Cable engineering intelligent cost method based on big data

Abstract

The invention relates to the technical field of cable engineering cost, in particular to a cable engineering intelligent cost method based on big data, which comprises the following steps of: S1, performing key data extraction on a cable engineering information extraction table by adopting an OCR (Optical Character Recognition) technology, forming design information extraction data, performing sequence acquisition and storage, and constructing a project basic database; and S2, based on the design information providing data, when the method is used, an information providing-amount calculation-pricing intelligent integrated intelligent pricing model is formed based on the civil engineering intelligent amount calculation model and the intelligent pricing model, communication and efficient utilization of cable line design and technical-economic data are realized, intelligent and precise improvement and transformation of cable line cost are realized, and the method is suitable for popularization and application. The method is beneficial to comprehensively improving the construction cost work quality and efficiency, facilitates the improvement of the automation and intelligent decision-making level of the construction cost management through the design of the whole-process intelligent integration of information extraction, the engineering calculation amount and the construction cost pricing, facilitates the reduction of the problems of manual checking and data splitting, and reduces the construction cost deviation and the project cost risk.

Description

Technical Field

[0001] This invention relates to the field of cable engineering cost estimation technology, specifically to an intelligent cost estimation method for cable engineering based on big data. Background Technology

[0002] Cable project cost refers to the total cost incurred throughout the entire process of cable project planning, design, procurement, construction to final acceptance. It is an important component of project management, spanning all stages of the project and playing a crucial role in project decision-making, implementation, and investment control.

[0003] Patent publication number CN112766914A describes in its specification a "method and cloud computing evaluation platform for power transmission and transformation engineering cost assessment based on big data." The method obtains the total length of laid cables, the slope angle of each tower foundation, tower images, tower volume, and soil moisture in each detection sub-area. Based on the tower images and soil moisture in each detection sub-area, it determines the type of each tower and the soil moisture level in each detection sub-area, and obtains the cost per square meter of cable and the standard cost of the power transmission and transformation project. This allows for the statistical calculation and display of the assessed cost of the power transmission and transformation project, demonstrating high intelligence and... The high reliability of the technology reduces labor costs, greatly improves the efficiency of evaluating the cost of power transmission and transformation projects, and enhances the economic and social benefits of investment. Although the above technology achieves an intelligent closed loop for evaluating the cost of power transmission and transformation projects through a three-layer architecture of data acquisition, dynamic modeling, and cost control, and has the advantages of engineering practicality and technological foresight, there is a lack of logical and computational connection between cable line design and technical and economic data in the process of calculating cable project costs. This leads to inefficient use of cable line design and technical and economic data, and is not conducive to the improvement and transformation of intelligent and accurate cable line cost estimation.

[0004] In conclusion, developing a big data-based intelligent cost estimation method for cable engineering remains a critical issue that urgently needs to be addressed in the field of cable engineering cost estimation technology. Summary of the Invention

[0005] The purpose of this invention is to solve the problem in the existing technology that there is a lack of logical and computational connection between cable line design and technical and economic data in the process of calculating cable engineering costs. This leads to the inefficient use of cable line design and technical and economic data, and is not conducive to the improvement and transformation of intelligent and accurate cable line cost estimation.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a smart cost estimation method for cable engineering based on big data, comprising the following steps: S1. Use OCR technology to extract key data from the cable engineering data submission form, form design data submission data, collect and store it sequentially, and build a basic database for the project. S2. Based on the design data, use big data processing technology to quantify the rules and generate the engineering quantity statistics logic and pricing rule mapping relationship; S3. Based on the aforementioned engineering quantity statistics logic, construct a civil engineering intelligent quantity calculation model; based on the aforementioned pricing rule mapping relationship, construct an intelligent pricing model. S4. Based on the aforementioned intelligent quantity calculation model and intelligent pricing model for civil engineering, an intelligent pricing model integrating cost estimation, quantity calculation, and pricing is formed. S5. Based on the intelligent pricing model, perform intelligent quantity calculation and pricing in both quota mode and list mode during the preliminary budget stage and the construction drawing budget stage.

[0007] Furthermore, in step S1, the method for extracting key data from the cable engineering data submission form using OCR technology, forming design data submission data, and sequentially collecting and storing it to construct the project's basic database is as follows: The method employs OCR technology to extract key data from the cable engineering data submission form. For the unstructured text sequence extracted as key data, a mapping function is used to convert it into structured data, expressed as: In the formula, This represents the final generated structured data. Represents a mapping function. It is an unstructured text sequence extracted by OCR. It indicates that it will start from the first One to the first The key data fields are merged into a single set. This represents the field name in the corresponding structured data, and the specific content of the corresponding field. It represents the total number of key-value pairs. Indicates the first Extracting values ​​from key data fields, Indicates the first One to the first Summing each character Indicates the first The first key data field The weight coefficient of each character, This is an indicator function that evaluates to 1 if the condition within the parentheses is true, and 0 otherwise. It is a text sequence recognized by OCR. The first in One character, This represents the total number of characters in the text sequence recognized by OCR. Indicates the first The key data fields correspond to character sets. The extracted key data are classified by project to form design data and collected and stored sequentially to build a project basic database. The key data includes, but is not limited to, project name, ductwork, cable trench, trenchless, cable-related data and key parameters of components of each line segment.

