed, compressed component features are obtained, the compressed component features are spatio-temporally registered with the spatial coordinates and time stamps of the construction schedule plan, dynamic weight markers are generated according to the registration result, and a lightweight component set with dynamic weight markers is output.

[0028] Based on the lightweight component set with dynamic weight markers, an engineering structure tree is generated with spatial position attributes and construction time sequence attributes as nodes.

[0029] Specifically, based on the lightweight component set with dynamic weight markers, spatial position attributes are extracted as three-dimensional coordinate points, construction time sequence attributes are extracted as time axis markers, spatial position attributes and construction time sequence attributes are organized to form two-dimensional nodes, the parent-child hierarchical relationship between the two-dimensional nodes and the lightweight component set with dynamic weight markers is associated, and an engineering structure tree is generated.

[0030] All nodes of the engineering structure tree are traversed, the spatial position attributes and construction time sequence attributes of each node are extracted, and the quantities and location range identifiers of the contract list are read; Specifically, the engineering structure tree is traversed, the spatial position attributes are extracted node by node, the construction time sequence attributes are read, the contract list items are searched, the quantities and location range identifiers are obtained, the spatial position attributes of each node and the location range identifiers of the contract list are merged, and data records integrating the spatial position attributes, construction time sequence attributes, quantities and location range identifiers of all nodes are output.

[0031] The spatial position attributes of the engineering structure tree are matched and associated with the location range identifiers of the contract list, and the construction time sequence attributes are fused to generate matching weight values; Specifically, spatial range comparison operations of the spatial position attributes of the engineering structure tree and the location range identifiers of the contract list are performed, it is determined whether the spatial position attributes are in the contract list location range identifier coverage area to generate a spatial association state, the spatial association state is fused with the construction time sequence attributes to produce a time sequence deviation, the time sequence deviation is input into an interpolation normalization method, a normalization coefficient is generated through numerical conversion, and the normalization coefficient is input into an exponential decay operation rule to generate a matching weight value combined with the influence weight of the spatial association state.

[0032] The quantities of the corresponding contract list items are corrected according to the matching weight values, engineering quantity correction records with construction time sequence influence factors are generated, and are integrated into a relationship mapping table.

[0033] Specifically, the matching weight value is integrated with the quantity of work in the contract list item. The matching weight value acts as an adjustment ratio factor for the quantity of work in the contract list item, and the quantity of work in the contract list item participates in the process as a benchmark quantity. Through non-linear weighting, the quantity correction result of the contract list item is generated. The quantity correction result of the contract list item is the corrected quantity. By combining the corrected quantity with the construction time sequence attribute, a construction time sequence influence factor is generated. The corrected quantity and the construction time sequence influence factor are combined to form a quantity correction record with the construction time sequence influence factor. All quantity correction records with the construction time sequence influence factor corresponding to the contract list item are summarized and integrated into a relational mapping table.

[0034] S2. Obtain ground-penetrating radar data and analyze it into karst development coefficients. Convert acceptance status data into quality risk values. Recognize handwritten annotation images through adversarial generative networks and output empirical risk coefficients.

[0035] Furthermore, the target tunnel section is located based on the spatial location attributes in the relational mapping table, and ground-penetrating radar data is collected and analyzed into karst development coefficients; Specifically, the spatial location attributes are extracted from the retrieval relation mapping table to determine the spatial coordinates of the target tunnel section. Based on the spatial coordinates, the ground-penetrating radar (GPR) scanning equipment is dispatched to collect GPR data. The GPR data is quantized through time-domain sampling to generate the original signal sequence of GPR reflected waves. The original signal sequence of GPR reflected waves is then decomposed using a wavelet multi-scale decomposition algorithm to separate the reflection characteristics of the karst interface. The reflection characteristics of the karst interface are then integralized with an envelope to output the karst anomaly energy value. The karst anomaly energy value is then input into the geological parameter normalization process for conversion to generate the karst development coefficient.

