Green low-carbon evaluation system and method for station-tunnel integration engineering

Abstract

This application discloses a green and low-carbon evaluation system and method for integrated station-tunnel engineering. The system includes an indicator layer, a module layer, an intelligent scoring module, and a green technology library linkage module. Each green and low-carbon evaluation indicator is categorized into corresponding professional evaluation modules under the design and equipment manufacturing stages, as well as the construction stage evaluation module. By matching the indicators to be improved in the green and low-carbon scoring of each stage, the system automatically recommends green technology solutions and simulates the expected scores after application. This application classifies green and low-carbon evaluation indicators into three key stages: design, equipment manufacturing, and construction, covering the entire project lifecycle. This ensures that decisions at each stage consider their impact on the overall green and low-carbon effect, achieving the overall green and low-carbon goals of the project. The green technology library linkage module automatically recommends green technology solutions, realizing a closed-loop management of "evaluation-technology matching-solution optimization-re-evaluation".

Description

Technical Field

[0001] This application relates to the field of green and low-carbon evaluation technology, specifically to a green and low-carbon evaluation system and method for integrated station-tunnel engineering. Background Technology

[0002] With the introduction of the "dual carbon" target, the green and low-carbon construction of subway projects has become a focus of industry attention. Chinese invention patent document CN114139846A discloses a subway carbon accounting and carbon neutrality evaluation system and method. It establishes carbon emission models for each stage of subway station civil engineering, including building material production, transportation, and construction installation, and summarizes the total carbon emissions during the construction phase. This evaluation system uses quantitative assessment methods to calculate and evaluate the carbon emissions of subway projects, providing data support for achieving carbon neutrality goals.

[0003] However, this scheme has the following shortcomings: First, the evaluation indicators mainly focus on the construction and implementation phases, failing to cover the design and equipment manufacturing phases. The design phase, as the starting point of the project, significantly impacts subsequent resource consumption and carbon emissions through its scheme selection; the equipment manufacturing phase involves the production and use of large equipment such as tunnel boring machines, directly affecting the overall green and low-carbon level of the project. This limitation in the evaluation scope prevents the achievement of refined management throughout the entire lifecycle.

[0004] In particular, when the evaluation results show that the indicators fail to meet the standards, the evaluation system can only provide an evaluation conclusion and cannot automatically match and recommend corresponding green technology improvement solutions. Engineers need to rely on experience to find alternative solutions, which is time-consuming, labor-intensive, and makes it difficult to guarantee the suitability of the solutions. Summary of the Invention

[0005] To address the aforementioned shortcomings, the technical problem this application aims to solve is to provide a green and low-carbon evaluation system and method for integrated station-tunnel engineering, in order to resolve the issues of incomplete evaluation scope in existing technologies, and the need for engineers to rely on experience to find alternative solutions when evaluation results show that indicators do not meet the standards, which is time-consuming, labor-intensive, and difficult to guarantee the adaptability of the solutions.

[0006] Therefore, this application provides a green and low-carbon evaluation system for integrated station-tunnel engineering, including: The input unit is used to receive data from the design, equipment manufacturing, and construction phases of the integrated station-tunnel project. The indicator layer includes multiple green and low-carbon evaluation indicators, which are respectively categorized into safety and durability, convenience and efficiency, health and comfort, resource conservation, environmental friendliness, and low-carbon construction. The module layer includes the design phase evaluation module, the equipment manufacturing phase evaluation module, and the construction phase evaluation module; The indicator intelligent association module uses a similarity algorithm to automatically calculate the similarity between the indicator features of green and low-carbon evaluation indicators and the professional feature vectors of each stage. Based on the similarity, the evaluation indicators are automatically classified, and an indicator-module mapping feature library is constructed and improved. The intelligent scoring module, based on the received data, converts the qualitative indicators into quantitative vectors using natural language processing algorithms, converts the quantitative indicators into scores using normalization algorithms, calculates the correlation coefficients between indicators to form a coupling matrix using correlation algorithms, and introduces coupling correction factors to calculate the scores of green and low-carbon evaluation indicators at each stage. The comprehensive evaluation module automatically calculates the dynamic weight values ​​of each stage using the analytic hierarchy process (AHP) algorithm, and calculates the comprehensive evaluation result and evaluation level based on the indicator scoring results of the intelligent scoring module. The green technology library linkage module stores green technology solutions and their corresponding core parameters. By matching the green and low-carbon evaluation indicators corresponding to the scores to be improved in each stage of green and low-carbon scoring, it automatically recommends green technology solutions and simulates the expected evaluation scores after application. The green and low-carbon evaluation indicators corresponding to the scores to be improved are those with scores lower than preset scores.

[0007] In the above system, preferably, the green technology library includes at least one technical solution record, which includes technology category, applicable stage, correspondence of green and low-carbon evaluation indicators, core parameters of energy saving and carbon emission reduction, applicable conditions, technology maturity level, and application case information; the green technology library is equipped with a feature vector extraction unit, a matching algorithm module, and a technology benefit prediction unit. The feature vector extraction unit is used to convert the technical solution into a multi-dimensional feature vector; the matching algorithm module matches green and low-carbon evaluation indicators with technical solutions based on a weighted similarity calculation method to generate a technology recommendation list; the technology benefit prediction unit predicts the expected evaluation score after the application of green technology solutions through historical data and simulation models.

[0008] In the above system, preferably, the feature vector extraction unit transforms the technical solution into an 18-dimensional feature vector; the matching algorithm module recommends the optimal technical solution based on the weighted cosine similarity algorithm; and the technical benefit prediction unit uses a multiple regression model to quantitatively predict the expected evaluation score after the application of the green technology solution.

[0009] In the above system, preferably, the design stage evaluation module includes an architectural sub-module, a structural sub-module, and an electromechanical sub-module; the equipment manufacturing stage evaluation module includes an equipment manufacturing sub-module and an equipment operation sub-module.

[0010] In the above system, preferably, the intelligent association module of indicators adopts the cosine similarity algorithm. When the similarity is ≥0.8, it is automatically classified into the corresponding safety and durability category, convenience and efficiency category, health and comfort category, resource saving category, environmentally friendly category or low-carbon construction category. When the similarity is <0.8, an abnormal prompt is triggered and a category with a high matching degree is recommended.

[0011] In the above system, preferably, it also includes a multi-stage early warning and evaluation module, which is used to generate pre-evaluation scores and optimization suggestions in the design stage, build a real-time data acquisition interface and preset early warning threshold algorithm to trigger graded early warning in the construction stage, and automatically summarize the full-cycle data to generate an evaluation report after completion. The multi-stage early warning and evaluation module automatically updates the indicator scores every hour during the construction phase. When the score of a single indicator is lower than the standard value by 30% or the overall score of the phase is lower than 60 points, a graded early warning is triggered. A yellow warning indicates that rectification is required, and a red warning indicates that construction is suspended. The module automatically pushes early warning information and matches rectification plans from the green technology library.

[0012] In the above system, preferably, the green technology library linkage module matches evaluation indicators and technical solutions using the K-nearest neighbor algorithm. When the score of a certain indicator is lower than a preset score, it automatically recommends a preset number of suitable technical solutions from the green technology library.

[0013] This application also provides a green and low-carbon evaluation method for integrated station-tunnel engineering, including the following steps: Based on the data collected during the design, equipment manufacturing, and construction phases of the station-tunnel integrated project, the green and low-carbon evaluation indicators for each phase are scored and calculated using an intelligent scoring module to obtain the corresponding green and low-carbon scores. The green and low-carbon scores obtained are used to identify those scores that are lower than a preset threshold and need improvement. Based on the identified indicators that need improvement, a demand feature vector V_need is generated. Iterate through all the technical solutions stored in the green technology library, and calculate the technical feature vector V_tech for each technical solution through the feature vector extraction unit; Using the weighted cosine similarity algorithm, the similarity between the generated demand feature vector V_need and the obtained technical feature vectors V_tech of each technical solution is calculated. All technical solutions are sorted from high to low according to the similarity and a recommended list of technical solutions is generated. For the recommended technical solution generated in the steps, based on the technical parameters of the technical solution, the project parameters, and the environmental parameters of the engineering environment, the technical benefit prediction unit is used to predict the technical benefits, calculate the expected improvement values ​​of each indicator to be improved after the application of the technical solution, and give the confidence interval of the prediction results.

[0014] In the above method, preferably, the green technology library includes at least one technical solution record, which includes technology category, applicable stage, correspondence of green and low-carbon evaluation indicators, core parameters of energy saving and carbon emission reduction, applicable conditions, technology maturity level, and application case information; the green technology library is equipped with a feature vector extraction unit, a matching algorithm module, and a technology benefit prediction unit. The feature vector extraction unit is used to convert the technical solution into a multi-dimensional feature vector; the matching algorithm module matches green and low-carbon evaluation indicators with technical solutions based on a weighted similarity calculation method to generate a technology recommendation list; the technology benefit prediction unit predicts the expected evaluation score after the application of green technology solutions through historical data and simulation models.

[0015] Preferably, the above method also includes a multi-stage early warning step: During the design phase, input the design parameters, and the system will automatically generate a pre-evaluation score and optimization suggestions based on the green technology library. During the construction phase, a real-time data acquisition interface is built to automatically update indicator scores every hour. When a single indicator score is 30% below the standard value or the overall score for a stage is below 60, a tiered warning is triggered. A yellow warning indicates that rectification is required, and a red warning indicates that construction is suspended. Warning information is automatically pushed and rectification plans are matched from the green technology library.