[0008] Furthermore, in step S2, the method for quantifying rules and generating the engineering quantity statistical logic and pricing rule mapping relationship based on the design data and big data processing technology is as follows: The method of using big data processing technology for rule quantification includes, but is not limited to, statistical rules for component parameters, steel reinforcement quantity, concrete quantity, cable laying quotas, pipe laying and pouring quotas, and surplus soil transportation. The expression for the cable laying quota is as follows: In the formula, The calculation result represents the cable laying quota. The basic labor quota benchmark for cable laying per unit length and unit cross-sectional area is an empirical coefficient obtained by fitting historical engineering data and industry standards. Represents the length of the cable. Represents the cross-sectional area of ​​the cable. It is the cross-sectional area correction factor. It is the number of construction difficulty factors. It is the first The weighting coefficients of each difficulty factor. It is the first The possible values ​​of each difficulty factor It is the overall difficulty adjustment coefficient.

[0009] Furthermore, in step S2, the method for quantifying rules and generating the engineering quantity statistical logic and pricing rule mapping relationship based on the design data and big data processing technology is as follows: The statistical rules are intelligently linked to the corresponding project segments. By defining project feature vectors, a similarity matrix between the rules and the projects is calculated, and then based on the similarity matrix... The Hungarian algorithm is used to perform the optimal allocation of the statistical rules, expressed as: In the formula, It is a double-layer summation symbol that represents the cumulative calculation of all combinations of items and rules. Indicates the first The project and the first The similarity value of the rules, This indicates that binary decision variables can only take either 0 or 1. Time represents the first The project allocation number Rule 1, in Time represents no allocation, Indicates the total number of projects. Indicates the total number of rules. It is the first constraint. The second constraint is to generate a mapping relationship between the project's quantity statistics logic and pricing rules.

[0010] Furthermore, in step S3, based on the engineering quantity statistics logic, a civil engineering intelligent quantity calculation model is constructed; based on the pricing rule mapping relationship, a smart pricing model is constructed as follows: Based on the aforementioned engineering quantity statistics logic, including but not limited to component parameters of ductwork, cable trenches, and manholes, a civil engineering intelligent quantity calculation model is constructed. This model standardizes the quantity calculations for, but is not limited to, cable ductwork, cable trenches, manholes, and auxiliary modules. Feature parameters are extracted, and the relationships between the parameters of each component are clarified. A multilayer perceptron model is then constructed to predict the engineering quantity. The expression is: In the formula, This indicates the predicted workload output. The mapping function representing the multilayer perceptron model. It is an activation function. These represent the weight matrices of the third, second, and first layers of a multilayer perceptron, respectively. It is the ReLU activation function. This represents the input engineering feature vector. These represent the bias vectors of the first, second, and third layers in the multilayer perceptron, respectively. Meanwhile, the multilayer perceptron model is trained using mean squared error combined with L1 regularization.

[0011] Furthermore, in step S3, based on the engineering quantity statistics logic, a civil engineering intelligent quantity calculation model is constructed; based on the pricing rule mapping relationship, a smart pricing model is constructed as follows: Based on the pricing rule mapping relationship, including but not limited to material parameters of cables and terminals, the intelligent pricing model is constructed. Combining the business logic of cable installation professional materials and quotas with the design material list format, parameterized intelligent pricing for the installation portion is performed. This parameterized intelligent pricing includes, but is not limited to, calculation of cable main material costs and cable laying labor costs. The expression is: In the formula, This indicates the cost of the main cable materials. Represents the benchmark price. Based on the natural constant An exponential function with base 0. It is a definite integral. It is the lower limit of integration. It is the maximum number of points. It is an integral variable. It is a price index function. It is the cross-sectional area of ​​the cable. Indicates the cable length. The tax rate is the percentage of the cost of main materials that are taxed. Represents labor costs, It is the basic working hours. The summation symbol comes from arrive Summing the subsequent terms, For the corresponding difficulty factor The weight, Is with The corresponding difficulty factor, It is the labor cost per unit of working hours. At the same time, the GARCH(1,1) model is used to predict the volatility of main material prices, and the parameterized intelligent pricing accuracy evaluation index is constructed.

[0012] Furthermore, in step S4, the method for forming an intelligent pricing model integrating cost estimation, quantity calculation, and pricing based on the aforementioned intelligent quantity calculation model and intelligent pricing model is as follows: Based on the outputs of the aforementioned intelligent quantity calculation model and intelligent pricing model for civil engineering, a tensor decomposition model is constructed to fuse multi-source data and generate nonlinear combination features, expressed as: In the formula, To merge tensors, The engineering quantity data output by the intelligent quantity calculation model for civil engineering are respectively Cost data output by the intelligent pricing model and design data The factor vector, Indicates to arrive Summing the combination of factor vectors Indicates the outer product. To decompose the rank, It is a generated nonlinear combination feature. It is the Sigmoid activation function. This is the weight matrix. express The first in One characteristic, express The first in One characteristic, express Zhongyu The index corresponds to certain characteristics. The study examines the logical relationship between architectural and installation professional data, quotas, main materials, and equipment, and defines entity sets. This includes, but is not limited to, quota items, main materials, equipment, and cost parameters, defining relation sets. This includes, but is not limited to, relationships of inclusion, association, and influence, and entities and relationships are represented based on the TransE model.

