Comparison evaluation method and system for environmental monitoring parameters based on completion process data

Abstract

The invention relates to the technical field of environment monitoring, and discloses a completion process data-based environment monitoring parameter comparison evaluation method, which comprises the following steps of: obtaining completion process data and environment monitoring data, and constructing a multi-dimensional data graph, mapping the completion process data and the environmental monitoring data to generate a standardized data set containing a construction feature vector and an environmental parameter vector; calculating a dynamic weight coefficient through a multi-objective optimization model based on the construction feature vector and the environment parameter vector in combination with the construction progress and the environment sensitivity, and outputting a weight matrix; the intelligent contract automatically triggers an evaluation process, generates a dynamic evaluation report containing an overproof event identifier, a compliance repair suggestion and a digital signature, and links an AR visualization system to be superposed to a BIM model to generate a three-dimensional thermodynamic diagram; and in combination with anomaly detection and a time sequence prediction model, identifying potential risks, generating a dynamic risk early warning map, and finally outputting a predictive evaluation report. According to the invention, the accuracy and real-time performance of environment monitoring evaluation can be improved.

Description

Technical Field

[0001] This application relates to the technical field of environmental monitoring, and in particular to a method and system for comparative evaluation of environmental monitoring parameters based on as-built process data. Background Technology

[0002] In the field of construction engineering, as-built process data such as the amount of construction materials used and process parameters are often stored in isolation, lacking a unified semantic framework for cross-source correlation with environmental parameters such as air quality and noise monitored by sensors. This makes it difficult to quantify the causal relationship between construction activities and environmental impacts, and the assessment results often rely on human experience, resulting in strong subjectivity and poor dynamic response capabilities.

[0003] The existing environmental monitoring systems have limited intelligence levels, making it difficult to achieve fully automated management of the entire process from risk identification to remediation. These issues collectively restrict the accuracy and real-time nature of environmental monitoring and assessment.

[0004] As can be seen from the above, how to improve the accuracy and real-time performance of environmental monitoring and assessment still needs to be addressed. Summary of the Invention

[0005] To improve the accuracy and real-time performance of environmental monitoring assessments, this application provides a method and system for comparing and evaluating environmental monitoring parameters based on as-built process data.

[0006] Firstly, this application provides a method for comparing and evaluating environmental monitoring parameters based on as-built process data, employing the following technical solution:

[0007] Environmental monitoring parameter comparison and evaluation methods based on completion process data include:

[0008] Acquire completion process data and environmental monitoring data. The completion process data includes the amount of construction materials used, process parameters, and configuration of environmental protection facilities. The environmental monitoring data includes real-time air quality, noise, and remote sensing image data monitored by sensors. Construct a multi-dimensional data map using semantic web technology to map the completion process data and environmental monitoring data to a unified semantic space, generating a standardized dataset containing semantic relationships. The standardized dataset includes construction feature vectors and environmental parameter vectors.

[0009] Based on the construction feature vector and the environmental parameter vector, a multi-objective optimization model is used to calculate the dynamic weight coefficients of the environmental monitoring parameters. The multi-objective optimization model combines construction progress and environmental sensitivity to generate pollution type weights and ecological impact level weights, and outputs an updated weight coefficient matrix. Among them, the construction progress includes the basic construction period and the decoration period. Different construction progress corresponds to different pollution types, and different environmental sensitivities correspond to different ecological impact level weights.

[0010] Based on the standardized dataset and the weight coefficient matrix, the data is stored on the blockchain in the form of a hash digest, and the evaluation process is automatically triggered by a smart contract. The smart contract embeds an evaluation rule engine, which generates a dynamic evaluation report based on the weight coefficient matrix and a preset comparison logic. The dynamic evaluation report includes an identifier of the non-compliance event, compliance remediation suggestions, and a digital signature. The dynamic evaluation report triggers an AR visualization system through the smart contract to overlay the non-compliance event with the BIM model, generate a three-dimensional heat map, and output rectification suggestions.

[0011] Lightweight edge computing nodes are deployed at the construction site to process sensor data in real time and perform sliding window analysis. Anomaly detection algorithms are used to identify potential environmental risks, and risk event markers are output based on these risks. Based on the construction feature vectors and the risk event markers, a time-series prediction model is used to predict the environmental parameter change trends for the next 24 hours. A dynamic risk warning map is generated based on these environmental parameter change trends. The dynamic risk warning map includes the spatial distribution of pollution hotspots and the impact paths of construction activities. The dynamic risk warning map is then linked with an AR visualization system to output a final predictive assessment report.

[0012] By adopting the above technical solutions, construction and environmental data are integrated through the Semantic Web, and influencing factors are quantitatively assessed using dynamic weight allocation. Blockchain and smart contracts are used to automate the assessment process, and AR visualization intuitively locates pollution hotspots and generates remediation suggestions. Edge computing analyzes sensor data in real time and combines it with predictive models to provide early warnings of risks, forming a closed loop of monitoring-prediction-remediation, which significantly improves the accuracy (quantitative correlation) and real-time performance (dynamic response) of environmental assessments.

[0013] Optionally, the construction feature vector and environmental parameter vector contained in the standardized dataset are generated through association rules between construction material usage and environmental protection facility configuration. The method further includes:

[0014] Extract key material types from construction material usage data and equipment models from environmental protection facility configuration data. Key material types include concrete and steel, and equipment models include dust collectors and sewage treatment ponds.

[0015] The pollutant categories in the environmental monitoring data are matched with the types of construction materials and the models of environmental protection facilities to generate a table showing the correspondence between construction activities and pollutant emissions.

[0016] Based on the aforementioned correspondence table, high-impact construction activities are selected through construction stage features, and priority ranking rules are generated in the construction feature vector. The construction stage features include the basic construction period and the decoration period.

[0017] By adopting the above technical solution, by extracting the types of construction materials (such as concrete and steel) and the models of environmental protection facilities (such as dust collectors and sewage treatment ponds), matching the pollutant categories in environmental monitoring data, establishing a correspondence table between construction activities and pollution emissions, and combining the characteristics of construction stages (such as the foundation construction period and the decoration period) to screen high-impact activities, generating priority ranking rules for construction feature vectors, thereby quantifying the correlation of the construction process's impact on the environment and improving the accuracy of assessment and the targeted nature of rectification.

[0018] Optionally, in the process of generating a 3D heatmap in an AR visualization system, the method further includes:

[0019] Spatially map the exceedance event identifiers in the dynamic assessment report to the locations of building components in the BIM model to generate coordinate information of pollution hotspot areas.

[0020] Based on the coordinate information of pollution hotspots, an AR device is used to overlay and display a pollutant concentration gradient map, where different colors represent different pollution levels.

[0021] Interactive labels with rectification suggestions are embedded in the 3D heat map. Users can directly view the corresponding compliance remediation measures and implementation timetable by clicking on the pollution hotspot area with AR devices.

[0022] By adopting the above technical solutions, the pollution levels are visualized intuitively with color gradients by mapping the pollution events to the BIM model space, combined with AR devices, and interactive rectification labels are embedded. This enables the accurate location of pollution hotspots and the real-time association of remediation measures, significantly improving the visualization efficiency and targeted nature of environmental supervision.

[0023] Optionally, the method further includes the following steps during the processing of sensor data at the edge computing node:

[0024] Deploy low-power sensor arrays at the construction site to collect noise decibel values, air quality index, and temperature and humidity data in real time.

[0025] By performing sliding window analysis on sensor data through edge computing nodes, the pollutant concentration change rate and noise fluctuation threshold over the past hour are calculated.

[0026] When the rate of change in pollutant concentration exceeds a preset threshold, a risk event marker is generated, and the marker is associated with the construction feature vector of the current construction stage and determined as the input value for the dynamic risk warning map.

[0027] By adopting the above technical solution, environmental data is collected in real time by deploying low-power sensor groups and combined with sliding window analysis of edge computing nodes to dynamically monitor pollutant concentration and noise fluctuations. Once the threshold is exceeded, a risk marker is generated and associated with the characteristics of the construction stage, providing accurate input for dynamic risk early warning and improving the real-time response and risk identification efficiency of environmental monitoring.

[0028] Optionally, the method further includes the following steps in generating the dynamic risk warning map:

[0029] By combining the material type and environmental protection facility configuration in the construction feature vector, the pollutant diffusion path and sensitive area range are predicted in the next 24 hours.

[0030] The predicted pollutant diffusion paths are spatially overlaid with the building layout in the BIM model to generate an animation of the evolution of pollution hotspots that includes a time dimension.

[0031] The AR visualization interface displays the evolution trend of pollution hotspots through dynamic heat maps and marks key impact points of construction activities on pollution spread.

[0032] By adopting the above technical solutions, the pollutant diffusion path is predicted by combining the type of construction materials and the configuration of environmental protection facilities. The resulting animation of the evolution of pollution hotspots is then overlaid with the BIM model and dynamically displayed in the AR interface, showing the trend and key impact nodes. This enables the spatiotemporal visualization analysis of pollution diffusion, providing an intuitive basis for risk prediction and precise intervention.

