Intelligent monitoring and predicting method and system for carbon emission in road construction

Abstract

The invention discloses an intelligent monitoring and predicting method and system for carbon emission in road construction, and particularly relates to the technical field of environmental protection and engineering construction, and the method comprises the following steps: S1, building a carbon emission database; s2, deploying Internet of Things monitoring equipment; s3, data transmission and storage; s4, data analysis and modeling: performing deep mining on the collected data by using a big data analysis technology, and establishing a carbon emission prediction model; s5, real-time monitoring and early warning: analyzing carbon emission data in real time through an intelligent algorithm, and immediately sending early warning information to a manager to timely adjust a construction plan to reduce carbon emission once an abnormal or standard exceeding condition is found; and S6, optimal control: dynamically adjusting the construction strategy to realize optimal control of carbon emission. And through deployment of Internet of Things monitoring equipment and real-time data transmission, the problems that an existing carbon emission monitoring means is low in efficiency and the real-time monitoring requirement is difficult to meet are solved, and the carbon emission condition in the construction process can be mastered in time.

Description

Technical Field

[0001] This application relates to the field of environmental protection and engineering construction technology, and more specifically, to a method and system for intelligent monitoring and prediction of carbon emissions from road construction. Background Technology

[0002] With increasing emphasis on environmental protection, monitoring and managing carbon emissions during engineering construction has become a trend. Current technologies typically estimate carbon emissions during construction through manual recording and statistical analysis, or by collecting and analyzing data from fixed monitoring equipment.

[0003] There are two main solutions in the existing technology: one is to manually record and statistically analyze data such as material usage and energy consumption during the construction process, and then combine this with theoretical models to calculate carbon emissions; the other is to deploy a sensor network at the construction site to collect environmental data during the construction process and then analyze it through post-processing software.

[0004] However, existing technologies have obvious drawbacks: manual recording methods are inefficient and prone to errors; while sensor network monitoring methods improve the accuracy of data acquisition, they lack intelligent analysis capabilities and cannot achieve the prediction and optimized control of carbon emissions. Therefore, a smart monitoring and prediction method and system for carbon emissions from road construction is proposed to address the above problems. Summary of the Invention

[0005] In order to overcome the above-mentioned deficiencies of the prior art, this application provides a method and system for intelligent monitoring and prediction of carbon emissions from road construction, so as to solve the problems mentioned in the background art.

[0006] on the one hand To achieve the above objectives, this application provides the following technical solution: a method for intelligent monitoring and prediction of carbon emissions from road construction, comprising the following steps: S1. Establish a carbon emission database: Collect and organize all information on materials, machinery and energy consumption involved in road construction to form a comprehensive carbon emission database as the basic data source for monitoring and prediction; S2. Deploy IoT monitoring equipment: Deploy IoT sensors at key locations in the construction area to collect environmental parameters and construction activity data in real time during the construction process; S3. Data transmission and storage: Data collected on-site is transmitted to the cloud platform via wireless communication technology for centralized storage and preliminary processing; S4. Data Analysis and Modeling: Utilize big data analytics to deeply mine the collected data and establish a carbon emission prediction model. This model predicts future carbon emission trends based on historical data and current construction conditions. S5. Real-time monitoring and early warning: Through intelligent algorithms, carbon emission data is analyzed in real time. Once an anomaly or exceeding the standard is detected, an early warning message is immediately sent to the management personnel to adjust the construction plan in a timely manner to reduce carbon emissions. S6. Optimization Control: Combining forecast results and construction progress, dynamically adjust construction strategies to achieve optimal control of carbon emissions.

[0007] Furthermore, the carbon emission database in S1 is established based on a coupled material balance and life cycle assessment model, specifically as follows: The formula for quantifying the input-output relationship of various materials in road construction based on the principle of material balance is: M input =M output +M residual , of which M input M represents the total amount of material input. output The amount of material used to form the road structure, M residual This refers to the amount of waste generated during the construction process; Carbon emissions are calculated using a life cycle assessment model, using the following formula: Where C total M represents total carbon emissions. i For the amount of material / energy used, EF i Let be the carbon emission factor corresponding to the i-th material / energy, and n be the number of types of materials / energy.

