Highway and waterway infrastructure whole-process carbon emission refined evaluation method and system

Abstract

The invention relates to a highway and waterway infrastructure whole-process carbon emission refined evaluation method and system. The method comprises the steps that carbon emission original data in the design, construction and operation and maintenance stages of highway and waterway infrastructures are acquired; carrying out preprocessing on the carbon emission original data; constructing a carbon emission factor library for carbon emission sources of highway and waterway infrastructures; evaluating the carbon emission of each stage based on the carbon emission factor library through life cycle evaluation; based on the original carbon emission data, obtaining a carbon emission prediction result under constraints through high-dimensional feature extraction, long-term and short-term feature modeling and inter-stage data interaction; and generating optimization suggestions based on carbon emission evaluation and carbon emission prediction results. Compared with the prior art, the method has the advantages of full-process coverage, high dynamic prediction capability and the like.

Description

Technical Field

[0001] The present invention relates to the field of infrastructure carbon emissions, and particularly to a refined assessment method and system for the whole-process carbon emissions of highway and waterway infrastructure. Background Art

[0002] With the intensification of global climate change, the carbon emissions problem in the whole life cycle of highway and waterway infrastructure has attracted much attention. Its main emission sources include the selection of high-carbon materials in the design stage, construction energy consumption and material transportation in the construction stage, and traffic flow and equipment operation emissions in the operation and maintenance stage. However, existing assessment methods are mostly based on static models, ignoring complex factors such as on-site energy consumption and dynamic changes in traffic flow, and the assessment scope is limited to a single stage, unable to comprehensively cover the whole life cycle. At the same time, existing technologies are difficult to capture the non-linear relationship between carbon emission factors and carbon emission amounts, and there is also a lack of low-carbon optimization guidance for highway and waterway scenarios.

[0003] Chinese Patent Application Publication No. CN119476560A discloses a method, system and electronic device for predicting the carbon emissions of a highway during the construction period. Based on the evaluated carbon emission factors, this application realizes the regular update of carbon emission factors as the project progresses, and uses an improved and optimized fusion model to accurately predict the carbon emission amount. However, it only uses static emission factors, which easily leads to the disconnection between the carbon emission calculation in the construction stage and the actual situation, thereby affecting the accuracy of carbon emission assessment.

[0004] In summary, there is currently a lack of a refined assessment method and system for the whole-process carbon emissions of road and waterway infrastructure. Summary of the Invention

[0005] The purpose of the present invention is to overcome the above-mentioned defects of the existing technology and provide a refined assessment method and system for the whole-process carbon emissions of highway and waterway infrastructure, so as to solve or partially solve the problems of poor dynamic performance of carbon emission assessment, incomplete coverage, and lack of optimization suggestions.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] In one aspect of the present invention, a refined assessment method for the whole-process carbon emissions of highway and waterway infrastructure is provided, including the following steps:

[0008] Obtain the original carbon emission data in the design, construction, and operation and maintenance stages of highway and waterway infrastructure;

[0009] Preprocess the original carbon emission data;

[0010] Construct a carbon emission factor library for the carbon emission sources of highway and waterway infrastructure;

[0011] Through life cycle assessment, the carbon emissions of each stage are evaluated based on the carbon emission factor library;

[0012] Based on the original carbon emission data, through high-dimensional feature extraction, long-term and short-term feature modeling, and data interaction between stages, a constrained carbon emission prediction result is obtained;

[0013] Based on the carbon emission assessment and the carbon emission prediction result, optimization suggestions are generated.

[0014] As a preferred technical solution, the original carbon emission data includes material usage data, construction equipment energy consumption data, traffic flow data, transportation routes and fuel consumption data, and port equipment operation energy consumption data.

[0015] As a preferred technical solution, the preprocessing includes cleaning, standardization, and normalization processing.

[0016] As a preferred technical solution, the carbon emission factor library includes production emission factors of highway materials and waterway materials, fuel emission factors of highway transport vehicles and ships, and operation emission factors of construction equipment and port equipment.

[0017] As a preferred technical solution, the process of obtaining the carbon emission prediction result includes the following steps:

[0018] Based on the original carbon emission data, a fusion model including a convolutional neural network and a long short-term memory network based on the attention mechanism is trained;

[0019] Use the convolutional neural network to extract multi-source high-dimensional features of the current carbon emission data;

[0020] Based on the high-dimensional features, use the long short-term memory network to model the long-term dependence and short-term volatility of the time series in the time dimension to obtain a preliminary carbon emission prediction result. During the modeling process, calculate the influence weights of different stages through the attention mechanism to enhance the interaction between data of each stage;

[0021] Based on the preliminary carbon emission prediction result, use the physical constraint model to correct the prediction result to obtain the final carbon emission prediction result.

