A road engineering low-carbon data credible interaction method and system

Abstract

The application belongs to the field of data processing, and provides a road engineering low-carbon data credible interaction method and system, which comprises the following steps: constructing a whole life cycle stage division model, generating corresponding low-carbon data according to the business activities of each stage, and adding stage identification, time stamp and space position label to the low-carbon data; based on the preset engineering logic, data causal relationship and carbon emission factor model, a data context tree containing multiple low-carbon data nodes is constructed, and the low-carbon data nodes in different space-time ranges and life cycle stages are organized into a life cycle data context tree with a phased conduction relationship; before data interaction, a multi-dimensional verification operation is performed on the target data based on the data context tree; the credible score of the target data is calculated based on the structure path verification result, the space-time consistency verification result and the carbon factor mapping verification result, and the interaction access permission of the low-carbon data is controlled according to the score result.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of data processing, and particularly relates to a low-carbon data credible interaction method and system for road engineering. BACKGROUND

[0002] The low-carbon development concept has been widely concerned in the field of infrastructure construction. As an energy-intensive and resource-consuming industry, road engineering is accompanied by significant carbon emission activities in the design, construction, operation and demolition stages. In order to realize whole-process carbon emission management, road engineering related units need to collect, transmit and share low-carbon emission data covering the whole life cycle for carbon footprint accounting, carbon emission evaluation and green performance assessment.

[0003] In the prior art, the management of low-carbon data is usually carried out in a mode of phased collection and subject attribution, specifically including construction site Internet of Things sensor collection, carbon emission estimation model, carbon checking platform and database-based data sharing interface and the like. In terms of data credibility verification, some platforms use digital signature, hash check or third-party audit means to guarantee the integrity and source authenticity of data.

[0004] However, the prior art generally ignores the life cycle dependence characteristics and phased logical transmission relationship of road engineering low-carbon data, lacks a systematic verification mechanism based on structural path, spatio-temporal dynamics and causal semantics, resulting in problems such as data island, causal rupture, spatio-temporal conflict and carbon emission deviation in the process of multi-stage cooperation and heterogeneous system interaction, which seriously affects the credibility and usability of low-carbon data in fine management and policy compliance. SUMMARY

[0005] In order to solve the problems in the prior art, the application provides a low-carbon data credible interaction method for road engineering, comprising the following steps:

[0006] S1, a whole life cycle stage division model is constructed, the life cycle of road engineering is divided into a design stage, a construction stage, an operation stage and a demolition stage, corresponding low-carbon data is generated according to the business activities of each stage, and a stage identifier, a timestamp and a spatial position label are attached to the low-carbon data;

[0007] S2, based on a preset engineering logic, a data causal relationship and a carbon emission factor model, a data context tree containing multiple low-carbon data nodes is constructed, the low-carbon data nodes in different spatio-temporal ranges and life cycle stages are organized into a life cycle data context tree with a phased transmission relationship, any data node in the data context tree includes a source stage, a spatial position, a timestamp, a responsible subject, a semantic label and a logical mapping relationship with an upstream node;

[0008] S3, before data interaction, performing multi-dimensional verification operation on target data based on the data context tree, the multi-dimensional verification operation comprising:

[0009] Structural path verification: judging whether the target data node has a complete upstream logical path according to the data context tree, and marking as incomplete structure data if the path is missing;

[0010] Temporal and spatial consistency verification: based on the timestamp and spatial label of the target data, comparing the semantic similarity of the same type of data in the spatial neighborhood and the temporal neighborhood of the target data, and triggering an abnormal mark if the comparison is inconsistent;

[0011] Carbon factor mapping verification: based on the carbon emission model, verifying the cause-effect calculation of the data content of the target data and its upstream nodes to determine whether it conforms to the carbon conversion logic;

[0012] S4, calculating the credibility score of the target data based on the structural path verification result, the temporal and spatial consistency verification result and the carbon factor mapping verification result, and controlling the interaction access permission of the low-carbon data according to the score result.

[0013] Further, the steps of the data context tree comprise:

[0014] Taking all design stage data nodes as a root node set, for each root node, performing a recursive process:

[0015] Traverse all construction stage nodes to find nodes that have a legal mapping relationship with the node;

[0016] For each matching node, establish a sub-node reference under the design node;

[0017] Repeat the process for each sub-node to continue searching for operation stage and demolition stage data nodes;

[0018] If a node has multiple upstream paths, the node is allowed to have multiple parent node references;

[0019] The constructed path forms a life cycle data context path, and multiple paths constitute a complete data context tree.

[0020] Further, the temporal and spatial consistency verification comprises:

[0021] Extracting the phase to which the target data belongs and setting an adaptive neighborhood, the adaptive neighborhood being a time and / or space field;

[0022] Retrieving the same type of data set in the neighborhood;

[0023] A multi-modal embedding representation is constructed, mapped to a unified dimensional vector using a multi-modal encoding model, and a similarity is calculated: sim(D, Di), where D is the target data, and Di is the i-th domain data point;

[0024] A neighborhood density driven anomaly index DDI = m / n is calculated, where: n is the total number of same type data points in the neighborhood; m is the number of data points that satisfy sim(D, Di) < θ; and θ is a similarity threshold;

[0025] The anomaly index is used to determine whether the target data is abnormal in space and time in its neighborhood.

[0026] Further, the structural path verification includes:

[0027] An upstream path of the target data node is extracted;

[0028] A system preset standard life cycle path set is called;

[0029] The node stage labels in the actual path are compared in semantics with the stage labels in the standard path, and if the semantic similarity is higher than a preset threshold, it is considered as a successful match;

[0030] The total number of successfully matched stages of the actual path and the standard path is counted, and a stage weight vector is introduced to express the importance of different stages to the credible path, and a structural path matching score is calculated;

[0031] The structural path matching score is used to determine whether the structural path is complete.

[0032] Further, the carbon factor mapping verification includes:

[0033] According to the upstream logical path of the target node in the data context tree, a carbon-related behavior node and a causal dependence relationship therebetween are extracted, and a directed graph is constructed, in which the nodes represent construction behaviors, and the edges represent direct or indirect effects on the target carbon emission;

[0034] According to the equipment model, the running load, the construction method, and the environmental temperature parameter, a corresponding working condition database is consulted, and the carbon factor value is corrected;

[0035] The theoretical carbon emission of the target node is calculated using the material quantity on all paths in the causal graph structure and the correction factor;

[0036] All upstream nodes that have a causal transmission path with the target node are compared one by one for the difference between their recorded values and the model estimated values, and if the deviations on all chains are within a tolerable error range, a collaborative consistency score is calculated;

[0037] If the collaborative consistency score is less than a preset score or the error is greater than a preset error, the system marks the node as a carbon causal anomaly.

[0038] The present application also provides a road engineering low-carbon data credible interaction system, characterized in that the system comprises the following modules:

[0039] A life cycle division module is configured to build a full life cycle stage division model, divide the road engineering life cycle into a design stage, a construction stage, an operation stage and a demolition stage, generate corresponding low-carbon data according to the business activities of each stage, and attach a stage identifier, a timestamp and a spatial location label to the low-carbon data;

[0040] A data context tree construction module is configured to build a data context tree comprising a plurality of low-carbon data nodes based on a preset engineering logic, a data causal relationship and a carbon emission factor model, organize the low-carbon data nodes in different spatio-temporal ranges and life cycle stages into a life cycle data context tree having a stage transmission relationship, and any data node in the data context tree comprises a source stage, a spatial location, a timestamp, a responsible subject, a semantic label and a logical mapping relationship with an upstream node;

[0041] A multi-dimensional verification module is configured to perform structural path verification, spatio-temporal consistency verification and carbon factor mapping verification on target data based on the data context tree before data interaction;

[0042] The multi-dimensional verification operation comprises:

[0043] Structural path verification: determining whether the target data node has a complete upstream logical path according to the data context tree, and marking as structural incomplete data if the path is missing;

[0044] Spatio-temporal consistency verification: comparing the semantic similarity of the same type of data in the spatial neighborhood and the temporal neighborhood of the target data based on the timestamp and the spatial label of the target data, and triggering an abnormal mark if they are inconsistent;

[0045] Carbon factor mapping verification: performing causal calculation verification on the data content of the target data and its upstream nodes based on the carbon emission model to determine whether it conforms to the carbon conversion logic;

[0046] A credible score and permission control module is configured to calculate the credible score of the target data based on the results of the above three verifications, and control the interaction access permission of the low-carbon data according to the score results.