[0016] As can be seen from the above technical solution, the green and low-carbon evaluation system and method for integrated station-tunnel engineering provided in this application solves the problems of incomplete evaluation scope and lack of intelligent technology recommendation mechanism in existing technologies. Compared with existing technologies, this application has the following beneficial effects: First, green and low-carbon evaluation indicators are categorized according to the implementation process of integrated station and tunnel construction into three key stages: design, equipment manufacturing, and construction, covering the entire life cycle of the project. This ensures that decisions at each stage take into account their impact on the overall green and low-carbon effect, thereby achieving the overall green and low-carbon goals of the project.

[0017] Second, the intelligent indicator association module uses a similarity algorithm to automatically calculate the similarity between indicator features and professional feature vectors at each stage, thereby achieving automatic indicator classification and improving the efficiency of evaluation system construction.

[0018] Third, the intelligent scoring module converts qualitative indicators into quantitative vectors using natural language processing algorithms, converts quantitative indicators into scores using normalization algorithms, and calculates the correlation coefficients between indicators using correlation algorithms to achieve scientific and accurate scoring.

[0019] Fourth, the green technology library linkage module automatically recommends technical solutions through matching algorithms and simulates the changes in indicators after the application of the solutions to output the expected evaluation score, thereby realizing closed-loop management of "evaluation-technology matching-solution optimization-re-evaluation".

[0020] Fifth, the multi-stage early warning and assessment module builds a real-time data acquisition interface during the construction phase and presets an early warning threshold algorithm to trigger tiered early warnings, enabling dynamic monitoring and timely intervention. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments of this application or the prior art will be briefly introduced and explained below. Obviously, the accompanying drawings described below are only some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0022] Figure 1 A schematic diagram of the green and low-carbon evaluation system architecture for station-tunnel integrated engineering provided for this application; Figure 2 A flowchart of the green and low-carbon evaluation method for integrated station-tunnel engineering provided in this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] To provide a clearer explanation and illustration of the technical solution and implementation of this application, several preferred specific embodiments for implementing the technical solution of this application are described below.

[0025] It should be noted that the directional terms such as "inner," "outer," "front," "back," "left," and "right" in this article are based on the product's usage status. Obviously, the use of these directional terms does not limit the scope of protection of this solution.

[0026] Example 1 Please see Figure 1 , Figure 1 This is a schematic diagram of the green and low-carbon evaluation system architecture for integrated station-tunnel engineering provided in this application.

[0027] like Figure 1As shown, this application provides a green and low-carbon evaluation system for integrated station-tunnel engineering. It adopts a hierarchical architecture design to systematically and intelligently evaluate the green and low-carbon performance of integrated station-tunnel engineering throughout its entire lifecycle, from design and equipment manufacturing to construction. This system overcomes the limitations of existing technologies that only focus on the construction phase, extending the evaluation scope to the design phase at the project's origin and the critical equipment manufacturing phase, thus achieving refined management throughout the entire lifecycle.

[0028] The evaluation system comprises the following components: input unit, indicator layer, module layer, intelligent processing unit, algorithm iteration unit, and output unit. Each layer and unit interacts with each other through standardized interfaces, forming a complete evaluation loop. The intelligent processing unit includes an intelligent indicator association module, an intelligent scoring module, a comprehensive evaluation module, a multi-stage early warning and assessment module, and a green technology library linkage module. These modules enable functions such as indicator association, intelligent scoring, dynamic weight calculation, comprehensive evaluation, and intelligent recommendation, providing intelligent decision support for project improvement.

[0029] The following sections provide a detailed explanation of each component of the green and low-carbon evaluation system for integrated station-tunnel engineering.

[0030] I. Input Unit.

[0031] The input unit is the data entry point for the evaluation system. It is responsible for receiving data from different stages and formats, such as the design, equipment manufacturing, and construction stages of the integrated station-tunnel project. It supports the input of structured and unstructured data.

[0032] Structured data includes: numerical parameters from design drawings (such as concrete usage, steel usage, building area, etc.), construction site monitoring data (such as PM2.5 concentration, PM10 concentration, noise levels, etc.), and equipment manufacturing data (such as raw material consumption, energy consumption, water consumption, etc.). This data is directly imported into the evaluation system through a standardized data interface, and automatic format verification and data cleaning are performed.

[0033] Unstructured data includes design specifications, construction logs, inspection reports, and images. For this type of data, an OCR module is used to recognize text information in scanned documents and images, and a natural language processing module is used to extract key information from the text, such as the compliance status of qualitative indicators like "using C50 high-strength concrete" and "configuring air conditioning system purification and disinfection devices."

[0034] Data input methods include: obtaining BIM models (Revit format) from design institutes, uploading them to the system, and the system automatically parsing the BIM models to extract component information, material information, etc.; reading production data from the ERP system of tunnel boring machine manufacturers; reading data from the energy management system; uploading design specifications, calculation sheets, and other documents; and manually entering some design parameters, construction parameters, and other data.

[0035] The input unit also includes a data validation module to check the completeness and reasonableness of the input data. For example, it checks whether required fields are complete, whether values ​​are within a reasonable range, and whether units are consistent. For abnormal data, the evaluation system will mark it and prompt the user to confirm or correct it.

[0036] II. Indicator Layer.

[0037] The indicator layer includes multiple green and low-carbon evaluation indicators, which have been scientifically screened and categorized into six categories: safety and durability, convenience and efficiency, health and comfort, resource conservation, environmental friendliness, and low-carbon construction.

[0038] The indicator selection process consists of two stages: Phase 1: Preliminary screening based on standards related to green and low-carbon evaluation. Reference standards include authoritative domestic and international standards such as the *Evaluation Standard for Green Construction of Building and Municipal Engineering* (GB / T 50640), the *Evaluation Standard for Green Buildings* (GB / T 50378), the *Code for Design of Metro* (GB 50157), and the US LEED green building assessment system. Using text mining technology, all evaluation indicators related to green and low-carbon construction were extracted from these standards, initially identifying approximately 128 candidate indicators, which were then preliminarily categorized according to factors such as safety and durability, convenience and efficiency, health and comfort, resource conservation, environmental friendliness, and low-carbon construction.

[0039] The second phase: Considering the adaptability of integrated station-tunnel construction, the ability of evaluation indicators to represent green and low-carbon attributes, the availability of indicator data, and the special construction requirements of metro variable-diameter shield tunneling projects, indicators that are not applicable to the construction scenario of metro variable-diameter shield tunneling projects, have excessively high data acquisition costs, or have no substantial representational significance for green and low-carbon attributes were eliminated. Ultimately, more than 40 core evaluation indicators covering the design, equipment manufacturing, and construction phases were determined. These indicators not only conform to international and domestic green and low-carbon evaluation standards but also fully consider the actual situation of integrated station-tunnel projects, possessing strong scientific validity and operability.

[0040] The specific details of each category of indicators are as follows: Safety and durability indicators include: the time required to evacuate and reach a safe zone in an emergency, the number of independent entrances and exits at the station, the proportion of high-strength concrete of grade C50 or higher used, and the proportion of high-strength steel bars of grade 400MPa or higher used. These indicators focus on the safety performance and service life of the project.

[0041] Convenience and efficiency indicators include walking distance from station entrances and exits to bus stops, rationality of transfer methods, and barrier-free design. These indicators focus on the ease of use and operational efficiency of the project.

[0042] Health and comfort indicators include the configuration of air conditioning system purification and disinfection devices, indoor thermal and humidity comfort, and building lighting quality. These indicators focus on the health and comfort experience of users.

[0043] Resource conservation indicators include the proportion of green building materials used, the percentage of recyclable and reusable materials used, the proportion of prefabricated components used, raw material saving rate, raw material recycling rate, water saving rate, water recycling rate, the proportion of building materials produced within 500km, and the loss rate of major building materials. These indicators focus on the efficient use and recycling of resources.

[0044] Environmentally friendly indicators include: construction waste recycling rate, construction waste discharge volume, wastewater recycling rate, PM2.5 and PM10 concentrations at the construction site boundary, construction dust control, noise emission control, exhaust gas emission control, light pollution control, and wastewater discharge control. These indicators focus on the project's impact on the surrounding environment.

[0045] Low-carbon construction indicators include carbon emissions per unit area, renewable energy utilization rate, and the application of energy-saving equipment. These indicators directly focus on carbon emissions and energy consumption.

[0046] III. Module Layer.

[0047] Following the construction process of the integrated station-tunnel project, the module layer divides the evaluation process into three main modules: the design stage evaluation module, the equipment manufacturing stage evaluation module, and the construction stage evaluation module.

[0048] (a) Evaluation module in the design phase.

[0049] The design phase evaluation module is further subdivided into architectural, structural, and mechanical and electrical sub-modules.

[0050] The indicators that the architecture sub-module focuses on include: the time required to evacuate and reach a safe zone in an emergency, the number of independent entrances and exits of the station, the walking distance between the station entrances and exits and the bus stops, anti-slip measures for special areas, barrier-free design, greening measures within the land acquisition area, comfort of the riding environment, and the rationality of transfer methods.