[0013] Furthermore, in step S4, the method for forming an intelligent pricing model integrating cost estimation, quantity calculation, and pricing based on the aforementioned intelligent quantity calculation model and intelligent pricing model is as follows: Combining the aforementioned design data and material list standard format, a built-in statistical rule base is implemented. This base includes a basic rule layer, a relational rule layer, and an optimization rule layer. The basic rule layer contains rules for quantity calculation, quota application, and cost calculation. The relational rule layer defines rules for relationships between entities. The optimization rule layer includes price adjustment rules and risk assessment rules. Production rule representation is used for rule formalization, and evidence fusion based on Dempster-Shafer theory is employed to resolve rule conflicts, forming an intelligent pricing model that integrates data provision, quantity calculation, and pricing. The expression is: In the formula, It is a multi-objective optimization loss function. To regress the loss, For classifying losses, For graph structure loss, Loss weighting coefficient.

[0014] Furthermore, in step S5, based on the intelligent pricing model, the method for intelligent quantity calculation and pricing under the quota mode and the list mode in the preliminary estimate stage and the construction drawing budget stage is as follows: Based on feedback data from the actual application of the intelligent pricing model and historical project data, the key cost models for each project are analyzed according to the technical and economic indicator system. The Pearson correlation coefficient method is used to calculate the main influencing factors and parameters of the key cost models. The expression is: In the formula, The Pearson correlation coefficient indicates that it measures the characteristics of a feature. With key cost model The range of linear correlation values Time is considered a strong correlation. express and covariance, express and standard deviation Indicates the first The first feature Each sample value Indicates the first The first key cost model Each sample value Represents the total number of samples. Indicates the first The sample mean of each feature, Indicates the first The sample mean of each key cost model.

[0015] Furthermore, in step S5, based on the intelligent pricing model, the method for intelligent quantity calculation and pricing under the quota mode and the list mode in the preliminary estimate stage and the construction drawing budget stage is as follows: Based on the aforementioned main influencing factors and parameters, an ELM mathematical prediction model is constructed. The prediction error is analyzed using a multiple linear regression model, and the coefficients of each key cost model are adjusted accordingly. The expression is: In the formula, The prediction error represents the true value. Compared with model predictions The difference, Indicates the first The true cost value for each sample Indicates the first The predicted value for each sample, This represents the fundamental error when all features are zero. Represents the feature weight coefficients. Indicates the first The feature vector of each sample To represent unexplained noise, iteratively optimize the intelligent valence model, expressed as: In the formula, The updated intelligent pricing model coefficients are used for the next round of prediction. This is the current coefficient. It's the learning rate. Represents the loss function for multi-objective optimization. With coefficient The changing slope is used to guide the direction of coefficient updates. It is a multi-objective optimization loss function. The input features of the intelligent pricing model prediction function Sum of coefficients Output the predicted cost value.

[0016] Beneficial effects Compared with known public technologies, the technical solution provided by this invention has the following beneficial effects: When in use, this invention forms an intelligent pricing model that integrates cost estimation, cost estimation, and pricing based on the aforementioned intelligent quantity calculation model and intelligent pricing model for civil engineering. This enables the efficient use of cable line design and technical and economic data, achieving intelligent and precise improvement and transformation in cable line cost estimation. This is conducive to comprehensively improving the quality and efficiency of cost estimation work. Through the intelligent integration of the entire process of design cost estimation, engineering quantity calculation, and cost estimation, it facilitates the improvement of the automation and intelligent decision-making level of engineering cost management, helps reduce manual verification and data fragmentation problems, and reduces cost deviations and project cost risks.

[0017] When in use, this invention improves the accuracy and efficiency of project pricing by mapping the statistical logic of project quantities and pricing rules, and reduces estimation errors caused by project complexity and parameter diversity. Through price fluctuation prediction and difficulty factor adjustment, it achieves risk warning and optimized control of material and labor costs, which can significantly reduce budget deviations caused by material price fluctuations or underestimation of construction complexity, and enhances the foresight and scientific nature of project cost management. Attached Figure Description