[0047] Further, the data context tree construction module comprises:

[0048] A node initialization sub-module is configured to use all data nodes in the design stage as a root node set;

[0049] The node recursive matching submodule is configured to recursively perform a stage node matching operation on each root node, traverse all construction stage nodes, operation stage nodes, and demolition stage nodes, find nodes having a legal mapping relationship with the current node, and establish a child node reference under the nodes.

[0050] The path construction submodule is configured to allow a node having multiple upstream paths to have multiple parent node references, and combine the multiple constructed paths into a complete life cycle data context tree structure.

[0051] Further, the spatiotemporal consistency verification unit in the multi-dimensional verification module includes:

[0052] The neighborhood setting unit is configured to extract a stage to which the target data belongs and set an adaptive neighborhood, including a time field and a space field.

[0053] The neighborhood data extraction unit is configured to retrieve a same-type data set within the neighborhood range.

[0054] The semantic similarity scoring unit is configured to construct a multi-modal embedding representation, map the multi-modal embedding representation to a unified dimension vector using a multi-modal encoding model, and calculate a similarity sim(D, Di), where D is the target data, and Di is an i-th data point in the neighborhood.

[0055] The anomaly index calculation unit is configured to calculate a neighborhood density-driven anomaly index DDI = m / n, where n is a total number of neighborhood data points, and m is a number of data points having a similarity lower than a preset threshold θ, and determine whether the target data has a spatiotemporal anomaly based on the index.

[0056] Further, the structure path verification unit in the multi-dimensional verification module includes:

[0057] The path extraction unit is configured to extract an upstream path of the target data node.

[0058] The standard path comparison unit is configured to call a system-preset standard life cycle path set, and compare stage labels of nodes in the actual path with the standard path in terms of semantic similarity.

[0059] The matching score calculation unit is configured to count a number of matching successful stages, and calculate a structure path matching score in combination with a stage weight vector.

[0060] The structure integrity determination unit is configured to determine whether a structure path of the target data is complete based on a matching score result.

[0061] Further, the carbon factor mapping verification unit in the multi-dimensional verification module includes:

[0062] The carbon causal diagram construction unit is configured to extract carbon-related behavior nodes and causal dependency relationships thereof according to an upstream logical path of a target node in the data context tree, and construct a carbon causal diagram structure.

[0063] A carbon factor correction unit is configured to correct the original carbon emission factor value according to parameters such as equipment model, operation load, construction mode and environmental temperature by querying a working condition database;

[0064] A theoretical carbon emission calculation unit is configured to calculate the theoretical carbon emission of the target node by using the material consumption on all causal paths and the corrected carbon factor;

[0065] A collaborative consistency analysis unit is configured to compare the actual recorded value and the theoretical estimated value of the target node and the upstream node, and if the deviation is within the tolerable error range, the collaborative consistency score is calculated, and if the score is lower than the threshold or the error is too large, it is marked as carbon causal abnormal data.

[0066] The present application realizes the whole life cycle organization and logical mapping of low-carbon data by constructing a life cycle data context tree, and can systematically identify the dependent path, upstream source and causal structure between each data node, thereby effectively solving the problem of loose traditional low-carbon data structure, path breakage, improving the traceability and semantic consistency of data.

[0067] The present application designs three complementary multi-dimensional verification mechanisms of structure path verification, space-time consistency verification and carbon factor mapping verification, which can comprehensively judge the credibility of the target data in structure rationality, space-time logic and carbon emission causal relationship, significantly improve the precision of data quality control and abnormal identification, and provide protection for data credible interaction between multiple subjects.

[0068] By introducing a trusted score model and an access permission control mechanism, the present application realizes a dynamic authorization strategy based on the verification result, which can ensure the stable operation of the system without leaking low-trust data, and supports flexible application in various scenarios such as government supervision, carbon verification and intelligent decision-making, thereby enhancing the security and scalability of the road engineering low-carbon data management system. BRIEF DESCRIPTION OF DRAWINGS

[0069] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0070] Figure 1 is the overall flowchart of the road engineering low-carbon data credible interaction method of the present application;

[0071] Figure 2 is a data context tree schematic diagram of the present application. DETAILED DESCRIPTION

[0072] The application is described in preferred embodiments in conjunction with the accompanying drawings and specific implementations.

[0073] In one embodiment, the application proposes a road engineering low-carbon data credible interaction method, which is suitable for realizing the space-time labeling, semantic modeling, logical mapping and multi-dimensional verification of low-carbon emission related data in each stage of road engineering, thereby constructing a credible interaction mechanism with structural traceability, semantic consistency and causal reasonableness, to support the data compliance management and responsibility traceability of road engineering in the scenarios of carbon accounting, carbon regulation and carbon trading.

[0074] In the present application, road engineering refers to a comprehensive engineering construction activity taking the planning, design, construction, operation and maintenance and reconstruction of various types of road infrastructure such as highways, urban roads and rural roads as the object. Its content covers route design, roadbed and pavement structure construction, bridge and tunnel supporting facility construction, drainage and lighting system integration, traffic safety engineering layout and later maintenance management, etc. It is an important part of the transportation infrastructure system. Road engineering has the characteristics of large investment scale, long construction period, multiple participants and extensive operation and maintenance period, and is an important support for national economic development and regional traffic operation.

[0075] Road engineering low-carbon data refers to a set of data related to carbon emission, energy consumption, carbon sink behavior, energy saving measures and green building material application, etc. directly or indirectly related to low-carbon elements in each stage of road engineering. These data cover the structure scheme and material list in the early design stage, the energy consumption of equipment operation and material transportation carbon emission in the construction stage, the lighting and maintenance energy consumption in the operation stage, and the carbon emission reduction records in the final demolition and recycling stage, which are associated with carbon emission factor models to form the basis for engineering carbon footprint quantification and monitoring. Road engineering low-carbon data not only reflects the resource use efficiency in the project construction process, but also carries the key indicators of green transformation of the project.

[0076] Road engineering low-carbon data has the following characteristics: first, the data has strong life cycle dependence, i.e. the data generated at different stages has a stage transmission relationship in logic and semantics, such as design parameters determining construction intensity and construction behavior affecting operation energy consumption; second, low-carbon data has dynamic evolution and space-time sensitivity, data changes dynamically with time and geographical location, showing non-stationary characteristics; third, the data is strongly bound to the responsible subject, and has high legal and compliance audit value. Therefore, road engineering low-carbon data is complex in structure, sensitive in content, and requires high authenticity, integrity and traceability.

[0077] Under the background of the dual-carbon strategy, the carbon emission performance of road engineering has become the basis for policy regulation, carbon trading, and green financing. The authenticity and verifiability of low-carbon data are of key value to the whole life cycle management of the project. However, the low-carbon data of road engineering is often scattered among different participants, and there are problems such as different data standards, inconsistent semantics, time sequence disorder, and unclear responsibilities, which seriously restrict the credible use and interactive sharing of the data. Therefore, constructing a low-carbon data credible interaction mechanism for road engineering can establish a semantic unified structural association among multi-source heterogeneous data, ensure the authenticity and integrity of the data through time, space, and stage verification, and realize the safe sharing and responsibility tracing of the data among government regulatory agencies, design and construction units, and carbon audit agencies, thereby improving the scientificity, transparency, and compliance of green construction of road engineering.

[0078] Based on the above background, referring to Figure 1 The low-carbon data credible interaction method for road engineering of the present application specifically comprises the following steps:

[0079] S1, a whole life cycle stage division model is constructed, the life cycle of road engineering is divided into a design stage, a construction stage, an operation stage, and a demolition stage, corresponding low-carbon data is generated according to the business activities of each stage, and stage identifiers, time stamps, and spatial position labels are attached to the low-carbon data.

[0080] In the whole life cycle management of road engineering, different stages of business activities correspond to different types of low-carbon emission behaviors and their data recording methods. The design stage mainly determines the material consumption and structural form, which is the setting link of the carbon emission source; the construction stage involves a large amount of energy consumption and material transportation, which is a high-density stage of carbon emission; the operation stage focuses on facility maintenance and energy use, which is the main source of continuous carbon emission; the demolition stage involves construction waste disposal and material recycling, which is related to carbon recovery and carbon offset. In order to effectively support the whole process regulation and data credible interaction of road engineering carbon emission, it is necessary to clearly divide the project stages according to the life cycle logic, and to attach multi-dimensional labels such as stage, time, and space in the data generation process, so as to ensure that the data source is clear, the context is consistent, and it is convenient for tracing and comparison.