[0033] Optionally, the method may further include the following in the output of the final predictive evaluation report:

[0034] By comparing pollution hotspots in the dynamic risk warning map with historical environmental monitoring data, a consistency analysis of pollution trends is generated.

[0035] Based on the consistency analysis results, the construction process parameters are adjusted to generate optimized construction plan suggestions. The construction process parameters include adjusting the construction period and increasing the operating frequency of the slag removal equipment.

[0036] The optimized construction plan and rectification suggestions are integrated, and a traceable execution plan document is generated through smart contracts and pushed to the construction management platform simultaneously.

[0037] By adopting the above technical solutions, and by comparing the dynamic risk warning map with historical data to generate a consistency analysis of pollution trends, construction parameters (such as adjusting time periods and enhancing dust removal equipment) can be optimized accordingly. This results in a traceable execution plan that is synchronized to the management platform, achieving closed-loop management from risk prediction to construction optimization, thereby improving rectification efficiency and execution transparency.

[0038] Optionally, the method further includes the following during blockchain storage:

[0039] The construction feature vectors in the standardized dataset are split into independent data blocks according to the construction stage. Each data block contains the construction stage identifier and the corresponding environmental parameter vector.

[0040] Each data block is hashed by a blockchain node, and the hash digest is bound to a construction phase identifier and stored in the blockchain's distributed ledger.

[0041] Set data integrity verification rules in the smart contract. When a break in the hash value link of data blocks in adjacent construction stages is detected, the data tracing process is automatically triggered, and an anomaly record containing a timestamp and repair suggestions are output.

[0042] By adopting the above technical solution, construction feature vectors are split into independent data blocks according to stages and bound with hash digests. Combined with blockchain distributed storage and smart contract verification, staged traceability management of construction data is achieved. When a data link break is detected, tracing and repair are automatically triggered to ensure data integrity and credibility, and improve the efficiency and security of data management for environmental monitoring and assessment.

[0043] Secondly, this application provides an environmental monitoring parameter comparison and evaluation system based on completion process data, which adopts the following technical solution:

[0044] An environmental monitoring parameter comparison and evaluation system based on completion process data includes:

[0045] A standardized dataset generation module acquires as-built process data and environmental monitoring data. The as-built process data includes the amount of construction materials used, process parameters, and configuration of environmental protection facilities. The environmental monitoring data includes real-time air quality, noise, and remote sensing image data monitored by sensors. A multi-dimensional data map is constructed using semantic web technology to map the as-built process data and environmental monitoring data to a unified semantic space, which is used to generate a standardized dataset containing semantic relationships. The standardized dataset includes construction feature vectors and environmental parameter vectors.

[0046] The weight coefficient matrix update output module calculates the dynamic weight coefficients of environmental monitoring parameters based on the construction feature vector and the environmental parameter vector through a multi-objective optimization model. The multi-objective optimization model combines construction progress and environmental sensitivity to generate pollution type weights and ecological impact level weights, which are then used to output the updated weight coefficient matrix. The construction progress includes the basic construction period and the decoration period. Different construction progress corresponds to different pollution types, and different environmental sensitivities correspond to different ecological impact level weights.

[0047] The dynamic assessment report generation module stores the standardized dataset and the weight coefficient matrix on the blockchain in the form of a hash digest, and automatically triggers the assessment process through a smart contract. An assessment rule engine is embedded in the smart contract, which generates a dynamic assessment report based on the weight coefficient matrix and a preset comparison logic. The dynamic assessment report includes an identifier of the non-compliance event, compliance remediation suggestions, and a digital signature. The dynamic assessment report triggers an AR visualization system through the smart contract, overlaying the non-compliance event with a BIM model to generate a 3D heatmap and output rectification suggestions.

[0048] The final predictive assessment report output module deploys lightweight edge computing nodes at the construction site to process sensor data in real time and perform sliding window analysis. It identifies potential environmental risks using anomaly detection algorithms and outputs risk event markers based on these risks. Based on the construction feature vectors and the risk event markers, it predicts the environmental parameter change trends for the next 24 hours using a time-series prediction model. A dynamic risk warning map is generated based on these environmental parameter change trends, including the spatial distribution of pollution hotspots and the impact paths of construction activities. This dynamic risk warning map is then linked with an AR visualization system to output the final predictive assessment report.

[0049] Thirdly, this application provides an environmental monitoring parameter comparison and evaluation system based on completion process data, which adopts the following technical solution:

[0050] An environmental monitoring parameter comparison and evaluation system based on completion process data includes a processor, wherein the processor runs a program of any one of the above-described environmental monitoring parameter comparison and evaluation methods based on completion process data.

[0051] Fourthly, this application provides a storage medium, which adopts the following technical solution:

[0052] A storage medium storing a program for the environmental monitoring parameter comparison and evaluation method based on completion process data as described in any one of the above.

[0053] In summary, this application includes at least one of the following beneficial technical effects:

[0054] By integrating multi-source construction and environmental data using Semantic Web technology and combining dynamic weight allocation to quantify influencing factors, the assessment process is automated using blockchain and smart contracts. AR visualization intuitively locates pollution hotspots and generates remediation suggestions, significantly improving the accuracy of environmental assessments (quantitatively linking construction activities with environmental impacts). Edge computing analyzes sensor data in real time and combines it with predictive models to provide 24-hour advance risk warnings, forming a closed loop of monitoring-prediction-remediation, greatly improving the real-time performance of environmental monitoring (dynamic response and risk prediction).

[0055] By storing construction data on the blockchain in stages and setting smart contract verification rules, the integrity and traceability of the data are ensured; the evolution trend of pollution is displayed by combining BIM model and AR dynamic heat map to assist precise intervention; at the same time, construction parameters are optimized based on historical data and executable plans are generated to achieve efficient closed-loop management from risk identification to construction adjustment, further enhancing the scientific nature and practicality of the assessment results. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating a method for comparing and evaluating environmental monitoring parameters based on completion process data, according to an exemplary embodiment.

[0057] Figure 2 This is a structural block diagram of an environmental monitoring parameter comparison and evaluation system based on completion process data, according to an exemplary embodiment. Detailed Implementation

[0058] The embodiments of this application are described in detail below, and examples of the embodiments are shown in the accompanying drawings.

[0059] In the description of this specification, the references to "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples" refer to specific features, structures, materials, or characteristics described in connection with the described embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0060] This application discloses a method for comparing and evaluating environmental monitoring parameters based on as-built process data, referring to... Figure 1 ,include:

[0061] S100: Acquire completion process data and environmental monitoring data. The completion process data includes the amount of construction materials used, process parameters, and configuration of environmental protection facilities. The environmental monitoring data includes real-time air quality, noise, and remote sensing image data monitored by sensors. Construct a multi-dimensional data map using semantic web technology to map the completion process data and environmental monitoring data to a unified semantic space, generating a standardized dataset containing semantic relationships. The standardized dataset includes construction feature vectors and environmental parameter vectors.

[0062] In the S100 phase, two types of core data need to be systematically collected through the project management system and on-site sensor equipment:

[0063] Completion process data: Data collection on construction material usage relies on the project management system or on-site weighing equipment. For example, the usage of key materials such as concrete and steel needs to be recorded in real time. Process parameters (such as construction temperature, humidity, and pouring time) are obtained through on-site sensors or manual recording to ensure the controllability of the construction process. Information on environmental protection facility configuration (such as dust collector models and wastewater treatment tank capacity) needs to be collected in conjunction with construction drawings and equipment lists to reflect the actual deployment of environmental protection measures.

[0064] Environmental monitoring data: Air quality data is monitored in real time using sensors deployed at the construction site, with key indicators including PM2.5 concentration and VOCs (volatile organic compounds) concentration. Noise data is recorded using decibel meters to capture the acoustic impact of construction activities on the surrounding environment. Remote sensing imagery data is acquired through satellite imagery or drone aerial photography to analyze the spatial distribution and environmental change trends of the construction area, such as vegetation coverage and the extent of water pollution.

[0065] After data collection is completed, heterogeneous data needs to be mapped to a unified semantic space using semantic web technology. The specific steps are as follows:

[0066] Data mapping: Semantic Web technologies (such as RDF triples) are used to establish correlations between as-built process data and environmental monitoring data. For example, a causal relationship is established between the construction material "concrete" and the environmental parameter "PM2.5 concentration," indicating that concrete pouring may lead to an increase in PM2.5 concentration; a temporal correlation is established between the process parameter "pouring time" and the noise monitoring data "decibels," indicating that pouring operations significantly increase noise levels. This mapping relationship is achieved through triples (subject-predicate-object), such as "concrete → leads to → an increase in PM2.5 concentration."