[0022] As a preferred technical solution, the physical constraints include non-negativity constraints on carbon emissions, seasonal fluctuation constraints, and total balance constraints.

[0023] As a preferred technical solution, the convolution kernel size of the convolutional neural network is 3×3, the stride is 1, and the number of hidden layer nodes of the long short-term memory network is 128.

[0024] As a preferred technical solution, the multi-source high-dimensional features include traffic flow pattern features and equipment operation features.

[0025] As a preferred technical solution, the convolutional neural network is used to extract the vehicle passing frequency pattern information in traffic flow data and the correlation information between the operation duration and fuel consumption in construction equipment data.

[0026] As a preferred technical solution, the long short-term memory network is used to extract seasonal traffic flow growth information and holiday peak emission information.

[0027] As a preferred technical solution, the prediction results include the carbon emission changes of short-term and long-term predictions, as well as key factor analysis.

[0028] As a preferred technical solution, the generation of optimization suggestions includes:

[0029] For the highway scenario, it is recommended to optimize the scheduling plan of construction equipment, adjust the transportation route to reduce fuel consumption, and introduce new energy vehicles in the operation stage to reduce traffic emissions;

[0030] For the waterway scenario, it is recommended to adopt LNG fuel ships or electric ships, optimize the energy efficiency management of port equipment, and improve the waterway design to shorten the voyage.

[0031] Another aspect of the present invention provides a refined carbon emission assessment system for the whole process of highway and waterway infrastructure, which is used to implement the refined carbon emission assessment method for the whole process of highway and waterway infrastructure as described above. The system includes:

[0032] A data acquisition module, which is used to collect the original carbon emission data in the design, construction and operation and maintenance stages of highway and waterway infrastructure;

[0033] A data processing module, which is used to preprocess the original carbon emission data;

[0034] A carbon emission assessment module, which is used to evaluate the carbon emissions of each stage based on the carbon emission factor library through life cycle assessment;

[0035] A carbon emission prediction module, which is used to obtain the carbon emission prediction results under constraints based on the original carbon emission data through high-dimensional feature extraction, long-term and short-term feature modeling, and data interaction between stages;

[0036] An optimization suggestion module, which is used to generate low-carbon optimization solutions for highway and waterway scenarios;

[0037] A display and interaction module, which is used to display the carbon emission assessment results, dynamic prediction and optimization suggestions in the form of charts according to the user's real-time input parameters.

[0038] Furthermore, for the input parameters, it includes:

[0039] Design phase parameters, including types and quantities of building materials, material sources and transportation distances, and carbon emission factors of materials.

[0040] Construction phase parameters, including operating hours of construction equipment, equipment fuel consumption rates, driving mileage of material transportation vehicles, and equipment scheduling plan parameters.

[0041] Operation and maintenance phase parameters, including traffic flow data, types and voyages of ships, operating parameters of port equipment, and fuel types and consumption.

[0042] Parameters related to the optimization scheme, including the proportion of new energy vehicles, ship fuel structure, equipment operation frequency, and airway route design parameters.

[0043] Compared with the prior art, the present invention has at least one of the following beneficial effects:

[0044] (1) Whole-process refined assessment: The present invention realizes the refined assessment of carbon emissions in the whole life cycle of highway and waterway infrastructure design, construction, and operation and maintenance. Through multi-source data collection and real-time monitoring, the accuracy and timeliness of the assessment results are ensured, overcoming the technical problems of limited assessment scope and insufficient accuracy in the prior art.

[0045] (2) High-precision dynamic prediction: The present invention innovatively combines convolutional neural network, long short-term memory network, and attention mechanism, and introduces a physical constraint model for correction, significantly improving the accuracy of carbon emission prediction. Verified by experiments, the prediction error is controlled within ±5%, providing reliable data support for the long-term planning of infrastructure carbon emissions.

[0046] (3) Intelligent optimization scheme: Based on the assessment and prediction results, the present invention can automatically generate targeted low-carbon optimization suggestions, covering multiple dimensions such as material selection, construction technology, equipment energy efficiency, and operation management. Practice shows that adopting the optimization scheme proposed by the present invention can achieve an average reduction of carbon emissions by more than 20%. (4) Versatility and scalability: The present invention adopts a modular design, which is applicable to both highway infrastructure and waterway infrastructure, with strong versatility. At the same time, the system architecture supports the addition of carbon emission factors and assessment models, has good scalability, and can be flexibly adjusted according to actual application requirements. Description of the Drawings

[0047] Figure 1 It is a schematic flow chart of the method for refined assessment of the whole-process carbon emissions of highway and waterway infrastructure in the embodiment;