[0008] Through the above technical solution, the total material input covers the raw materials used in each stage of construction. The amount of material forming the road structure is the actual material value that constitutes the road's physical components. The amount of waste includes discarded and damaged materials generated during construction. By applying the principle of material balance, the quantitative relationship of materials from input to output can be clearly quantified, laying an accurate material foundation for subsequent carbon emission calculations. In this process, the entire process data, including material procurement, transportation, and usage, is accurately recorded to ensure the accuracy of material balance calculations and to ensure that the quantitative results truly reflect the material flow in road construction.

[0009] Furthermore, the environmental parameters and construction activity data collected in S2 are processed using a Kalman filter model, which includes state updates and measurement updates. The formula for the state update phase is: The formula for the measurement update phase is: in, Let F be the predicted state, representing the prediction of the state at time k based on information from time k-1. k Here is the state transition matrix. It is the optimal estimate from the previous time step, i.e., the optimal state estimate after the update at time step k-1, B. k For the control matrix, u k Let the control vector represent the external control input applied to the system, P k|k-1 To predict covariance, Q k For process noise covariance, K k H is the Kalman gain, used to weigh the weights of predicted and observed values ​​in state updates. k Let z be the observation matrix. k For the observed value, R k To observe the noise covariance, For the updated state estimate, P k|k This is the updated covariance.

[0010] By adopting the above technical solutions, the accuracy and reliability of monitoring data can be effectively improved, providing a high-quality data foundation for subsequent carbon emission analysis.

[0011] Furthermore, in S3, during the initial data processing, a data cleaning algorithm is used to denoise the collected data and remove duplicate, erroneous, or severely missing data records. For data with a small number of missing values, methods such as mean imputation and regression prediction are used to complete the data, ensuring the accuracy and completeness of the data. Then, the data is normalized using Z-score.

[0012] Furthermore, the carbon emission prediction model established in S4 adopts a multiple linear regression model, with the formula C = α1M1 + α2M2 + ... + α n M n +β1E1+β2E2+…+β m E m +γT+ε, where C is carbon emissions, M1,M2,...,M n For different amounts of material, α1, α2, ..., α n For the corresponding material emission factors, E1, E2, ..., E m For the energy consumption of different machines, β1, β2, ..., β m ε represents the carbon emission coefficient of the corresponding machinery, T represents the construction time, γ represents the carbon emission coefficient related to the construction time, and ε represents the error term.

[0013] By adopting the above technical solutions and incorporating core influencing factors such as material usage, mechanical energy consumption, and construction duration into the model, the complex causes of carbon emissions during road construction can be comprehensively reflected. For example, material usage and mechanical energy consumption are directly related to resource consumption, while construction duration reflects process management efficiency. By integrating these variables, the changing patterns of carbon emissions can be accurately captured. The introduction of the error term ε effectively covers random factors and measurement errors not considered by the model, such as difficult-to-quantify external influences like weather changes and sudden equipment failures. This makes the model more robust in real-world complex construction scenarios, and the prediction results are closer to reality, providing a reliable basis for carbon emission control.

[0014] Furthermore, in step S4, when training the carbon emission prediction model using machine learning algorithms, the random forest algorithm is employed. This algorithm trains the model on the data by constructing multiple decision trees and combines the prediction results of multiple decision trees to obtain the final prediction value. The specific formula is as follows: in, The predicted carbon emissions are given by k, the number of decision trees, and f. i (x) represents the prediction result of the i-th decision tree, and x is the input feature vector.