[0067] Entity alignment: Semantic ambiguity between different data sources is eliminated through Natural Language Processing (NLP) and knowledge graph technologies. For example, "dust collector" and "dust removal equipment" are unified as the same entity to ensure data consistency and operability. Furthermore, for "sewage treatment pond" in construction drawings and "sewage treatment unit" in sensor data, context matching and synonym recognition are used to map them as the same entity, avoiding data fragmentation caused by differences in description.

[0068] After completing data mapping and entity alignment, a structured, standardized dataset needs to be generated, which includes the following three categories:

[0069] Construction Feature Vectors: Key features are extracted from the completion process data, such as material type (concrete, steel), process complexity (e.g., the precision of the pouring process), and environmental protection facility coverage (dust collector installation rate), forming structured vectors. These feature vectors need to quantify the environmental impact potential of construction activities. For example, "concrete usage (tons)" is directly related to PM2.5 emissions, and "dust collector coverage (%)" reflects particulate matter control capabilities.

[0070] Environmental parameter vector: Extracts pollutant categories (e.g., PM2.5, noise) and concentration thresholds (e.g., PM2.5 exceeding 50 μg / m³) from environmental monitoring data. 3 The data can be structured into a vector by considering factors such as temporal and spatial distribution characteristics (e.g., pollution hotspots). For example, PM2.5 concentration data can be encoded according to time (hourly) and spatial (grid area) dimensions to generate a three-dimensional "time-space-concentration" vector for subsequent dynamic weight calculations.

[0071] Semantic Relationship Table: The causal relationship between construction activities and environmental impacts is encoded into a semantic relationship table. For example, the construction activity "concrete pouring" leads to the environmental impact "increased PM2.5 concentration," and the correlation strength (e.g., high, medium, low) is recorded in the table; a strong correlation is established between the process parameter "nighttime construction" and the noise impact "decibels exceeding the standard." This table is stored in key-value pair format, such as "concrete pouring → increased PM2.5 concentration (intensity: high)," providing a quantitative basis for subsequent assessment models.

[0072] By constructing a unified multidimensional data map using semantic web technology, the S100 deeply integrates construction and environmental data, eliminates data silos, quantifies the causal relationship between construction activities and environmental impacts, and generates structured datasets to support blockchain evidence storage and dynamic risk warning, significantly improving the data credibility, analysis dimensions, and processing efficiency of environmental monitoring and assessment.

[0073] S200 calculates the dynamic weight coefficients of environmental monitoring parameters based on construction feature vectors and environmental parameter vectors through a multi-objective optimization model. The multi-objective optimization model combines construction progress and environmental sensitivity to generate pollution type weights and ecological impact level weights, and outputs an updated weight coefficient matrix. Among them, the construction progress includes the basic construction period and the decoration period. Different construction progress corresponds to different pollution types, and different environmental sensitivities correspond to different ecological impact level weights.

[0074] After defining the objective function, multiple types of input parameters need to be integrated to support the model's operation. Specifically, this includes:

[0075] Construction feature vector: Construction features, such as material type (concrete, steel), process complexity (precision of pouring), and environmental protection facility coverage (dust collector installation ratio), are extracted from the standardized dataset generated by S100. Environmental parameter vector: Pollutant concentration (PM2.5, VOCs), noise level, and remote sensing image features (vegetation coverage, water body distribution) are extracted from environmental monitoring data. Construction progress and environmental sensitivity: Construction stages (such as foundation construction period and decoration period) and environmental sensitivity (such as regional ecological value and resident density) are used as dynamically adjusted parameters input into the model. These parameters together constitute the input space of the model, providing a data foundation for subsequent optimization calculations.

[0076] After integrating the input parameters, a multi-objective optimization model needs to be constructed and constraints need to be set.

[0077] Constraints are set as follows: Construction schedule constraints, with significant differences in the intensity of environmental impact at different construction stages. For example, dust emissions are high during the foundation construction phase, requiring priority control of PM2.5; noise pollution is prominent during the decoration phase, requiring a focus on noise management; environmental sensitivity constraints, with pollution tolerance in highly sensitive areas (such as around schools and hospitals) needing to be strictly limited, for example, the pollution weight of water bodies needs to be significantly higher than that of ordinary areas.

[0078] Optimization algorithm selection: Non-dominated sorting genetic algorithm or particle swarm optimization is employed to generate Pareto optimal solution sets through iterative calculation. These algorithms can find a balance between pollution type and ecological impact, avoiding over-optimization of a single objective that could lead to the neglect of other dimensions.

[0079] Based on the calculation results of the multi-objective optimization model, dynamic weight coefficients need to be generated:

[0080] Pollution type weight allocation: The weight coefficients are dynamically adjusted based on the actual degree of harm of pollutants (such as the direct health impact of PM2.5) and the correlation strength of construction activities (such as the increase of PM2.5 concentration due to concrete pouring). For example, the weight of PM2.5 during the foundation construction period may be as high as 0.6, while the weight of VOCs is lower.

[0081] Ecological impact level weight allocation: Combining regional ecological value (e.g., wetland protected areas have higher priority than ordinary areas) and the potential destructiveness of construction activities (e.g., the compaction effect of large machinery operations on soil), a stratified weight is generated. For example, the weight of water pollution may reach 0.8 in highly sensitive areas, while the weight of soil degradation is 0.2.

[0082] Finally, the pollution type weights and ecological impact level weights are categorized according to construction stage and environmental sensitivity, outputting an updated weight coefficient matrix. For example: during the basic construction phase: PM2.5 has a weight of 0.6, noise has a weight of 0.3, and ecological impact has a weight of 0.1; in highly sensitive areas: water pollution has a weight of 0.8, and soil degradation has a weight of 0.2. The weight coefficient matrix provides a quantitative basis for subsequent dynamic assessment reports and risk warnings, ensuring that the weight allocation is highly matched with the actual construction scenario.

[0083] By dynamically adapting the multi-objective optimization model to construction scenarios and environmental sensitivity, the model scientifically quantifies the weight allocation of pollution types and ecological impacts (e.g., focusing on PM2.5 during the basic construction phase and strengthening water pollution control in highly sensitive areas). The output weight coefficient matrix provides a quantitative basis for subsequent assessment and early warning, supporting precise intervention decisions. The model covers multiple weight combinations through Pareto solutions, improving flexibility and adaptability. At the same time, automated weight allocation reduces human intervention, significantly improving the objectivity, efficiency, and resource optimization capabilities of environmental assessment.

[0084] S300 stores standardized datasets and weight coefficient matrices on the blockchain in the form of hash digests and automatically triggers the evaluation process through smart contracts. The smart contract embeds an evaluation rule engine that generates a dynamic evaluation report based on the weight coefficient matrix and preset comparison logic. The dynamic evaluation report includes an identifier of the non-compliance event, compliance remediation suggestions, and a digital signature. The dynamic evaluation report triggers an AR visualization system through the smart contract to overlay the non-compliance event with the BIM model, generate a 3D heat map, and output rectification suggestions.

[0085] The execution process of S300 specifically includes the following steps:

[0086] First, the standardized dataset generated by S100 (containing construction feature vectors and environmental parameter vectors) and the weight coefficient matrix output by S200 are processed using hash digest technology. For example, the hash value of the data is calculated using the SHA-256 algorithm to generate a unique identifier. Then, these hash digests are bound to the corresponding data identifiers (such as construction phase identifiers and timestamps) and stored in the distributed ledger through blockchain nodes. For example, the hash value of the data block for the construction phase "basic construction period" is associated with the construction feature vector and weight matrix of that phase, forming an immutable data link to ensure transparent and traceable data sources.

[0087] Deploy smart contracts in the blockchain and embed an evaluation rule engine. The rule logic includes: determining if a pollution exceeds the standard, based on the pollution type weight in the weighted coefficient matrix (e.g., PM2.5 weight 0.6) and a preset threshold (e.g., PM2.5 concentration ≥ 50 μg / m³). 3The system automatically identifies events exceeding standards and provides compliance remediation suggestions. Based on the type of exceedance (e.g., PM2.5 exceedance) and the characteristics of the construction stage (e.g., foundation construction period), it calls upon a pre-set remediation strategy library (e.g., increasing the operating frequency of dust removal equipment) to generate targeted suggestions.

[0088] After detecting an event exceeding the limit, the smart contract automatically generates a dynamic evaluation report, which includes:

[0089] Exceedance event identification: Automatically mark events that exceed standards (such as "PM2.5 concentration exceeds standard during foundation construction period") and record the timestamp and spatial coordinates;

[0090] Compliance remediation recommendations: Based on the weight matrix and rule logic, propose specific rectification measures (such as "immediately start the dust removal equipment and increase the frequency of water spraying");

[0091] Digital signature: The report is digitally signed using the private key of the blockchain account to ensure its immutability and traceability.