[0051] The structural engineering submodule focuses on the safety, durability, and material efficiency of structural design. The evaluation indicators include: the application ratio of high-strength concrete of not less than C50, the application ratio of high-strength steel bars of 400MPa grade and above, the application ratio of green building materials, and the use of ready-mixed concrete and ready-mixed mortar.

[0052] The electromechanical sub-module focuses on the energy efficiency, environmental friendliness, and comfort of electromechanical systems (including ventilation and air conditioning, water supply and drainage, power supply and distribution, etc.). The evaluation indicators include: configuration of air conditioning system purification and disinfection devices, indoor thermal and humidity comfort, lighting power density value meeting the current national standards, application of elevator energy-saving technology, leakage rate of water supply and drainage network, energy efficiency level of electrical equipment, wastewater recycling rate, etc.

[0053] The characteristic of the design phase evaluation is that it is a pre-evaluation based on design drawings and design specification documents, which can identify potential green and low-carbon issues before the project is implemented and provide a basis for design optimization.

[0054] (ii) Equipment manufacturing stage evaluation module.

[0055] The equipment manufacturing stage evaluation module is further divided into an equipment manufacturing sub-module and an equipment operation sub-module.

[0056] The Equipment Manufacturing sub-module focuses on the green and low-carbon performance of large equipment such as tunnel boring machines during the manufacturing process. The evaluation indicators include: raw material savings, raw material recycling rate, remanufacturing rate, water saving rate, water recycling rate, and renewable energy utilization rate, all of which are quantitative indicators.

[0057] The Equipment Operation sub-module focuses on the green and low-carbon performance of tunnel boring machines during operation. Evaluation indicators include: power factor of low-voltage power distribution system, protection level of electrical equipment, and concentration of harmful gas CO in the working environment.

[0058] The equipment manufacturing phase evaluation fills a gap in existing technologies by including the production and use of large equipment in the evaluation scope, thus achieving a more comprehensive life cycle assessment.

[0059] (III) Construction phase evaluation module.

[0060] The construction phase is the implementation phase of a project, and the management level of the construction process directly affects resource consumption, environmental impact, and carbon emissions. The construction phase evaluation module focuses on the green and low-carbon performance of the construction process.

[0061] The construction phase evaluation module is further divided into four professional sub-modules: construction management, resource utilization, environmental protection, and carbon emission control.

[0062] The construction management sub-module focuses on green and low-carbon management of construction organization, construction technology, and construction techniques. Evaluation indicators include: the proportion of prefabricated components used, the proportion of building materials produced within 500km, the loss rate of major building materials, and the establishment of a green construction management system at the construction site.

[0063] The resource utilization sub-module focuses on the efficiency of resource utilization such as water, electricity, and materials during construction. Evaluation indicators include: wastewater recycling rate, proportion of recyclable and reusable materials, and reuse of temporary facilities and turnover materials.

[0064] The environmental protection sub-module focuses on the control of environmental impacts such as dust, noise, exhaust gas, wastewater, and construction waste during the construction process. Evaluation indicators include: construction waste recycling rate, construction waste discharge, PM2.5 and PM10 concentrations at the construction site boundary, construction dust control, noise emission control, and light pollution control.

[0065] The carbon emission control submodule is the core of the construction phase evaluation, directly focusing on the carbon emission level during the construction process. Evaluation indicators include carbon emissions per unit area.

[0066] IV. Intelligent Indicator Association Module.

[0067] The intelligent association module for indicators is used to build an indicator-module mapping feature library. It uses a similarity algorithm to automatically calculate the similarity between indicator features and professional feature vectors at each stage, so as to realize the automatic classification of indicators and form a self-learning mapping library.

[0068] Specifically, the intelligent association module for indicators uses a cosine similarity algorithm. First, each green and low-carbon evaluation indicator is transformed into a feature vector. The feature dimensions include: indicator category (safety and durability, convenience and efficiency, health and comfort, resource conservation, environmental friendliness, low-carbon construction), applicable stage (design, equipment manufacturing, construction), data type (qualitative, quantitative), data source (BIM model, monitoring system, documents, etc.), and evaluation object (building, structure, electromechanical, equipment, etc.).

[0069] At the same time, each stage and professional sub-module is also transformed into feature vectors, with the feature dimensions consistent with the indicator feature vectors.

[0070] Then, the cosine similarity between the feature vector of each indicator and the feature vector of each professional sub-module is calculated. When the similarity is ≥0.8, the system automatically classifies the indicator into the corresponding professional sub-module; when the similarity is <0.8, an anomaly prompt is triggered and a module with a high matching degree is recommended for manual confirmation by experts.

[0071] For example, the indicator "Time taken to evacuate and reach a safe zone in an emergency" has the following feature vector: [Safety and durability category, design phase, quantitative, BIM model + simulation, architectural specialty]. Its feature vector similarity to the architectural specialty sub-module is 0.92, and it is automatically categorized into the architectural specialty sub-module of the design phase evaluation module.

[0072] Through this intelligent association mechanism, when new evaluation indicators are added, the system can automatically complete the classification without manual intervention, which greatly improves the scalability and maintenance efficiency of the evaluation system.

[0073] V. Intelligent Scoring Module.

[0074] Based on the received data, the intelligent scoring module converts qualitative indicators into quantitative vectors using natural language processing algorithms, converts quantitative indicators into scores using normalization algorithms, calculates the correlation coefficients between indicators to form a coupling matrix using correlation algorithms, and introduces coupling correction factors to calculate the scores of green and low-carbon evaluation indicators at each stage.

[0075] (a) Quantification of qualitative indicators.

[0076] For qualitative indicators, such as "configuration of air conditioning system purification and disinfection devices" and "barrier-free design", the intelligent scoring module uses natural language processing algorithms to convert text descriptions into quantitative vectors.

[0077] The specific steps are as follows: Retrieve relevant documents or text descriptions from the input unit; Use natural language processing algorithms to extract key information and identify whether it meets the standard requirements; The recognition results are converted into quantitative values: complete compliance is converted to 1, partial compliance is converted to 0.5, and non-compliance is converted to 0. Automatic rejection logic is implemented by combining threshold algorithms: for key qualitative indicators (such as safety indicators), if the requirements are not met, the evaluation of this stage is directly marked as unsuccessful.

[0078] For example, for the indicator "configuration of air conditioning system purification and disinfection device", the system extracts the air conditioning system configuration information from the design drawings, identifies that "ultraviolet disinfection device and electrostatic dust removal device are configured", judges it as fully compliant, the quantitative value is 1, and gets the full score of 5 points.

[0079] (ii) Normalization of quantitative indicators.

[0080] For quantitative indicators, such as "the application ratio of high-strength concrete of not less than C50" and "the recycling rate of construction waste", the intelligent scoring module uses the Min-Max standardization algorithm to convert the actual values ​​into scores.

[0081] Standardized score = (actual value - minimum value) / (maximum value - minimum value) × 100.

[0082] The minimum and maximum values ​​are determined based on industry standards or project objectives.

[0083] For example, the industry standard for the "construction waste recycling rate" indicator requires a minimum of 10%, with an excellent level of over 30%. If a project's actual recycling rate is 80%, then: Standardized score = (80% - 10%) / (30% - 10%) × 100 = 350.

[0084] Because it exceeded the excellent level, the final score was a perfect 5 points.

[0085] (III) Correlation analysis between indicators.

[0086] In actual engineering projects, the evaluation indicators are not completely independent and there are certain correlations between them. For example, there is a positive correlation between "the proportion of green building materials used" and "the content of harmful substances in building materials", while there is a negative correlation between "renewable energy utilization rate" and "carbon emissions per unit area".

[0087] The intelligent scoring module calculates the correlation coefficients between indicators using the grey relational analysis algorithm, forming a coupling matrix. The basic steps of the grey relational analysis algorithm are as follows: Determine the reference sequence and the comparison sequence; Perform dimensionless processing on the sequence; Calculate the correlation coefficient; Calculate the correlation degree.

[0088] The correlation coefficient matrix between the indicators was obtained through calculation.

[0089] (iv) Coupling correction.

[0090] A coupling correction factor is introduced to adjust the scoring results. The corrected scoring formula is as follows: The scoring result = basic standardized score × [1 + Σ(correlation index score × correlation coefficient) / n].

[0091] Where n is the number of related indicators.

[0092] For example, if the base score for the "proportion of green building materials application" indicator is 5 points, and the related indicator "content of harmful substances in building materials" also scores 5 points, with a correlation coefficient of 0.6, then: The corrected score is 5 × [1 + (5 × 0.6) / 1] = 5 × 1.3 = 6.5 points.

[0093] However, since the maximum score is 5 points, the final score remains 5 points. This coupled correction mechanism can more accurately reflect the overall performance of each indicator.

[0094] (v) Handling missing data.

[0095] In cases where qualitative indicators are missing, the intelligent scoring module triggers a rejection algorithm, directly marking the evaluation for that stage as unsuccessful and outputting a list of missing indicators, prompting the user to supplement the data.

[0096] If a quantitative indicator is missing, the system will automatically assign a score of 0 and mark the missing status in the scoring report to remind the user to pay attention.

[0097] VI. Determining the weights.