[0015] Furthermore, in S5, a dynamic time series analysis algorithm is used, combined with a threshold comparison mechanism, to monitor carbon emission data in real time. Anomaly detection is based on the following formula: Anomaly Detection = The fluctuation coefficient ranges from 0.1 to 0.2. When the monitored data triggers the abnormal judgment condition, the system pushes an early warning information to the management personnel.

[0016] Through the above technical solutions, real-time monitoring enables managers to grasp the dynamics of carbon emissions during road construction in real time. Compared with periodic inspections or post-event analysis, it greatly shortens the problem discovery cycle, making management more timely. Once an anomaly occurs, an immediate warning is issued, which can quickly trigger response mechanisms to prevent the carbon emission situation from continuing to worsen, control construction carbon emissions within a reasonable range, and reduce the environmental risks and economic losses caused by exceeding the standards. Based on real-time data and warnings, managers can adjust construction plans in a timely manner, such as rationally allocating construction equipment and optimizing construction processes, so as to achieve a win-win situation of energy conservation, emission reduction and construction efficiency while ensuring construction progress.

[0017] Furthermore, in S6, the dynamic adjustment of the construction strategy introduces a genetic algorithm as an intelligent optimization algorithm to automatically find the optimal construction plan and resource allocation scheme. Its objective function is: minC=f(M,E,T), where f is the carbon emission function, M is the material consumption vector, E is the mechanical energy consumption vector, T is the construction time, and the constraints include construction progress requirements and resource supply limitations.

[0018] The above technical solution uses carbon emissions (C) as the optimization object of the objective function minC = f(M, E, T), directly focusing on the most critical environmental indicators in road construction. By quantifying the impact of material usage (M), mechanical energy consumption (E), and construction time (T) on carbon emissions, the optimization direction is clear, enabling a systematic reduction of carbon emissions during construction. This aligns with the concept of green construction and comprehensively covers the main factors affecting carbon emissions in road construction. The extraction, transportation, and use of materials, the energy consumption of machinery operation, and the continuous consumption of resources during the construction period are all closely related to carbon emissions. Through the synergistic optimization of these factors, multi-dimensional energy conservation and emission reduction can be achieved. By introducing constraints such as construction schedule requirements and resource supply limitations, the optimized construction plan and resource allocation scheme meet both low-carbon goals and actual project needs. This avoids delays in construction schedules or interruptions in resource supply due to an excessive pursuit of low carbon, ensuring the smooth progress of construction projects and balancing environmental and economic benefits.

[0019] Furthermore, in S6, when optimizing control to improve the accuracy and reliability of carbon emission assessment, a weighted fusion algorithm is used to combine multiple external data sources, including satellite remote sensing data and meteorological data, with field monitoring data for fusion analysis. The formula is: D = ω1D1 + ω2D2 + ... + ω p D p Where D represents the fused composite data, D1, D2, ..., D p For data from different data sources, ω1,ω2,...,ω p The weights of each data source are given, and

[0020] Through the above technical solutions, satellite remote sensing data can obtain the spatial distribution characteristics of carbon emissions in large-scale road construction areas at a macro level, meteorological data (such as wind speed and temperature) can reflect the impact of environmental factors on carbon emission diffusion and transformation, while on-site monitoring data provides localized, precise, real-time emission information. Through a weighted fusion algorithm, the advantages of different data sources are complemented, avoiding information loss caused by limitations in coverage, monitoring frequency, or environmental interference from a single data source. The weights ω of each data source are... i Its adjustable characteristics enable the system to dynamically adjust the fusion strategy based on the actual construction environment and data quality. For example, when meteorological conditions are complex, the weight of meteorological data is increased to more accurately reflect the impact of the environment on carbon emissions; when the resolution of satellite remote sensing data is improved, its weight is increased to strengthen macro-trend analysis and significantly enhance the adaptability of the assessment model to complex construction scenarios.