[0036] Specifically, based on the karst development coefficient, the corresponding inspection batch records are matched with the associated construction time sequence attributes, and the quality risk value is calculated using a fuzzy comprehensive evaluation algorithm. The expression is as follows: ; ; In the formula, Indicates the quality risk value. Represents pi (π). Represents the integral normalization coefficient. This represents a definite integral operation with the karst development coefficient as the upper limit. Indicates the karst development coefficient. Represents the geological attenuation term. Indicates the geological attenuation factor. Represents the integral variable. Represents the natural logarithm function. This represents the difference operator. Indicates the planned construction sequence. This represents the absolute value of the deviation from the construction schedule. Indicates the integral variable The differential, Represents the multiplication operator. This represents the natural exponential function. This represents the sensitivity coefficient of the inspection batch. This indicates that the summation is performed by iterating through all inspection batch records. This indicates the total number of inspection batch records. This represents the index of inspection batch records. Represents the risk density function, Indicates inspection batch record Average defect value Indicates inspection batch record The standard deviation of the defect value, This represents the quantifiable specific defect characteristic value in the inspection batch record. Indicates inspection batch record The actual pass rate This represents the correction constant.

[0037] It should be noted that the integral normalization coefficient is a mathematical transformation coefficient used to eliminate differences in the range of definite integral values. For example, it is fixed at the reciprocal of pi ≈ 0.318. The karst development coefficient is derived from ground-penetrating radar data through a porosity-karst intensity mapping model. For example, the measured value in limestone areas is 0.42. The geological attenuation factor is a rock layer stability attenuation parameter fitted based on historical collapse data. For example, it is 1.8 in strongly weathered shale areas. The inspection batch sensitivity coefficient is adjusted according to the risk level of the railway tunnel construction quality acceptance standard. For example, it is 0.9 in high-risk areas. The correction constant mathematical constraint condition (to avoid division by zero error) is an engineering approximation. For example, it is fixed at 0.001.

[0038] Using quality risk value as a weighting benchmark, the construction drawing area associated with the node in the project structure tree is located, the handwritten annotation image is extracted, and the semantic content is parsed through adversarial generative network to identify the annotation semantics; Specifically, using quality risk value as a weighting benchmark, nodes in the engineering structure tree whose quality risk value exceeds a preset risk threshold are removed. The coordinate boundaries of the associated construction drawing area are determined based on the node's spatial location attribute offset range. Handwritten annotation images are extracted from the image region within the construction drawing area coordinate boundaries. These handwritten annotation images are input into the generative network of the generative adversarial network to generate an initial text sequence. The discriminator network evaluates the authenticity probability of the initial text sequence. When the authenticity probability is lower than a preset probability threshold, the generative network performs convolutional kernel optimization operations on the feature extraction unit, outputting an optimized text sequence. The discriminator network then verifies the authenticity probability of the optimized text sequence again. This iterative loop continues until the authenticity probability reaches the preset probability threshold, outputting the final text sequence as the semantic recognition of the annotations.

[0039] It should be noted that the preset risk threshold is determined based on statistical analysis of historical landslide accident data, with an example value of 0.5; the preset probability threshold is based on the industry acceptance standard for the semantic parsing accuracy of engineering text, with an example value of 0.95.

[0040] The pre-training process of the adversarial generative network (GAN) involves organizing 100,000 handwritten annotation image samples from railway engineering projects, covering scenarios such as crack markings, seepage markers, and support deformation records. The text content is annotated according to railway construction log recording specifications. The generator is configured using a U-Net architecture, and the discriminator contains a seven-layer convolutional network and two fully connected layers to output true / false results. After freezing the generator, the discriminator is trained for 50 cycles to optimize its true / false recognition capabilities. After switching to freezing the discriminator, the generator is trained for 30 cycles to optimize text generation quality. This process is iterated for 100 rounds until the discriminator accuracy stabilizes within the 50% range. A channel attention mechanism is incorporated, and the performance of crack width recognition accuracy and seepage description accuracy is verified on a railway tunnel engineering test set, ultimately outputting the adversarial generative network.