[0081] In this step, the stage identifier refers to the structured information used to indicate which stage of the project life cycle the data belongs to, such as the design stage, the construction stage, the operation stage, or the demolition stage. The time stamp refers to the specific time information recorded for the generation or recording of the data, usually in a standard time format, used to ensure the time consistency and sorting logic of the data. The spatial position label refers to the identification method representing the physical space position involved in the data, common forms include stake number coding, K section number, GPS coordinates, geographical zoning number, etc., used to realize spatial attribution and spatial aggregation analysis.

[0082] A life cycle stage division model of road engineering is established. The life cycle stage model is composed of a plurality of engineering stages arranged in time sequence, and at least includes a design stage, a construction stage, an operation stage and a demolition stage. Each stage corresponds to one or more data types and business activity modules. The design stage includes structure selection, material list preparation and carbon emission estimation; the construction stage includes construction organization plan, equipment scheduling, material transportation and construction energy consumption collection; the operation stage includes pavement maintenance, lighting electricity, greening maintenance, etc.; the demolition stage includes structure demolition, construction waste classification and recycling statistics. Optional implementation schemes include: establishing a standardized stage structure according to national or industry road engineering specifications, or mapping and establishing in combination with pre-defined stage objects of a BIM (Building Information Modeling) system.

[0083] A stage-data type mapping table is constructed. On the basis of the life cycle model, a one-to-one or many-to-many mapping relationship between the stage and the data type is established, which is used for quick classification during data collection. For example, the design stage is associated with data types including structure scheme data and material quota data; the construction stage is associated with equipment oil consumption records, construction logs and transportation records; the operation stage is associated with lighting energy consumption data and greening energy consumption data; the demolition stage is associated with disassembly plan and construction waste classification statistics data. The mapping table is used for subsequent automatic data attribution.

[0084] A stage identification field is defined and implemented in the data collection system. Each piece of low-carbon data should be automatically or manually attached with a stage identification field when generated or entered. Optional implementation schemes include: embedding a stage label field in the upload protocol of the Internet of Things collection device, which is predefined in the system by the stage to which the device belongs; or providing a stage selector in the manual entry process, which is selected by the entry personnel based on the current work task. For example, a phasetag field is added to the data packet structure reported by the sensor, and the field value is taken from the preset dictionary set {design stage, construction stage, operation stage, demolition stage}.

[0085] A standard timestamp field is added to record the occurrence or upload time of the data. The timestamp field records time information accurate to seconds in standard UTC format or local standard time format. Optional implementation schemes include: automatically calling the system clock of the collection device to automatically generate a timestamp when the data is reported; or the data collection platform labels each piece of data with the receiving time when receiving the data. If the data is manually supplemented, the entry interface allows manual selection of the time value, and the system must perform validity check on the time range to prevent out-of-range or conflict.

[0086] A spatial location identification system is constructed, and a spatial location label is assigned to each piece of data. The spatial location label is used to identify the physical area, construction unit, or measurement point corresponding to the data. Implementation methods include: defining spatial coding rules based on the road stake number system, such as K10+500 representing the position 500 meters forward from the starting point at 10 kilometers; or using GPS latitude and longitude coordinates to record spatial positions, suitable for mobile collection equipment; or defining the correspondence between structure number and spatial position in the BIM model, using structure ID as the spatial label. For fixed monitoring point data, the device ID can be bound to the spatial position one by one during device initialization, realizing automatic injection of data spatial labels.

[0087] The stage identifier, timestamp, and spatial location label are embedded as structured fields in the data model structure of each piece of low-carbon data, and stored in the database or real-time transmission protocol. The data model structure should include original business fields, three types of context label fields (stage, time, space), and optional semantic extension fields, for subsequent data context modeling, interactive verification, and credibility assessment. JSON, XML, or relational database table structure can be used to organize data formats, ensuring uniform structure and semantic context among data.

[0088] After the above steps are implemented, each piece of low-carbon data not only has content ontology information (such as quantity, type, and source), but also has context attributes highly coupled with engineering logic, time flow, and geographical section, thereby supporting subsequent data positioning, attribution, and comparison operations in structural path verification, spatio-temporal consistency verification, and carbon factor causality verification, effectively improving data manageability, comparability, and traceability.

[0089] In a specific example:

[0090] A provincial highway project is numbered G2020, with a starting stake number of K0+000 and an ending stake number of K20+000, with a total length of 20 kilometers. The project design phase determines that the 10th to 12th kilometer section uses cement stabilized gravel base, with a thickness of 18 cm, and the design uses materials such as gravel, cement, and water. The system generates a material estimation list based on the construction drawings and outputs the following design data records:

[0091] Material name: cement stabilized gravel

[0092] Designed quantity: 10800 cubic meters

[0093] Estimated cement consumption: 1620 tons

[0094] Design phase identifier: design phase

[0095] Timestamp: 2020-04-12 14:00:00

[0096] Spatial position tag: K10+000 to K12+000

[0097] The data is exported by the design unit in the design BIM system, the platform system automatically parses the stage attribute and adds the stage identifier as the design stage, the system obtains the export time as the timestamp, and automatically extracts the route range from the structural component mapping, and labels the spatial position tag as K10+000 to K12+000.

[0098] Entering the construction stage, from August 10, 2020 to August 17, 2020, the construction unit carries out the paving operation of the base layer in the range of K10+800 to K11+200. The equipment monitoring system deployed on site automatically collects the operation data, fuel consumption data and material transportation records of the paving equipment, and generates the following three low-carbon data records respectively:

[0099] Equipment operation record:

[0100] Equipment number: PL-012

[0101] Operation time: 6.0 hours

[0102] Fuel consumption: 18 liters of diesel

[0103] Stage identifier: construction stage

[0104] Timestamp: 2020-08-12 09:30:00

[0105] Spatial position tag: K10+900

[0106] Material entry record:

[0107] Material name: cement

[0108] Actual entry amount: 120 tons

[0109] Batch number: SJK-20200812

[0110] Stage identifier: construction stage

[0111] Timestamp: 2020-08-12 07:45:00

[0112] Spatial position tag: K10+800

[0113] Paving carbon emission estimation result (calculated based on the former two):

[0114] Operation section: K10+800 to K11+200

[0115] Estimated carbon emissions: 24.3 tons

[0116] Factor: Carbon factor of fuel consumption is 2.68 kg Material factor is 0.89 kg / kg

[0117] Phase label: Construction phase

[0118] Timestamp: 2020-08-12 17:30:00

[0119] Spatial position label: K10+800 to K11+200

[0120] The above three data are reported by the automatic collection system in real time, and after the platform system is uniformly structured, the system adds the phase label to the construction phase, uses the collection equipment to record the time as the timestamp, and uses the equipment binding or construction section code to automatically analyze the spatial position label.

[0121] The data structure is as follows (represented by JSON):

[0122] {

[0123] dataid: G2020-PL012-20200812,

[0124] stagelabel: Construction phase,

[0125] timestamp: 2020-08-12 09:30:00,

[0126] location: K10+900,

[0127] datatype: Equipment operation record,

[0128] content: {

[0129] deviceid: PL-012,

[0130] runhours: 6.0,

[0131] diesellitre: 18.0

[0132] }

[0133] }

[0134] When carbon accounting is performed subsequently, the system automatically matches the data with the material list data of the previous design stage according to the above timestamp, spatial location tag and stage identifier; the system then converts the oil consumption data into carbon emissions using carbon factor mapping logic. Carbon factor refers to the relationship between various carbon emission-related behaviors (such as material use, equipment operation, energy consumption) in road engineering and their corresponding carbon emission equivalents, i.e., the carbon dioxide equivalent corresponding to a predetermined amount of carbon emission-related behavior. Carbon factor can be determined according to the following sources: national or local carbon accounting standards, such as the “Guidelines for Voluntary Greenhouse Gas Reduction Projects”; IPCC guidelines, such as “IPCC Guidelines for National Greenhouse Gas Inventories”; engineering industry databases or carbon emission search platforms, such as “China Carbon Measurement Network” and “Carbon Emission Factor Database”; and reports on working condition correction factors published by third-party carbon verification agencies or industry yearbooks. Carbon factor mapping refers to calculating the specific carbon amount (for example, if the oil consumption data is 18 liters and the oil consumption carbon factor is 2.68 kg / L, then the carbon emission equivalent corresponding to the oil consumption is 18 liters x 2.68 kg / L = 48.24 kg ), and performing spatio-temporal consistency verification with the total carbon emission record of the construction section. If there is a spatial tag offset, a timestamp overlap conflict or a design-construction stage chain break, the system will issue an abnormality mark and block the subsequent interaction process.