[0021] on the other hand A smart monitoring and prediction system for carbon emissions from road construction, the system comprising: Carbon emission database construction module 1 is used to build a carbon emission database. It collects and organizes information on all materials, machinery and energy consumption involved in road construction to build a comprehensive carbon emission database as the basic data source for monitoring and prediction. IoT monitoring equipment deployment module 2: used to deploy IoT sensors at key locations in the construction area to collect environmental parameters and construction activity data in real time during the construction process; Data transmission and storage module 3: Used for communication connections between various modules, it transmits data collected on-site to the cloud platform via wireless communication technology for centralized storage and preliminary processing; Data Analysis and Modeling Module 4: Used to conduct in-depth mining of collected data using big data analytics techniques to build carbon emission prediction models; Real-time monitoring and early warning module 5: It is used to analyze carbon emission data in real time through intelligent algorithms. Once an anomaly or exceeding the standard is detected, it will immediately send an early warning message to the management personnel to adjust the construction plan in a timely manner to reduce carbon emissions. Optimized Control Module 6: Used to combine prediction results and construction progress to dynamically adjust construction strategies to achieve optimal control of carbon emissions.

[0022] The technical effects and advantages of this application are as follows: Compared with existing technologies, this intelligent monitoring and prediction method and system for carbon emissions from road construction achieves real-time monitoring: by deploying IoT monitoring equipment and transmitting data in real time, it solves the problems of low efficiency and inability to meet the real-time monitoring needs of existing carbon emission monitoring methods, and can promptly grasp the carbon emission situation during the construction process.

[0023] It possesses intelligent prediction and optimization control capabilities: By utilizing big data and artificial intelligence technologies, and by establishing accurate carbon emission prediction models and introducing intelligent optimization algorithms, it solves the problem of lacking carbon emission prediction and optimization control capabilities based on big data and artificial intelligence, improves prediction accuracy and control effect, and can achieve optimal control of carbon emissions.

[0024] Improved accuracy and reliability of assessment: By adopting multi-source data fusion technology and calibration mechanism, combining multiple external data sources and performing periodic model calibration, the problem of low accuracy and reliability of carbon emission assessment caused by manual recording and simple data analysis methods was solved, and the comprehensive analysis capability of the data was enhanced. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the method flow of this application; Figure 2 This is a schematic diagram of the system flow of this application.

[0026] The attached diagram is labeled as follows: 1. Carbon emission database construction module; 2. IoT monitoring equipment deployment module; 3. Data transmission and storage module; 4. Data analysis and modeling module; 5. Real-time monitoring and early warning module; 6. Optimization control module. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] Example 1 like Figure 1 The method shown is an intelligent monitoring and prediction method for carbon emissions from road construction. The method includes the following steps: S1. Establish a carbon emission database: Collect and organize all information on materials, machinery and energy consumption involved in road construction to form a comprehensive carbon emission database as the basic data source for monitoring and prediction; The carbon emission database in S1 is established based on a coupled model of material balance and life cycle assessment, specifically: The formula for quantifying the input-output relationship of various materials in road construction based on the principle of material balance is: M input =M output +M residual , of which M input M represents the total amount of material input. output The amount of material used to form the road structure, M residual This refers to the amount of waste generated during the construction process; Carbon emissions are calculated using a life cycle assessment model, using the following formula: Where C total M represents total carbon emissions. i For the amount of material / energy used, EF i Let be the carbon emission factor corresponding to the i-th material / energy, and n be the number of types of materials / energy.