[0048] Figure 2 It is a schematic diagram of the system for refined assessment of the whole-process carbon emissions of highway and waterway infrastructure in the embodiment;

[0049] Figure 3Box plot of prediction error distribution in the embodiment;

[0050] Figure 4 Trend chart of prediction error in the embodiment;

[0051] Figure 5 Correlation chart of actual and predicted carbon emissions in the embodiment;

[0052] Figure 6 Comparison chart of predicted value and actual value in the embodiment;

[0053] Figure 7 Histogram of absolute value distribution of prediction error in the embodiment;

[0054] Figure 8 Schematic diagram of training loss and validation loss in the embodiment;

[0055] Figure 9 Time series comparison chart in the embodiment;

[0056] Figure 10 Schematic diagram of the electronic device in the embodiment. Specific implementation manners

[0057] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0058] Embodiment 1

[0059] In view of the problems existing in the foregoing prior art, this embodiment provides a refined assessment method for the whole-process carbon emissions of highway and waterway infrastructure. Multidimensional data such as material usage, equipment energy consumption, and traffic flow are collected in real time through a sensor network. Combining a dedicated carbon emission factor library and the life cycle assessment (LCA) method, the system realizes phased carbon emission quantification and dynamic assessment. A convolutional neural network (CNN) is used to extract high-dimensional features, a long short-term memory network (LSTM) is used to model the time series changes, an attention mechanism is used to enhance multi-stage data interaction, and a physical model is combined to correct the prediction results, accurately capturing the carbon emission trend and the contribution of key links. Based on the prediction results, the system generates low-carbon solutions such as material optimization, construction process improvement, and traffic management optimization, providing comprehensive technical support for the low-carbon transformation of highway and waterway infrastructure.

[0060] See Figure 1 , the method includes the following steps:

[0061] S1. Data collection: During the entire process of the design, construction, and operation and maintenance of road and waterway infrastructure, collect carbon emission-related data, including material usage data, construction equipment energy consumption data, traffic flow data, transportation routes and fuel consumption data, and port equipment operation energy consumption data.

[0062] Through sensor networks and monitoring devices, collect carbon emission-related data in the design, construction, operation, and maintenance stages of road and waterway infrastructure, covering material usage, construction energy consumption, transportation emissions, and traffic flow and equipment energy consumption during the operation stage.

[0063] Specifically, in the road scenario, collect traffic flow, vehicle trajectories, and construction equipment energy consumption data through traffic monitoring devices; in the waterway scenario, collect ship speed, fuel consumption, and port equipment operation data through on-board sensors and port equipment monitoring systems.

[0064] For example:

[0065] Design stage: Record the usage, origin, and emission factors of building materials such as reinforced concrete and asphalt. For example, the concrete usage is 5000 cubic meters, and the emission factor is 0.2 tons of CO2 / cubic meter.

[0066] Construction stage: Deploy fuel sensors and operation time monitoring devices at the construction site to collect energy consumption data of construction machinery such as pavers and rollers; record the mileage and fuel consumption of material transportation vehicles through the GPS system.

[0067] Operation and maintenance stage: Collect the number of vehicles passing through per hour, types (such as gasoline vehicles, diesel vehicles, electric vehicles), and their corresponding emission factors through traffic flow monitoring devices.

[0068] S2. Data preprocessing: Clean, standardize, and normalize the collected raw data to ensure data consistency and comparability.

[0069] Clean, normalize, and standardize the collected raw data, and construct a carbon emission factor library in combination with the carbon emission characteristics of different materials, energy sources, and equipment for quantifying carbon emissions in each link.

[0070] For example, clean and normalize the data, and process outliers (such as excessively high or low energy consumption data) and missing data. For example, for outliers in the operation time of pavers, use the historical mean for replacement. Unify all data into the standard unit of carbon emissions per hour for subsequent evaluation and modeling.

[0071] S3. Construction of carbon emission factor library: Based on the unique carbon emission sources of road and waterway infrastructure, construct a dedicated carbon emission factor library, including but not limited to:

[0072] Production emission factors of highway materials (such as asphalt, concrete, etc.) and waterway materials (such as waterproof coatings, steel bars, etc.);

[0073] Fuel emission factors of highway transport vehicles and ships;

[0074] Operating emission factors of construction equipment and port equipment;

[0075] S4. Multi-stage carbon emission assessment: Based on the life cycle assessment (LCA) method, combined with the carbon emission factor library, quantitatively analyze the carbon emission contributions in the design, construction, and operation and maintenance stages, and generate a whole-process carbon emission assessment report.

[0076] Based on the life cycle assessment (LCA) method, quantify the carbon emission contributions in the design, construction, and operation and maintenance stages, and generate a whole-process total carbon emission assessment report and stage distribution analysis results.