[0018] Figure 1 This is a flowchart of a smart cost estimation method for cable engineering based on big data, according to the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but includes other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] The present invention will now be described in further detail with reference to the accompanying drawings: Example: like Figure 1 As shown, this invention provides an intelligent cost estimation method for cable engineering based on big data, including the following steps: S1. Use OCR technology to extract key data from the cable engineering data submission form, form design data submission data, collect and store it sequentially, and build a basic database for the project. Furthermore, in step S1, the method for extracting key data from the cable engineering data submission form using OCR technology, forming design data submission data, and sequentially collecting and storing it to construct the project's basic database is as follows: OCR technology was used to extract key data from the cable engineering data submission form. For the unstructured text sequence of key data extraction, a mapping function was used to convert it into structured data. The expression is as follows: In the formula, This represents the final generated structured data. Represents a mapping function. It is an unstructured text sequence extracted by OCR. It indicates that it will start from the first One to the first The key data fields are merged into a single set. This represents the field name in the corresponding structured data, and the specific content of the corresponding field. It represents the total number of key-value pairs. Indicates the first Extracting values ​​from key data fields, Indicates the first One to the first Summing each character Indicates the first The first key data field The weight coefficient of each character, This is an indicator function that evaluates to 1 if the condition within the parentheses is true, and 0 otherwise. It is a text sequence recognized by OCR. The first in One character, This represents the total number of characters in the text sequence recognized by OCR. Indicates the first The key data fields correspond to character sets. The extracted key data are classified by project, forming design data and sequentially collected and stored to build the project's basic database. Key data includes, but is not limited to, project name, ductwork, cable trench, trenchless, cable-related data, and key parameters of components for each line segment.

[0022] In this embodiment, the method for extracting key OCR data and constructing a basic project database from cable engineering cost estimates can be applied to engineering fields such as power cable laying and communication pipeline construction. First, OCR technology is used to identify unstructured text data in the cost estimates form. Then, a mapping function is used to convert the extracted character sequences into standardized structured data, such as core parameters like project name, duct layout, cable trench, trenchless method, cable type and length. After being classified by project section, the data is serialized, collected, and stored. Finally, a complete basic project database is established, which helps improve the efficiency and accuracy of cable engineering cost-related work, while reducing errors and cumbersome processes associated with manual data processing.

[0023] S2. Based on the design data, use big data processing technology to quantify the rules and generate the engineering quantity statistics logic and pricing rule mapping relationship; Furthermore, in step S2, the method for quantifying rules based on design data and generating the engineering quantity statistical logic and pricing rule mapping relationship using big data processing technology is as follows: The rules are quantified using big data processing technology. This quantification includes, but is not limited to, statistical rules for component parameters, steel reinforcement quantity, concrete quantity, cable laying quotas, duct pouring quotas, and surplus soil transportation. The expression for the cable laying quota is as follows: In the formula, The calculation result represents the cable laying quota. The basic labor quota benchmark for cable laying per unit length and unit cross-sectional area is an empirical coefficient obtained by fitting historical engineering data and industry standards. Represents the length of the cable. Represents the cross-sectional area of ​​the cable. It is the cross-sectional area correction factor. It is the number of construction difficulty factors. It is the first The weighting coefficients of each difficulty factor. It is the first The possible values ​​of each difficulty factor It is the overall difficulty adjustment coefficient.

[0024] Furthermore, in step S2, the method for quantifying rules based on design data and generating the engineering quantity statistical logic and pricing rule mapping relationship using big data processing technology is as follows: The statistical rules are intelligently linked to the corresponding project segments. This is done by defining project feature vectors, calculating the similarity matrix between the rules and the projects, and then basing the results on the similarity matrix. The Hungarian algorithm is used for optimal allocation based on statistical rules. The expression is: In the formula, It is a double-layer summation symbol that represents the cumulative calculation of all combinations of items and rules. Indicates the first The project and the first The similarity value of the rules, This indicates that binary decision variables can only take either 0 or 1. Time represents the first The project allocation number Rule 1, in Time represents no allocation, Indicates the total number of projects. Indicates the total number of rules. It is the first constraint. The second constraint is to generate a mapping relationship between the project's quantity statistics logic and pricing rules.

[0025] In this embodiment, the design data for the cable laying project in the new residential area includes a cable length of 800m, a cross-sectional area of ​​120mm², and construction scenario information such as crossing roads and green belts. When using big data technology to quantify the rules, the cable laying quota calculation will refer to historical data to determine the basic labor quota benchmark, and combine various parameters to obtain the result. At the same time, the statistical rules such as component parameters and surplus soil transportation are matched with the cable laying project in the residential area. By calculating the similarity matrix and the Hungarian algorithm, the appropriate engineering quantity statistical logic and pricing rule mapping relationship are determined, which helps to improve the accuracy and efficiency of engineering pricing and reduce the estimation error caused by project complexity and parameter diversity.