[0029] The total material input encompasses all raw materials used in each stage of construction. The amount of material forming the road structure is the actual material quantity constituting the road's physical components. Waste includes discarded and damaged materials generated during construction. Through the principle of material balance, the quantitative relationship between materials from input to output can be clearly quantified, laying a precise material foundation for subsequent carbon emission calculations. During this process, precise recording of data throughout the entire process of material procurement, transportation, and usage ensures the accuracy of material balance calculations and guarantees that the quantitative results truly reflect the material flow during road construction. S2. Deploy IoT monitoring equipment: Deploy IoT sensors at key locations in the construction area to collect environmental parameters and construction activity data in real time during the construction process. The environmental parameters and construction activity data collected in S2 are processed using a Kalman filter model, which includes state updates and measurement updates. The formula for the state update phase is: The formula for the measurement update phase is: in, Let F be the predicted state, representing the prediction of the state at time k based on information from time k-1. k Here is the state transition matrix. It is the optimal estimate from the previous time step, i.e., the optimal state estimate after the update at time step k-1, B. k For the control matrix, u k Let the control vector represent the external control input applied to the system, P k|k-1 To predict covariance, Q k For process noise covariance, K k H is the Kalman gain, used to weigh the weights of predicted and observed values ​​in state updates. k Let z be the observation matrix. k For the observed value, R k To observe the noise covariance, For the updated state estimate, P k|k The updated covariance can effectively improve the accuracy and reliability of monitoring data, providing a high-quality data foundation for subsequent carbon emission analysis; S3. Data Transmission and Storage: Data collected on-site is transmitted to a cloud platform via wireless communication technology for centralized storage and preliminary processing. During preliminary processing, data cleaning algorithms are used to denoise the collected data, removing duplicate, erroneous, or severely missing records. For data with a small number of missing values, methods such as mean imputation and regression prediction are used to complete the data, ensuring accuracy and completeness. Finally, Z-score normalization is applied to the data. S4. Data Analysis and Modeling: Utilize big data analytics to deeply mine the collected data and establish a carbon emission prediction model. This model predicts future carbon emission trends based on historical data and current construction conditions. The carbon emission prediction model established in S4 adopts a multiple linear regression model, with the formula C = α1M1 + α2M2 + ... + α n M n +β1E1+β2E2+…+β m E m +γT+ε, where C is carbon emissions, M1,M2,...,M n For different amounts of material, α1, α2, ..., α n For the corresponding material emission factors, E1, E2, ..., E m For the energy consumption of different machines, β1, β2, ..., β m Let T be the carbon emission coefficient of the corresponding machinery, γ be the construction duration, γ be the carbon emission coefficient related to construction duration, and ε be the error term. By incorporating core influencing factors such as material usage, machinery energy consumption, and construction duration into the model, it can comprehensively reflect the complex causes of carbon emissions during road construction. For example, material usage and machinery energy consumption are directly related to resource consumption, while construction duration reflects process management efficiency. By integrating these variables, the changing patterns of carbon emissions can be accurately captured. The introduction of the error term ε effectively covers random factors and measurement errors not considered by the model, such as weather changes and sudden equipment failures—external influences that are difficult to quantify. This makes the model more robust in complex actual construction scenarios, and the prediction results are closer to reality, providing a reliable basis for carbon emission control.

[0030] In S4, when training the carbon emission prediction model using machine learning algorithms, the random forest algorithm is employed. This algorithm trains the model on the data by constructing multiple decision trees and then combines the prediction results from these multiple decision trees to obtain the final prediction value. The specific formula is as follows: in, The predicted carbon emissions are given by k, the number of decision trees, and f. i (x) represents the prediction result of the i-th decision tree, where x is the input feature vector; S5. Real-time monitoring and early warning: Through intelligent algorithms, carbon emission data is analyzed in real time. Once an anomaly or exceeding the standard is detected, an early warning message is immediately sent to the management personnel to adjust the construction plan in a timely manner to reduce carbon emissions. S5 employs a dynamic time series analysis algorithm, combined with a threshold comparison mechanism, to monitor carbon emission data in real time. Anomaly detection is based on the following formula: The fluctuation coefficient ranges from 0.1 to 0.2. When the monitored data triggers the abnormal judgment condition, the system pushes an early warning information to the management personnel.