[0041] The semantic input of the recognition annotation is quantitatively mapped to the preset engineering experience risk knowledge base to generate experience risk coefficients.

[0042] It should be noted that the pre-set engineering experience risk knowledge base refers to a structured database established based on historical railway engineering accident reports, experience manuals and industry standards, which includes key words (such as "excessive seepage rate" and "support cracks") and corresponding risk quantification values. The preset risk quantification value refers to the predefined numerical risk weight in the preset engineering experience risk knowledge base, with a range of [0,1], and is calibrated according to the accident cause analysis; Example: Identify the annotation semantics containing the keyword "exceeding the limit of seepage rate" and match the knowledge base entry annotation with a risk value of 0.3 and the keyword "support crack" and match with a risk value of 0.6. The weighted aggregation generates an empirical risk coefficient of 0.45.

[0043] Specifically, the semantic recognition annotations are segmented into words to extract keywords. Preset keywords from the engineering experience risk knowledge base are retrieved and matched with the keywords. Preset risk quantification values ​​corresponding to the matched keywords are obtained. The preset risk quantification values ​​of all matched feature keywords are weighted and aggregated to generate an experience risk coefficient.

[0044] S3. Construct a geological risk knowledge graph, using karst development coefficient, quality risk value and empirical risk coefficient as nodes, and integrate them to generate a quantitative risk coefficient.

[0045] Furthermore, we define geological risk nodes, quality risk nodes, and experience risk nodes, establish spatial correlations, temporal evolution, and causal influence relationships, and generate a basic framework for a geological risk knowledge graph. Specifically, the system defines geological risk nodes containing spatial coordinate data of karst development coefficients, quality risk nodes containing time series records of quality risk values, and empirical risk nodes containing classification labels of empirical risk coefficients. It connects geological risk nodes and quality risk nodes to create spatial correlation attributes to express location coupling relationships, connects quality risk nodes and empirical risk nodes to establish temporal evolution attribute records of risk evolution paths, and connects geological risk nodes and empirical risk nodes to form causal impact attribute identifiers of risk transmission logic. It integrates spatial correlation attributes, temporal evolution attributes, and causal impact attributes to generate the basic framework of a geological risk knowledge graph.

[0046] Based on spatial location attributes, the karst development coefficient is bound to the coordinates of geological risk nodes, and the quality risk value is associated as a node chain according to the construction time sequence attribute. At the same time, the empirical risk coefficient is injected into the empirical risk node to form a geological risk knowledge graph. Specifically, the karst development coefficient is written into the coordinate field of the geological risk node, the quality risk nodes are linked together to form a node chain through the construction time sequence attribute, the empirical risk coefficient is injected into the coefficient storage area of ​​the empirical risk node, the three entities of geological risk node, quality risk node chain and empirical risk node are combined, spatial correlation attribute is added to connect geological risk node and quality risk node chain, temporal evolution attribute is added to link quality risk node chain and empirical risk node, and causal influence attribute is attached to link geological risk node and empirical risk node to form a geological risk knowledge graph; Specifically, the risk interaction matrix is ​​extracted from the geological risk knowledge graph, and a three-dimensional orthogonal convolution coupling operation is performed to output the updated geological risk knowledge graph, expressed as: ; ; ; In the formula, This represents the updated geological risk knowledge graph. This represents the spatial index of the x-axis of the graph. This represents the spatial index of the y-axis of the graph. This represents the spatial index of the z-axis of the graph. Represents the risk domain of the risk interaction matrix. The triple integral operation, This represents the risk domain of the risk interaction matrix. This represents the risk volatility decay function. Indicates the time lag of risk propagation. The standard deviation of the risk propagation time lag variable, This represents the risk decay benchmark rate. Represents the risk interaction matrix. Represents the index of the integral variable along the x-axis of the matrix. This represents the index of the integral variable along the y-axis of the matrix. Indicates the z-axis integral variable index of the matrix. Indicates the characteristic scale of karst potential field. Represents the characteristic scale of karst potential field The partial derivative operation, Represents the Laplace operator. Represents the karst potential field function. Represents the baseline constant of the karst potential field. Indicates the correlation factor between karst porosity and permeability. This represents the resistance coefficient to compressive deformation of rock strata. Indicates the spatial scale attenuation index of karst. Indicates the geological permeability adjustment index. Represents the gradient operator. Indicates the gradient magnitude. The vector modulus operator is used to represent the vector length. This indicates a dynamic assignment operation. Represents the feature scale value, Represents a three-dimensional integral volume element. This represents the risk tensor canonical transformation. Indicates the transform function identifier. This represents the tensor convolution operator. Indicates rotation-invariant convolution kernel, Represents the basic convolution kernel tensor. Indicates rotation invariance identifier, It represents the rotation angle in three-dimensional space.