[0092] Furthermore, the indicators of pollution exceedances in the dynamic assessment report (such as areas exceeding PM2.5 standards) are spatially mapped to the locations of building components in the BIM model (such as concrete pouring areas) to generate coordinate information of pollution hotspots. Using AR devices (such as tablets and helmets), pollutant concentration gradient maps (such as red high-concentration areas and yellow medium-concentration areas) are overlaid onto the BIM model of the construction site to form a three-dimensional heat map. In addition, interactive labels are embedded in the heat map; users can click on pollution hotspots to view corresponding compliance remediation measures (such as "Immediately start dust removal equipment") and implementation schedules (such as "Complete adjustments within 30 minutes").

[0093] After the rectification measures are implemented, sensor data is transmitted back to the blockchain in real time via edge computing nodes, updating environmental monitoring data and triggering a new assessment process. Based on the historical rectification results, smart contracts dynamically adjust the weight coefficient matrix (e.g., reducing the weight of PM2.5 and increasing the weight of noise), promote the optimization of construction processes (e.g., adjusting construction periods), and form a closed-loop management system covering the entire process from risk identification to implementation.

[0094] By leveraging blockchain technology to store standardized datasets and weight matrices in hash digest form, the immutability and traceability of data are ensured. Combined with an evaluation rule engine within smart contracts, the system automates the identification of pollution exceedances and generates remediation recommendations, reducing traditional manual assessment time from hours to minutes. An AR system overlays pollution exceedances onto BIM models to generate 3D heatmaps, intuitively locating pollution hotspots and providing interactive remediation suggestions, significantly improving problem identification efficiency. Dynamic evaluation reports generate targeted strategies based on construction stage characteristics and pollution weights, avoiding resource waste. Finally, edge computing provides real-time feedback on remediation results, driving smart contracts to dynamically adjust the weight matrix, forming a closed-loop management system of monitoring, evaluation, and remediation. This promotes continuous optimization of construction processes and comprehensively enhances the scientific rigor and enforceability of environmental supervision.

[0095] The S400 deploys lightweight edge computing nodes at the construction site to process sensor data in real time and perform sliding window analysis. It identifies potential environmental risks through anomaly detection algorithms and outputs risk event markers based on these risks. Based on construction feature vectors and risk event markers, it predicts the environmental parameter change trends for the next 24 hours using a time-series prediction model. Based on these environmental parameter change trends, it generates a dynamic risk warning map, which includes the spatial distribution of pollution hotspots and the impact paths of construction activities. The dynamic risk warning map is then linked with an AR visualization system to output a final predictive assessment report.

[0096] The execution process of S400 specifically includes the following steps:

[0097] 1. Lightweight edge computing node deployment and data processing:

[0098] Real-time sensor data acquisition: Deploy multiple types of sensors (such as PM2.5 sensors, noise sensors, temperature and humidity sensors) at the construction site to collect environmental parameter data (such as pollutant concentration, noise level, temperature, etc.) in real time.

[0099] Edge computing node processing: Sensor data is transmitted to lightweight edge computing nodes (deployed near the construction site) for initial processing using their low latency characteristics. Edge nodes ensure data security through encrypted communication protocols, avoiding the high latency issues of centralized servers.

[0100] 2. Sliding window analysis and anomaly detection:

[0101] Sliding window construction: Using the sliding window technique, the latest data is dynamically updated in a fixed time window (e.g., 10 minutes), and statistical indicators (e.g., mean, standard deviation) within the window are calculated.

[0102] Anomaly detection algorithm execution: Based on statistical models (such as Z-score) or machine learning algorithms (such as LSTM time-series anomaly detection), it determines in real time whether the current data deviates from the normal range. For example, if the Z-score of PM2.5 concentration exceeds a threshold (such as 3) within the sliding window, it is determined to be an anomaly event.

[0103] Potential environmental risk identification: Mark abnormal events as "potential environmental risks" (such as "PM2.5 exceeding the standard") and record the timestamp and spatial coordinates (such as sensor location).

[0104] 3. Correlation between risk event markers and construction feature vectors:

[0105] Construction feature vector extraction: Extract construction features (such as material type, process complexity, and environmental protection facility coverage) from the standardized dataset generated by S100.

[0106] Risk event and construction feature mapping: Risk event labels (such as "PM2.5 exceeding the standard") are associated with construction feature vectors (such as "concrete pouring stage") to construct a risk-construction feature matrix for subsequent prediction model training.

[0107] 4. Time-series forecasting models predict trends in environmental parameter changes:

[0108] Model Selection and Training: A multivariate time series prediction model (such as LSTM) is used, based on historical environmental parameter data (such as PM2.5 concentration and noise levels over the past 7 days) and construction feature vectors, to train the model to predict the trend of environmental parameter changes over the next 24 hours. For example, it can predict whether the PM2.5 concentration will exceed a threshold (such as 50 μg / m³) in the next hour. 3 ).

[0109] Dynamic risk warning map generation: Pollution hotspot area identification, based on prediction results, marking areas that may exceed pollution standards within the next 24 hours (such as concrete mixing areas); Construction activity impact path simulation, combined with the construction schedule (such as "foundation construction period"), predicting the impact path of construction activities (such as mechanical operations) on pollution diffusion (such as the direction of dust diffusion with the wind); Map visualization, overlaying pollution hotspot areas and impact paths onto the BIM model to generate a dynamic risk warning map (such as heat map color gradients representing the severity of pollution).

[0110] 5. Output of AR visualization system linkage and predictive evaluation report:

[0111] AR system integration: This involves linking dynamic risk warning maps with AR devices (such as tablets and helmets) and overlaying them onto the BIM model of the construction site in real time. For example, users can view a 3D heat map of a "high PM2.5 concentration area" using AR devices and see the pollution diffusion path.

[0112] Interactive rectification suggestions: Interactive tags are embedded in the AR heat map. Clicking on a pollution hotspot will bring up targeted rectification suggestions (such as "start dust removal equipment and adjust construction time") and provide an execution schedule (such as "complete the adjustment within 30 minutes").

[0113] The final predictive assessment report is generated by integrating risk event markers, predicted trends, AR maps, and rectification recommendations. The report includes timestamps, spatial coordinates, and data sources, and is digitally signed using blockchain technology to ensure its immutability.

[0114] By deploying lightweight edge computing nodes and sliding window analysis technology, real-time detection of environmental parameter anomalies (such as PM2.5 exceeding standards) is achieved. Combined with construction feature vectors, pollution sources (such as the mixing plant area) are accurately located, overcoming the lag of traditional manual inspections. Based on time-series prediction models such as LSTM, dynamic trends and risk warning maps of environmental parameter changes over the next 24 hours (such as prediction of noise exceeding standards) are generated to assist construction parties in adjusting processes or time periods in advance, achieving proactive risk prevention and control. The dynamic warning maps are overlaid with BIM models through an AR system, intuitively presenting the spatial distribution and diffusion paths of pollution hotspots (such as red high-concentration areas), and providing interactive rectification suggestions (such as "start dust removal equipment"), improving decision-making efficiency and resource utilization. Combined with predictive assessment reports stored on blockchain and edge computing feedback mechanisms, the weight matrix is ​​dynamically optimized (such as reducing PM2.5 weight and increasing noise weight), promoting continuous improvement of construction processes. Finally, through automated anomaly detection, 3D visualization linkage, and closed-loop management, the cost of manual inspections is reduced by 80%, the rectification response time is shortened to minutes, and the losses from work stoppages caused by environmental risks are significantly reduced, achieving an upgrade in environmental management from passive response to proactive governance.

[0115] In this embodiment of the application, the construction feature vector and environmental parameter vector contained in the standardized dataset are generated through association rules between construction material usage and environmental protection facility configuration. The method further includes:

[0116] First, extract key construction material usage data (such as concrete and steel usage) and environmental protection facility configuration data (such as dust collector and wastewater treatment plant models) from construction plan documents, procurement lists, and equipment ledgers. Based on environmental impact assessment standards (such as the "Green Construction Guidelines"), identify material types that significantly impact environmental parameters (e.g., concrete is marked as a key material due to its high dust emissions). Simultaneously, standardize the coding of environmental protection facility models (e.g., classify dust collectors as "baghouse dust collectors" and "electrostatic precipitators") and record their processing capacity parameters (such as processing air volume and purification efficiency).

[0117] This involves matching pollutant categories from environmental monitoring data with construction material types and environmental protection facility models to generate a table showing the correspondence between construction activities and pollutant emissions. Specifically, it involves extracting pollutant categories and their concentration ranges from environmental monitoring data (such as PM2.5, noise, and wastewater COD) to construct association rules between pollutant categories and construction activities. This includes two aspects:

[0118] Material-Pollution Mapping: Establish the correspondence between construction materials and pollutants through historical data or expert rule bases (e.g., "concrete mixing → PM2.5" "steel cutting → noise").

[0119] Equipment-pollution inhibition relationship: Analyze the ability of different types of environmental protection facilities to treat pollutants (e.g., "bag filter → reduces PM2.5 concentration by 30%").