[0077] For example, based on the life cycle assessment (LCA) method, quantify the carbon emissions in each stage by combining the carbon emission factor library.

[0078] Design stage: Calculate the design emissions by multiplying the material usage by the corresponding emission factor. For example, the concrete emission is 5000m 3 ×0.2 tons CO2 / m 3 = 1000 tons CO2.

[0079] Construction stage: Calculate the emissions by combining the operating time of construction equipment and the material transportation data. For example, the operating time of the paver is 500 hours, the fuel consumption is 20 liters / hour, and the total emissions are 500 hours × 20 liters / hour × 2.68 kg CO2 / liter = 26.8 tons CO2;

[0080] Operation and maintenance stage: Calculate the annual operation emissions based on the traffic flow and vehicle types. For example, 1000 vehicles pass through per hour, the average emission factor is 0.2 kg CO2 / vehicle, and the daily emissions are 1000 vehicles / hour × 24 hours × 0.2 kg CO2 / vehicle = 4.8 tons CO2.

[0081] S5. Dynamic analysis and prediction of carbon emissions: Use the advanced hybrid model algorithm to dynamically predict carbon emissions.

[0082] Combine real-time data and historical data, use the hybrid model to dynamically analyze and trend predict carbon emissions, identify the key influencing factors of carbon emissions, and quantify the future emission trends.

[0083] Specifically, the hybrid model includes:

[0084] (1) Convolutional neural network (CNN) is used to extract high-dimensional features of multi-source data.

[0085] Extract high-dimensional features of multi-source data using a Convolutional Neural Network (CNN). For example, extract the vehicle passing frequency pattern from traffic flow data and the correlation features between operation duration and fuel consumption from construction equipment data. After the time series image of traffic flow data is input into the CNN, the contribution pattern of traffic flow to emissions during peak hours (8:00 and 17:00 daily) is extracted.

[0086] (2) The Long Short-Term Memory Network (LSTM) is used to model the long-term dependence and short-term volatility of time series data.

[0087] By introducing the Long Short-Term Memory Network (LSTM), time series modeling is performed on the equipment operation duration during the construction phase and the traffic flow during the operation and maintenance phase. The LSTM captures long-term dependencies (such as seasonal traffic flow growth) and short-term fluctuations (such as peak emissions during holidays). The LSTM model predicts that during the next two years, every time there is a holiday, the traffic flow will increase by 15%, corresponding to an increase in emissions of 1.5 tons of CO2 / day.

[0088] (3) The Attention Mechanism enhances the deep interaction of multi-stage data.

[0089] Use the Attention Mechanism to quantify the mutual influence of data at each stage and dynamically adjust the key weights. The model calculates that the emission contribution during the construction phase accounts for 35% of the total emissions, while the operation and maintenance phase accounts for 50%, and the design phase accounts for 15%. Priority is given to optimizing the operation and maintenance emissions.

[0090] (4) The physical constraint model corrects the prediction results to ensure that the predictions conform to actual physical laws.

[0091] Among them, the Attention Mechanism is used to weight the outputs of each time step of the LSTM network. The specific calculation process is as follows:

[0092] Calculate the attention score for the hidden state h(t) of each time step of the LSTM, denoted as alpha(t).

[0093] The attention score is obtained through a feed-forward neural network, including a linear transformation layer W and an attention vector v

[0094] Use tanh as the activation function and normalize the scores through the softmax function

[0095] The final context vector c is the weighted average of the outputs of each time step, and the weights are the corresponding attention scores

[0096] In this way, the model can adaptively focus on the features of important time steps. For example, when predicting carbon emissions during holidays, the model will assign higher attention weights to historical data of the same period.

[0097] The prediction process of the hybrid model is achieved through the following steps:

[0098] S5.1. Data feature extraction: Use CNN to extract high-dimensional features of multi-source data such as material usage, equipment energy consumption, and traffic flow from the original data.

[0099] S5.2. Time series modeling: Use LSTM to process the time series of the extracted features and model the trend of carbon emissions changing over time.

[0100] S5.3. Data interaction enhancement: Through the Attention Mechanism, perform in-depth interaction on data at different stages (such as design, construction, and operation and maintenance), and quantify the influence weights of each stage on the total carbon emissions.

[0101] S5.4. Physical constraint correction: Combine the carbon emission factor library and use the physical model to correct the output of the hybrid algorithm.