[0047] It should be noted that the karst porosity-permeability correlation factor is determined by the ratio of the average cross-sectional area of ​​the pore throat to the volume of the rock mass unit through CT scanning of borehole core samples, with an example value of 0.08; the rock layer compressibility deformation resistance coefficient is obtained by converting the rock mass elastic modulus and Poisson's ratio from the geological exploration report, with an example value of 1.7; and the karst spatial scale attenuation index is obtained by fractal dimension analysis of the karst cave size distribution from the three-dimensional scanning data of ground-penetrating radar, with an example value of 0.02.

[0048] Specifically, multi-scale wavelet packet decomposition is performed on the updated geological risk knowledge graph to extract frequency band energy features. An entropy weight fusion algorithm is then used to generate a quantitative risk coefficient, expressed as: ; ; In the formula, This indicates the measurement risk coefficient. This indicates that the summation is performed by iterating through the wavelet packet decomposition scale levels. This indicates the number of wavelet packet decomposition scale levels. This represents the wavelet packet decomposition scale level index. Indicates the first Layer wavelet packet decomposition scale entropy weights This represents the extraction of the real part of the complex wavelet coefficients. Indicates the first Wavelet coefficient matrix at the layer wavelet packet decomposition scale. Represents the wavelet coefficient matrix. It represents the Hadamah accumulation. Represents the frequency band density function. Indicates the first The bandwidth of the layer wavelet packet decomposition scale. Indicates the first The center frequency of the layer wavelet packet decomposition scale. This represents the square of the standard deviation of the frequency band energy. Indicates the average energy of the frequency band. Represents the frequency domain decay index. Represents risk perception entropy. Represents the tensor direct sum operation. Represents the spatiotemporal correlation gain coefficient. Represents the inverse hyperbolic tangent function. Denotes the Frobenius norm of the Jacobian interaction matrix. Represents the Jacobian interaction matrix. This represents the Frobenius norm operator.

[0049] S4. Call the relationship mapping table, apply the measurement risk coefficient to the corresponding contract list quantity, and calculate the current period's work measurement value.

[0050] Furthermore, the quantity correction records with construction time sequence influencing factors in the relational mapping table are called, and a four-dimensional tensor coupling matrix is ​​constructed by combining the measurement risk coefficient. Specifically, the relational mapping table is read, and the construction time sequence value, quantity correction value, and spatial coordinate data contained in the quantity correction record with construction time sequence influencing factors are extracted. Combined with the measurement risk coefficient, the three-dimensional components of the spatial coordinate data are combined as the first three dimensions of the four-dimensional tensor coupling matrix, the construction time sequence value is spliced ​​as the fourth dimension, the quantity correction value is integrated and written into the four-dimensional coordinate storage area, the measurement risk coefficient is embedded into the coupling tensor feature channel, and a four-dimensional tensor coupling matrix is ​​generated.