[0026] S3. Based on the engineering quantity statistics logic, construct an intelligent quantity calculation model for civil engineering; based on the pricing rule mapping relationship, construct an intelligent pricing model. Furthermore, in step S3, the method for constructing an intelligent quantity calculation model for civil engineering based on the engineering quantity statistics logic and an intelligent pricing model based on the pricing rule mapping relationship is as follows: Based on the statistical logic of engineering quantities, including but not limited to component parameters of ductwork, cable trenches, and manholes, an intelligent quantity calculation model for civil engineering is constructed. This model standardizes the quantity calculations for, but is not limited to, cable ductwork, cable trenches, manholes, and ancillary modules. Feature parameters are extracted, and the relationships between the parameters of each component are clarified. A multilayer perceptron model is then constructed to predict the engineering quantities. The expression is: In the formula, This indicates the predicted workload output. The mapping function representing the multilayer perceptron model. It is an activation function. These represent the weight matrices of the third, second, and first layers of a multilayer perceptron, respectively. It is the ReLU activation function. This represents the input engineering feature vector. represents the bias vectors of the first, second, and third layers in the multilayer perceptron, respectively. Meanwhile, the multilayer perceptron model is trained using mean square error combined with L1 regularization.

[0027] Furthermore, in step S3, the method for constructing an intelligent quantity calculation model for civil engineering based on the engineering quantity statistics logic and an intelligent pricing model based on the pricing rule mapping relationship is as follows: Based on the pricing rule mapping relationship, including but not limited to material parameters of cables and terminals, an intelligent pricing model is constructed. Combining the business logic of cable installation professional materials and quotas with the design material list format, parameterized intelligent pricing for the installation portion is performed. Parameterized intelligent pricing includes, but is not limited to, calculation of cable main material costs and cable laying labor costs. The expression is: In the formula, This indicates the cost of the main cable materials. Represents the benchmark price. Based on the natural constant An exponential function with base 0. It is a definite integral. It is the lower limit of integration. It is the maximum number of points. It is an integral variable. It is a price index function. It is the cross-sectional area of ​​the cable. Indicates the cable length. The tax rate is the percentage of the cost of main materials that are taxed. Represents labor costs, It is the basic working hours. The summation symbol comes from arrive Summing the subsequent terms, For the corresponding difficulty factor The weight, Is with The corresponding difficulty factor, It represents the labor cost per unit of working hours. Meanwhile, the GARCH(1,1) model is used to predict the volatility of main material prices, and a parameterized intelligent pricing accuracy evaluation index is constructed.

[0028] In this embodiment, the cable trench for the industrial park cable renovation project is known to be 500m long, 1.2m wide, and 1.5m deep, with a cable cross-sectional area of ​​185mm² and a length of 1000m, and it needs to cross the factory road. When constructing the intelligent quantity calculation model for civil engineering, these parameters are extracted and input into a multilayer sensor to accurately predict the quantities of concrete and steel reinforcement used in the cable trench. When constructing the intelligent pricing model, the main material costs (considering the definite integral result of recent copper price fluctuations) and labor costs (including the difficulty factor weight of crossing the road) are calculated according to the expression, combined with the material and quota logic. At the same time, the price trend of main materials is predicted through the GARCH model to ensure the accuracy of pricing, which is conducive to improving the accuracy and dynamic response capability of project cost calculation. In addition, through price fluctuation prediction and difficulty factor adjustment, risk warning and optimization control of material and labor costs are realized, which can significantly reduce budget deviations caused by material price fluctuations or underestimation of construction complexity, and enhance the foresight and scientific nature of project cost control.

[0029] S4. Based on the intelligent quantity calculation model and intelligent pricing model for civil engineering, an intelligent pricing model integrating capital raising, quantity calculation and pricing is formed. Furthermore, in step S4, the method for forming an intelligent pricing model that integrates cost estimation, quantity calculation, and pricing based on the intelligent quantity calculation model and the intelligent pricing model for civil engineering is as follows: Based on the outputs of the intelligent quantity calculation model and intelligent pricing model for civil engineering, a tensor decomposition model is constructed to fuse multi-source data and generate nonlinear combination features, expressed as: In the formula, To merge tensors, The engineering quantity data output by the intelligent quantity calculation model for civil engineering are respectively Cost data output by the intelligent pricing model and design data The factor vector, Indicates to arrive Summing the combination of factor vectors Indicates the outer product. To decompose the rank, It is a generated nonlinear combination feature. It is the Sigmoid activation function. This is the weight matrix. express The first in One characteristic, express The first in One characteristic, express Zhongyu The index corresponds to certain characteristics. The study examines the logical relationship between architectural and installation professional data, quotas, main materials, and equipment, and defines entity sets. This includes, but is not limited to, quota items, main materials, equipment, and cost parameters, defining relation sets. This includes, but is not limited to, relationships of inclusion, association, and influence, and entities and relationships are represented based on the TransE model.

[0030] Furthermore, in step S4, the method for forming an intelligent pricing model that integrates cost estimation, quantity calculation, and pricing based on the intelligent quantity calculation model and the intelligent pricing model for civil engineering is as follows: Combining design data and standard material list formats, a built-in statistical rule base is implemented. This base includes a basic rule layer, a relational rule layer, and an optimization rule layer. The basic rule layer contains rules for quantity calculation, quota application, and cost calculation. The relational rule layer defines rules for relationships between entities. The optimization rule layer includes price adjustment rules and risk assessment rules. The rules are formalized using production rules, and rule conflicts are resolved through evidence fusion based on Dempster-Shafer theory, forming an intelligent pricing model that integrates data submission, quantity calculation, and pricing. The expression is: In the formula, It is a multi-objective optimization loss function. To regress the loss, For classifying losses, For graph structure loss, Loss weighting coefficient.