[0120] Ultimately, construction activities (such as "concrete pouring"), material types (such as "concrete"), pollutant categories (such as "PM2.5"), and environmental protection facility models (such as "baghouse dust collectors") are integrated into a single table to clarify the causal chain between construction activities and pollutant emissions, thus forming a direct mapping relationship between construction activities and environmental impact.

[0121] Based on the characteristics of the construction stage (e.g., "foundation construction period" or "decoration period"), corresponding feature parameters are loaded (e.g., the foundation construction period mainly involves earthwork excavation and concrete pouring). Using the corresponding relationship table, construction activities with high pollutant emission intensity in specific stages are selected (e.g., "concrete mixing" during the foundation construction period is marked as a high-impact activity due to high PM2.5 emissions). Based on this, the priority ranking rules for construction feature vectors are generated through the following steps:

[0122] Weighting calculation: The environmental impact weight of each construction activity is calculated based on pollutant emissions (such as PM2.5 concentration), the treatment capacity of environmental protection facilities (such as dust collector efficiency) and the frequency of construction (such as 5 times of concrete mixing per day).

[0123] Sorting output: Priority sorting rules for generating construction feature vectors in descending order of weight (e.g., "concrete mixing > rebar tying > formwork installation") are used for dynamic weight allocation in subsequent risk assessment to ensure that the model focuses on high-impact aspects.

[0124] By establishing correlation rules between materials and pollution and equipment and pollution, the causal relationship between construction activities (such as concrete mixing) and pollutant categories (such as PM2.5) can be accurately located. The priority of feature vectors can be dynamically optimized in combination with the characteristics of construction stages, focusing on high-impact links (such as PM2.5 emissions during foundation construction), quantifying the treatment efficiency of environmental protection facilities (such as dust collectors), effectively reducing data redundancy and improving the accuracy of prediction models (such as S400 time series prediction). Finally, the priority ranking rules guide the construction party to optimize resource allocation (such as deploying dust removal equipment), realizing closed-loop management from risk identification to green construction decision-making, significantly reducing environmental risks and promoting continuous optimization of construction technology.

[0125] In this embodiment of the application, the method further includes the following steps in the process of generating a three-dimensional heat map using an AR visualization system:

[0126] 1. Spatial Mapping of Exceeding Standards Events to the BIM Model: First, extract the exceeding standard event identifiers (e.g., "PM2.5 exceeding standard area") and their spatial coordinates (e.g., sensor location or pollution source coordinates) from the dynamic assessment report, and load the location data of building components in the BIM model (e.g., concrete pouring areas, pipe layout locations). Align the coordinates of the exceeding standard events with the coordinate system of the BIM model using WGS84 or UTM coordinate transformation. Then, use spatial analysis tools (e.g., ArcGIS's Near tool) to match the nearest neighbor relationships between the exceeding standard events and BIM components to determine the precise location of pollution hotspot areas (e.g., "ventilation vents near the concrete mixing plant"). Finally, output the coordinate set of pollution hotspot areas (e.g., [longitude, latitude, altitude]) as the base data for subsequent AR overlay.

[0127] 2. AR Overlay Display of Pollutant Concentration Gradient Maps: Based on pollutant concentration thresholds, levels are categorized. Using ArcGIS heatmap creation methods (such as kernel density analysis or interval segmentation), color gradient schemes (e.g., red → yellow → green) are assigned to different concentration levels. The coordinate information and color coding rules of pollution hotspots are imported into the AR device, and the pollutant concentration gradient map is overlaid in real-time onto the BIM model of the construction site using AR software. For example, high-concentration red areas will float above the concrete mixing plant, visually presenting the distribution of pollution intensity.

[0128] 3. Embedded Interactive Rectification Suggestion Tags: Interactive tags (such as floating icons or semi-transparent boxes) are embedded in the pollution hotspots (e.g., red areas) of the 3D heat map. Each tag is associated with a corresponding compliance remediation measure (e.g., "Start dust removal equipment") and an execution timeline (e.g., "Complete the adjustment within 30 minutes"). Users can trigger tag interaction via AR devices (e.g., finger taps or voice commands) to view detailed rectification suggestions (e.g., measure: "Increase the operating frequency of dust removal equipment," execution time: "Start immediately, complete within 30 minutes," and the responsible person designated as "Construction Team A"). Simultaneously, the tag information supports dynamic updates (e.g., adjusting suggestion content based on sensor feedback) to ensure that rectification suggestions remain synchronized with real-time data.

[0129] By deeply integrating AR devices with BIM models, precise location of pollution sources (such as ventilation openings in concrete mixing plants) and intuitive visualization of diffusion paths (color gradient maps) can be achieved. Interactive labels can be used to link remedial measures (such as activating dust removal equipment) with high-risk areas, significantly improving the operability of remedial suggestions and the efficiency of resource utilization. At the same time, the real-time overlay function of AR devices reduces the time from problem discovery to execution response to minutes, reducing communication costs and environmental risks. Furthermore, by using sensor data to drive dynamic updates of heat maps and storing historical records on the blockchain, a closed-loop management system of monitoring, assessment, and remediation is formed, ultimately achieving proactive prevention and control of environmental risks and continuous optimization of construction processes.

[0130] In this embodiment of the application, the method further includes the following steps during the processing of sensor data by the edge computing node:

[0131] 1. Deploy low-power wireless sensor arrays in key areas of the construction site (such as the batching plant and concrete pouring area) to ensure coverage of environmental parameters such as noise, air quality (e.g., PM2.5, VOCs), and temperature and humidity. Specific data collection modes are as follows:

[0132] Noise monitoring: The noise decibel value is collected in real time through a decibel sensor, and the data is recorded once per minute to capture noise fluctuations generated by construction activities (such as concrete mixing and steel bar cutting).

[0133] Air quality monitoring: The system uses multi-gas sensors to collect the Air Quality Index (AQI), covering indicators such as PM2.5 and carbon dioxide. Data is recorded every 5 minutes to monitor changes in the concentration of dust and harmful gases.

[0134] Temperature and humidity monitoring: Real-time changes in ambient temperature and humidity are recorded using temperature and humidity sensors, with sampling every 5 minutes to assist in the analysis of the impact of the construction environment on the diffusion of pollutants.

[0135] It should be noted that, in order to reduce energy consumption and extend sensor battery life, data transmission uses low-power communication protocols (such as MQTT or LoRaWAN technology) and only transmits critical data (such as outliers) rather than the original full data.

[0136] 2. Edge computing nodes process sensor data in real time using a sliding window analysis method:

[0137] Time window configuration: Set the sliding window duration to 1 hour and the step size to 10 minutes to ensure the continuity and real-time nature of data aggregation.

[0138] Data aggregation: Calculate statistical indicators (such as mean, maximum, and minimum values) for sensor data within every 10-minute window to identify short-term fluctuation trends in environmental parameters.

[0139] Pollutant concentration change rate calculation: This calculates the percentage change in concentration by comparing the pollutant concentration values ​​in the current window with those in the previous window. For example, if the PM2.5 concentration changes from 40 μg / m³... 3 Increased to 50 μg / m 3 The rate of change is 25%.

[0140] Noise fluctuation threshold determination: A preset alarm threshold for noise fluctuation is set (e.g., triggering an alarm when the decibel value fluctuates by more than 5dB per minute), and the root mean square (RMS) value of the noise fluctuation within the current window is calculated to determine whether the threshold is exceeded.

[0141] 3. When the rate of change in pollutant concentration or noise fluctuation is detected to exceed a preset threshold, the edge node generates a risk event marker and performs the following operations:

[0142] Matching with construction stage: Load the corresponding construction feature vector (such as the priority sorting rules defined in S100) according to the current construction stage (such as "foundation construction period" or "decoration period") to clarify the main construction activities of this stage (such as concrete mixing and formwork installation).

[0143] Causal chain mapping: Combining material-pollution association rules, risk event tags are linked to specific construction activities, revealing the direct correlation between pollution sources and construction processes.

[0144] Input to the dynamic risk warning map: The tagged events are integrated with the construction feature vectors as input values ​​for the S400 dynamic risk warning map. For example, "PM2.5 exceeding the standard during the concrete mixing stage" will trigger a red warning area and push rectification suggestions (such as "start the dust removal equipment") in real time through the AR system (such as S300).

[0145] By deploying low-power sensor arrays, minute-level monitoring of noise, air quality, temperature, and humidity at the construction site is achieved. Combined with sliding window analysis, abnormal pollution fluctuations are accurately identified. Local computing at edge nodes reduces data transmission costs and improves processing efficiency. Simultaneously, risk events are dynamically correlated with construction feature vectors to clarify the causal relationship between pollution sources and construction activities, significantly enhancing the pertinence and speed of early warnings. Rectification suggestions can be pushed in real time through the AR system, reducing response time to minutes. Finally, relevant data can be connected to the blockchain system to support the traceability of risk events and continuous optimization of construction processes, forming a complete closed loop for environmental monitoring and management.