[0051] By performing a non-Abelian gauge field transformation on the four-dimensional tensor coupling matrix using the Dirac matrix operator, a gauge field action functional under the current spatiotemporal boundary constraints is generated. Specifically, a non-Abelian gauge field transformation is performed by applying the Dirac matrix operator to the four-dimensional tensor coupling matrix, mapping the engineering quantity correction value and the measurement risk coefficient to generate gauge connection field components and form gauge potential data; the gamma matrix multiplication operation of the Dirac matrix operator is applied to fuse the difference operation and commutation operation of the spatiotemporal coordinate derivative to output the gauge field strength tensor curvature data, which is then substituted into the coordinates of the engineering space range defined by the current spatiotemporal boundary constraints and the start and end time marks of the contract period to output the gauge field action quantity functional.

[0052] Specifically, the functional of the gauge field action is input into the conformal deformation mapping engine, and Riemann surface parametric integration is performed along the engineering partition boundary to output the current construction measurement value, expressed as: ; In the formula, This represents the current period's work measurement value. Denotes the normalization constant of the conformal integral. Denotes the Riemann surface parametric integral along the boundary of the engineering partition. The Riemann surface parameters representing the boundaries of the engineering partitions. Represents the conformal deformation kernel function. The curvature density represents the conformal mapping. This represents a second-order partial differential. Represents the first-order partial differential of the complex coordinate variables of the Riemann surface. The first partial differential of the derivative of the complex coordinate variables of a Riemann surface is given by... Represents the complex phase rotation factor. Represents the imaginary unit. Indicates the phase angle for measuring risk. Represents the Riemannian surface element. Represents the determinant of the gauge field-degree tensor. Represents the frequency domain integral of the gauge field. The Fourier spectrum of the functional of gauge field action is represented. This represents the frequency power-law decay term. This represents a frequency micro-element.

[0053] It should be noted that the conformal integral normalization constant is a mathematical transformation coefficient used to eliminate the influence of differences in the size of different engineering zones on the integration results, and the example value is 0.318; the complex phase rotation factor is the phase offset derived from the spatial gradient direction of the measurement risk coefficient, and the example value is 0.785.

[0054] S5. Compare the cumulative value of the current period's work measurement with the total amount in the contract list. If the cumulative value does not exceed the total amount in the contract list, enter it into the approval process; otherwise, freeze the approval process and generate a high-risk marker.

[0055] Furthermore, historical cumulative values ​​are retrieved and corrected using the time-dependent coefficient of karst geology to generate a correction result. This correction result is then overlaid with the current period's construction measurement value to generate a cumulative value of the current period's construction measurement value. Specifically, the stored cumulative sum of past inspection measurement values ​​is called as the historical cumulative value. The historical cumulative value is input into the time-effect characteristics of karst geology and the attenuation correction amount is output. The historical cumulative value is combined with the attenuation correction amount to perform an accumulation operation to generate the historical cumulative value correction result. The historical cumulative value correction result is arithmetically superimposed with the current inspection measurement value generated in the current implementation cycle to generate the cumulative value of the current inspection measurement value.

[0056] The cumulative value of the current period's work inspection measurement is compared with the total amount of the contract list. If the cumulative value of the current period's work inspection measurement does not exceed the total amount of the contract list, the multi-level approval and signing process is automatically initiated; otherwise, the approval process is frozen, and a high-risk marker is generated in the associated spatial coordinates of the BIM model.

[0057] Specifically, the cumulative value of the current period's work inspection measurement is compared with the total amount of the contract list. If the cumulative value of the current period's work inspection measurement does not exceed the total amount of the contract list, a multi-level approval and signing process is initiated. If the cumulative value of the current period's work inspection measurement exceeds the total amount of the contract list, the approval process execution thread is frozen. After freezing the approval process execution thread, the spatial coordinates corresponding to the current period's work inspection measurement are located and written into high-risk marker data. The high-risk marker data is then injected into the associated spatial coordinate field of the BIM model to generate high-risk markers.