[0098] This application employs the Analytic Hierarchy Process (AHP) to determine the weights of each evaluation indicator, each professional sub-module, and each stage module in the overall evaluation. The basic steps of the Analytic Hierarchy Process include: (a) Establish a hierarchical structure model.

[0099] The decision problem is decomposed into an objective layer, a criterion layer, and a solution layer. Regarding the evaluation system for this application: Target layer: Green and low-carbon comprehensive evaluation of station-tunnel integrated project.

[0100] The first level of the criteria layer includes the design phase, equipment manufacturing phase, and construction phase.

[0101] Level 2 of the criteria layer: Professional sub-modules for each stage.

[0102] Level 3 of the criteria layer: Evaluation indicators for each professional sub-module.

[0103] (ii) Construct the judgment matrix.

[0104] For each factor at the same level, pairwise comparisons are made with respect to a factor at the next higher level to construct a judgment matrix. The judgment matrix uses a 1-9 scale. 1: Both factors are equally important.

[0105] 3: The former is slightly more important than the latter.

[0106] 5: The former is significantly more important than the latter.

[0107] 7: The former is more strongly important than the latter.

[0108] 9: The former is extremely important than the latter.

[0109] 2, 4, 6, 8: These are between the adjacent judgments mentioned above.

[0110] Reciprocal: When the latter is more important than the former, take the reciprocal of the former's importance to the latter.

[0111] For example, the judgment matrix for the three stages of design, equipment manufacturing, and construction is shown in Table 1: Table 1: Judgment matrix for design stage, equipment manufacturing stage, and construction stage.

[0112]

[0113] The judgment matrix indicates that the design phase is slightly more important than the equipment manufacturing phase (scale 2), the construction phase is slightly more important than the equipment manufacturing phase (scale 2), and the design phase and the construction phase are equally important (scale 1).

[0114] (iii) Calculate the weight vector.

[0115] The eigenvector corresponding to the largest eigenvalue of the judgment matrix, after normalization, becomes the weight vector. The calculation method uses the square root method: multiply the elements of each row of the judgment matrix, then take the nth root (where n is the matrix order), and then normalize to obtain the weight vector.

[0116] Taking the above judgment matrix as an example, the weights are calculated using the square root method: Design phase: = ≈ 1.260.

[0117] Equipment manufacturing stage: = ≈ 0.630.

[0118] Construction phase: = ≈ 1.260.

[0119] Normalization: Design phase weight: 1.260 / (1.260 + 0.630 + 1.260) ≈ 0.40.

[0120] Weight of equipment manufacturing stage: 0.630 / (1.260 + 0.630 + 1.260) ≈ 0.20.

[0121] Construction phase weight: 1.260 / (1.260 + 0.630 + 1.260) ≈ 0.40.

[0122] (iv) Consistency check.

[0123] Because human subjective judgments may be inconsistent, a consistency check is needed on the judgment matrix. The consistency check is performed by calculating the consistency ratio (CR). CR = CI / RI, where: CI is the consistency index, CI = (λmax - n) / (n - 1), where λmax is the largest eigenvalue of the judgment matrix and n is the matrix order.

[0124] RI is a random consistency index, which is obtained by looking up a table based on the matrix order.

[0125] When CR < 0.10, the consistency of the judgment matrix is ​​considered acceptable; when CR ≥ 0.10, the judgment matrix needs to be adjusted and the judgment needs to be redone.

[0126] VII. Comprehensive Evaluation Module.

[0127] The comprehensive evaluation module automatically calculates the dynamic weight values ​​of each stage using the analytic hierarchy process (AHP) algorithm, and calculates the comprehensive evaluation result and evaluation level based on the index scoring results of the intelligent scoring module.

[0128] (a) Dynamic weight calculation.

[0129] The comprehensive evaluation module presets the basic weight range for each stage: Design phase: 0.3-0.4.

[0130] Equipment manufacturing stage: 0.2-0.3.

[0131] Construction phase: 0.3-0.4.

[0132] In the actual evaluation process, based on project characteristics and data completeness, the dynamic weight values ​​of each stage are automatically calculated using the analytic hierarchy process (AHP) algorithm. For example, for projects that emphasize design optimization, the weight of the design stage may reach 0.4; for projects with high construction difficulty, the weight of the construction stage may reach 0.4.

[0133] (II) Calculation of Overall Score The formula for calculating the overall score is: Overall Score = Σ(Score of each stage × Dynamic weight of each stage).

[0134] For example, the evaluation result of a certain project is: Design phase score: 4.86 points, weight 0.35.

[0135] Equipment manufacturing stage score: 5.00 points, weight 0.25.

[0136] Construction phase score: 4.90 points, weight 0.40.

[0137] The overall score is 4.86 × 0.35 + 5.00 × 0.25 + 4.90 × 0.40 = 1.701 + 1.250 + 1.960 = 4.911 points.

[0138] (III) Evaluation level determination.

[0139] The evaluation level is output based on the overall score: Excellent (≥4.5 points).

[0140] Good (3.5-4.5 points).

[0141] Medium (2.5-3.5 points).

[0142] Poor (<2.5 points).

[0143] The above items received a total score of 4.911, and the evaluation level was "Excellent".

[0144] 8. Green Technology Library Linkage Module.

[0145] The green technology library linkage module is one of the core innovations of this application. It is used to store green technology solutions and their corresponding core parameters. By matching the green and low-carbon evaluation indicators corresponding to the scores to be improved in each stage of green and low-carbon scoring, it automatically recommends green technology solutions and simulates the expected evaluation scores after application.

[0146] (a) Green technology library structure.

[0147] The green technology library includes at least one technical solution record, and each technical solution record includes the following: Technology categories: Level 1 (e.g., green building materials, energy-saving technologies, environmental protection technologies, etc.), Level 2 (e.g., recycled concrete materials, photovoltaic power generation technologies, etc.), Level 3 (e.g., high-performance recycled aggregate concrete).

[0148] Applicable stages: main stages (design, equipment manufacturing, construction), secondary stages, applicable engineering parts (station main structure, tunnel sections, entrance and exit passages, etc.).

[0149] Correspondence between green and low-carbon evaluation indicators: List the evaluation indicators that the technical solution can improve, the degree of impact (high, medium, low), and the direction of improvement (the percentage range of improvement or reduction).

[0150] Core parameters for energy conservation and carbon emission reduction: quantitative parameters (such as carbon emission reduction, construction waste utilization, compressive strength, cost coefficient, impact of construction period, durability index, water saving rate, etc.) and proportioning parameters (such as cement dosage, aggregate dosage, water-cement ratio, admixture dosage, etc.).

[0151] Applicable conditions: Environmental conditions (temperature, humidity, freeze-thaw cycles, etc.), engineering conditions (structural importance level, design strength level, structural type, etc.), construction conditions (equipment requirements, personnel training, quality inspection frequency, etc.), and material sourcing conditions (source of recycled materials, transportation distance, raw material quality requirements, etc.).

[0152] Technology Readiness Level: The TRL (Technology Readiness Level) rating system is adopted, ranging from Level 1 (basic principle observation) to Level 9 (system practical application verification) to indicate the maturity level of the technology.

[0153] Application case information: Project name, application scale, actual results, user satisfaction, etc.

[0154] (ii) Feature vector extraction unit.

[0155] The feature vector extraction unit is used to transform the technical solution into a multi-dimensional feature vector. This application uses an 18-dimensional feature vector, with the following dimensions: v1: Carbon emission reduction benefits; v2: Resource recovery rate; v3: Cost-effective; v4: Technology maturity level; v5: Strength level; v6: Construction difficulty; v7: Wide range of applications; v8: Environmental adaptability; v9: Durability; v10: Number of cases; v11: User satisfaction; v12: Standard compliance; v13: Applicability during the design phase; v14: Applicability during the construction phase; v15: Equipment manufacturing applicability; v16: Water-saving benefits; v17: Impact of construction period; v18: Transportation distance requirements.

[0156] The values ​​in each dimension are normalized and transformed into values ​​between 0 and 1.

[0157] For example, the feature vector of "high-performance recycled aggregate concrete technology" is: V_tech = [0.74, 0.75, 0.92, 0.89, 0.83, 0.65, 0.60, 0.70, 0.83, 0.60, 0.85, 1.00, 0.60, 0.90, 0.20, 0.40, 0.50, 0.75].

[0158] (III) Matching Algorithm Module.

[0159] The matching algorithm module uses a weighted similarity calculation method to match green and low-carbon evaluation indicators with technical solutions, and generates a list of recommended technologies.

[0160] The specific steps are as follows: Identify indicators for improvement: Identify indicators that score below the preset score (e.g., below 60% of the full score) from the green and low-carbon scores at each stage as indicators for improvement.

[0161] Constructing the demand feature vector: Based on the indicators to be improved, the system automatically generates a demand feature vector V_need. The dimensions of the demand feature vector are consistent with those of the technical solution feature vector, and the values ​​reflect the intensity of demand for each dimension.

[0162] Calculate the weighted similarity: Use the weighted cosine similarity algorithm to calculate the similarity between the feature vector of the requirement and the feature vector of the technical solution. Similarity = (V_need · V_tech · W) / (||V_need|| × ||V_tech||).

[0163] Here, W is the weight matrix, which is dynamically adjusted according to the importance of the indicators. In the current scenario, resource recycling and carbon emission reduction have relatively high weights.