[0031] In this embodiment, the urban underground cable network project includes design data such as a duct length of 800m and a cable cross-sectional area of ​​240mm². The intelligent quantity calculation model for civil engineering outputs concrete and steel reinforcement quantities, while the intelligent pricing model outputs the costs of main materials and labor. When constructing the integrated model, the tensor decomposition model integrates these data to generate fused tensors and nonlinear features. The logic of the inclusion relationship between quota items and main cable materials is clearly defined. The built-in rule library covers rules for engineering quantity calculation and price adjustment. When rule conflicts occur, evidence fusion is used to resolve them. The final model can achieve coherent processing from data submission to quantity calculation and pricing. Through the intelligent integration of the entire process of design data submission, engineering quantity calculation, and cost pricing, it is convenient to improve the automation and intelligent decision-making level of engineering cost management, which helps to reduce manual verification and data fragmentation problems, and reduce cost deviations and project cost risks.

[0032] S5. Based on the intelligent pricing model, perform intelligent quantity calculation and pricing in both quota mode and list mode during the preliminary budget stage and the construction drawing budget stage. Furthermore, in step S5, based on the intelligent pricing model, the method for intelligent quantity calculation and pricing under the quota mode and the list mode in the preliminary estimate stage and the construction drawing budget stage is as follows: Based on feedback data from the practical application of the intelligent cost model and historical engineering data, the key cost models for each project are analyzed according to the technical and economic indicator system. The Pearson correlation coefficient method is used to calculate the main influencing factors and parameters of the key cost models. The expression is: In the formula, The Pearson correlation coefficient indicates that it measures the characteristics of a feature. With key cost model The range of linear correlation values Time is considered a strong correlation. express and covariance, express and standard deviation Indicates the first The first feature Each sample value Indicates the first The first key cost model Each sample value Represents the total number of samples. Indicates the first The sample mean of each feature, Indicates the first The sample mean of each key cost model.

[0033] Furthermore, in step S5, based on the intelligent pricing model, the method for intelligent quantity calculation and pricing under the quota mode and the list mode in the preliminary estimate stage and the construction drawing budget stage is as follows: An ELM mathematical prediction model is constructed based on the main influencing factors and parameters. The prediction error is analyzed and the coefficients of each key cost model are adjusted using a multiple linear regression model. The expression is: In the formula, The prediction error represents the true value. Compared with model predictions The difference, Indicates the first The true cost value for each sample Indicates the first The predicted value for each sample, This represents the fundamental error when all features are zero. Represents the feature weight coefficients. Indicates the first The feature vector of each sample To represent unexplained noise, an iterative optimization smart pricing model is used, expressed as: In the formula, The updated intelligent pricing model coefficients are used for the next round of prediction. This is the current coefficient. It's the learning rate. Represents the loss function for multi-objective optimization. With coefficient The changing slope is used to guide the direction of coefficient updates. It is a multi-objective optimization loss function. The input features of the intelligent pricing model prediction function Sum of coefficients Output the predicted cost value.

[0034] In this embodiment, the cable supporting project in the residential community adopts a quota model in the preliminary budget stage. Using an intelligent pricing model combined with historical data, the Pearson correlation coefficient method is used to find that the cable cross-sectional area, laying length and civil engineering costs are strongly correlated. Based on this, an ELM model is constructed to predict the cost. In the construction drawing budget stage, a list model is adopted. After analyzing the prediction error of the model, the coefficients are adjusted and the model is iteratively optimized to make the quantity calculation and pricing of the list items more accurate. The Pearson correlation coefficient method can accurately locate the key factors affecting the project cost. The application of the ELM model and the multiple linear regression model improves the accuracy of cost prediction. The iterative optimization mechanism allows the intelligent pricing model to continuously adapt to the actual situation, which helps to overcome the problems of reliance on experience, long cycle and insufficient accuracy in traditional engineering quantity calculation and pricing. It is conducive to realizing the automation, data-driven and intelligent optimization of cost calculation.

[0035] 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 the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart cost estimation method for cable engineering based on big data, characterized in that, Includes the following steps: S1. Use OCR technology to extract key data from the cable engineering data submission form, form design data submission data, collect and store it sequentially, and build a basic database for the project. S2. Based on the design data, use big data processing technology to quantify the rules and generate the engineering quantity statistics logic and pricing rule mapping relationship; S3. Using the aforementioned engineering quantity statistics logic, construct a civil engineering intelligent quantity calculation model, and based on the aforementioned pricing rule mapping relationship, construct an intelligent pricing model; S4. Based on the aforementioned intelligent quantity calculation model and intelligent pricing model for civil engineering, an intelligent pricing model integrating cost estimation, quantity calculation, and pricing is formed. S5. Based on the intelligent pricing model, perform intelligent quantity calculation and pricing in both quota mode and list mode during the preliminary budget stage and the construction drawing budget stage.