[0058] For example, if the cumulative value of the current period's work measurement is 9,000 cubic meters and less than the total contract list of 10,000 cubic meters, a multi-level approval and signing process will be initiated; if the cumulative value of the current period's work measurement is 10,500 cubic meters and exceeds the total contract list, the approval process will be frozen and a red high-risk mark will be placed at the BIM model coordinates of 102.3 degrees east longitude, 28.6 degrees north latitude, and 300 meters elevation.

[0059] S6. Based on the high-risk markers, mark the high-risk sections corresponding to the ground-penetrating radar data, associate the measurement risk coefficient with the spatial coordinates of the high-risk sections and store them in the blockchain to generate a work acceptance and pricing sheet.

[0060] Furthermore, based on the spatial coordinates of high-risk markers, intelligent segmentation processing of ground-penetrating radar data is triggered to generate a set of spatial coordinates for high-risk sections; Specifically, based on the spatial coordinates of high-risk markers, the ground-penetrating radar scanning equipment is used to collect ground-penetrating radar data. The ground-penetrating radar data is then input into the karst anomaly segmentation model to perform intelligent segmentation processing, generating a mask image of the high-risk section. The coordinates of the high-risk pixel set in the high-risk section mask image are extracted, and the coordinates of the high-risk pixel set are converted to geospatial coordinates to generate a spatial coordinate set of the high-risk section.

[0061] It should be noted that the pre-training process of the karst anomaly segmentation model is as follows: organize 100,000 ground-penetrating radar grayscale image samples to cover the complete karst anomaly morphology spectrum, label high-risk sections with mask labels, configure the U-Net architecture, initialize the karst anomaly segmentation model parameters, perform the first stage of training, freeze the upsampling layer and optimize only the downsampling feature extractor with a 50-cycle learning rate, switch to the second stage of training, freeze the downsampling layer and optimize the upsampling segmentation head with a 30-cycle learning rate, alternately iterate to the validation set, and output the karst anomaly segmentation model.

[0062] Based on the spatial coordinate set of high-risk areas, a spatiotemporal risk vector is constructed by integrating the measurement risk coefficient, and a spatiotemporal digital fingerprint is generated through blockchain notarization. Specifically, a spatiotemporal risk vector is generated by fusing the spatial coordinate set of high-risk areas with the measured risk coefficient. The spatial coordinate components and the measured risk coefficient are combined to form multidimensional vector data. The spatiotemporal risk vector is written into the blockchain notarization interface to execute the smart contract notarization operation, triggering the blockchain consensus mechanism to verify the notarized data and generate a transaction hash value. The transaction hash value is then extracted as a spatiotemporal digital fingerprint.

[0063] Dynamic risk weights are extracted using spatiotemporal digital fingerprints, and risk correction measurement is performed on the quantities of related contract list works to generate correction measurement results. Specifically, the spatiotemporal risk vector associated with the spatiotemporal digital fingerprint is retrieved to extract the measurement risk coefficient as the dynamic risk weight. The relationship mapping table is called to obtain the quantities of the related contract list. The dynamic risk weight is multiplied by the quantities of the contract list to perform risk correction measurement calculation, and the calculation result is output as the correction measurement result.

[0064] For example, the spatiotemporal digital fingerprint a1b2c3d4 corresponds to a measurement risk coefficient of 0.85 as a dynamic risk weight. The quantity of the project in the contract list is 100 cubic meters. After correction measurement, the result is 85 cubic meters.

[0065] Based on the corrected measurement results and spatiotemporal digital fingerprints, an inspection and valuation sheet with blockchain verification marks and risk coordinates is generated.