[0164] Sorting and generating a recommendation list: Sort by similarity from high to low and generate a top 3 technology solution recommendation list.

[0165] This application uses the K-Nearest Neighbors (KNN) algorithm as a supplement to the matching algorithm. When the score of a certain indicator is lower than the preset score, it automatically recommends a preset number (e.g., 3) of the matching technology solutions in the green technology library.

[0166] (iv) Technical benefit prediction unit.

[0167] The technology benefit prediction unit uses historical data and simulation models to predict the expected evaluation score after the application of green technology solutions.

[0168] The prediction model uses a multiple regression model: ΔScore_i = α + β1×P_tech + β2×P_project + β3×P_env + ε.

[0169] in: ΔScore_i: The expected improvement score for metric i; P_tech: Technical solution parameter vector; P_project: Project feature parameter vector; P_env: Vector of environment constraint parameters; α, β1, β2, β3: regression coefficients (trained using historical data); ε: Error term.

[0170] The predicted output includes: Expected improvement values ​​for each indicator; Predicted score; Confidence interval (95%); Comprehensive benefit forecast (improvement of total score during construction phase, change in project comprehensive evaluation level, economic benefits, environmental benefits, and social benefits); Risk warnings (extended construction period, increased costs, personnel training needs, etc.); Implementation recommendations (trial program, parameter adjustment, quality traceability mechanism, etc.).

[0171] This intelligent recommendation and benefit prediction mechanism enables closed-loop management of "evaluation-technology matching-solution optimization-re-evaluation", which greatly improves the efficiency and accuracy of green and low-carbon improvements.

[0172] IX. Multi-stage early warning and assessment module.

[0173] The multi-stage early warning and evaluation module is used to generate pre-evaluation scores and optimization suggestions during the design phase, build a real-time data acquisition interface during the construction phase and preset early warning threshold algorithms to trigger graded early warnings, and automatically summarize the full-cycle data to generate an evaluation report after completion.

[0174] (a) Pre-evaluation during the design phase.

[0175] During the design phase, users input design parameters (such as BIM models, design specification documents, etc.), and the system automatically generates pre-evaluation scores and optimization suggestions based on the green technology library.

[0176] Pre-evaluation process: Analyze the design parameters and extract the design values ​​of each evaluation index; The intelligent scoring module calculates scores for each indicator. The overall score for the design phase is calculated based on the comprehensive evaluation module. Identify metrics with low scores and match optimization suggestions from the green technology library; Generate a pre-evaluation report, including the current score, potential problems, optimization suggestions, and expected improvement effects.

[0177] Pre-evaluation during the design phase can identify problems before project implementation, allowing for timely optimization of the design and preventing rework and resource waste later on.

[0178] (ii) Real-time early warning during the construction phase.

[0179] During the construction phase, the system establishes a real-time data acquisition interface, automatically updating indicator scores hourly. Data sources include: Environmental monitoring system (PM2.5, PM10, noise, etc.); Energy management system (electricity consumption, water consumption, etc.); Materials management system (material consumption, waste generation, etc.); Construction management system (construction progress, machinery usage, etc.).

[0180] The system has a preset warning threshold algorithm that triggers a tiered warning when one of the following conditions is met: A score for a single indicator falling below 30% of the standard value triggers a yellow alert, indicating that rectification is required. If the overall score for a stage is below 60 points, a red alert is triggered, indicating that construction will be suspended.

[0181] After the warning is triggered, the system will automatically perform the following operations: Push early warning information to project management personnel (via SMS, email, APP push, etc.); The system matches rectification plans with green technologies from its database and generates rectification suggestions. Record early warning events and include them in project files; After the data is collected and rectified, it will be automatically re-scored until the warning is lifted.

[0182] Through a real-time early warning mechanism, green and low-carbon issues during the construction process can be detected in a timely manner, allowing for rapid response and rectification, ensuring that the project remains under control at all times.

[0183] (iii) Post-completion full-cycle evaluation.

[0184] Upon project completion, the system automatically aggregates all lifecycle data and generates a multi-dimensional evaluation report. The report includes: Project Overview: Project name, construction unit, project scale, evaluation date, etc.; Evaluation results: total score, evaluation level, scores for each stage, scores for each professional sub-module, and scores for each indicator; Highlights Analysis: High-scoring indicators and summarizing successful experiences; Problem Analysis: For indicators with low scores, analyze the reasons and propose improvement suggestions; Benchmarking analysis: Compare with similar projects to identify gaps and strengths; Economic benefit analysis: Cost savings brought about by green and low-carbon measures; Environmental benefit analysis: carbon emission reduction, resource conservation, pollutant emission reduction, etc.; Social benefit analysis: impact on the surrounding environment and residents; Summary of technology application: green technology solutions applied, actual results, lessons learned; Recommendations for improvement: suggestions for similar projects in the future.

[0185] The evaluation report supports visualization, including radar charts, bar charts, trend charts, etc., to present the evaluation results intuitively.

[0186] 10. Algorithm Iteration Unit.

[0187] The algorithm iteration unit periodically optimizes the mapping algorithm, coupling algorithm, and prediction model based on historical evaluation data and expert assessment.

[0188] Specifically, the algorithm iteration unit includes the following functions: Data accumulation: The input data, evaluation results, technology recommendations, actual effects, and other information for each evaluation are stored in the database to form a historical dataset.

[0189] Model training: Regularly (e.g., quarterly) retrain the prediction model using historical datasets to optimize regression coefficients and improve prediction accuracy.

[0190] Algorithm optimization: Based on expert feedback and actual application results, adjust parameters such as the weight matrix of the similarity algorithm and the correlation coefficient of the coupling algorithm.

[0191] Feature library update: Update the indicator-module mapping feature library and green technology library based on the newly added evaluation indicators and technical solutions.

[0192] Performance evaluation: Regularly evaluate algorithm performance, including metrics such as prediction accuracy, recommendation hit rate, and user satisfaction, and identify areas for improvement.

[0193] Through algorithm iteration mechanisms, the evaluation system can continuously learn and optimize, adapt to new standard requirements and technological developments, and maintain its advanced nature and practicality.

[0194] XI. Output Unit.

[0195] The output unit automatically generates evaluation reports, early warning notices, and solution suggestions, and supports visual display.

[0196] The main functions of the output unit include: Report generation: Automatically generates evaluation reports in Word and PDF formats based on the evaluation results, including text descriptions, tables, charts, etc.

[0197] Visualization: Evaluation results are displayed through a web interface or mobile app, including radar charts, bar charts, trend charts, heat maps, etc., and interactive queries and drill-down are supported.

[0198] Warning push notification: When a warning is triggered, a warning notification will be automatically sent to relevant personnel via SMS, email, APP push, etc.

[0199] Solution recommendations output: Based on the matching results from the green technology library, generate a technical solution recommendation document, including technical description, applicable conditions, expected effects, implementation steps, etc.

[0200] Data Export: Supports exporting evaluation data to Excel, CSV and other formats for further analysis and archiving.

[0201] API Interface: Provides standardized API interfaces to support integration with other systems (such as BIM platforms, project management systems, etc.).

[0202] This is a conventional technique in this field and will not be elaborated further.

[0203] 12. Carbon Emission Calculation.

[0204] Carbon emission calculation is a standard technique in this field and will not be elaborated upon here.

[0205] XIII. Application Examples.

[0206] The following is a complete application example demonstrating the actual application process and effect of the evaluation system in this application.

[0207] Project Name: A Station and Section Project of Metro Line 5 in a Certain City Construction Unit: Metro Group Co., Ltd. of a Certain City Project Scale: Station building area 25,000m², section length 1,500m.

[0208] (a) Evaluation during the design phase.

[0209] Data collection: Obtain the BIM model (Revit format) from the design institute and upload it to the system. The system automatically parses the BIM model and extracts component information, material information, etc. Upload design specifications, calculation sheets, and other documents. Manually enter some design parameters.

[0210] Indicator Calculation and Scoring: Architecture: Evacuation time in emergency situations: The system extracts the station's geometric parameters from the BIM model and uses Pathfinder software to simulate personnel evacuation. The calculated evacuation time is 5.2 minutes, which is less than the standard requirement of 6 minutes, so it scores 5 points.

[0211] Number of independent entrances / exits of the station: According to the design drawings, the station has 4 independent entrances / exits, which meets the requirements and earns 5 points.

[0212] Walking distance from station entrance / exit to bus stop: Calculated using GIS technology, the nearest entrance / exit is 80m away from the bus stop, which is less than 100m, so 5 points are awarded.

[0213] Anti-slip measures for special areas: Anti-slip materials or anti-slip treatments are used in areas such as platforms, passages, and entrances / exits, earning 5 points.

[0214] Accessibility Design: The design drawings include facilities such as accessible elevators, accessible restrooms, tactile paving, and wheelchair ramps, which meet the requirements of the "Accessibility Design Code" and earn 5 points.

[0215] Greening measures within the land acquisition area: The landscape design drawings adopted a greening design that combines trees, shrubs and ground cover plants, which is worth 5 points.

[0216] Comfortable riding environment: Seats are provided in the waiting area, the air conditioning system is designed to be 24-28℃, and the noise control target is 70dB(A), scoring 4 points.