[0135] In step S1, by assigning each low-carbon data a clear life cycle stage identifier, standardized timestamp and spatial location tag, the upstream and downstream dependency relationship of the data in the project life cycle, the evolution position of the data on the timeline and the attribution range of the data in the physical space can be accurately determined, providing structured support for subsequent data interaction, carbon emission mapping, path verification and responsibility attribution, and realizing the verifiability, traceability and trusted sharing ability of road engineering low-carbon data in the full life cycle range.

[0136] S2, based on the preset engineering logic, data causal relationship and carbon emission factor model, a data context tree containing multiple low-carbon data nodes is constructed, and low-carbon data nodes in different spatio-temporal ranges and life cycle stages are organized into a life cycle data context tree with stage transmission relationship. Any data node in the data context tree includes source stage, spatial location, timestamp, responsibility subject, semantic tag and logical mapping relationship with upstream nodes.

[0137] In the process of low-carbon management in the whole life cycle of road engineering, there is an obvious stage transmission relationship and logical dependence relationship between the low-carbon data generated in different stages. The design stage determines the types and quantities of materials, the construction stage records the material consumption and equipment operation, the operation stage reflects the energy consumption and maintenance behavior, and the demolition stage involves carbon emission recycling and reuse. In order to effectively realize the logical linking and semantic mapping of these data in the life cycle dimension, and support subsequent data verification, responsibility tracking and carbon accounting, it is necessary to structure the data nodes of various types according to the time sequence, spatial location and logical causal relationship, so as to build a data context structure with stage evolution attributes. By building a life cycle data context tree, the low-carbon data of different sources, formats and time points can be collected and associated in a unified logical structure, ensuring their integrity, consistency and traceability in subsequent data interaction, verification and analysis.

[0138] In this step, the data context tree refers to a data structure organized by life cycle stages, with engineering low-carbon data as nodes, and the logical relationship between upstream and downstream, time and space sequence, and semantic mapping as connection methods, forming a chain evolution structure from the design stage to the demolition stage. The data node refers to a low-carbon data instance at a certain time point, spatial location and responsibility subject, usually containing carbon emission related numerical values, sources and label information. The logical mapping relationship refers to the causal deduction or functional correspondence between nodes, such as the mapping between material design quantity and construction consumption, or the transformation relationship between equipment oil consumption data and carbon emission, i.e. carbon factor.

[0139] In an optional implementation, the implementation process of step S2 is as follows:

[0140] Step S21, collect and structure low-carbon data of each life cycle stage

[0141] The collection sources include material list and structure scheme in the design stage, construction log, material entry record and equipment oil consumption data in the construction stage, road maintenance record, lighting and green energy consumption in the operation stage, and waste disposal data and material recycling rate data in the demolition stage. After structured processing, all original data are converted into uniform data node objects.

[0142] Each data node has the following basic fields:

[0143] Source stage (such as design stage)

[0144] Data type (such as structure design, oil consumption record, lighting energy consumption)

[0145] Timestamp (accurate to seconds, standard format such as 2020-08-12 09:30:00)

[0146] Spatial location tags (e.g. K10+800, or GPS coordinates)

[0147] Responsible subject identification (e.g. SG-Construction Unit 01)

[0148] Semantic tags (e.g. carbon source: mechanical energy consumption or carbon benchmark: design plan)

[0149] The process can be automatically completed by the data access module or the structured conversion module, and the output is a set of standard data nodes.

[0150] Step S22, establish the engineering logic mapping relationship between nodes

[0151] After obtaining the standardized nodes, a conductive mapping path between the data nodes needs to be established according to the engineering logic, time sequence, spatial attribution, and semantic consistency. The system preloads a set of mapping rule templates, including but not limited to:

[0152] Time sequence mapping rule: if the timestamp of data node A is earlier than node B, and its life cycle stage is earlier than B, then the time sequence is satisfied;

[0153] Spatial inclusion mapping rule: if the spatial tag of node B belongs to the spatial range of node A, then a spatial subordination relationship is established;

[0154] Data type logic mapping rule: for example, the design material list can be mapped to the material arrival record, and the equipment oil consumption record can be mapped to the carbon emission estimation;

[0155] Semantic tag matching rule: if the semantic tags of upstream and downstream nodes have functional mapping (such as carbon factor calculation model), then a causal path is established.

[0156] The system executes the above rule comparison for each pair of nodes, and if all matches are matched, a directed edge is generated between the nodes, and the mapping rule source is recorded. The direction of the edge is always from the early stage to the later stage, for example, the design node points to the construction node, and the construction node points to the operation node.

[0157] Step S23, organize to form a life cycle data context tree structure

[0158] The system takes the data nodes of all design stages as the root node set. For each root node, a recursive process is executed:

[0159] Traverse all construction stage nodes to find nodes that have a legal mapping relationship with the design node;

[0160] For each matching node, a sub-node reference is established under the design node;

[0161] Repeat the process for each sub-node to continue to find data nodes in the operation stage and the demolition stage;

[0162] If a node has multiple upstream paths, the node is allowed to have multiple parent node references;

[0163] The constructed paths form a life cycle data thread path, and multiple paths constitute a complete data thread tree.

[0164] The tree structure can be stored in one of the following three ways:

[0165] Graph database representation, such as Neo4j, each node is a graph vertex, and the edge is a mapping relationship, supporting graph query;

[0166] Tree-like nested structure, such as expressing parent-child reference relationship with JSON object nesting;

[0167] Table storage form, the main table records node ID, and the associated table records mapping edge relationship (start ID, end ID, mapping type).

[0168] Step S24, check the integrity of the data thread path and mark the node state

[0169] After completing the construction of the data thread tree, the system checks the integrity of each path. Including:

[0170] Whether there is a complete path covering the design-construction-operation phase;

[0171] Whether there is a time reverse order or a space boundary mapping;

[0172] Whether there is an isolated node (no upstream or downstream);

[0173] Whether there is an unmapped logical node (such as there is carbon emission estimation but no fuel consumption data);

[0174] Whether there is a repeated mapping path (such as the same construction behavior is pointed to by two design nodes);

[0175] The system assigns a state label to each node according to the verification result, including structure integrity, structure breakage, time conflict, space misplacement, and semantic absence, and outputs the path coverage index , The calculation method is:

[0176] = Effective mapping path number / Theoretical phase chain number

[0177] Among them, the effective mapping path number represents the number of existing design-construction-operation paths, and the theoretical chain number is the number of life cycle phases defined by the project. If Less than the set threshold (such as 0.7), the system marks the data as incomplete structure.

[0178] The life cycle data context tree constructed by the above process can clearly describe the evolution path and causal logic of low-carbon data in different life cycles based on engineering stages, making the upstream and downstream structure, spatial attribution, and temporal evolution relationship between data nodes clear and visible, supporting subsequent data credibility verification, semantic consistency verification, and responsibility subject attribution analysis, solving the traditional carbon data isolation, mismatch, and broken chain problems, and improving the structural organization, traceability, and credible interaction ability of engineering carbon data.

[0179] As shown in Figure 2 a simplified specific example, in a certain G2020 road engineering project, the K10+500 to K12+000 section implements base structure overlay and surface layer updating operations. The full life cycle low-carbon data of this project includes key node information in the design, construction, operation and demolition stages, as follows:

[0180] In the design stage, Design Institute-A formed the structural design document on April 1, 2020, which defined the use of cement stabilized gravel base for the section of subgrade, with a thickness of 18 cm, and a pile number range of K10+500 to K12+000. Subsequently, on April 2, 2020, the design unit issued a material consumption list, planning to use 1600 tons of cement and 8600 tons of gravel.

[0181] Entering the construction stage, on August 10, 2020, at 8:00 am, construction unit SG01 recorded 210 tons of gravel entering at K10+800, and then recorded 75 tons of cement entering at 8:15 am. On the same day at 14:00, in the range of K10+800 to K10+900, the compaction equipment started to operate, recording diesel consumption of 20 liters, and the construction time was 6 hours. On August 15, 2020, at 10:30, the surface layer asphalt concrete paving was completed, with a calibrated thickness of 5 cm, covering the pile number section of K10+900 to K11+100. On August 16, 2020, the third-party carbon accounting unit HX01 conducted carbon emission estimation based on the above equipment operation and material use data, with a result of 42.6 tons of carbon dioxide equivalent ( ).

[0182] The operation stage started energy consumption data collection on August 1, 2021, and the maintenance unit MAINT01 recorded the energy consumption of the section of green irrigation as 9.8 kWh and the lighting electricity consumption as 41.2 kWh.