[0146] In this embodiment of the application, the method further includes the following steps in generating the dynamic risk warning map:

[0147] 1. Predict pollutant diffusion paths and sensitive areas by combining construction feature vectors:

[0148] By integrating key information from the construction feature vector (such as material types like concrete and steel, and environmental protection facility configurations such as dust collector models and coverage areas), and combining historical sensor data (such as PM2.5 concentration changes and noise fluctuation thresholds) and real-time meteorological data (such as wind speed and wind direction), the initial conditions for pollutant diffusion are constructed.

[0149] Subsequently, diffusion prediction methods such as computational fluid dynamics (CFD) or Gaussian plume models are used to simulate the diffusion paths of pollutants over the next 24 hours (e.g., the trend of PM2.5 diffusion from the concrete mixing plant to the surrounding area). Simultaneously, based on material release characteristics (e.g., the size of dust particles generated from concrete mixing) and the suppression effect of environmental protection facilities (e.g., dust collector coverage), the scope of sensitive areas (e.g., residential areas, areas surrounding schools) is further predicted. Finally, a dataset of pollutant diffusion paths containing timestamps is generated (e.g.,

[0150] The peak PM2.5 concentration occurred in Area A at 08:00:00 on July 28, 2025, providing basic data support for subsequent spatial overlay and dynamic display.

[0151] 2. Overlay with BIM model to generate animation of pollution hotspot evolution:

[0152] During the BIM model loading phase, the 3D model of the construction site, including the building layout and pipeline distribution, is imported, and its coordinate system is aligned with the predicted data. Subsequently, the pollutant diffusion path data is spatially matched with the physical components (such as walls and ventilation openings) in the BIM model to accurately identify pollution hotspots (e.g., "PM2.5 concentration reaches its peak at the east wall of the concrete mixing plant").

[0153] By dynamically updating the location and concentration values ​​of pollution hotspots (e.g., updating the diffusion status hourly), the spatial distribution of pollutants over time is gradually constructed. Finally, using 3D modeling software (such as Navisworks) or AR development tools (such as Unity), an animation of the evolution of pollution hotspots with a time dimension is generated (e.g., a red heatmap spreading from the mixing plant to the surrounding area over time), visually demonstrating the spread trend of pollutants and their impact on the surrounding environment.

[0154] 3. Display dynamic heatmaps and key impact node annotations on the AR interface:

[0155] Using AR devices, the evolution animation of pollution hotspots is rendered in real time onto the BIM model of the construction site, forming a dynamic heat map. The color coding rules of the heat map are consistent with the pollutant concentration gradient (e.g., red represents high concentration, yellow represents medium concentration, and green represents normal), intuitively reflecting the change in pollution intensity over time.

[0156] Building upon this foundation, interactive labels (such as floating icons) are added to key nodes along the pollution diffusion path (e.g., locations of sudden increases in PM2.5 concentration) to indicate the specific impact of construction activities on pollution diffusion (e.g., "Extended concrete mixing time leads to an expansion of dust diffusion range"). Users can click on the labels using AR devices to view detailed impact analyses (e.g., "PM2.5 concentration increases by 20% in areas not covered by the mixing plant") and optimization suggestions (e.g., "Adjust mixing time to periods of lower wind speed"), thereby achieving precise intervention and dynamic management of pollution sources and construction processes.

[0157] In this embodiment of the application, the method further includes the following steps in the output of the final predictive evaluation report:

[0158] 1. Compare the pollution hotspots in the dynamic risk warning map with historical environmental monitoring data to generate pollution trend consistency analysis results:

[0159] Key pollution areas, such as the PM2.5 peak area on the east side of the concrete mixing plant's perimeter wall, are extracted from the pollution hotspot evolution animation generated by S300. Historical environmental monitoring data from the past week to month is also retrieved, including sensor-recorded PM2.5 concentrations, noise levels in decibels, and meteorological data (such as wind speed and direction). Using a spatial matching method, the coordinates of the pollution hotspots in the dynamic map are aligned with the high-pollution areas in the historical data to verify overlap. For example, if the currently predicted PM2.5 peak area matches the location of the high-pollution area on the east side of the perimeter wall recorded in historical data, it indicates that the prediction results have high spatial reliability.

[0160] Next, the trends in pollutant concentrations in the same area were compared with historical data. For example, the PM2.5 peaks during the concrete mixing period in the past week all occurred on the east side of the wall, and the current predicted peak area is consistent with this pattern. However, the sudden increase in PM2.5 concentration at the north ventilation opening was not seen in historical data, which may indicate a new source of pollution or changes in meteorological conditions.

[0161] Further analysis using regression analysis or machine learning models (such as LSTM) is conducted to assess the deviation between predicted and historical values. For example, a current predicted PM2.5 peak value 15% higher than the historical average may indicate a decline in the efficiency of environmental protection facilities or an increase in construction intensity. Finally, a pollution trend consistency analysis report is generated, clearly identifying areas of high consistency (e.g., the PM2.5 peak value area matching rate of 90% on the east wall) and abnormal areas (e.g., the sudden increase in PM2.5 concentration at the north ventilation opening that did not appear in historical data).

[0162] 2. Adjust the construction process parameters based on the consistency analysis results to generate optimized construction plan suggestions:

[0163] For areas with high consistency (such as the area with peak PM2.5 concentrations on the east side of the perimeter wall), optimize existing construction parameters to reduce risks. For example, adjust concrete mixing times to nighttime when wind speeds are lower to reduce the range of dust dispersion. For abnormal areas (such as areas with sudden increases in PM2.5 concentrations at the north ventilation opening), increase environmental protection intervention. For example, install dust removal equipment at the north ventilation opening and increase its operating frequency to three times per hour to suppress sudden pollution. Specific adjustments include: optimizing construction time based on meteorological data (such as wind speed and humidity), for example, avoiding outdoor welding operations during midday when ozone levels are high, and choosing nighttime when wind speeds are lower for high-dust construction; increasing the operating frequency of dust removal equipment (such as adjusting the dust collector at the mixing plant from twice per hour to three times per hour), or replacing filters with more efficient ones (such as using HEPA filters); and adopting low-dust construction technologies (such as using wet spraying instead of dry mixing) to reduce pollutant emissions at the source.

[0164] The final optimized construction plan recommendations are generated, such as "adjusting the concrete mixing period to 22:00-06:00 and increasing the dust removal equipment operation frequency to 3 times per hour," along with an explanation of the expected effects (e.g., "the peak PM2.5 concentration is expected to decrease by 20%"). This plan needs to be tailored to the specific construction scenario to ensure a balance between feasibility and economy.

[0165] 3. Integrate the optimized construction plan suggestions and rectification suggestions, generate a traceable execution plan document through smart contracts, and simultaneously push it to the construction management platform:

[0166] The optimized construction plan (such as adjusting the mixing time) is combined with the rectification suggestions generated by S400 (such as starting dust removal equipment) to form a complete implementation strategy. For example, during the nighttime mixing period, the dust removal equipment should be started simultaneously and the running time should be recorded to ensure that construction and environmental protection measures are implemented in sync.

[0167] Subsequently, the execution strategy is translated into smart contract terms. For example, "When the mixing period is from 22:00 to 06:00, automatically trigger the dust removal equipment operation command." Contract terms and execution records (such as "Dust removal equipment started at 22:00 on July 28, 2025") are stored using blockchain technology to ensure data immutability and provide a basis for subsequent audits. An execution plan document is generated, including execution steps, responsible persons, and time nodes, for example, "From 22:00 to 06:00 on July 29, the person in charge of mixing plant A needs to start the dust removal equipment and record its operating status every hour." The document must clearly define the division of responsibilities and time nodes to facilitate on-site implementation.

[0168] Finally, the execution plan document is pushed to the construction management platform, linking it with the BIM model and schedule. For example, the operating periods of dust removal equipment are marked in the BIM model, ensuring that construction personnel can intuitively obtain execution information. The platform automatically verifies the execution status through sensor data (such as the operating status of dust removal equipment). For example, if it detects that the dust removal equipment has not started as planned, it triggers an alarm and notifies the person in charge, ensuring that the execution strategy is not ignored.

[0169] Through the above three steps, the system achieves closed-loop management from pollution trend analysis to construction plan optimization and execution supervision. Pollution trend consistency analysis ensures the reliability of prediction results, construction parameter adjustments specifically reduce environmental risks, and the linkage between smart contracts and the construction management platform ensures the transparency and traceability of execution strategies, ultimately forming a scientific and efficient environmental management solution.

[0170] In this embodiment of the application, the method further includes the following during blockchain storage:

[0171] 1. Split the construction feature vectors in the standardized dataset into independent data blocks according to the construction stage.

[0172] Standardized datasets are obtained from the construction management platform. These datasets include construction feature vectors (such as material type and process parameters) and corresponding environmental parameter vectors (such as PM2.5 concentration and noise levels). For example, the construction feature vector might record that "the concrete mixing plant operates from 8:00 to 12:00 daily," while the environmental parameter vector might include that "the peak PM2.5 time is 10:30 daily."