[0217] Reasonable transfer method: This station is not a transfer station, so this indicator does not apply.

[0218] Architecture major score = (5×0.15 + 5×0.15 + 5×0.12 + 5×0.08 + 5×0.10 + 5×0.15 + 4×0.12 + 5×0.08) / 0.95 = 4.92 points.

[0219] Structural Engineering: Application ratio of high-strength concrete of not less than C50: According to statistics from the BIM model, the amount of C50 and above concrete is 8,000 m³, the total amount of concrete is 20,000 m³, the proportion is 40%, and it gets 4 points.

[0220] Green building materials application ratio: Identify green building materials from the material list. The proportion of green building materials used is 55%, which scores 5 points.

[0221] Application ratio of high-strength steel bars of 400MPa grade and above: The application ratio of steel bars of strength grade of 400MPa and above reaches 85%, which is 4 points.

[0222] Use of ready-mixed concrete and ready-mixed mortar: The design documents clearly state that all ready-mixed concrete and ready-mixed mortar will be used, which is 5 points.

[0223] Structural Engineering Score = 4 × 0.25 + 5 × 0.35 + 4 × 0.20 + 5 × 0.20 = 4.55 points.

[0224] Mechanical and Electrical Engineering: Air conditioning system purification and disinfection device configuration: The air conditioning system in the design drawings is equipped with an ultraviolet disinfection device and an electrostatic dust removal device, which is worth 5 points.

[0225] The indoor thermal and humidity environment is comfortable: the air conditioning system is designed to maintain a temperature of 24-28℃ and a relative humidity of 40%-70%, which meets the comfort requirements and scores 5 points.

[0226] The lighting power density value meets the current national standard: the lighting power density value of the platform, station hall, and entrance / exit hall does not exceed the limit specified in the "General Specification for Energy Conservation and Renewable Energy Utilization in Buildings" GB55015, and 5 points are awarded.

[0227] Elevator energy-saving technology application: The design of an energy-saving elevator using a permanent magnet synchronous motor and energy feedback technology is worth 5 points.

[0228] Leakage rate of water supply and drainage network: If the leakage rate of water supply and drainage network does not exceed 3%, 5 points will be awarded.

[0229] Energy efficiency rating of electrical equipment: Core electrical equipment such as transformers, lighting fixtures, and elevators all meet Level 1 energy efficiency, earning 5 points.

[0230] Wastewater recycling rate: The rate of reuse of domestic sewage and cooling wastewater generated by the station after treatment is higher than 80%, which earns 5 points.

[0231] Score for Mechanical and Electrical Engineering major = 5 × (1 / 7) × 7 = 5.00 points.

[0232] Design phase score = 4.92×0.30 + 4.55×0.40 + 5.00×0.30 = 4.80 points.

[0233] (ii) Evaluation of the equipment manufacturing stage.

[0234] Data collection: Production data was retrieved from the ERP system of the tunnel boring machine manufacturer. Energy consumption data was retrieved from the energy management system. Some data was manually entered.

[0235] Indicator Calculation and Scoring: Equipment Manufacturing Major: Raw material savings: Theoretical usage 1,200t, actual usage 1,140t, savings rate = (1200-1140) / 1200 = 5%, 5 points.

[0236] Raw material recycling rate: 80t of scrap and waste is generated, and 72t is recycled. The recycling rate is 72 / 80 = 90%, which is 5 points.

[0237] Remanufacturing rate: Total number of parts: 10,000; 500 parts are remanufactured; Remanufacturing rate = 500 / 10000 = 5%, 5 points.

[0238] Water saving rate: The baseline water consumption is 50,000 m³, and the actual water consumption is 45,000 m³. The water saving rate is (50,000 - 45,000) / 50,000 = 10%, which is 5 points.

[0239] Water recycling rate: The recycled water volume is 40,000 m³, the total water volume is 45,000 m³, the water recycling rate = 40,000 / 45,000 = 88.9%, and you get 5 points.

[0240] Renewable energy utilization rate: Renewable energy power generation is 500,000 kWh, total energy consumption is 5,000,000 kWh, renewable energy utilization rate = 500,000 / 5,000,000 = 10%, 5 points.

[0241] Score for Equipment Manufacturing Major = 5 × (1 / 6) × 6 = 5.00 points.

[0242] Equipment Operation Specialty: Low-voltage power distribution system power factor: Optimize equipment circuit system, control system and facilities to improve the automatic compensation of the overall power factor, and achieve an automatic compensation of the overall power factor of 0.95, which is 5 points.

[0243] CO concentration in the work environment: The carbon monoxide (CO) concentration (PC-TWA) in the work environment air is less than 20 mg / m³, and 5 points are awarded.

[0244] Protection level of electrical equipment: Electrical equipment should have protective structures and measures adapted to environmental conditions. The protection level of electrical equipment shall not be lower than IP55, of which the protection level of the main drive motor shall not be lower than IP65, which is 5 points.

[0245] Equipment operation specialty score = 5 × (1 / 3) × 3 = 5.00 points.

[0246] Score for the equipment manufacturing stage = 5.00 × 0.40 + 5.00 × 0.60 = 5.00 points (III) Evaluation of the construction phase.

[0247] Data collection: Real-time environmental monitoring data is read from the monitoring system. Data on material consumption and machinery usage is read from the construction management system. On-site data is collected via a mobile application. Construction records and photos are uploaded.

[0248] Indicator Calculation and Scoring: Construction Management Major: The proportion of prefabricated components used is as follows: the amount of prefabricated components used is 5,000 m³, the total amount of components used is 25,000 m³, the proportion = 5000 / 25000 = 20%, and 3 points are awarded.

[0249] Percentage of building materials produced within 500km: 70% are locally sourced, earning 5 points.

[0250] The main building material loss rates are as follows: concrete loss rate is 2%, and steel loss rate is 1.5%, both of which are lower than the quota loss rates, earning 5 points.

[0251] A green construction management system has been established at the construction site: there are green construction plans, management systems, division of responsibilities, and training records, which earns 5 points.

[0252] Construction Management Score = 3×0.25 + 5×0.25 + 5×0.25 + 5×0.25 = 4.50 points.

[0253] Resource Utilization Major: Wastewater recycling rate: Recycled water volume is 8,000 m³, total water volume is 10,000 m³, recycling rate = 8000 / 10000 = 80%, 5 points.

[0254] Percentage of recyclable and reusable materials: Recyclable materials account for 60%, 5 points.

[0255] Reuse of temporary facilities and turnover materials: 90% reuse rate of temporary facilities and 85% reuse rate of turnover materials, 5 points.

[0256] Score for Resource Utilization = 5 × (1 / 3) × 3 = 5.00 points.

[0257] Environmental Protection Major: Construction waste recycling rate: 2,000 tons of construction waste are generated, and 1,600 tons are recycled. The recycling rate is 1600 / 2000 = 80%, which is 5 points.

[0258] Construction waste discharge: 400t of construction waste, building area 25,000m², discharge per unit area = 400 / 25000 = 0.016t / m² = 16kg / m², which is lower than the standard of 20kg / m², so 5 points are awarded.

[0259] PM2.5 and PM10 concentrations at the construction site boundary: Daily average PM2.5 concentration 65 μg / m³, daily average PM10 concentration 130 μg / m³, both meet the standards, earning 5 points.

[0260] Construction dust control: Dust control measures are fully implemented and monitoring data meet the standards, earning 5 points.

[0261] Noise emission control: Noise control measures are fully implemented and monitoring data meet the standards, earning 5 points.

[0262] Light pollution control: Light pollution control measures were implemented effectively, with no complaints, earning 5 points.

[0263] Score for Environmental Protection Major = 5 × (1 / 6) × 6 = 5.00 points.

[0264] Carbon emission control major: Calculation of carbon emissions per unit area: Carbon emissions from materials production: Cement: 5,000t × 0.87 = 4,350 tCO2eq; Steel: 3,000t × 2.2 = 6,600tCO2eq; Concrete: 20,000 m³ × 2.4 t / m³ × 0.3 = 14,400 tCO2eq; Sand: 8,000t × 0.003 = 24 tCO2eq; Gravel: 12,000t × 0.004 = 48 tCO2eq; Other materials: 2,000 tCO2eq; Subtotal carbon emissions from materials production: 27,422 tCO2eq.

[0265] Carbon emissions from construction machinery: Excavator: 500 shifts × 108.36 kgCO2eq / shift = 54.18 tCO2eq; Loader: 300 shifts × 139.32 kgCO2eq / shift = 41.80 tCO2eq; Concrete mixer truck: 200 shifts × 77.4 kgCO2eq / shift = 15.48 tCO2eq; Other machinery: 150 tCO2eq; Subtotal carbon emissions from construction machinery: 261.46 tCO2eq.

[0266] Carbon emissions from construction electricity use: Construction power consumption: 500,000 kWh × 0.5810 kgCO2eq / kWh = 290.5 tCO2eq.

[0267] Carbon emissions from material transportation: The average material transportation distance is 200km, and the transportation volume is 30,000t; Carbon emissions from transportation: 30,000t × 200km × 0.12 kgCO2eq / (t·km) = 720 tCO2eq.

[0268] Total carbon emissions: Total carbon emissions = 27,422 + 261.46 + 290.5 + 720 = 28,693.96 tCO2eq.