2. The intelligent cost estimation method for cable engineering based on big data according to claim 1, characterized in that, In step S1, key data is extracted from the cable engineering data submission form using OCR technology to form design data submission data, which is then sequentially collected and stored to construct the project's basic database. The method employs OCR technology to extract key data from the cable engineering data submission form. For the unstructured text sequence extracted as key data, a mapping function is used to convert it into structured data, expressed as: In the formula, This represents the final generated structured data. Represents a mapping function. It is an unstructured text sequence extracted by OCR. It indicates that it will start from the first One to the first The key data fields are merged into a single set. This represents the field name in the corresponding structured data, and the specific content of the corresponding field. It represents the total number of key-value pairs. Indicates the first Extracting values ​​from key data fields, Indicates the first One to the first Summing each character Indicates the first The first key data field The weight coefficient of each character, This is an indicator function that evaluates to 1 if the condition within the parentheses is true, and 0 otherwise. It is a text sequence recognized by OCR. The first in One character, This represents the total number of characters in the text sequence recognized by OCR. Indicates the first The key data fields correspond to character sets. The extracted key data are classified by project to form design data and collected and stored sequentially to build a project basic database. The key data includes, but is not limited to, project name, ductwork, cable trench, trenchless, cable-related data and key parameters of components of each line segment.

3. The intelligent cost estimation method for cable engineering based on big data according to claim 2, characterized in that, In step S2, the method for quantifying rules and generating the engineering quantity statistical logic and pricing rule mapping relationship based on the design data is as follows: The method of using big data processing technology for rule quantification includes, but is not limited to, statistical rules for component parameters, steel reinforcement quantity, concrete quantity, cable laying quotas, pipe laying and pouring quotas, and surplus soil transportation. The expression for the cable laying quota is as follows: In the formula, The calculation result represents the cable laying quota. The basic labor quota benchmark for cable laying per unit length and unit cross-sectional area is an empirical coefficient obtained by fitting historical engineering data and industry standards. Represents the length of the cable. Represents the cross-sectional area of ​​the cable. It is the cross-sectional area correction factor. It is the number of construction difficulty factors. It is the first The weighting coefficients of each difficulty factor. It is the first The possible values ​​of each difficulty factor It is the overall difficulty adjustment coefficient.

4. The intelligent cost estimation method for cable engineering based on big data according to claim 3, characterized in that, In step S2, the method for quantifying rules and generating the engineering quantity statistical logic and pricing rule mapping relationship based on the design data is as follows: The statistical rules are intelligently linked to the corresponding project segments. By defining project feature vectors, a similarity matrix between the rules and the projects is calculated, and then based on the similarity matrix... The Hungarian algorithm is used to perform the optimal allocation of the statistical rules, expressed as: In the formula, It is a double-layer summation symbol that represents the cumulative calculation of all combinations of items and rules. Indicates the first The project and the first The similarity value of the rules, This indicates that binary decision variables can only take either 0 or 1. Time represents the first The project allocation number Rule 1, in Time represents no allocation, Indicates the total number of items. Indicates the total number of rules. It is the first constraint. The second constraint is to generate a mapping relationship between the project's quantity statistics logic and pricing rules.

5. The intelligent cost estimation method for cable engineering based on big data according to claim 4, characterized in that, In step S3, based on the engineering quantity statistics logic, a civil engineering intelligent quantity calculation model is constructed. Based on the pricing rule mapping relationship, the method for constructing an intelligent pricing model is as follows: Based on the aforementioned engineering quantity statistics logic, including but not limited to component parameters of ductwork, cable trenches, and manholes, a civil engineering intelligent quantity calculation model is constructed. This model standardizes the quantity calculations for, but is not limited to, cable ductwork, cable trenches, manholes, and auxiliary modules. Feature parameters are extracted, and the relationships between the parameters of each component are clarified. A multilayer perceptron model is then constructed to predict the engineering quantity. The expression is: In the formula, This indicates the predicted workload output. The mapping function representing the multilayer perceptron model. It is an activation function. These represent the weight matrices of the third, second, and first layers in a multilayer perceptron, respectively. It is the ReLU activation function. This represents the input engineering feature vector. These represent the bias vectors of the first, second, and third layers in the multilayer perceptron, respectively. Meanwhile, the multilayer perceptron model is trained using mean squared error combined with L1 regularization.