[0183] The demolition stage was established by the government unit JTG on May 1, 2024, with a demolition plan covering the K10+500 to K10+800 section of the road surface, with an estimated demolition area of 2500 square meters, and asphalt recycling was carried out. On May 10, 2024, construction unit SG01 counted the recycling data, with a result of 15.3 tons of asphalt and 21.6 tons of cement stabilized gravel base material.

[0184] In the above example, starting from D1, through multiple construction nodes, finally converging to operation and demolition data, forming a complete life cycle path.

[0185] And the logical conduction is clear:

[0186] Design structure and dosage determine material access;

[0187] Material-driven equipment operation, in turn, produces carbon emissions;

[0188] Surface layer construction data not only affects carbon accounting, but also determines future demolition behavior;

[0189] The demolition plan is based on design and construction history, while forming a closed loop of recycling data.

[0190] Each node is marked with engineering stage, time, space and data type, which helps to verify the structure path, analyze the causality and establish the credibility scoring model.

[0191] S3, before data interaction, based on the data context tree, performing multi-dimensional verification operation on the target data, the multi-dimensional verification operation includes:

[0192] Structural path verification: according to the data context tree, it is judged whether the target data node has a complete upstream logical path, if the path is missing, it is marked as structural incomplete data.

[0193] Temporal and spatial consistency verification: based on the time stamp and space label of the target data, the same type of data in its spatial neighborhood and time neighborhood is compared for semantic similarity, if it is inconsistent, an abnormal mark is triggered.

[0194] Carbon factor mapping verification: based on the carbon emission model, the data content of the target data and its upstream nodes is verified for causality calculation, to determine whether it meets the carbon emission conversion logic.

[0195] In low-carbon-oriented road engineering, data has strong causal chain, time sequence and spatial attribution in different life cycle stages. Since low-carbon data is used for supervision, verification, accountability and later optimization design, its credibility directly determines the accuracy of management behavior and the effectiveness of decision-making behavior. Therefore, before data interaction (including data calling, exchanging, sharing and fusion), multi-dimensional verification operation must be performed to ensure that the target data node is traceable in structure, consistent in time and space, and meets the carbon emission model logic requirements in semantics and causality, so as to support credible and safe data circulation mechanism.

[0196] To achieve the above goal, the present application defines three core verification dimensions, namely structural path verification, temporal and spatial consistency verification, and carbon factor mapping verification, the concept definitions are as follows:

[0197] Structural path verification refers to judging whether the target data node has an upstream data chain from design to the current stage by using the established data transmission path in the life cycle data context tree, to ensure the completeness and traceability of its generation process.

[0198] Temporal-spatial consistency verification refers to searching for similar data in the adjacent time window and spatial neighborhood of the target data according to its timestamp and spatial location label, analyzing semantic similarity and deviation degree, and judging whether it conforms to the spatial attribution and time effectiveness logic.

[0199] Carbon factor mapping verification refers to performing causal calculation and quantitative comparison on the target data and its upstream data nodes based on the preset carbon emission causal model, judging whether the target data is consistent with the physical process and process model, and excluding false, discrete or logically incorrect data injection.

[0200] In an optional specific implementation, the multi-dimensional verification step specifically includes the following sub-steps:

[0201] Step S31, performing structural path verification

[0202] For example, the life cycle path analysis is performed on the target data node, and all upstream node paths of the target data node are extracted according to the constructed data context tree structure. If any life cycle stage is missing a node, or the path is broken or mapping is incorrect, the target node is marked as structural path incomplete.

[0203] The verification method can call the structural path analysis method to return the completeness data of the path.

[0204] Step S32, performing temporal-spatial consistency verification

[0205] For example, the timestamp T and spatial location S of the target data are extracted, respectively, to extract the similar data set Dneighbor in the time neighborhood ] and the spatial neighborhood ] range, respectively. The semantic similarity value sim(D, Di) of the target data D and the neighborhood data is calculated, and if most of the similarity is lower than the preset threshold , the target data is marked as temporal-spatial abnormal.

[0206] Wherein sim(D, Di) represents a semantic similarity scoring function, which can adopt a cosine similarity calculation method based on a word embedding model, is a time window, is a spatial buffer distance, is a minimum semantic consistency threshold (such as 0.6).

[0207] Specifically, semantic text fields (e.g., data type description, behavior label, responsibility subject description, etc.) in data D and Di are input into a word embedding encoding model to generate corresponding embedding vectors VD, VDi. The word embedding model can be a trained BERT, Word2Vec, RoBERTa, or a custom encoding network.

[0208] Then, the similarity score of the two embedding vectors is calculated using the cosine similarity function, and the formula is as follows:

[0209] , wherein, represents the dot product, represents the Euclidean norm of the vector.

[0210] Step S33, performing carbon factor mapping verification

[0211] For example, the carbon emission causality function is called wherein Qi represents the amount of material or energy consumption, and EFi is the corresponding carbon emission factor. The carbon emission data Creal recorded in the target node is compared with the theoretical carbon emission Ccalc calculated according to the upstream nodes. If (error tolerance), it is judged that the data violates the causal consistency logic.

[0212] The formula variables are as follows:

[0213] Creal represents the actual recorded carbon emission value of the target data node;

[0214] Ccalc represents the theoretical carbon emission value derived according to the upstream material and equipment behavior;

[0215] Qi represents the i-th consumption item, such as diesel consumption, cement consumption, etc.;

[0216] EFi represents the i-th carbon factor;

[0217] represents the acceptable error limit, such as 3%;

[0218] f(C) is the carbon emission causality mapping function.

[0219] Through the above three types of verification steps, the target data can be analyzed from the aspects of structure chain, time and space label, and carbon emission causality logic, effectively excluding low-quality data input such as unknown source, logical conflict, time and space misplacement, or data fabrication, ensuring that data interaction behavior is based on reliable data, thereby supporting the precision and credibility of full life cycle carbon supervision.

[0220] In a specific example, in the G2020 project, assuming a trusted verification of node C5 (the carbon emission estimation node), the system first traces its structural path back:

[0221] Linking C5 upwards to C3 (equipment operation record) and C4 (paving record), and then tracing back to C1 and C2 (material arrival), it can ultimately connect to D1 and D2 (design stage nodes). The structural path is complete, and S31 verification is passed.

[0222] Subsequently, the system extracted the time neighborhood of C5 as 2020-08-16±1 days and the spatial neighborhood as K10+500 to K11+100. The average semantic similarity of adjacent carbon emission nodes was 0.89, which is higher than the threshold of 0.7, and the S32 verification was passed.

[0223] Finally, based on a C3 diesel consumption of 20 liters (carbon factor 2.68 kg) Based on the quantities of C1 and C2 materials ( / L), the system calculates the theoretical carbon emissions as 20 × 2.68 + the total carbon emissions of cement and crushed stone, estimating a total of 41.9 tons. The actual value for C5 is 42.6 tons, with an error of only 1.7%, less than the set value. S33 verification passed.

[0224] In a preferred embodiment, to verify whether low-carbon data nodes in road engineering possess a complete structural path, the system performs the following structural path verification steps. This verification method combines the logic of engineering phases, semantic relationships, and node path graph structure to confirm the structural legitimacy of multi-source low-carbon data.

[0225] The structural path verification step specifically includes the following sub-steps:

[0226] Step S311: Extract the upstream path of the target data node.

[0227] Perform a reverse path analysis operation on the target data node N (e.g., carbon emission estimation node C5 of the G2020 project). In the constructed data network tree, trace back from node N upwards and sequentially search for the direct upstream nodes that have a logical mapping relationship with it, until tracing back to the node at the beginning of the life cycle (e.g., design phase D1).

[0228] In the C5 example, the system can obtain the following path chain in sequence:

[0229] C5 (Carbon Emissions Estimation)

[0230] ←C3 (Equipment Operation)

[0231] ←C1 (Crushed stone arrival), C2 (Cement arrival)

[0232] ←D2 (Material Quantity List)

[0233] ←D1 (structure design)

[0234] The obtained upstream path chain is denoted as Pactual = [D1, D2, C1, C2, C3, C5].

[0235] Step S312, call the standard life cycle path set

[0236] The system presets a standard path set, which represents a typical life cycle structure path of road engineering low-carbon data, denoted as Pstandard. Each standard path represents a compliant data generation sequence.

[0237] In this example, the standard path is set as:

[0238] Pstd = [design stage → material arrival → equipment operation → carbon estimation]

[0239] for subsequent comparison with the actual path.

[0240] Step S313, perform semantic mapping of stage names

[0241] The semantic comparison of stage labels in the actual path Pactual and the stage labels in the standard path Pstd is performed. To adapt to different labels caused by differences in data record format, granularity, or system naming, the system introduces a semantic similarity analysis method based on word embedding models (such as Word2Vec or BERT).