[0031] Real-time monitoring allows managers to grasp the dynamics of carbon emissions during road construction in real time. Compared with periodic inspections or post-event analysis, it greatly shortens the problem discovery cycle, making management more timely. Once an anomaly occurs, an immediate warning is issued, which can quickly trigger response mechanisms to prevent the carbon emission situation from continuing to worsen. This keeps construction carbon emissions within a reasonable range, reducing the environmental risks and economic losses caused by exceeding the standards. Based on real-time data and warnings, managers can adjust construction plans in a timely manner, such as rationally allocating construction equipment and optimizing construction processes, so as to achieve a win-win situation of energy conservation, emission reduction and construction efficiency while ensuring construction progress.

[0032] S6. Optimization Control: Combining prediction results and construction progress, the construction strategy is dynamically adjusted to achieve optimal control of carbon emissions. The dynamic adjustment of the construction strategy introduces a genetic algorithm as an intelligent optimization algorithm to automatically find the optimal construction plan and resource allocation scheme. Its objective function is: minC=f(M,E,T), where f is the carbon emission function, M is the material consumption vector, E is the mechanical energy consumption vector, T is the construction time, and the constraints include construction progress requirements and resource supply limitations.

[0033] By using carbon emissions C as the optimization target of the objective function minC = f(M, E, T), this approach directly focuses on the most critical environmental indicators in road construction. Quantifying the impact of material usage (M), mechanical energy consumption (E), and construction duration (T) on carbon emissions clarifies the optimization direction, enabling a systematic reduction of carbon emissions during construction. This aligns with the concept of green construction and comprehensively covers the main factors influencing carbon emissions in road construction. The extraction, transportation, and use of materials, the energy consumption of machinery operation, and the continuous resource consumption during the construction cycle are all closely related to carbon emissions. Through the synergistic optimization of these factors, multi-dimensional energy conservation and emission reduction can be achieved. Introducing constraints such as construction schedule requirements and resource supply limitations ensures that the optimized construction plan and resource allocation scheme meet both low-carbon goals and actual project needs. This avoids delays in construction schedules or interruptions in resource supply due to an excessive pursuit of low carbon emissions, ensuring the smooth progress of construction projects and balancing environmental and economic benefits.

[0034] In S6, to improve the accuracy and reliability of carbon emission assessment, a weighted fusion algorithm is used to combine multiple external data sources, including satellite remote sensing data and meteorological data, with field monitoring data for fusion analysis. The formula is: D=ω1D1+ω2D2+…+ω p D p Where D represents the fused composite data, D1, D2, ..., D p For data from different data sources, ω1,ω2,...,ω p The weights of each data source are given, and

[0035] Satellite remote sensing data can provide macroscopic information on the spatial distribution of carbon emissions in large-scale road construction areas, while meteorological data (such as wind speed and temperature) reflects the impact of environmental factors on carbon emission diffusion and transformation. Meanwhile, on-site monitoring data provides precise, real-time emission information for specific locations. Through a weighted fusion algorithm, the advantages of different data sources are complemented, avoiding information gaps caused by limitations in coverage, monitoring frequency, or environmental interference from single data sources. The weights ω for each data source are calculated. i Its adjustable characteristics enable the system to dynamically adjust the fusion strategy based on the actual construction environment and data quality. For example, when meteorological conditions are complex, the weight of meteorological data is increased to more accurately reflect the impact of the environment on carbon emissions; when the resolution of satellite remote sensing data is improved, its weight is increased to strengthen macro-trend analysis and significantly enhance the adaptability of the assessment model to complex construction scenarios.