[0102] Specifically, the physical constraints include:

[0103] (1) Non-negative constraint on carbon emissions: All predicted carbon emissions must be greater than or equal to 0;

[0104] (2) Seasonal fluctuation constraint:

[0105] The seasonal change range of traffic flow does not exceed ±20% of the historical maximum fluctuation range,

[0106] The seasonal change of equipment energy consumption needs to conform to the correlation law between temperature and energy consumption,

[0107] The seasonal fluctuation of port throughput needs to conform to the historical statistical law,

[0108] (3) Total balance constraint:

[0109] The sum of carbon emissions in each stage is equal to the total carbon emissions,

[0110] The total contribution ratio of different emission sources is 100%,

[0111] The upstream and downstream carbon emissions need to satisfy the material flow balance,

[0112] (4) Equipment physical limit constraint:

[0113] The energy consumption of the equipment shall not exceed its rated power,

[0114] The equipment utilization rate is between 0 - 100%,

[0115] The carbon emissions during the equipment life cycle show an increasing trend,

[0116] (5) Traffic flow constraint:

[0117] The traffic volume does not exceed the road design capacity,

[0118] The ship traffic volume is limited by the navigable capacity of the waterway,

[0119] The flow fluctuations during peak hours conform to statistical laws.

[0120] S6. Optimization suggestion generation: According to the evaluation and prediction results, generate low-carbon optimization suggestions for road and waterway infrastructure, including material optimization, construction process improvement, equipment energy efficiency enhancement, and traffic or shipping management optimization plans. The carbon emission evaluation results, prediction trends, and optimization suggestions are displayed in real time through a visual interface, supporting users to adjust input parameters and updating the evaluation and optimization results in real time.

[0121] According to the evaluation and prediction results, generate low-carbon suggestions for design optimization, construction optimization, and operation optimization, including material selection, construction process improvement, equipment energy efficiency enhancement, and operation management optimization plans.

[0122] Specifically, for the road scenario, it is recommended to optimize the dispatching plan of construction equipment, adjust the transportation route to reduce fuel consumption, and introduce new energy vehicles during the operation stage to reduce traffic emissions; for the waterway scenario, it is recommended to use LNG fuel ships or electric ships, optimize the energy efficiency management of port equipment, and improve the waterway design to shorten the voyage.

[0123] For example, the output suggestions include short-term prediction, long-term prediction, and key factor analysis, as follows:

[0124] Short-term prediction: Combining the operating status of construction equipment, predict the emission changes during the construction stage in the next 6 months. For example, due to the increase in equipment use during the construction peak period, it is expected that the emissions will increase by 10%.

[0125] Long-term prediction: Based on the historical traffic flow data and vehicle electrification policies, predict that the carbon emissions during the operation stage will be reduced by 25% in the next 10 years.

[0126] Key factor analysis: The model identifies that the main driving factors for emission changes are traffic flow growth (30% contribution) and construction equipment fuel efficiency (20% contribution).

[0127] Preferably, according to the evaluation and prediction results, propose the following low-carbon optimization measures to form an optimization plan:

[0128] Material optimization: Replace 20% of the traditional steel in reinforced concrete with recycled steel, reducing emissions during the design stage by 15%.

[0129] Construction optimization: Optimize equipment dispatching, reduce the standby time of pavers by 20%, and adjust the material transportation route to reduce the single transportation mileage by 10 km, reducing emissions during the construction stage by 10%.

[0130] Operation and maintenance optimization: Promote new energy vehicles and adjust toll station design to reduce congestion, with an expected 20% reduction in operation and maintenance emissions per year.

[0131] Preferably, it also includes result verification and feedback:

[0132] Re-evaluate the total carbon emissions after optimization:

[0133] Before optimization: The total emissions were 3,600 tons of CO2, with design, construction, and operation and maintenance accounting for 1,000 tons, 1,500 tons, and 1,100 tons respectively.

[0134] After optimization: The total emissions dropped to 3,000 tons of CO2, with the three stages reduced to 850 tons, 1,300 tons, and 850 tons respectively, achieving an overall emission reduction of 16.7%.

[0135] Model prediction shows that the growth trend of traffic flow and the promotion of new energy vehicles in the next 10 years can further reduce operation and maintenance emissions by about 25%.

[0136] To verify that the hybrid model provided in this embodiment has better performance, a model performance test was conducted.

[0137] See Figure 3 It is a box plot of error distribution, which shows the distribution of prediction errors. The relative error is mainly concentrated in the lower range, indicating that the accuracy of the model prediction is relatively high. A very small number of data points have large deviations, which may be caused by abnormal data or complex characteristics that the model fails to fully capture.

[0138] See Figure 4 It is an error trend chart. The chart shows the rolling mean absolute error as the sample index changes. From the overall trend, the error remains low and relatively stable during the prediction process, indicating that the performance of the model is consistent across different time periods and samples.

[0139] See Figure 5 It is a correlation chart. This chart shows the linear relationship between the actual values and the predicted values through a scatter plot and a fitted line. The actual values and the predicted values are basically highly positively correlated, and the fitted line is close to the diagonal, indicating that the accuracy of the model prediction is relatively high and it can effectively reflect the actual carbon emission trend.