[0242] Taking D2 as an example, its stage label is the material quantity list, which is compared with the material arrival in Pstd, and the system judges that it is a synonymous expression and is matched.

[0243] If the semantic similarity is higher than a preset threshold (such as 0.85), it is considered as a successful match.

[0244] Step S314, calculate the path stage coverage

[0245] The total number of successfully matched stages of Pactual and Pstd is counted, and a stage weight vector W is introduced to express the importance of different stages to the credible path.

[0246] The stage weights are set as follows:

[0247] W = {design stage: 0.3, material arrival: 0.2, equipment operation: 0.3, carbon estimation: 0.2}

[0248] In this example, C5 corresponds to a path that completely covers the four stages of Pstd, and each stage is successfully matched.

[0249] The system calculates the structure path matching score Score:

[0250] Score = (0.3 + 0.2 + 0.3 + 0.2) / 1.0 = 1.0

[0251] Step S315, determine whether the score exceeds the trust threshold

[0252] Let the trust threshold be If , then determine that the target data node structure path verification is passed, otherwise mark it as an incomplete structure path.

[0253] In this example, Score = 1.0 > 0.75, the structure path is complete, and the verification is passed.

[0254] The verification method is based on a semantic enhanced path matching mechanism, which effectively overcomes the misjudgment problem caused by differences in stage label naming, diversification of path forms or loss of non-key stages, and significantly improves the robustness and accuracy of structure path verification. By introducing a stage weight model and a path coverage score mechanism, the contribution of key stages to structure credibility can be reasonably reflected, effectively supporting path tracing and structure review requirements in the process of low-carbon data interaction in road engineering.

[0255] When performing spatio-temporal consistency verification, traditional methods mainly rely on fixed time windows , spatial buffer distance and static semantic similarity function sim(D, Di) for adjacent data comparison, which cannot effectively adapt to the following complex scenarios:

[0256] In different life cycle stages, the evolution speed and spatial diffusion pattern of data are different, and fixed neighborhood range is difficult to accurately capture reasonable similar data;

[0257] Some nodes may contain unstructured text, images, sensor sequences and other multi-modal information, and a single text embedding model cannot be compared comprehensively;

[0258] In areas with sparse spatial distribution or information density mutation, the average similarity index does not have effective distinguishing ability, and misjudgment or omission is easy to occur.

[0259] To solve the above problems, in a preferred embodiment, the spatio-temporal consistency verification method comprises the following sub-steps:

[0260] Step S321, extract the stage of the target data and set an adaptive neighborhood

[0261] According to the stage label of the target data, set its time window and spatial buffer radius Δs.

[0262] Take the carbon emission node C5 of the G2020 project as an example. C5 belongs to the construction phase, and the system sets Δt = 2 days and Δs = 500 meters to form the spatiotemporal neighborhood:

[0263] Time neighborhood = [2020-08-14, 2020-08-18]

[0264] Space neighborhood = [K10+500, K11+000]

[0265] Step S322, search for similar data sets in the neighborhood

[0266] Within the above spatiotemporal range, extract a set of low-carbon data nodes of the same type as carbon estimation, which includes, for example:

[0267] C5a: K10+600, 2020-08-15, value = 41.8 tons

[0268] C5b: K10+800, 2020-08-16, value = 42.2 tons

[0269] C5c: K11+200, 2020-08-17, value = 38.9 tons

[0270] Step S323, construct a multi-modal embedding representation

[0271] For the target data D and each neighborhood data Di, extract the following fields:

[0272] Text description (such as carbon emission estimation after the operation of compaction equipment)

[0273] Numerical parameters (such as diesel consumption, cement consumption)

[0274] Node phase semantic label (such as carbon accounting)

[0275] Use a multi-modal encoding model (such as MLP for structured fields, BERT for text, and normalization for numerical values) to map to a unified dimensional vector VD, VDi, and calculate the composite similarity:

[0276] sim(D, Di) = cosinesimilarity(VD, VDi)

[0277] Step S324, calculate the neighborhood density-driven anomaly index

[0278] Define the density-driven anomaly index DDI = m / n, where:

[0279] n is the total number of similar data points in the neighborhood;

[0280] m is the number of data points that satisfy sim(D, Di) < θ;

[0281] θ is a similarity threshold, for example 0.65.

[0282] If DDI ≥ γ (such as 0.5), it is determined that the target data is a spatio-temporal anomaly in its neighborhood.

[0283] Taking C5 as an example, if the similarity of two of the three neighborhood points is lower than 0.65, then DDI = 2 / 3 ≈ 0.67, triggering the anomaly marking.

[0284] The method comprehensively considers the stage difference, information modal complexity and spatial density change, and can accurately judge the spatio-temporal consistency in a multi-source, multi-type road engineering low-carbon data environment. Compared with the traditional method, the present application significantly improves the identification ability of semantic shift type anomaly and density anomaly, is suitable for multi-stage data scenes of construction high-frequency disturbance and operation long-period stability, and enhances the intelligence, accuracy and fault tolerance of the spatio-temporal verification mechanism.

[0285] When performing carbon factor mapping verification, the traditional method only relies on a simplified linear product formula to establish a carbon emission estimation model, which cannot accurately simulate the carbon emission deviation caused by material variety fluctuation, energy efficiency difference, construction intensity, climate conditions and other factors in real engineering. In addition, this method usually only verifies a single node, ignoring the upstream and downstream collaborative carbon conduction characteristics between nodes.

[0286] To improve the dynamics and accuracy of verification, in a preferred embodiment, the carbon factor mapping verification method specifically comprises the following steps:

[0287] Step S331, constructing a carbon causal graph of the target node

[0288] According to the upstream logical path of the target node in the data context tree, the carbon-related behavior nodes and the causal dependency relationship therebetween are extracted, and a directed graph Gcausal is constructed, wherein the nodes in the graph represent materials, equipment operation, construction tasks and the like, and the edges represent the direct or indirect influence of the same on the target carbon emission.

[0289] Taking C5 (carbon emission estimation) of G2020 project as an example:

[0290] Gcausal includes:

[0291] C1 (gravel transportation) → C3 (compaction equipment) → C5

[0292] C2 (cement mixing) → C3 → C5

[0293] Step S332, working condition correction of carbon factor

[0294] According to the parameters such as equipment model, running load, construction method and environmental temperature, the corresponding working condition database is consulted, and the carbon factor value EFi is corrected, and the dynamic carbon factor EFi' is calculated.

[0295]

[0296] wherein αi represents the working condition adjustment factor (for example, the diesel equipment in highland working condition due to low combustion efficiency,

[0297] Step S333, calculating the theoretical carbon emission value Ccalc

[0298] Using the material quantity Qi and the correction factor EFi' on all paths in the causal graph structure, the theoretical carbon emission of the target node is calculated:

[0299]

[0300] Taking C5 as an example:

[0301] The crushed stone transportation quantity Q1=5 tons, EF1=0.06t / ton, α1=1.00

[0302] The cement quantity Q2=3 tons, EF2=0.72t / ton, α2=1.00

[0303] The diesel equipment Q3=25 liters, EF3=2.68kg / liter, α3=1.10

[0304] Ccalc=5×0.06+3×0.72+(25×2.68×1.10) / 1000≈2.53 tons

[0305] Step S334, performing upstream and downstream collaborative consistency scoring

[0306] For all upstream nodes that have a causal transmission path with the target node C5, the difference between the recorded value and the model estimated value is compared one by one, and if the deviation on all chains is within the tolerable error range, the collaborative consistency score CCS is calculated:

[0307] ,…, each path error

[0308] If CCS≥ (for example, 0.85), it is determined that the target data carbon factor verification is passed.

[0309] Step S335, outputting the verification result

[0310] If CCS<γ or the error is greater than , the system marks the node as a carbon causal anomaly and refuses to participate in trusted interaction.

[0311] In this further improvement, full-path tracing and upstream and downstream collaborative verification of carbon emissions are achieved by carbon causal diagram modeling. A dynamic working condition correction factor αi is introduced to improve the adaptability of the model to actual engineering environment fluctuations. A collaborative consistency score index CCS is designed to not only test the target node itself, but also to verify the consistency and continuity of the upstream chain. This method greatly improves the anomaly detection capability and logical consistency guarantee capability, and provides high-precision support for low-carbon data credible interaction in road engineering.

[0312] S4, based on the structure path verification result, the spatio-temporal consistency verification result and the carbon factor mapping verification result, calculating the credible score of the target data, and controlling the interactive access right of the low-carbon data according to the score result.