[0036] Example 2 A smart monitoring and prediction system for carbon emissions from road construction, the system comprising: Carbon emission database construction module 1 is used to build a carbon emission database. It collects and organizes information on all materials, machinery and energy consumption involved in road construction to build a comprehensive carbon emission database as the basic data source for monitoring and prediction. IoT monitoring equipment deployment module 2: used to deploy IoT sensors at key locations in the construction area to collect environmental parameters and construction activity data in real time during the construction process; Data transmission and storage module 3: Used for communication connections between various modules, it transmits data collected on-site to the cloud platform via wireless communication technology for centralized storage and preliminary processing; Data Analysis and Modeling Module 4: Used to conduct in-depth mining of collected data using big data analytics techniques to build carbon emission prediction models; Real-time monitoring and early warning module 5: It is used to analyze carbon emission data in real time through intelligent algorithms. Once an anomaly or exceeding the standard is detected, it will immediately send an early warning message to the management personnel to adjust the construction plan in a timely manner to reduce carbon emissions. Optimized Control Module 6: This module combines forecast results and construction progress to dynamically adjust construction strategies and achieve optimal control of carbon emissions. By deploying IoT monitoring devices and transmitting data in real time, it solves the problem of low efficiency and inability to meet real-time monitoring needs of existing carbon emission monitoring methods, and can promptly grasp the carbon emission situation during the construction process.

Claims

1. A method for intelligent monitoring and prediction of carbon emissions from road construction, characterized in that, The method includes the following steps: S1. Establish a carbon emission database: Collect and organize all information on materials, machinery and energy consumption involved in road construction to form a comprehensive carbon emission database as the basic data source for monitoring and prediction; S2. Deploy IoT monitoring equipment: Deploy IoT sensors at key locations in the construction area to collect environmental parameters and construction activity data in real time during the construction process; S3. Data transmission and storage: Data collected on-site is transmitted to the cloud platform via wireless communication technology for centralized storage and preliminary processing; S4. Data Analysis and Modeling: Utilize big data analytics to deeply mine the collected data and establish a carbon emission prediction model. This model predicts future carbon emission trends based on historical data and current construction conditions. S5. Real-time monitoring and early warning: Through intelligent algorithms, carbon emission data is analyzed in real time. Once an anomaly or exceeding the standard is detected, an early warning message is immediately sent to the management personnel to adjust the construction plan in a timely manner to reduce carbon emissions. S6. Optimization Control: Combining forecast results and construction progress, dynamically adjust construction strategies to achieve optimal control of carbon emissions.

2. The intelligent monitoring and prediction method for carbon emissions from road construction according to claim 1, characterized in that: The carbon emission database in S1 is established based on a coupled material balance and life cycle assessment model, specifically as follows: The formula for quantifying the input-output relationship of various materials in road construction based on the principle of material balance is: M input =M output +M residual , of which M input M represents the total amount of material input. output The amount of material used to form the road structure, M residual This refers to the amount of waste generated during the construction process; Carbon emissions are calculated using a life cycle assessment model, using the following formula: Where C total M represents total carbon emissions. i For the amount of material / energy used, EF i Let be the carbon emission factor corresponding to the i-th material / energy, and n be the number of types of materials / energy.

3. The intelligent monitoring and prediction method for carbon emissions from road construction according to claim 2, characterized in that: The environmental parameters and construction activity data collected in S2 are processed using a Kalman filter model, which includes state updates and measurement updates. The formula for the state update phase is: The formula for the measurement update phase is: in, Let F be the predicted state, representing the prediction of the state at time k based on information from time k-1. k Here is the state transition matrix. It is the optimal estimate from the previous time step, i.e., the optimal state estimate after the update at time step k-1, B. k For the control matrix, u k Let the control vector represent the external control input applied to the system, P k|k-1 To predict covariance, Q k For process noise covariance, K k H is the Kalman gain, used to weigh the weights of predicted and observed values ​​in state updates. k Let z be the observation matrix. k For the observed value, R k To observe the noise covariance, For the updated state estimate, P k|k This is the updated covariance.

4. The intelligent monitoring and prediction method for carbon emissions from road construction according to claim 3, characterized in that: In the initial data processing of S3, a data cleaning algorithm is used to denoise the collected data and remove duplicate, erroneous or severely missing data records. For data with a small number of missing values, methods such as mean imputation and regression prediction are used to complete the data to ensure the accuracy and completeness of the data. Then, the data is normalized by Z-score.