[0269] Carbon emissions per unit area: Carbon emissions per unit area = 28,693.96 / 25,000 = 1.148 tCO2eq / m².

[0270] The industry average is 1.3 tCO2eq / m², and this project is below the average, so it gets 5 points.

[0271] Construction phase score = 4.50×0.25 + 5.00×0.25 + 5.00×0.30 + 5.00×0.20 = 4.88 points.

[0272] (iv) Comprehensive evaluation.

[0273] Total score = 4.80×0.35 + 5.00×0.25 + 4.880×0.40 = 1.680 + 1.250 + 1.952 = 4.882 points.

[0274] According to the grading criteria, a total score of 4.882 points (≥ 4.5 points) is considered excellent.

[0275] (v) Recommendations for the Green Technology Library.

[0276] Although the project received an excellent overall rating, the system identified a low score (3 points) for the "proportion of prefabricated components" indicator, triggering the green technology library recommendation mechanism.

[0277] The system automatically generates a demand feature vector V_need, traverses all technical solutions in the green technology library, calculates similarity, and generates a Top 3 recommendation list.

[0278] Prefabricated assembly component technology (similarity: 0.89).

[0279] Matching indicators: proportion of prefabricated components, construction efficiency, and material loss rate.

[0280] Reason for recommendation: It can significantly increase the proportion of prefabricated components and reduce on-site wet work.

[0281] Expected results: The proportion of prefabricated components will increase from 20% to 50%, and the score will increase from 3 to 5.

[0282] Prefabricated steel structure technology (similarity: 0.82).

[0283] Matching indicators: proportion of prefabricated components, construction period, and carbon emissions.

[0284] Reason for recommendation: The steel structure has a high degree of prefabrication and fast construction speed.

[0285] Expected results: The proportion of prefabricated components will increase to 45%, and the construction period will be shortened by 15%.

[0286] Precast concrete composite slab technology (similarity: 0.78).

[0287] Matching indicators: proportion of prefabricated components and structural performance.

[0288] Recommendation reason: Composite slab technology is mature and suitable for subway station floor slabs.

[0289] Expected outcome: The proportion of prefabricated components will increase to 35%.

[0290] The system recommends using "prefabricated assembly component technology" and simulates the expected evaluation score after its application: The score for the construction management major increased from 4.50 to 4.80.

[0291] Construction phase score: improved from 4.88 to 4.95.

[0292] Overall score: increased from 4.882 to 4.929.

[0293] Economic benefit analysis: Reducing on-site wet work will lower labor costs by approximately 500,000 yuan.

[0294] The construction period was shortened by 10 days, saving approximately 200,000 yuan in management costs.

[0295] The total cost savings amounted to approximately 700,000 yuan.

[0296] Environmental benefit analysis: Reduce on-site dust emissions by approximately 30%.

[0297] Reduce construction noise by approximately 20%.

[0298] This will reduce carbon emissions by approximately 150 tCO2eq.

[0299] Risk warning: Precast components need to be customized in advance, which increases the difficulty of design changes.

[0300] It requires high-precision hoisting equipment and construction precision.

[0301] The initial investment increased by approximately 800,000 yuan.

[0302] Implementation Recommendations: The modularization and standardization of prefabricated components are fully considered during the design phase.

[0303] Choose an experienced precast component supplier.

[0304] Strengthen the training of construction personnel to ensure installation quality.

[0305] Establish a strict quality inspection system.

[0306] (vi) Evaluation report generation.

[0307] The system automatically generates an evaluation report, the main contents of which include: Project Name: A station and section of Metro Line 5 in a certain city.

[0308] Construction unit: A certain city's subway group company.

[0309] Project scale: Station building area 25,000m², section length 1,500m Evaluation results: Total score: 4.882 points, rating: excellent.

[0310] Design phase score: 4.80 points; Equipment manufacturing stage score: 5.00 points; Construction phase score: 4.88 points.

[0311] Suggestions for improvement: To increase the proportion of prefabricated components, it is recommended to adopt prefabricated assembly component technology. Further optimize the application ratio of C50 high-strength concrete to improve structural durability; This complete application example demonstrates the application process and effect of the evaluation system in actual engineering, and verifies the scientific nature, practicality and effectiveness of the system.

[0312] Example 2.

[0313] Based on the aforementioned green and low-carbon evaluation system for integrated station-tunnel engineering, the second aspect of this application provides a green and low-carbon evaluation method for integrated station-tunnel engineering, comprising the following steps: Step S110: Data collection and score calculation.

[0314] Based on the data collected during the design, equipment manufacturing, and construction phases of the integrated station-tunnel project, the green and low-carbon evaluation indicators for each phase are scored and calculated using an intelligent scoring module to obtain the corresponding green and low-carbon scores.

[0315] Specifically, for qualitative indicators, text descriptions are transformed into quantitative vectors using natural language processing algorithms; for quantitative indicators, actual values ​​are transformed into scores using normalization algorithms; the correlation coefficients between indicators are calculated using grey relational analysis algorithms to form a coupling matrix, and a coupling correction factor is introduced to correct the scoring results, ultimately obtaining the green and low-carbon scores for each indicator at each stage.

[0316] Step S120: Identification of indicators to be improved and generation of requirement feature vectors.

[0317] Identify the scores that need improvement from the green and low-carbon scores obtained in step S110, i.e., the scores of indicators that are lower than a preset threshold. The preset threshold can be set according to project requirements, preferably 60% of the full score.

[0318] Based on the identified indicators requiring improvement, a demand feature vector V_need is automatically generated. This demand feature vector is a multi-dimensional vector, with dimensions consistent with the technical solution feature vector, and its values ​​reflect the intensity of demand for each dimension. The generation of the demand feature vector comprehensively considers factors such as the category of the indicator requiring improvement, its applicable stage, and the urgency of improvement.

[0319] Step S130: Extraction of feature vectors for the technical solution.

[0320] Iterate through all the technical solutions stored in the green technology library and calculate the technical feature vector V_tech for each technical solution through the feature vector extraction unit.

[0321] The technical feature vector V_tech is a multi-dimensional feature vector, preferably 18-dimensional, with dimensions including but not limited to: carbon emission reduction benefits, resource recovery rate, cost-effectiveness, technology maturity, strength level, construction difficulty, breadth of applicability, environmental adaptability, durability, number of cases, user satisfaction, standard compliance, applicability in the design phase, applicability in the construction phase, applicability in equipment manufacturing, water-saving benefits, impact of construction cycle, and transportation distance requirements. The values ​​of each dimension are normalized to a value between 0 and 1.

[0322] Step S140: Technical solution matching and recommendation list generation.

[0323] Using the weighted cosine similarity algorithm, the similarity between the demand feature vector V_need generated in step S120 and the technical feature vector V_tech of each technical solution obtained in step S130 is calculated. All technical solutions are sorted from high to low according to the similarity and a recommended list of technical solutions is generated.

[0324] Preferably, the top three technical solutions with the highest similarity are selected to generate a Top 3 technical solution recommendation list. The recommendation list includes basic information about the technical solutions, similarity scores, and matching indicators to be improved.

[0325] In the calculation of weighted cosine similarity, the weight matrix is ​​dynamically adjusted according to the project context. For projects that focus on carbon emission reduction, the weight of the carbon emission reduction benefit dimension is increased accordingly; for projects that focus on cost control, the weight of the cost-benefit dimension is increased accordingly.

[0326] Step S150: Technical benefit prediction and output.

[0327] Based on the recommended technical solution generated in step S140, the technical benefits are predicted through the technical benefit prediction unit according to the technical parameters of the technical solution, the project parameters, and the environmental parameters of the environment in which the project is located.

[0328] The technology benefit prediction unit uses a multiple regression model, which is a prediction model trained based on historical evaluation data. It calculates the expected improvement values ​​of each indicator to be improved after the application of the technology solution and gives the confidence interval of the prediction results.

[0329] The output includes, but is not limited to: the expected improvement value, predicted score, confidence interval, comprehensive benefit prediction (including the improvement of the total score during the construction phase, the change of the project's comprehensive evaluation level, economic benefits, environmental benefits, and social benefits), risk warnings (including the extension of the construction period, the increase in costs, the need for personnel training, etc.), and implementation suggestions (including trial programs, parameter adjustments, quality traceability mechanisms, etc.).

[0330] Preferably, the method further includes step S160: based on the technical benefit prediction results, simulate the comprehensive project score after the application of the technical solution, and compare it with the current score to provide a quantitative basis for project decision-making.

[0331] Preferably, the method further includes step S170: automatically generating an evaluation report from the evaluation results, recommended schemes, predicted benefits and other information, supporting visualization, including radar charts, bar charts, trend charts and other chart formats.

[0332] The above methods enable green and low-carbon evaluation of the entire life cycle of the station-tunnel integrated project. Through an intelligent technology recommendation mechanism, scientific, accurate, and operable decision support is provided for project improvement, effectively enhancing the green and low-carbon performance of the project.

[0333] The technical details of the above methods have been explained in detail in the aforementioned system and will not be repeated here.

[0334] The above methods enable green and low-carbon evaluation of the entire life cycle of the station-tunnel integrated project, and provide strong support for project improvement through an intelligent technology recommendation mechanism.