[0066] Specifically, the basic form of the work verification and pricing form is generated by combining and correcting the measurement result values, spatiotemporal digital fingerprints, and spatial coordinate sets of high-risk areas. A blockchain verification mark field is added to write the blockchain transaction hash value corresponding to the spatiotemporal digital fingerprint. An additional risk coordinate field is added to store the geographic information of the spatial coordinate set of high-risk areas, and the work verification and pricing form is output.

[0067] This embodiment also provides a computer device applicable to the method of constructing a railway engineering acceptance and pricing management system based on BIM, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the method of constructing a railway engineering acceptance and pricing management system based on BIM as proposed in the above embodiment.

[0068] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0069] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the method for constructing a BIM-based railway engineering inspection and pricing management system as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0070] In summary, this invention achieves deep fusion and real-time decision response of multi-source heterogeneous data by constructing a geological risk knowledge graph, overcoming the limitations of traditional linear weighted fusion. The topological structure of the geological risk knowledge graph captures the nonlinear correlation between geological, quality inspection, and experience data; the iterative update mechanism of the GCN layer enables the measurement risk coefficient to have dynamic response capabilities, supporting accurate correction of measurement values ​​and blockchain-based evidence storage and auditing.

[0071] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for constructing a BIM-based railway engineering inspection and pricing management system, characterized in that: The application relates to a tunnel quality inspection method based on a geological risk knowledge graph. The BIM model of a railway engineering is decomposed to generate an engineering structure tree, the structural units of the engineering structure tree are bound to a contract list, and a relationship mapping table is generated; Geological radar data is acquired and analyzed to obtain a karst development coefficient, acceptance state data is converted into a quality risk value, handwritten annotation images are identified through a generative adversarial network, and an experience risk coefficient is output; A geological risk knowledge graph is constructed, the karst development coefficient, the quality risk value and the experience risk coefficient are taken as nodes, and a measurement risk coefficient is generated by fusion; The relationship mapping table is called, the measurement risk coefficient is applied to the corresponding contract list quantity of the engineering, and a current inspection measurement value is calculated; The cumulative value of the current inspection measurement value is compared with the total quantity of the contract list, if the cumulative value does not exceed the total quantity of the contract list, the approval process is input, otherwise the approval process is frozen and a high-risk mark is generated; According to the high-risk mark, a high-risk section corresponding to the geological radar data is marked, the measurement risk coefficient is associated with the spatial coordinates of the high-risk section and stored in a blockchain, and an inspection measurement bill is generated.

2. The method for constructing a BIM-based railway engineering acceptance pricing management system according to claim 1, characterized in that: The engineering structure tree is generated, and the specific steps are as follows, The component geometric properties and dynamic load characteristics in the railway engineering BIM model are read, component semantic reorganization is performed, and a component topology structure with classification marking engineering priority is output; The component topology structure with the engineering priority mark is subjected to frequency domain compression processing, and is spatiotemporally bound to a construction schedule, and a lightweight component set with a dynamic weight mark is output; Based on the lightweight component set with the dynamic weight mark, the spatial position attribute and the construction time sequence attribute are taken as nodes, and the engineering structure tree is generated.

3. The method for constructing a BIM-based railway engineering acceptance pricing management system according to claim 2, characterized in that: The relationship mapping table is generated, and the specific steps are as follows, All nodes of the engineering structure tree are traversed, the spatial position attribute and the construction time sequence attribute of each node are extracted, and the quantities and position range marks of the contract list are read; The spatial position attribute of the engineering structure tree is matched and associated with the position range mark of the contract list, and the construction time sequence attribute is fused to generate a matching weight value; According to the matching weight value, the quantity of the corresponding contract list item is corrected, an engineering quantity correction record with a construction time sequence influence factor is generated, and the record is integrated into the relationship mapping table.