6. The intelligent cost estimation method for cable engineering based on big data according to claim 5, characterized in that, In step S3, based on the engineering quantity statistics logic, a civil engineering intelligent quantity calculation model is constructed. Based on the pricing rule mapping relationship, the method for constructing an intelligent pricing model is as follows: Based on the pricing rule mapping relationship, including but not limited to material parameters of cables and terminals, the intelligent pricing model is constructed. Combining the business logic of cable installation professional materials and quotas with the design material list format, parameterized intelligent pricing for the installation portion is performed. This parameterized intelligent pricing includes, but is not limited to, calculation of cable main material costs and cable laying labor costs. The expression is: In the formula, This indicates the cost of the main cable materials. Represents the benchmark price. Based on the natural constant An exponential function with base 0. It is a definite integral. It is the lower limit of integration. It is the maximum number of points. It is an integral variable. It is a price index function. It is the cross-sectional area of ​​the cable. Indicates the cable length. The tax rate is the percentage of the cost of main materials that are taxed. Represents labor costs, It is the basic working hours. The summation symbol comes from arrive Summing the subsequent terms, For the corresponding difficulty factor The weight, Is with The corresponding difficulty factor, It is the labor cost per unit of working hours. At the same time, the GARCH(1,1) model is used to predict the volatility of main material prices, and the parameterized intelligent pricing accuracy evaluation index is constructed.

7. The intelligent cost estimation method for cable engineering based on big data according to claim 6, characterized in that, In step S4, the method for forming an intelligent pricing model integrating cost estimation, quantity calculation, and pricing based on the aforementioned intelligent quantity calculation model and intelligent pricing model is as follows: Based on the outputs of the aforementioned intelligent quantity calculation model and intelligent pricing model for civil engineering, a tensor decomposition model is constructed to fuse multi-source data and generate nonlinear combination features, expressed as: In the formula, To merge tensors, The engineering quantity data output by the intelligent quantity calculation model for civil engineering are respectively Cost data output by the intelligent pricing model and design data The factor vector, Indicates to arrive Summing the combination of factor vectors Indicates the outer product. To decompose the rank, It is a generated nonlinear combination feature. It is the Sigmoid activation function. This is the weight matrix. express The first in One characteristic, express The first in One characteristic, express Zhongyu The index corresponds to certain characteristics. The study examines the logical relationship between architectural and installation professional data, quotas, main materials, and equipment, and defines entity sets. This includes, but is not limited to, quota items, main materials, equipment, and cost parameters, defining relation sets. This includes, but is not limited to, relationships of inclusion, association, and influence, and entities and relationships are represented based on the TransE model.

8. The intelligent cost estimation method for cable engineering based on big data according to claim 7, characterized in that, In step S4, the method for forming an intelligent pricing model integrating cost estimation, quantity calculation, and pricing based on the aforementioned intelligent quantity calculation model and intelligent pricing model is as follows: Combining the aforementioned design data and material list standard format, a built-in statistical rule base is implemented. This base includes a basic rule layer, a relational rule layer, and an optimization rule layer. The basic rule layer contains rules for quantity calculation, quota application, and cost calculation. The relational rule layer defines rules for relationships between entities. The optimization rule layer includes price adjustment rules and risk assessment rules. Production rule representation is used for rule formalization, and evidence fusion based on Dempster-Shafer theory is employed to resolve rule conflicts, forming an intelligent pricing model that integrates data provision, quantity calculation, and pricing. The expression is: In the formula, It is a multi-objective optimization loss function. To regress the loss, For classifying losses, For graph structure loss, Loss weighting coefficient.

9. The intelligent cost estimation method for cable engineering based on big data according to claim 8, characterized in that, In step S5, based on the intelligent pricing model, the method for intelligent quantity calculation and pricing in both quota mode and list mode during the preliminary budget stage and the construction drawing budget stage is as follows: Based on feedback data from the actual application of the intelligent pricing model and historical project data, the key cost models for each project are analyzed according to the technical and economic indicator system. The Pearson correlation coefficient method is used to calculate the main influencing factors and parameters of the key cost models. The expression is: In the formula, The Pearson correlation coefficient indicates that it measures the characteristics. With key cost model The range of linear correlation values Time is considered a strong correlation. express and covariance, express and standard deviation Indicates the first The first feature Each sample value Indicates the first The first key cost model Each sample value Represents the total number of samples. Indicates the first The sample mean of each feature, Indicates the first The sample mean of each key cost model.

10. The intelligent cost estimation method for cable engineering based on big data according to claim 8, characterized in that, In step S5, based on the intelligent pricing model, the method for intelligent quantity calculation and pricing in both quota mode and list mode during the preliminary budget stage and the construction drawing budget stage is as follows: Based on the aforementioned main influencing factors and parameters, an ELM mathematical prediction model is constructed. The prediction error is analyzed using a multiple linear regression model, and the coefficients of each key cost model are adjusted accordingly. The expression is: In the formula, The prediction error represents the true value. Compared with model predictions The difference, Indicates the first The true cost value for each sample Indicates the first The predicted value for each sample, This represents the fundamental error when all features are zero. Represents the feature weight coefficients. Indicates the first The feature vector of each sample To represent unexplained noise, iteratively optimize the intelligent valence model, expressed as: In the formula, The updated intelligent pricing model coefficients are used for the next round of prediction. This is the current coefficient. It's the learning rate. Represents the loss function for multi-objective optimization. With coefficient The changing slope is used to guide the direction of coefficient updates. It is a multi-objective optimization loss function. The input features of the intelligent pricing model prediction function Sum of coefficients Output the predicted cost value.

Images