[0313] In this step, the credible score refers to the quantitative expression result of the comprehensive credibility of the target data node in multi-dimensional verification, denoted as Scoretotal, and the numerical range is usually 0 to 1; the permission level refers to the data interaction permission type divided according to the credible score, which can include three levels of full access, limited access and denied access; the weight coefficient refers to the parameter used to represent the importance of each verification result in the score, which is respectively set as 、 And corresponding to structure path verification, spatio-temporal consistency verification and carbon factor mapping verification.

[0314] In an optional specific implementation, the credible score and access control step includes the following sub-steps:

[0315] Step S41, three verification results are summarized

[0316] The system receives and summarizes the score results of the target data node in the structure path verification, spatio-temporal consistency verification and carbon factor mapping verification, denoted as Scorestruct, Scoretemporal and Scorecarbon respectively.

[0317] Step S42, applying a weighted scoring model to calculate the total score

[0318] According to the preset weight coefficient, the total credible score Scoretotal is calculated, and the following weighted formula is adopted:

[0319]

[0320] Among them

[0321] Scorestruct represents the structure path verification score

[0322] Scoretemporal represents the spatio-temporal consistency verification score

[0323] Scorecarbon represents the carbon factor mapping verification score.

[0324] , , The weight coefficients for the three dimensions are respectively, satisfying... .

[0325] Optional weight allocation schemes include: , , Or, it may be dynamically adjusted at different stages based on the data source.

[0326] Step S43: Assign access permissions based on the scoring results.

[0327] The system determines the access permission level for the data based on the range of values ​​in Scoretotal.

[0328] like Then set to full access.

[0329] like Then set it to restricted access.

[0330] If Scoretotal < 0.60, then access is denied.

[0331] Under restricted access conditions, the system can limit the frequency of data access, block sensitive fields, or restrict access to internally only.

[0332] Through the above steps, this method realizes a quantitative assessment and permission classification mechanism for the credibility of low-carbon data, effectively reducing the risks of misjudgment, mistransmission, and trust attacks caused by low-credibility data.

[0333] This method boasts advantages such as multi-dimensional fusion, adaptive adjustment, and transparent results. It not only comprehensively considers the structural integrity, dynamic evolution, and causal consistency of data, but also flexibly adjusts weighting coefficients according to different scenarios, achieving refined trust management of the data interaction process. Compared to traditional judgment modes based on single verification, this invention significantly improves the scientific rigor, robustness, and application adaptability of data access control.

[0334] In a specific example, continuing with the G2020 project example above, suppose the three verification scores of the target node C5 are as follows:

[0335] (Complete structural path)

[0336] (Good neighborhood consistency)

[0337] (Carbon causality)

[0338] adopting weight coefficients , , , the calculation is:

[0339]

[0340] Therefore, C5 is determined as the full access level, and full-quantity interaction permission is opened in the road engineering digitization platform for various business links such as scheduling analysis, carbon assessment report, and supervision backtracking.

[0341] In another embodiment, the present application also provides a road engineering low-carbon data credible interaction system, comprising:

[0342] A life cycle division module is configured to build a full life cycle stage division model, divide the road engineering life cycle into a design stage, a construction stage, an operation stage, and a demolition stage, generate corresponding low-carbon data according to the business activities of each stage, and attach a stage identifier, a timestamp, and a spatial location label to the low-carbon data;

[0343] A data context tree construction module is configured to build a data context tree containing multiple low-carbon data nodes based on a preset engineering logic, data causal relationship, and carbon emission factor model, organize the low-carbon data nodes in different spatiotemporal ranges and life cycle stages into a life cycle data context tree with a stage transmission relationship, and any data node in the data context tree includes a source stage, a spatial location, a timestamp, a responsible subject, a semantic label, and a logical mapping relationship with an upstream node;

[0344] A multi-dimensional verification module is configured to perform structural path verification, spatiotemporal consistency verification, and carbon factor mapping verification on target data based on the data context tree before data interaction;

[0345] The multi-dimensional verification operation includes:

[0346] Structural path verification: determining whether the target data node has a complete upstream logical path according to the data context tree, and marking as structural incomplete data if the path is missing;

[0347] Spatiotemporal consistency verification: comparing the semantic similarity of the same type of data in the spatial and temporal neighborhoods of the target data based on the timestamp and spatial label of the target data, and triggering an abnormal mark if they are inconsistent;

[0348] Carbon factor mapping verification: performing causal calculation verification on the data content of the target data and its upstream nodes based on the carbon emission model to determine whether it conforms to the carbon conversion logic;

[0349] A trust score and permission control module is configured to calculate a trust score of the target data based on the three verification results, and control the interactive access permission of the low-carbon data according to the score result.

[0350] In a further improvement, the data context tree construction module comprises:

[0351] A node initialization submodule is configured to use data nodes of all design stages as a root node set;

[0352] A node recursive matching submodule is configured to perform a stage node matching operation recursively for each root node, traverse all construction stage nodes, operation stage nodes and demolition stage nodes, find nodes having a legal mapping relationship with the current node, and establish a child node reference thereunder;

[0353] A path construction submodule is configured to allow a node having multiple upstream paths to have multiple parent node references, and combine the multiple constructed paths into a complete life cycle data context tree structure.

[0354] In a further improvement, the spatiotemporal consistency verification unit in the multi-dimensional verification module comprises:

[0355] A neighborhood setting unit is configured to extract a stage to which the target data belongs and set an adaptive neighborhood, including a time field and a space field;

[0356] A neighborhood data extraction unit is configured to retrieve a same-type data set within the neighborhood range;

[0357] A semantic similarity scoring unit is configured to construct a multi-modal embedding representation, map it to a unified dimensional vector using a multi-modal encoding model, and calculate a similarity sim(D, Di), where D is the target data and Di is the i-th data point within the neighborhood;

[0358] An anomaly index calculation unit is configured to calculate a neighborhood density-driven anomaly index DDI = m / n, where n is the total number of neighborhood data points and m is the number of data points having a similarity lower than a preset threshold θ, and determine whether the target data has a spatiotemporal anomaly based on the index.

[0359] In a further improvement, the structural path verification unit in the multi-dimensional verification module comprises:

[0360] A path extraction unit is configured to extract an upstream path of the target data node;

[0361] A standard path comparison unit is configured to call a system-preset standard life cycle path set, and compare the stage labels of the nodes in the actual path with the standard paths in terms of semantic similarity;

[0362] The matching score calculation unit is configured to count the number of matching successful stages and calculate the structure path matching score in combination with the stage weight vector.

[0363] The structure integrity determination unit is configured to determine whether the structure path of the target data is complete according to the matching score result.

[0364] In a further improvement, the carbon factor mapping verification unit in the multi-dimensional verification module comprises:

[0365] The carbon causal diagram construction unit is configured to extract carbon-related behavior nodes and their causal dependency relationships according to the upstream logical path of the target node in the data context tree, and construct a carbon causal diagram structure.

[0366] The carbon factor correction unit is configured to query the working condition database according to parameters such as device model, running load, construction method and environmental temperature, and correct the original carbon emission factor value.

[0367] The theoretical carbon emission calculation unit is configured to calculate the theoretical carbon emission of the target node by using the material consumption on all causal paths and the corrected carbon factor.

[0368] The collaborative consistency analysis unit is configured to compare the actual recorded value and the theoretical estimated value of the target node and the upstream node, and if the deviation is within the tolerable error range, calculate the collaborative consistency score, and if the score is lower than the threshold or the error is too large, mark it as carbon causal abnormal data. It should be noted that the above-mentioned explanation and description of the road engineering low-carbon data credible interaction method embodiment is also applicable to the device of the present application embodiment, which will not be described here.

[0369] Those skilled in the art can realize that the units and algorithm steps described in the embodiments disclosed herein can be realized by electronic hardware, computer software and combination of electronic hardware and computer software. Whether the functions are realized by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0370] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-mentioned system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0371] In several embodiments provided in the present application, any function, if realized in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or in part or parts of the technical solutions that make contributions to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (Read-Only Memory; hereinafter referred to as: ROM), a random access memory (Random Access Memory; hereinafter referred to as: RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0372] The above is only a specific embodiment of the present application, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. The module structure of the part of the present application not specifically mentioned is subject to the content recorded in the prior art. The prior art mentioned in the foregoing background section and the specific embodiment section of the present application can be used as a part of the present application to understand the meaning of some technical features or parameters.