4. The method for constructing a BIM-based railway engineering acceptance pricing management system according to claim 3, characterized in that: The experience risk coefficient is output, and the specific steps are as follows, According to the spatial position attribute in the relationship mapping table, a tunnel target section is located, geological radar data is collected and analyzed to obtain a karst development coefficient; Based on the karst development coefficient, the corresponding inspection batch record is matched by associating the construction time sequence attribute, and a quality risk value is calculated by using a fuzzy comprehensive evaluation algorithm; Taking the quality risk value as a weight reference, the construction drawing region associated with the node in the engineering structure tree is located, a handwritten annotation image is intercepted, and the semantic content is analyzed through a generative adversarial network to identify the annotation semantics; The identified annotation semantics are input into a preset engineering experience risk knowledge base for quantitative mapping to generate an experience risk coefficient.

5. The method for constructing a BIM-based railway engineering acceptance pricing management system according to claim 4, characterized in that: The measurement risk coefficient is generated by fusion, and the specific steps are as follows, Geological risk nodes, quality risk nodes and experience risk nodes are defined, relationship attributes of spatial association, time sequence evolution and causal influence are established, and a basic framework of a geological risk knowledge graph is generated; According to the spatial position attribute, the karst development coefficient is bound to the geological risk node coordinates, the quality risk value is associated as a node chain according to the construction time sequence attribute, and the experience risk coefficient is injected into the experience risk node to form a geological risk knowledge graph; The risk interaction matrix in the geological risk knowledge graph is extracted, and three-dimensional orthogonal convolution coupling operation is implemented to output the updated geological risk knowledge graph; The updated geological risk knowledge graph is subjected to multi-scale wavelet packet decomposition, the frequency band energy features are extracted, and the measurement risk coefficient is generated through an entropy weight fusion algorithm.

6. The method for constructing a BIM-based railway engineering acceptance pricing management system according to claim 5, characterized in that: The current measurement value of the construction is calculated, and the specific steps are as follows, The engineering quantity correction record with the construction time sequence influence factor in the relationship mapping table is called, and the measurement risk coefficient is combined to construct a four-dimensional tensor coupling matrix; The four-dimensional tensor coupling matrix is subjected to non-Abelian gauge field transformation through Dirac matrix operators to generate a gauge field action functional under the constraint of the current space-time boundary; The gauge field action functional is input into the conformal deformation mapping engine, and the Riemann surface parameter integral is performed along the engineering partition boundary to output the current measurement value of the construction.

7. The method for constructing a BIM-based railway engineering acceptance pricing management system according to claim 6, characterized in that: The high-risk label is generated, and the specific steps are as follows, The historical cumulative value is called, and the correction result is generated by correcting the time coefficient of the karst area geology, and the correction result is superimposed with the current measurement value of the construction to generate the cumulative value of the current measurement value of the construction; The cumulative value of the current measurement value of the construction is compared with the total quantity of the contract list, if the cumulative value of the current measurement value of the construction does not exceed the total quantity of the contract list, the multi-level approval and signature process is automatically started, otherwise the approval process is frozen, and the high-risk label is generated in the associated spatial coordinates of the BIM model.

8. The method for constructing a BIM-based railway engineering acceptance pricing management system according to claim 7, characterized in that: The construction valuation sheet is generated, and the specific steps are as follows, According to the spatial coordinates of the high-risk label, the intelligent segmentation processing of the geological radar data is triggered to generate the spatial coordinate set of the high-risk section; Based on the spatial coordinate set of the high-risk section, the time-space risk vector is constructed by fusing the measurement risk coefficient, and the time-space digital fingerprint is generated by block chain notarization; The dynamic risk weight is extracted by using the time-space digital fingerprint, the risk correction measurement of the associated contract list engineering quantity is performed, and the correction measurement result is generated; According to the correction measurement result and the time-space digital fingerprint, the construction valuation sheet with the block chain verification label and the risk coordinates is generated. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the method for constructing the BIM-based railway engineering construction valuation management system according to any one of claims 1-8.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the method for constructing the BIM-based railway engineering construction valuation management system according to any one of claims 1-8.

Images