[0140] See Figure 6 It is a comparison chart of predicted values and actual values. This chart further verifies the fitting effect between the actual values and the predicted values. The scatter distribution is in good agreement with the diagonal, indicating that the model has strong prediction ability for data.

[0141] See Figure 7It is a histogram of error distribution. This histogram shows the absolute value distribution of prediction errors. Most of the error values are concentrated in the lower range, indicating that the model has small prediction errors and good robustness.

[0142] See Figure 8 It is a training history graph. The curves of training loss and validation loss changing with the number of training epochs reflect the training effect of the model. The two curves gradually decline and tend to be stable, indicating that the model training process converges well and there is no obvious overfitting or underfitting phenomenon.

[0143] See Figure 9 It is a time series comparison graph. This graph shows the time series change trends of the actual values and predicted values of the first 100 samples. The two curves are highly coincident, indicating that the model has excellent prediction ability in the time series.

[0144] Example 2

[0145] Based on Example 1, see Figure 2 , this example provides a refined carbon emission assessment system for the whole process of highway and waterway infrastructure, including:

[0146] (1) Data acquisition module, used to collect data related to the whole process of carbon emissions of highway and waterway infrastructure;

[0147] (2) Data processing module, used to clean, standardize and normalize the collected raw data;

[0148] (3) Carbon emission assessment module, used to calculate the total carbon emissions in the design, construction and operation and maintenance stages based on the LCA method;

[0149] (4) Carbon emission prediction module, used to predict future carbon emission trends using a hybrid model (CNN + LSTM + Attention + physical correction);

[0150] (5) Optimization suggestion module, used to generate low-carbon optimization solutions for highway and waterway scenarios;

[0151] (6) Display and interaction module, used to display the carbon emission assessment results, dynamic prediction and optimization suggestions in the form of charts, and support users to adjust input parameters in real time.

[0152] This example is applied to the whole process carbon emission assessment and optimization of waterway infrastructure (such as ports, waterways). Through data collection and dynamic analysis, low-carbon management in the port construction, shipping and operation stages is achieved.

[0153] 1. Data collection

[0154] Data collection covers three stages of port design, construction and operation and maintenance:

[0155] 1.1 Design stage: Record the types, quantities, and sources of building materials required for the expansion project. For example, the usage of waterproof coating is 200 tons, the transportation distance from the source is 300 kilometers, and the carbon emission factor is 3 tons CO2 / ton. The usage of steel bars is 500 tons, the transportation distance from the source is 500 kilometers, and the carbon emission factor is 2 tons CO2 / ton.

[0156] 1.2 Construction stage: Monitor the operating hours, fuel consumption, and material transportation routes of construction equipment through on-site sensors. For example, the operating time of the crane is 1000 hours, and the fuel consumption rate is 20 liters / hour. The average driving mileage of material transportation vehicles is 300 kilometers, and the fuel consumption per 100 kilometers is 30 liters.

[0157] 1.3 Operation and maintenance stage: Collect the following data using on-board sensors and port monitoring systems: ship type, voyage, and fuel consumption. For example, the fuel consumption of a diesel ship is 50 liters per hour. The energy consumption of port equipment, such as the electricity consumption of a crane is 1000 kWh per day.

[0158] 2. Data preprocessing.

[0159] Perform the following processing on the collected raw data: Clean the data and remove outliers (such as abnormally high equipment operating time). Standardize data such as fuel consumption and electricity usage to the unit of "carbon emissions per hour" for easy calculation.

[0160] 3. Carbon emission assessment.

[0161] Based on the life cycle assessment (LCA) method, combined with the carbon emission factor library, calculate the carbon emissions in each stage.

[0162] 3.1 Design stage:

[0163] The carbon emission of the waterproof coating is: 200 tons × 3 tons CO2 / ton = 600 tons.

[0164] The carbon emission of the steel bars is: 500 tons × 2 tons CO2 / ton = 1000 tons CO2.

[0165] 3.2 Construction stage:

[0166] The carbon emission of the crane is: 1000 hours × 20 liters / hour × 2.68 kg CO2 / liter (carbon emission factor of the corresponding fuel for the crane) = 53.6 tons CO2.

[0167] The carbon emission of the material transportation vehicle is: (300 km / 100 km) × 30 liters / 100 km × 2.68 kg CO2 / liter = 24.12 tons CO2.

[0168] 3.3 Operation and maintenance stage:

[0169] The annual emissions of a ship are: 1,000 hours of sailing time per year, and fuel consumption is 1,000 hours × 50 liters / hour × 2.68 kg CO2 / liter = 134 tons of CO2.