[0173] The splitting logic is based on construction phases, dividing the dataset into multiple independent data blocks. Each data block contains two core components: a construction phase identifier and an environmental parameter vector. The construction phase identifier is a unique identifier, such as "Foundation Construction - 2025Q3" or "Main Structure - 2025Q4," used to clearly identify the construction phase and time range to which the data block belongs. The environmental parameter vector records all monitored pollutant concentrations, equipment operating status, and other data within that phase. For example, in the foundation construction phase, a data block might contain information such as "The mixing plant operates from 8:00 to 12:00 daily, the dust removal equipment operates twice per hour, and the peak PM2.5 concentration occurs at 10:30 daily"; while in the main structure phase, a data block might contain information such as "Welding operations occur from 14:00 to 18:00 daily, the peak noise level is 75 decibels, and the ozone concentration trend is an afternoon increase."

[0174] This data splitting method structures the data into modules directly related to the construction progress, facilitating subsequent phased verification and traceability. For example, when analyzing whether there are PM2.5 exceedances during the foundation construction phase, the corresponding data block can be located directly without traversing the entire dataset.

[0175] 2. Each data block is hashed by blockchain nodes, and the hash digest is bound to the construction phase identifier and stored in the distributed ledger of the blockchain.

[0176] Blockchain nodes perform hash calculations on the content of each data block (including construction phase identifiers and environmental parameter vectors). For example, the SHA256 algorithm is used to generate a fixed-length hash digest for the data block of the "Basic Construction - 2025Q3" phase. The hash digest is immutable: if the content of the data block is modified (for example, by manually changing the peak time of PM2.5), the hash value will change significantly, thus being identified as an anomaly by the system.

[0177] Next, the hash digest will form a "stage identifier → hash value" mapping with the construction stage identifier and be stored in the blockchain's distributed ledger. The characteristics of a distributed ledger are data redundancy and immutability: each node maintains a complete copy of the ledger, and once stored, it cannot be modified by a single point of view. For example, the hash digest of the "basic construction - 2025Q3" stage will be bound and stored in block A, while the hash digest of the "main structure - 2025Q4" stage will be bound and stored in block B.

[0178] In this way, data integrity at each construction stage is guaranteed. For example, if it is necessary to verify whether the data in the "Foundation Construction - 2025Q3" stage has been tampered with, it is only necessary to recalculate the hash value of that data block and compare it with the hash digest of the "Foundation Construction - 2025Q3" stage stored in the blockchain. If they do not match, it means that the data has been modified.

[0179] 3. Set data integrity verification rules in the smart contract. When a break in the hash value link of data blocks in adjacent construction stages is detected, the data tracing process is automatically triggered.

[0180] The smart contract pre-defines a set of data integrity verification rules: the hash values ​​of data blocks from adjacent construction stages must be linked in a chain. For example, the hash value of block A (foundation construction stage) is the basis for generating block B (main structure stage), forming a hash link between the two. If the hash link between adjacent stages is broken (for example, the hash value of block B cannot be calculated from the content of block A), a data tracing process is triggered.

[0181] When blockchain nodes verify the hash links of adjacent stages in real time, if an anomaly is detected (for example, the hash value of block B does not match the expected value), the smart contract will automatically initiate a source tracing analysis. The source tracing analysis will call historical data in the blockchain ledger, such as the original hash values ​​of blocks A and B, to locate the time point when the anomaly occurred (e.g., "2025-07-28T14:00").

[0182] The output consists of two parts: anomaly logs and remediation suggestions. Anomaly logs generate detailed reports with timestamps, such as "2025-07-28T14:00, main structure phase data block hash link broken." Remediation suggestions propose specific solutions based on the anomaly type, such as "re-upload environmental parameter data for the main structure phase and regenerate hash digests." Remediation suggestions are simultaneously pushed to the construction management platform to ensure timely problem resolution.

[0183] For example, if the hash value of a data block in the main structure phase is tampered with, causing a data link break, the smart contract will suggest "restoring the original environmental parameter data and regenerating the hash digest," and push this to the construction management platform for relevant personnel to perform the repair operation. This process ensures the continuity of the data link and the transparency of the construction process.

[0184] The following will illustrate this with examples:

[0185] During the final acceptance phase of a construction project in an industrial park, a dynamic comparison model was constructed by integrating environmental monitoring data from the construction process (such as PM2.5 concentration during foundation construction and noise levels during welding operations) with real-time monitoring data from the operation phase (such as wastewater discharge and ozone concentration changes). For example, during construction, a peak PM2.5 concentration of 150 μg / m³ was detected on the east side of the mixing plant's perimeter wall. 3 During the operational phase, the average PM2.5 level in the same area decreased to 80 μg / m³. 3 This indicates that optimized dust removal equipment operating parameters during the construction phase (such as operation during low wind speed periods at night) effectively reduced pollution. Through this comparison, park regulators can intuitively identify the effectiveness of environmental protection measures and propose targeted rectification suggestions for parameters that do not meet standards (such as a sudden increase in ozone concentration at the north ventilation opening), such as installing activated carbon adsorption devices or adjusting the layout of ventilation openings, thereby improving the precision of environmental governance.

[0186] During project operation, the system uses blockchain technology to store hash digests of environmental parameters during the construction phase (such as the PM2.5 hash value for "Foundation Construction - 2025Q3") and verifies the link with real-time data from the operation phase. For example, if a break in the hash value link of noise monitoring data during the main structure phase is detected (such as an abnormal increase in noise levels during welding operations), the system automatically triggers a source tracing process, locates the abnormal record for the period "2025-07-28T14:00", and pushes remediation suggestions to the management platform. Simultaneously, it dynamically adjusts construction periods based on meteorological data (such as wind speed and humidity), for example, shifting high-noise operations to low-wind-speed periods at night to reduce the impact on the surrounding environment. This real-time response mechanism significantly improves the efficiency of environmental problem detection and resolution, avoiding pollution spread due to delayed feedback.

[0187] By constructing an environmental function model, the system predicts the pollution risk of adjacent plots. For example, if a residential area is planned to the west of the project, the system predicts the potential soil pollution risk the residential area may face in the future based on soil monitoring data (such as heavy metal residue levels) at the completion stage and historical pollution data of surrounding plots, and suggests isolation measures or soil remediation solutions in advance. Furthermore, by analyzing the consistency of trends in construction and operation data (such as whether the operating efficiency of wastewater treatment facilities declines over time), the system provides early warnings of potential environmental hazards, avoiding reactive rectification due to data lag.

[0188] This application discloses an environmental monitoring parameter comparison and evaluation system based on as-built process data, referring to... Figure 2 ,include:

[0189] The standardized dataset generation module 001 acquires completion process data and environmental monitoring data. The completion process data includes the amount of construction materials used, process parameters, and configuration of environmental protection facilities. The environmental monitoring data includes real-time air quality, noise, and remote sensing image data monitored by sensors. A multi-dimensional data graph is constructed using semantic web technology to map the completion process data and environmental monitoring data to a unified semantic space, which is used to generate a standardized dataset containing semantic relationships. The standardized dataset contains construction feature vectors and environmental parameter vectors.

[0190] The weight coefficient matrix update output module 002 calculates the dynamic weight coefficients of environmental monitoring parameters based on the construction feature vector and environmental parameter vector through a multi-objective optimization model. The multi-objective optimization model combines construction progress and environmental sensitivity to generate pollution type weights and ecological impact level weights, which are then used to output the updated weight coefficient matrix. Among them, the construction progress includes the basic construction period and the decoration period. Different construction progress corresponds to different pollution types, and different environmental sensitivities correspond to different ecological impact level weights.

[0191] The dynamic assessment report generation module 003 stores the standardized dataset and weight coefficient matrix on the blockchain in the form of a hash digest, and automatically triggers the assessment process through a smart contract. The smart contract embeds an assessment rule engine, which generates a dynamic assessment report based on the weight coefficient matrix and preset comparison logic. The dynamic assessment report includes an identifier of the non-compliance event, compliance remediation suggestions, and a digital signature. The dynamic assessment report triggers an AR visualization system through the smart contract, overlaying the non-compliance event with the BIM model to generate a 3D heat map and output rectification suggestions.

[0192] The final predictive assessment report output module 004 deploys lightweight edge computing nodes at the construction site to process sensor data in real time and perform sliding window analysis. It identifies corresponding potential environmental risks through anomaly detection algorithms and outputs risk event markers based on these risks. Based on construction feature vectors and risk event markers, it predicts the corresponding environmental parameter change trends for the next 24 hours through a time-series prediction model. Based on these environmental parameter change trends, it generates a dynamic risk warning map, which includes the spatial distribution of pollution hotspots and the impact paths of construction activities. The dynamic risk warning map is then linked with an AR visualization system to output the final predictive assessment report.