5. The intelligent monitoring and prediction method for carbon emissions from road construction according to claim 2, characterized in that: The carbon emission prediction model established in S4 adopts a multiple linear regression model, with the formula C = α1M1 + α2M2 + ... + α n M n +β1E1+β2E2+…+β m E m +γT+ε, where C is carbon emissions, M1,M2,...,M n For different amounts of material, α1, α2, ..., α n For the corresponding material emission factors, E1, E2, ..., E m For the energy consumption of different machines, β1, β2, ..., β m ε represents the carbon emission coefficient of the corresponding machinery, T represents the construction time, γ represents the carbon emission coefficient related to the construction time, and ε represents the error term.

6. The intelligent monitoring and prediction method for carbon emissions from road construction according to claim 5, characterized in that: In step S4, when training the carbon emission prediction model using machine learning algorithms, the random forest algorithm is employed. This algorithm trains the model on the data by constructing multiple decision trees and then combines the prediction results from these multiple decision trees to obtain the final prediction value. The specific formula is as follows: in, The predicted carbon emissions are given by k, the number of decision trees, and f. i (x) represents the prediction result of the i-th decision tree, and x is the input feature vector.

7. The intelligent monitoring and prediction method for carbon emissions from road construction according to claim 6, characterized in that: The S5 process employs a dynamic time series analysis algorithm, combined with a threshold comparison mechanism, to monitor carbon emission data in real time. Anomaly detection is based on the following formula: The fluctuation coefficient ranges from 0.1 to 0.

2. When the monitored data triggers the abnormal judgment condition, the system pushes an early warning information to the management personnel.

8. The intelligent monitoring and prediction method for carbon emissions from road construction according to claim 7, characterized in that: In S6, the dynamic adjustment of the construction strategy introduces a genetic algorithm as an intelligent optimization algorithm to automatically find the optimal construction plan and resource allocation scheme. Its objective function is: minC=f(M,E,T), where f is the carbon emission function, M is the material consumption vector, E is the mechanical energy consumption vector, T is the construction time, and the constraints include construction progress requirements and resource supply restrictions.

9. The intelligent monitoring and prediction method for carbon emissions from road construction according to claim 2, characterized in that: In S6, when optimizing control to improve the accuracy and reliability of carbon emission assessment, a weighted fusion algorithm is used to combine multiple external data sources, including satellite remote sensing data and meteorological data, with field monitoring data for fusion analysis. The formula is: D = ω1D1 + ω2D2 + ... + ω p D p Where D represents the fused composite data, D1, D2, ..., D p For data from different data sources, ω1,ω2,...,ω p The weights of each data source are given, and 10. A smart monitoring and prediction system for carbon emissions from road construction, characterized in that, The system includes: a carbon emission database construction module (1), which is used to construct a carbon emission database. By collecting and organizing all material, machinery and energy consumption information involved in the road construction process, a comprehensive carbon emission database is constructed as the basic data source for monitoring and prediction. IoT monitoring equipment deployment module (2): used to deploy IoT sensors at key locations in the construction area to collect environmental parameters and construction activity data in real time during the construction process; Data transmission and storage module (3): Used for communication connection between various modules, transmitting the data collected on site to the cloud platform through wireless communication technology for centralized storage and preliminary processing; Data Analysis and Modeling Module (4): Used to conduct in-depth mining of collected data using big data analysis technology and establish carbon emission prediction models; Real-time monitoring and early warning module (5): Used to analyze carbon emission data in real time through intelligent algorithms. Once an abnormality or exceeding the standard is detected, an early warning message is immediately sent to the management personnel to adjust the construction plan in a timely manner to reduce carbon emissions. Optimization control module (6): It is used to combine the prediction results and construction progress to dynamically adjust the construction strategy to achieve optimal control of carbon emissions.

Images