[0170] The annual emissions of port equipment are: 1000 kWh / day × 365 days × 0.6 kg CO2 / kWh = 219 tons CO2.

[0171] 4. Dynamic prediction of carbon emissions.

[0172] A hybrid algorithm (CNN+LSTM+Attention) is used to dynamically predict carbon emissions, focusing on analyzing the future trends of energy consumption in shipping and port equipment:

[0173] 4.1 Data modeling: CNN extracts the correlation pattern between ship speed and fuel consumption, as well as the energy consumption characteristics during equipment operation peak hours; LSTM models time series data, such as predicting that ship shipping volume will increase by 10% year by year in the next five years; the attention mechanism dynamically assigns weights to each factor and calculates the key contributions of ship emissions (accounting for 60% of total emissions) and port equipment emissions (accounting for 30% of total emissions).

[0174] 4.2 Prediction results:

[0175] Short-term forecast: In the next year, ship speed will increase by 5%, and port equipment energy consumption will increase by 3%, resulting in an increase of about 2% in total emissions.

[0176] Long-term forecast: Over the next five years, the number of ships and shipping volume will increase by 50%, and total emissions are expected to increase by 20%.

[0177] 5. Generate optimization plan.

[0178] Based on the evaluation and prediction results, the following optimization suggestions are put forward:

[0179] 5.1 Construction phase optimization:

[0180] Use green building materials (low-carbon waterproof coatings) to reduce material production emissions by 20%. Optimize construction equipment scheduling and reduce crane standby time by 15%.

[0181] 5.2 Optimization in the operation and maintenance phase:

[0182] Promote LNG (liquefied natural gas) fuel ships to reduce ship emissions by 25%. Gradually switch port equipment to solar power supply systems, which is expected to reduce port equipment emissions by 20%. Improve waterway design to shorten voyages and reduce ship fuel consumption by 15%.

[0183] 6. Result verification and feedback.

[0184] Before the implementation of optimization measures, the total carbon emissions of the project were 2,030.72 tons of CO2 / year, and the emission distribution in each stage was as follows:

[0185] Design stage: 600 tons of CO2, accounting for 29.55%.

[0186] Construction stage: 1000 + 53.6 + 24.12 = 1077.72 tons of CO2, accounting for 53.09%.

[0187] Operation and maintenance stage: 134 + 219 = 353 tons of CO2, accounting for 17.36%.

[0188] After the implementation of optimization measures, the total emissions decreased to 1,621.76 tons of CO2 / year, and the emission reduction rate was 20.15%. The emissions in each stage after optimization are as follows:

[0189] Design stage: By introducing low-carbon waterproof coatings, it was reduced to 480 tons of CO2 (600 tons before optimization, with a 20% reduction).

[0190] Construction stage: After optimizing the equipment scheduling, the emissions from the transportation of construction equipment and materials were reduced to 45.56 + 20.5 = 66.06 tons of CO2 respectively, and the total emissions in the construction stage were reduced to 866 tons of CO2 (1,077.72 tons before optimization, with a 19.65% reduction).

[0191] Operation and maintenance stage: After replacing diesel with LNG fuel for ships, the emissions were reduced to 100.5 tons of CO2; after introducing a solar power supply system for port equipment, the emissions were reduced to 175.2 tons of CO2. The total emissions in the operation and maintenance stage were reduced to 275.7 tons of CO2 (353 tons before optimization, with a 21.88% reduction).

[0192] The results show that the optimization measures have achieved significant emission reduction effects in each stage, especially in the construction stage and the operation and maintenance stage.

[0193] Long-term trend verification:

[0194] Through the prediction and verification of the optimization measures by the hybrid model, the total carbon emission trend in the next 5 years was obtained. The results show:

[0195] Before optimization, it was predicted that the emissions in the next 5 years would increase to 2,436 tons of CO2 / year due to the increase in ship shipping volume;

[0196] After optimization, it is predicted that the emissions in the next 5 years will be stable at about 1,700 tons of CO2 / year, with an emission reduction rate of 30.2%.

[0197] Visualization of optimization results:

[0198] Using the display and interaction module of the present invention, bar charts, trend line charts, and distribution charts of carbon emissions changes at each stage before and after optimization are generated to visually display the emission reduction effect. Users can adjust input parameters (such as ship fuel ratio, equipment operation frequency, etc.) to view the new prediction results and the emission reduction potential of the optimization plan in real time. The preferred specific embodiments of the present invention have been described in detail above.

[0199] Embodiment 3

[0200] This embodiment provides an electronic device, including: one or more processors and a memory. The memory stores one or more programs, and the one or more programs include instructions for executing the method for refined assessment of carbon emissions throughout the process of road and waterway infrastructure as described in Embodiment 1.