[0193] This application also discloses an environmental monitoring parameter comparison and evaluation system based on completion process data, including a processor, wherein the processor runs a program of any one of the above-described environmental monitoring parameter comparison and evaluation methods based on completion process data.

[0194] This application also discloses a storage medium storing a program for the environmental monitoring parameter comparison and evaluation method based on completion process data as described in any one of the above embodiments.

[0195] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for comparing and evaluating environmental monitoring parameters based on as-built process data, characterized in that, include: Acquire completion process data and environmental monitoring data. The completion process data includes the amount of construction materials used, process parameters, and configuration of environmental protection facilities. The environmental monitoring data includes real-time air quality, noise, and remote sensing image data monitored by sensors. Construct a multi-dimensional data map using semantic web technology to map the completion process data and environmental monitoring data to a unified semantic space, generating a standardized dataset containing semantic relationships. The standardized dataset includes construction feature vectors and environmental parameter vectors. Based on the construction feature vector and the environmental parameter vector, a multi-objective optimization model is used to calculate the dynamic weight coefficients of the environmental monitoring parameters. The multi-objective optimization model combines construction progress and environmental sensitivity to generate pollution type weights and ecological impact level weights, and outputs an updated weight coefficient matrix. The construction progress includes the basic construction period and the decoration period. Different construction progress corresponds to different pollution types, and different environmental sensitivities correspond to different ecological impact level weights. Based on the standardized dataset and the weight coefficient matrix, the data is stored on the blockchain in the form of a hash digest, and the evaluation process is automatically triggered by a smart contract. The smart contract embeds an evaluation rule engine, which generates a dynamic evaluation report based on the weight coefficient matrix and a preset comparison logic. The dynamic evaluation report includes an identifier of the non-compliance event, compliance remediation suggestions, and a digital signature. The dynamic evaluation report triggers an AR visualization system through the smart contract to overlay the non-compliance event with the BIM model, generate a three-dimensional heat map, and output rectification suggestions. Lightweight edge computing nodes are deployed at the construction site to process sensor data in real time and perform sliding window analysis. Anomaly detection algorithms are used to identify potential environmental risks, and risk event markers are output based on these risks. Based on the construction feature vectors and the risk event markers, a time-series prediction model is used to predict the environmental parameter change trends for the next 24 hours. A dynamic risk warning map is generated based on these environmental parameter change trends. The dynamic risk warning map includes the spatial distribution of pollution hotspots and the impact paths of construction activities. The dynamic risk warning map is then linked with an AR visualization system to output a final predictive assessment report.

2. The method for comparing and evaluating environmental monitoring parameters based on completion process data according to claim 1, characterized in that, The standardized dataset contains construction feature vectors and environmental parameter vectors generated through association rules between construction material usage and environmental protection facility configuration. The method also includes: Extract key material types from construction material usage data and equipment models from environmental protection facility configuration data. Key material types include concrete and steel, and equipment models include dust collectors and sewage treatment ponds. The pollutant categories in the environmental monitoring data are matched with the types of construction materials and the models of environmental protection facilities to generate a table showing the correspondence between construction activities and pollutant emissions. Based on the aforementioned correspondence table, high-impact construction activities are selected through construction stage features, and priority ranking rules are generated in the construction feature vector. The construction stage features include the basic construction period and the decoration period.

3. The method for comparing and evaluating environmental monitoring parameters based on completion process data according to claim 1, characterized in that, In the process of generating 3D heatmaps in AR visualization systems, the methods also include: Spatially map the exceedance event identifiers in the dynamic assessment report to the locations of building components in the BIM model to generate coordinate information of pollution hotspot areas. Based on the coordinate information of pollution hotspots, an AR device is used to overlay and display a pollutant concentration gradient map, where different colors represent different pollution levels. Interactive labels with rectification suggestions are embedded in the 3D heat map. Users can directly view the corresponding compliance remediation measures and implementation timetable by clicking on the pollution hotspot area with AR devices.

4. The method for comparing and evaluating environmental monitoring parameters based on completion process data according to claim 1, characterized in that, The method also includes the following in the process of edge computing nodes processing sensor data: Deploy low-power sensor arrays at the construction site to collect noise decibel values, air quality index, and temperature and humidity data in real time. By performing sliding window analysis on sensor data through edge computing nodes, the pollutant concentration change rate and noise fluctuation threshold over the past hour are calculated. When the rate of change in pollutant concentration exceeds a preset threshold, a risk event marker is generated, and the marker is associated with the construction feature vector of the current construction stage and determined as the input value for the dynamic risk warning map.

5. The method for comparing and evaluating environmental monitoring parameters based on completion process data according to claim 1, characterized in that, The method also includes the following steps in generating the dynamic risk warning map: By combining the material type and environmental protection facility configuration in the construction feature vector, the pollutant diffusion path and sensitive area range are predicted in the next 24 hours. The predicted pollutant diffusion paths are spatially overlaid with the building layout in the BIM model to generate an animation of the evolution of pollution hotspots that includes a time dimension. The AR visualization interface displays the evolution trend of pollution hotspots through dynamic heat maps and marks key impact points of construction activities on pollution spread.

6. The method for comparing and evaluating environmental monitoring parameters based on completion process data according to claim 1, characterized in that, The methodology also includes the following in the output of the final predictive assessment report: By comparing pollution hotspots in the dynamic risk warning map with historical environmental monitoring data, a consistency analysis of pollution trends is generated. Based on the consistency analysis results, the construction process parameters are adjusted to generate optimized construction plan suggestions. The construction process parameters include adjusting the construction period and increasing the operating frequency of dust removal equipment. The optimized construction plan and rectification suggestions are integrated, and a traceable execution plan document is generated through smart contracts and pushed to the construction management platform simultaneously.

7. The method for comparing and evaluating environmental monitoring parameters based on completion process data according to claim 1, characterized in that, In the blockchain storage process, the methods also include: The construction feature vectors in the standardized dataset are split into independent data blocks according to the construction stage. Each data block contains the construction stage identifier and the corresponding environmental parameter vector. Each data block is hashed by a blockchain node, and the hash digest is bound to a construction phase identifier and stored in the blockchain's distributed ledger. Set data integrity verification rules in the smart contract. When a break in the hash value link of data blocks in adjacent construction stages is detected, the data tracing process is automatically triggered, and an anomaly record containing a timestamp and repair suggestions are output.

8. A comparative evaluation system for environmental monitoring parameters based on as-built process data, characterized in that, include: A standardized dataset generation module acquires as-built process data and environmental monitoring data. The as-built process data includes the amount of construction materials used, process parameters, and configuration of environmental protection facilities. The environmental monitoring data includes real-time air quality, noise, and remote sensing image data monitored by sensors. A multi-dimensional data map is constructed using semantic web technology to map the as-built process data and environmental monitoring data to a unified semantic space, which is used to generate a standardized dataset containing semantic relationships. The standardized dataset includes construction feature vectors and environmental parameter vectors. The weight coefficient matrix update output module calculates the dynamic weight coefficients of environmental monitoring parameters based on the construction feature vector and the environmental parameter vector through a multi-objective optimization model. The multi-objective optimization model combines construction progress and environmental sensitivity to generate pollution type weights and ecological impact level weights, which are then used to output the updated weight coefficient matrix. The construction progress includes the basic construction period and the decoration period. Different construction progress corresponds to different pollution types, and different environmental sensitivities correspond to different ecological impact level weights. The dynamic assessment report generation module stores the standardized dataset and the weight coefficient matrix on the blockchain in the form of a hash digest, and automatically triggers the assessment process through a smart contract. An assessment rule engine is embedded in the smart contract, which generates a dynamic assessment report based on the weight coefficient matrix and a preset comparison logic. The dynamic assessment report includes an identifier of the non-compliance event, compliance remediation suggestions, and a digital signature. The dynamic assessment report triggers an AR visualization system through the smart contract, overlaying the non-compliance event with a BIM model to generate a 3D heatmap and output rectification suggestions. The final predictive assessment report output module deploys lightweight edge computing nodes at the construction site to process sensor data in real time and perform sliding window analysis. It identifies potential environmental risks using anomaly detection algorithms and outputs risk event markers based on these risks. Based on the construction feature vectors and the risk event markers, it predicts the environmental parameter change trends for the next 24 hours using a time-series prediction model. A dynamic risk warning map is generated based on these environmental parameter change trends, including the spatial distribution of pollution hotspots and the impact paths of construction activities. This dynamic risk warning map is then linked with an AR visualization system to output the final predictive assessment report.

9. A comparative evaluation system for environmental monitoring parameters based on as-built process data, characterized in that, Includes a processor, wherein the processor runs a program for the environmental monitoring parameter comparison and evaluation method based on completion process data as described in any one of claims 1-7.

10. A storage medium, characterized in that, The program stores the method for comparing and evaluating environmental monitoring parameters based on completion process data as described in any one of claims 1-7.

Images