Claims

1. A road engineering low-carbon data credible interaction method, characterized in that, The method comprises the following steps: S1, a full life cycle stage division model is constructed, the road engineering life cycle is divided into a design stage, a construction stage, an operation stage and a demolition stage, low-carbon data corresponding to business activities in each stage is generated, and the low-carbon data is additionally provided with a stage identifier, a timestamp and a spatial position label; S2, based on a preset engineering logic, a data causal relationship and a carbon emission factor model, a data context tree comprising a plurality of low-carbon data nodes is constructed, the low-carbon data nodes in different space-time ranges and life cycle stages are organized into a life cycle data context tree having a stage transmission relationship, any data node in the data context tree comprises a source stage, a spatial position, a timestamp, a responsible subject, a semantic label and a logical mapping relationship with an upstream node; S3, before data interaction, a multi-dimensional verification operation is performed on target data based on the data context tree, the multi-dimensional verification operation comprises: structure path verification: whether the target data node has a complete upstream logical path is judged according to the data context tree, if the path is missing, the data is marked as structurally incomplete; spatio-temporal consistency verification: based on the timestamp and the spatial label of the target data, the same type of data in the spatial neighborhood and the time neighborhood of the target data is compared for semantic similarity, if the comparison result is inconsistent, an abnormality is triggered; carbon factor mapping verification: based on the carbon emission model, the data content of the target data and its upstream nodes is verified by causal calculation, whether the carbon conversion logic is met is judged; S4, a credibility score of the target data is calculated based on the structure path verification result, the spatio-temporal consistency verification result and the carbon factor mapping verification result, and the interaction access permission of the low-carbon data is controlled according to the score result; The carbon factor mapping verification comprises: According to the upstream logical path of the target node in the data context tree, the carbon-related behavior nodes and the causal dependence relationship therebetween are extracted, a directed graph is constructed, the nodes in the graph represent construction behaviors, and the edges represent the direct or indirect influence of the construction behaviors on the target carbon emission; According to the equipment model, the operation load, the construction method and the environmental temperature parameter, the corresponding working condition database is consulted, and the carbon factor value is corrected; The theoretical carbon emission amount of the target node is calculated by using the material quantity on all paths in the causal graph structure and the correction factor; The difference between the recorded value and the model estimated value of all upstream nodes having a causal transmission path with the target node is compared one by one, if the deviation on all chains is within a tolerable error range, a collaborative consistency score is calculated; If the collaborative consistency score is less than a preset score or the error is greater than a preset error, the system marks the node as a carbon causal abnormality.

2. The road engineering low-carbon data trustful interaction method according to claim 1, characterized in that, The step of constructing the data context tree comprises: All data nodes in the design stage are taken as a root node set, for each root node, a recursive process is performed: All construction stage nodes are traversed to find nodes having a legal mapping relationship with the node; For each matched node, a sub-node reference is established under the design node; The process is repeated for each sub-node to continue to find data nodes in the operation stage and the demolition stage; If a node has a plurality of upstream paths, the node is allowed to have a plurality of parent node references; The constructed path forms a life cycle data context path, and multiple paths form a complete data context tree.

3. The road engineering low-carbon data trustful interaction method according to claim 1, characterized in that, The spatiotemporal consistency verification comprises: Extracting a stage to which the target data belongs and setting an adaptive neighborhood, the adaptive neighborhood being a time and / or space field; Retrieving a same-type data set in the neighborhood; Constructing a multi-modal embedding representation, mapping to a uniform dimension vector using a multi-modal encoding model, and calculating a similarity sim(D, Di), where D is the target data and Di is the i-th field data point; Calculating a neighborhood density-driven anomaly index DDI = m / n, where: n is the total number of same-type data points in the neighborhood; m is the number of data points satisfying sim(D, Di) < θ; and θ is a similarity threshold; Determining whether the target data is abnormal in space and time in its neighborhood according to the anomaly index.

4. The road engineering low-carbon data trustful interaction method according to claim 1, characterized in that, The structure path verification comprises: Extracting an upstream path of the target data node; Calling a system-preset standard life cycle path set; Comparing the semantic similarity of the stage labels in the actual path with those in the standard path, and if the semantic similarity is higher than a preset threshold, the comparison is considered successful; Counting the total number of stages that match successfully between the actual path and the standard path, and introducing a stage weight vector to express the importance of different stages to the trusted path, and calculating a structure path matching score; Determining whether the structure path is complete according to the structure path matching score.

5. A road engineering low-carbon data credible interaction system, characterized in that, The system comprises the following modules: A life cycle division module for constructing a full life cycle stage division model, dividing the road engineering life cycle into a design stage, a construction stage, an operation stage, and a demolition stage, generating corresponding low-carbon data according to the business activities of each stage, and attaching stage labels, timestamps, and spatial location labels to the low-carbon data; A data context tree construction module for constructing a data context tree comprising multiple low-carbon data nodes based on a preset engineering logic, data causal relationship, and carbon emission factor model, organizing low-carbon data nodes in different spatiotemporal ranges and life cycle stages into a life cycle data context tree having a stage-conductive relationship, any data node in the data context tree comprising a source stage, a spatial location, a timestamp, a responsible subject, a semantic label, and a logical mapping relationship with an upstream node; A multi-dimensional verification module for performing structure path verification, spatiotemporal consistency verification, and carbon factor mapping verification on target data based on the data context tree before data interaction; The multi-dimensional verification operation comprises: Structure path verification: determining whether the target data node has a complete upstream logical path according to the data context tree, and marking the data as structurally incomplete if the path is missing; Spatiotemporal consistency verification: comparing the semantic similarity of same-type data in the spatial neighborhood and the temporal neighborhood of the target data based on the timestamp and the spatial label of the target data, and triggering an abnormality mark if the data is inconsistent; Carbon factor mapping verification: verifying the causal relationship between the data content of the target data and its upstream nodes based on a carbon emission model, and determining whether the data conforms to the carbon conversion logic; A trusted score and permission control module for calculating the trusted score of the target data based on the three verification results, and controlling the interactive access permission of the low-carbon data according to the score result. The carbon factor mapping verification unit in the multi-dimensional verification module comprises: a carbon causal diagram construction unit configured to extract carbon-related behavior nodes and their causal dependency relationships according to the upstream logical paths of the target node in the data context tree, and construct a carbon causal diagram structure; a carbon factor correction unit configured to correct the original carbon emission factor value according to the equipment model, operation load, construction method and environmental temperature parameter to query a working condition database; a theoretical carbon emission calculation unit configured to calculate the theoretical carbon emission of the target node by using the material consumption on all causal paths and the corrected carbon factor; a collaborative consistency analysis unit configured to compare the actual recorded value and the theoretical estimated value of the target node and the upstream node, and calculate a collaborative consistency score if the deviation is within the tolerable error range, and mark the carbon causal abnormal data if the score is lower than the threshold or the error is too large.

6. The road engineering low-carbon data trustful interaction system according to claim 5, characterized in that, The data context tree construction module comprises: a node initialization submodule configured to use all data nodes in the design stage as a root node set; a node recursive matching submodule configured to perform a stage node matching operation recursively on each root node, traverse all construction stage nodes, operation stage nodes and demolition stage nodes, find nodes having a legal mapping relationship with the current node, and establish a child node reference under the nodes; a path construction submodule configured to allow nodes having multiple upstream paths to have multiple parent node references, and combine the multiple constructed paths into a complete life cycle data context tree structure.

7. The road engineering low-carbon data trustful interaction system according to claim 5, characterized in that, The spatio-temporal consistency verification unit in the multi-dimensional verification module comprises: a neighborhood setting unit configured to extract the stage to which the target data belongs and set an adaptive neighborhood, including a time field and a space field; a neighborhood data extraction unit configured to retrieve a same-type data set within the neighborhood range; a semantic similarity score unit configured to construct a multi-modal embedding representation, map the multi-modal encoding model to a unified dimension vector, and calculate the similarity sim(D, Di), wherein D is the target data and Di is the i-th data point in the neighborhood; an anomaly index calculation unit configured to calculate the neighborhood density-driven anomaly index DDI = m / n, wherein n is the total number of neighborhood data points and m is the number of data points with a similarity lower than a preset threshold θ, and determine whether the target data has a spatio-temporal anomaly based on the index.

8. The road engineering low-carbon data trustful interaction system according to claim 5, characterized in that, The structure path verification unit in the multi-dimensional verification module comprises: a path extraction unit configured to extract the upstream path of the target data node; a standard path comparison unit configured to call a system-preset standard life cycle path set, and compare the stage labels of the nodes in the actual path with the standard path in terms of semantic similarity; a matching score calculation unit configured to count the number of matched stages, and calculate a structure path matching score in combination with a stage weight vector; a structure integrity determination unit configured to determine whether the structure path of the target data is complete according to the matching score result.

Images