[0201] As Figure 10 described, at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory. Of course, it may also include other hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the above Figure 1 described method. Of course, in addition to the software implementation method, the present invention does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, and can also be hardware or a logic device.

[0202] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM), and / or non-volatile memory in the form of, for example, read-only memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.

[0203] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassette tapes, magnetic disk storage, or other magnetic storage devices, or any other non-transmission media that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0204] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A refined assessment method for the whole-process carbon emissions of highway and waterway infrastructure, characterized in that It includes the following steps: Obtain the original carbon emission data in the design, construction, and operation and maintenance stages of highway and waterway infrastructure; Perform preprocessing on the original carbon emission data; Construct a carbon emission factor library for the carbon emission sources of highway and waterway infrastructure; Through life cycle assessment, evaluate the carbon emissions in each stage based on the carbon emission factor library; Based on the original carbon emission data, obtain the constrained carbon emission prediction results through high-dimensional feature extraction, long-term and short-term feature modeling, and data interaction between stages; Generate optimization suggestions based on the carbon emission assessment and the carbon emission prediction results; 2. The refined carbon emission assessment method for the whole process of highway and waterway infrastructure according to claim 1, characterized in that The original carbon emission data includes material usage data, construction equipment energy consumption data, traffic flow data, transportation routes and fuel consumption data, and port equipment operation energy consumption data; 3. A method for refined assessment of carbon emissions throughout the whole process of highway and waterway infrastructure, as claimed in claim 1, wherein The preprocessing includes cleaning, standardization, and normalization; 4. A method for refined assessment of carbon emissions throughout the whole process of highway and waterway infrastructure, as claimed in claim 1, wherein The carbon emission factor library includes production emission factors for highway materials and waterway materials, fuel emission factors for highway transport vehicles and ships, and operation emission factors for construction equipment and port equipment; 5. The refined carbon emission assessment method for the whole process of highway and waterway infrastructure according to claim 1, characterized in that The process of obtaining the carbon emission prediction results includes the following steps: Based on the original carbon emission data, train a fusion model including a convolutional neural network and a long short-term memory network based on the attention mechanism; Use the convolutional neural network to extract multi-source high-dimensional features of the current carbon emission data; Based on the high-dimensional features, use the long short-term memory network to model the long-term dependence and short-term volatility of the time series in the time dimension to obtain a preliminary carbon emission prediction result. During the modeling process, calculate the influence weights of different stages through the attention mechanism to enhance the interaction between data in each stage; Based on the preliminary carbon emission prediction result, use a physical constraint model to correct the prediction result to obtain the final carbon emission prediction result; 6. The refined carbon emission assessment method for the whole process of highway and waterway infrastructure according to claim 5, characterized in that, The convolutional neural network is used to extract the vehicle passing frequency pattern information in the traffic flow data and the correlation information between the operation duration and fuel consumption in the construction equipment data; 7. The refined carbon emission assessment method for the whole process of highway and waterway infrastructure according to claim 5, characterized in that The long short-term memory network is used to extract seasonal traffic flow growth information and holiday peak emission information; 8. The refined carbon emission assessment method for the whole process of highway and waterway infrastructure according to claim 1, characterized in that The prediction results include the carbon emission changes in short-term and long-term predictions, as well as key factor analysis; 9. The fine-grained carbon emission assessment method for the whole process of highway and waterway infrastructure according to claim 1, characterized in that The generation of optimization suggestions includes: For the highway scenario, it is recommended to optimize the scheduling plan of construction equipment, adjust the transportation route to reduce fuel consumption, and introduce new energy vehicles in the operation stage to reduce traffic emissions; For the waterway scenario, it is recommended to use LNG fuel ships or electric ships, optimize the energy efficiency management of port equipment, and improve the waterway design to shorten the voyage; 10. A refined carbon emission assessment system for the whole process of highway and waterway infrastructure, characterized in that, For implementing the method for refined assessment of the whole-process carbon emissions of highway and waterway infrastructure as described in any one of claims 1-9, the system includes: A data collection module for collecting the original carbon emission data in the design, construction, and operation and maintenance stages of highway and waterway infrastructure; A data processing module for performing preprocessing on the original carbon emission data; A carbon emission assessment module for evaluating the carbon emissions in each stage based on the carbon emission factor library through life cycle assessment; A carbon emission prediction module, which is used to obtain a carbon emission prediction result under constraints based on the original carbon emission data through high-dimensional feature extraction, long-term and short-term feature modeling, and inter-stage data interaction; An optimization suggestion module, which is used to generate low-carbon optimization solutions for highway and waterway scenarios; A display and interaction module, which is used to display the carbon emission assessment results, dynamic predictions, and optimization suggestions in the form of charts according to the real-time input parameters of the user.

Images