[0335] Based on the above description of specific embodiments, the green and low-carbon evaluation system and method for integrated station-tunnel engineering provided in this application have the following advantages compared with the prior art: First, the selected green and low-carbon evaluation indicators are categorized into three key stages—design, equipment manufacturing, and construction—according to the implementation process of integrated station and tunnel construction. This covers the entire life cycle of the project, ensuring that decisions at each stage take into account their impact on the overall green and low-carbon effect, and achieving the overall green and low-carbon goals of the project.

[0336] Second, the intelligent indicator association module uses a similarity algorithm to automatically calculate the similarity between indicator features and professional feature vectors at each stage, thereby achieving automatic indicator classification, improving the efficiency of evaluation system construction, and reducing manual costs.

[0337] Third, the intelligent scoring module converts qualitative indicators into quantitative vectors using natural language processing algorithms, and converts quantitative indicators into scores using normalization algorithms. It also calculates the correlation coefficients between indicators using correlation algorithms, thus achieving scientific and accurate scoring and avoiding the subjectivity and one-sidedness of traditional evaluation methods.

[0338] Fourth, the green technology library linkage module automatically recommends technical solutions through matching algorithms and simulates the changes in indicators after the application of the solutions to output expected evaluation scores, realizing closed-loop management of "evaluation-technology matching-solution optimization-re-evaluation", which greatly improves the efficiency and accuracy of green and low-carbon improvements.

[0339] Fifth, the multi-stage early warning and assessment module establishes a real-time data acquisition interface during the construction phase and presets early warning threshold algorithms to trigger tiered early warnings, enabling dynamic monitoring and timely intervention. This ensures that the project remains under control and avoids the passive situation of remediation after the fact.

[0340] Sixth, by using algorithm iteration units to periodically optimize the mapping algorithm, coupling algorithm, and prediction model based on historical evaluation data and expert assessments, the evaluation system can continuously learn and optimize, adapt to new standard requirements and technological developments, and maintain its advanced nature and practicality.

[0341] Seventh, the life cycle assessment (LCA) method is used to calculate carbon emissions, covering the entire process of material production, transportation, construction machinery use, and energy consumption, which enables accurate accounting of carbon emissions and provides a scientific basis for carbon emission reduction.

[0342] Eighth, the evaluation system has good scalability and compatibility, supports integration with other systems such as BIM platforms and project management systems, and is easy to promote and apply in actual engineering projects.

[0343] Finally, it should be noted that the terms "comprising," "including," or any other variations thereof as used herein are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0344] This application is not limited to the above-described preferred embodiments. Anyone should know that any structural changes made under the guidance of this application, and any technical solutions that are the same as or similar to those in this application, fall within the protection scope of this application.

Claims

1. A green and low-carbon evaluation system for integrated station-tunnel engineering, characterized in that, include: The input unit is used to receive data from the design, equipment manufacturing, and construction phases of the integrated station-tunnel project. The indicator layer includes multiple green and low-carbon evaluation indicators, which are respectively categorized into safety and durability, convenience and efficiency, health and comfort, resource conservation, environmental friendliness, and low-carbon construction. The module layer includes the design phase evaluation module, the equipment manufacturing phase evaluation module, and the construction phase evaluation module; The indicator intelligent association module uses a similarity algorithm to automatically calculate the similarity between the indicator features of green and low-carbon evaluation indicators and the professional feature vectors of each stage. Based on the similarity, the evaluation indicators are automatically classified, and an indicator-module mapping feature library is constructed and improved. The intelligent scoring module, based on the received data, converts the qualitative indicators into quantitative vectors using natural language processing algorithms, converts the quantitative indicators into scores using normalization algorithms, calculates the correlation coefficients between indicators to form a coupling matrix using correlation algorithms, and introduces coupling correction factors to calculate the scores of green and low-carbon evaluation indicators at each stage. The comprehensive evaluation module automatically calculates the dynamic weight values ​​of each stage using the analytic hierarchy process (AHP) algorithm, and calculates the comprehensive evaluation result and evaluation level based on the indicator scoring results of the intelligent scoring module. The green technology library linkage module stores green technology solutions and their corresponding core parameters. By matching the green and low-carbon evaluation indicators corresponding to the scores to be improved in each stage of green and low-carbon scoring, it automatically recommends green technology solutions and simulates the expected evaluation scores after application. The green and low-carbon evaluation indicators corresponding to the scores to be improved are those with scores lower than preset scores.

2. The system according to claim 1, characterized in that, The green technology database includes at least one technical solution record. Each technical solution record includes the technology category, applicable stage, correspondence with green and low-carbon evaluation indicators, core parameters for energy conservation and carbon reduction, applicable conditions, technology maturity level, and application case information. The green technology database is equipped with a feature vector extraction unit, a matching algorithm module, and a technology benefit prediction unit. The feature vector extraction unit converts the technical solution into a multi-dimensional feature vector. The matching algorithm module matches green and low-carbon evaluation indicators with technical solutions based on a weighted similarity calculation method, generating a technology recommendation list. The technology benefit prediction unit predicts the expected evaluation score after the application of green technology solutions using historical data and simulation models.

3. The system according to claim 2, characterized in that, The feature vector extraction unit transforms the technical solution into an 18-dimensional feature vector; the matching algorithm module recommends the optimal technical solution based on the weighted cosine similarity algorithm; and the technical benefit prediction unit uses a multiple regression model to quantitatively predict the expected evaluation score after the application of the green technology solution.

4. The system according to claim 1, characterized in that, The design phase evaluation module includes architectural, structural, and electromechanical sub-modules; the equipment manufacturing phase evaluation module includes equipment manufacturing and equipment operation sub-modules.

5. The system according to claim 1, characterized in that, The intelligent association module for indicators uses a cosine similarity algorithm. When the similarity is ≥0.8, it automatically classifies the corresponding category into safety and durability, convenience and efficiency, health and comfort, resource conservation, environmental friendliness, or low-carbon construction. When the similarity is <0.8, it triggers an anomaly prompt and recommends a category with a high matching degree.

6. The system according to claim 1, characterized in that, It also includes a multi-stage early warning and evaluation module, which is used to generate pre-evaluation scores and optimization suggestions during the design stage, build a real-time data acquisition interface and preset early warning threshold algorithms to trigger graded early warnings during the construction stage, and automatically summarize the full-cycle data to generate an evaluation report after completion. The multi-stage early warning and evaluation module automatically updates the indicator scores every hour during the construction phase. When the score of a single indicator is lower than the standard value by 30% or the overall score of the phase is lower than 60 points, a graded early warning is triggered. A yellow warning indicates that rectification is required, and a red warning indicates that construction is suspended. The module automatically pushes early warning information and matches rectification plans from the green technology library.

7. The system according to claim 6, characterized in that, The green technology library linkage module uses the K-nearest neighbor algorithm to match evaluation indicators with technical solutions. When the score of a certain indicator is lower than the preset score, it automatically recommends a preset number of suitable technical solutions from the green technology library.

8. A green and low-carbon evaluation method for integrated station-tunnel engineering, characterized in that, Includes the following steps: Based on the data collected during the design, equipment manufacturing, and construction phases of the station-tunnel integrated project, the green and low-carbon evaluation indicators for each phase are scored and calculated using an intelligent scoring module to obtain the corresponding green and low-carbon scores. The green and low-carbon scores obtained are used to identify those scores that are lower than a preset threshold and need improvement. Based on the identified indicators that need improvement, a demand feature vector V_need is generated. Iterate through all the technical solutions stored in the green technology library, and calculate the technical feature vector V_tech for each technical solution through the feature vector extraction unit; Using the weighted cosine similarity algorithm, the similarity between the generated demand feature vector V_need and the technical feature vector V_tech of each technical solution is calculated. All technical solutions are sorted from high to low according to the similarity and a recommended list of technical solutions is generated. For the generated recommended technical solution, based on the technical parameters of the technical solution, the project parameters, and the environmental parameters of the engineering environment, the technical benefit prediction unit is used to predict the technical benefits, calculate the expected improvement values ​​of each indicator to be improved after the application of the technical solution, and give the confidence interval of the prediction results.

9. The method according to claim 8, characterized in that, The green technology database includes at least one technical solution record. Each technical solution record includes the technology category, applicable stage, correspondence with green and low-carbon evaluation indicators, core parameters for energy conservation and carbon reduction, applicable conditions, technology maturity level, and application case information. The green technology database is equipped with a feature vector extraction unit, a matching algorithm module, and a technology benefit prediction unit. The feature vector extraction unit converts the technical solution into a multi-dimensional feature vector. The matching algorithm module matches green and low-carbon evaluation indicators with technical solutions based on a weighted similarity calculation method, generating a technology recommendation list. The technology benefit prediction unit predicts the expected evaluation score after the application of green technology solutions using historical data and simulation models.

10. The method according to claim 8, characterized in that, It also includes multi-stage early warning procedures: During the design phase, input the design parameters, and the system will automatically generate a pre-evaluation score and optimization suggestions based on the green technology library. During the construction phase, a real-time data acquisition interface is built to automatically update indicator scores every hour. When a single indicator score is 30% below the standard value or the overall score for a stage is below 60, a tiered warning is triggered. A yellow warning indicates that rectification is required, and a red warning indicates that construction is suspended. Warning information is automatically pushed and rectification plans are matched from the green technology library.

Images